Large Language Models (LLMs) face severe challenges in long-context processing, including quadratic computational costs, information forgetting, and the context fragmentation inherent in Retrieval-Augmented Generation (RAG).We introduce LycheeMemory, a cognitively inspired framework that enables efficient long-context inference via chunk-wise compression and selective memory recall, rather than processing all raw tokens.LycheeMemory segments the input into chunks and encodes each into compressed KV-cache-style representations using a Compressor.A Gate then dynamically selects relevant memory blocks, which a Reasoner iteratively processes with an evolving working memory to solve downstream tasks.The Compressor and Reasoner are jointly optimized via end-to-end reinforcement learning, while the Gate is trained separately as a classifier.Experimental results demonstrate that LycheeMemory achieves competitive accuracy (up to 82% in ablation variants) on multi-hop reasoning benchmarks (e.g., RULER-HQA), successfully extrapolates context length from 7K to 1.75M, and provides a favorable accuracy-efficiency trade-off against strong long-context baselines.Notably, compared to MemAgent, LycheeMemory achieves an average 2× reduction in peak GPU memory usage and a 6× speedup during inference.

Existing long-document question answering systems typically process texts as flat sequences or use heuristic chunking, which overlook the discourse structures that naturally guide human comprehension.We present a discourse-aware hierarchical framework that leverages rhetorical structure theory (RST) for long document question answering.Our approach converts discourse trees into sentence-level representations and employs LLM-enhanced node representations to bridge structural and semantic information.The framework involves three key innovations: language-universal discourse parsing for lengthy documents, LLM-based enhancement of discourse relation nodes, and structure-guided hierarchical retrieval.Extensive experiments on four datasets demonstrate consistent improvements over existing approaches through the incorporation of discourse structure, across multiple genres and languages.Moreover, the proposed framework exhibits strong robustness across diverse document types and linguistic settings.

Yumeng Fu, Weitao Huang, Junjie Wu, Hao Teng, Shouduo Shang, Meishan Zhang, Bingquan Liu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Emotion Recognition in Conversation (ERC), the task of identifying the emotion of each utterance in a conversation, is crucial for humanmachine interaction.Existing LLM-based ERC methods focus on standard prompting and slow thinking for emotion analysis.However, they suffer from the lack of human-like emotion reasoning and discrimination between similar emotions, thus limiting accurate emotion predictions.To this end, we present JoPR, jointing perception-curriculum learning and emotional reasoning for conversational emotion recognition.Specifically, we devise a multi-dimension curriculum with long CoT fine-tuning to clone human-like emotion reasoning.We further design an emotion-specific reward function in a novel reinforcement learning framework, thereby enhancing the discernment between similar emotions.We conduct extensive experiments on three widely used benchmark datasets, and the results demonstrate that our JoPR achieves consistent and significant improvements over baselines.

Large Language Models (LLMs) enhance their problem-solving capability by utilizing external tools.However, in open-world scenarios with massive and evolving tool repositories, existing methods relying on static embedding retrieval or parameter memorization of tools struggle to align user intent with tool semantics or generalize to unseen tools, respectively, leading to suboptimal accuracy of open-world tool retrieval and execution.To address these, we present ToolOmni, a unified agentic framework that enables LLMs for open-world tool use by proactive retrieval and grounded execution within a reasoning loop.First, we construct a cold-start multi-turn interaction dataset to instill foundational agentic capabilities via Supervised Fine-Tuning (SFT).Then, we introduce open-world tool learning based on a Decoupled Multi-Objective GRPO algorithm, which simultaneously optimizes LLMs for both tool retrieval accuracy and execution efficacy in online environments.Extensive experiments demonstrate that ToolOmni achieves state-ofthe-art performance both in retrieval and execution, surpassing strong baselines by a significant margin of +10.8% in end-to-end execution success rate, while exhibiting exceptional robustness and generalization capabilities.

Bobo Li, Wu Rui, Zibo Ji, Meishan Zhang, Hao Fei, Min Zhang, Mong-Li Lee, Wynne Hsu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Chenyuan Zhang, Qiguang Chen, Xie Chen, Zhuotao Tian, Bowen Xing, Meishan Zhang, Libo Qin, Baotian Hu, Min Zhang. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Despite increasing research attention to text ranking, most studies focus on monolingual scenarios, with a particular emphasis on English-language contexts. This narrow focus limits the applicability of ranking models in cross-lingual contexts, such as ranking Chinese documents based on English queries. Recent advances in large language models (LLMs) have significantly reduced inter-language barriers through pre-training on extensive multilingual corpora, thus facilitating the study of multilingual text ranking (MTR). In this work, we explore the potential of LLMs in MTR tasks. Specifically, we first introduce an MTR benchmark encompassing both monolingual and cross-lingual scenarios. Then, we propose a two-stage training pipeline to alleviate the misalignment between LLMs and text ranking. Lastly, we adapt this training pipeline to multilingual scenarios from the perspective of training data and methods. Our experiments on the MTR benchmark demonstrate that the proposed multilingual two-stage training pipeline significantly improves LLM ranking performance in both monolingual and cross-lingual scenarios, particularly in out-domain settings. We complement these findings with a thorough analysis to deepen the understanding of our approach.

Cross-domain constituency parsing is still an unsolved challenge in computational linguistics since the available multi-domain constituency treebank is limited.We investigate automatic treebank generation by large language models (LLMs) in this paper.The performance of LLMs on constituency parsing is poor, therefore we propose a novel treebank generation method, LLM back generation, which is similar to the reverse process of constituency parsing.LLM back generation takes the incomplete cross-domain constituency tree with only domain keyword leaf nodes as input and fills the missing words to generate the cross-domain constituency treebank.Besides, we also introduce a span-level contrastive learning pretraining strategy to make full use of the LLM back generation treebank for cross-domain constituency parsing.We verify the effectiveness of our LLM back generation treebank coupled with contrastive learning pre-training on five target domains of MCTB.Experimental results show that our approach achieves state-of-theart performance on average results compared with various baselines.

Xin Zhang, Ziqi Dai, Yongqi Li, Yanzhao Zhang, Dingkun Long, Pengjun Xie, Meishan Zhang, Jun Yu, Wenjie Li, Min Zhang. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Longhui Zhang, Jiahao Wang, Meishan Zhang, GaoXiong Cao, Ensheng Shi, Mayuchi Mayuchi, Jun Yu, Honghai Liu, Jing Li, Min Zhang. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Emotion recognition in conversation (ERC), the task of discerning human emotions for each utterance within a conversation, has garnered significant attention in human-computer interaction systems. Previous ERC studies focus on speaker-specific information that predominantly stems from relationships among utterances, which lacks sufficient information around conversations. Recent research in ERC has sought to exploit pre-trained large language models (LLMs) with speaker modelling to comprehend emotional states. Although these methods have achieved the encouraging results, the extracted speaker-specific information struggles to indicate emotional dynamics. In this paper, motivated by the fact that speaker characteristics play a crucial role and LLMs have rich world knowledge, we present LaERC-S, a novel framework that stimulates LLMs to explore speaker characteristics involving the mental state and behavior of interlocutors, for accurate emotion predictions. To endow LLMs with these knowledge information, we adopt the two-stage learning to make the models reason speaker characteristics and track the emotion of the speaker in complex conversation scenarios. Extensive experiments on three benchmark datasets demonstrate the superiority of LaERC-S, reaching the new state-of-the-art.

Universal Multimodal Retrieval (UMR) aims to enable search across various modalities using a unified model, where queries and candidates can consist of pure text, images, or a combination of both. Previous work has attempted to adopt multimodal large language models (MLLMs) to realize UMR using only text data. However, our preliminary experiments demonstrate that more diverse multimodal training data can further unlock the potential of MLLMs. Despite its effectiveness, the existing multimodal training data is highly imbalanced in terms of modality, which motivates us to develop a training data synthesis pipeline and construct a large-scale, high-quality fused-modal training dataset. Based on the synthetic training data, we develop the General Multimodal Embedder (GME), an MLLM-based dense retriever designed for UMR. Furthermore, we construct a comprehensive UMR Benchmark (UMRB) to evaluate the effectiveness of our approach. Experimental results show that our method achieves state-of-the-art performance among existing UMR methods. Last, we provide in-depth analyses of model scaling and training strategies, and perform ablation studies on both the model and synthetic data.

Enabling digital humans to express rich emotions has significant applications in dialogue systems, gaming, and other interactive scenarios.While recent advances in talking head synthesis have achieved impressive results in lip synchronization, they tend to overlook the rich and dynamic nature of facial expressions.To fill this critical gap, we introduce an endto-end text-to-expression model that explicitly focuses on emotional dynamics.Our model learns expressive facial variations in a continuous latent space and generates expressions that are diverse, fluid, and emotionally coherent.To support this task, we introduce EmoAva, a large-scale and high-quality dataset containing 15,000 text-3D expression pairs.Extensive experiments on both existing datasets and EmoAva demonstrate that our method significantly outperforms baselines across multiple evaluation metrics, marking a significant advancement in the field.

Open Information Extraction (OIE) aims at extracting the relational triplets from open-domain texts. Existing methods, unfortunately, mostly fall prey to the complex OIE setting, due to the failure to extract unseen words and underutilize syntactic features. In this work, we propose a novel system tailored for complex OIE, where a generative PLM with non-autoregressive generative decoding is adopted for abstractive OIE generation. We introduce a heterogeneous syntactic forests as features for the task, merging constituency and dependency forests into a unified syntax graph, aiding in better detecting potential boundaries and relationships within complex terms. Also an Implicit-Explicit Contrastive Learning mechanism is devised to boost the perception of implicit relations and spans. Our system outperforms the current state-of-the-art model across seven OIE datasets.

Photo-Sharing Multi-modal dialogue generation requires a dialogue agent not only to generate text responses but also to share photos at the proper moment. Using image text caption as the bridge, a pipeline model integrates an image caption model, a text generation model, and an image generation model to handle this complex multi-modal task. However, representing the images with text captions may lose important visual details and information and cause error propagation in the complex dialogue system. Besides, the pipeline model isolates the three models separately because discrete image text captions hinder end-to-end gradient propagation. We propose the first end-to-end model for photo-sharing multi-modal dialogue generation, which integrates an image perceptron and an image generator with a large language model. The large language model employs the vision encoder to perceive visual images in the input end. For image generation in the output end, we propose a dynamic vocabulary transformation matrix and use straight-through and gumbel-softmax techniques to align the large language model and stable diffusion model and achieve end-to-end gradient propagation. We perform experiments on PhotoChat and DialogCC datasets to evaluate our end-to-end model. Compared with pipeline models, the end-to-end model gains state-of-the-art performances on various metrics of text and image generation.

Large language models (LLMs) have made significant strides in code translation tasks. However, ensuring both the correctness and readability of translated code remains a challenge, limiting their effective adoption in real-world software development. In this work, we propose F2STrans, a function-to-style guiding paradigm designed to progressively improve the performance of LLMs in code translation. Our approach comprises two key stages: (1) Functional learning, which optimizes translation correctness using high-quality source-target code pairs mined from online programming platforms, and (2) Style learning, which improves translation readability by incorporating both positive and negative style examples. Additionally, we introduce a novel code translation benchmark that includes up-to-date source code, extensive test cases, and manually annotated ground-truth translations, enabling comprehensive functional and stylistic evaluations. Experiments on both our new benchmark and existing datasets demonstrate that our approach significantly improves code translation performance. Notably, our approach enables Qwen-1.5B to outperform prompt-enhanced Qwen-32B and GPT-4 on average across 20 diverse code translation scenarios.

Large Language Models (LLMs) have demonstrated outstanding performance in mathematical reasoning capabilities. However, we argue that current large-scale reasoning models primarily rely on scaling up training datasets with diverse mathematical problems and long thinking chains, which raises questions about whether LLMs genuinely acquire mathematical concepts and reasoning principles or merely remember the training data. In contrast, humans tend to break down complex problems into multiple fundamental atomic capabilities. Inspired by this, we propose a new paradigm for evaluating mathematical atomic capabilities. Our work categorizes atomic abilities into two dimensions: (1) field-specific abilities across four major mathematical fields, algebra, geometry, analysis, and topology, and (2) logical abilities at different levels, including conceptual understanding, forward multi-step reasoning with formal math language, and counterexample-driven backward reasoning. We propose corresponding training and evaluation datasets for each atomic capability unit, and conduct extensive experiments about how different atomic capabilities influence others, to explore the strategies to elicit the required specific atomic capability. Evaluation and experimental results on advanced models show many interesting discoveries and inspirations about the different performances of models on various atomic capabilities and the interactions between atomic capabilities. Our findings highlight the importance of decoupling mathematical intelligence into atomic components, providing new insights into model cognition and guiding the development of training strategies toward a more efficient, transferable, and cognitively grounded paradigm of "atomic thinking".

This paper introduces **VimoRAG**, a novel video-based retrieval-augmented motion generation framework for motion large language models (LLMs). As motion LLMs face severe out-of-domain/out-of-vocabulary issues due to limited annotated data, **VimoRAG** leverages large-scale in-the-wild video databases to enhance 3D motion generation by retrieving relevant 2D human motion signals. While video-based motion RAG is nontrivial, we address two key bottlenecks: (1) developing an effective motion-centered video retrieval model that distinguishes human poses and actions, and (2) mitigating the issue of error propagation caused by suboptimal retrieval results. We design the Gemini Motion Video Retriever mechanism and the Motion-centric Dual-alignment DPO Trainer, enabling effective retrieval and generation processes. Experimental results show that **VimoRAG** significantly boosts the performance of motion LLMs constrained to text-only input.

Searching over semi-structured data with natural language (NL) queries has attracted sustained attention, enabling broader audiences to access information easily. As more applications, such as LLM agents and RAG systems, emerge to search and interact with semi-structured data, two major challenges have become evident: (1) the increasing diversity of domains and schema variations, making domain-customized solutions prohibitively costly; (2) the growing complexity of NL queries, which combine both exact field matching conditions and fuzzy semantic requirements, often involving multiple fields and implicit reasoning. These challenges make formal language querying or keyword-based search insufficient. In this work, we explore neural retrievers as a unified non-formal querying solution by directly index semi-structured collections and understand NL queries. We employ LLM-based automatic evaluation and build a large-scale semi-structured retrieval benchmark (SSRB) using LLM generation and filtering, containing 14M semi-structured objects from 99 different schemas across 6 domains, along with 8,485 test queries that combine both exact and fuzzy matching conditions. Our systematic evaluation of popular retrievers shows that current state-of-the-art models could achieve acceptable performance, yet they still lack precise understanding of matching constraints. While by in-domain training of dense retrievers, the performance can be significantly improved. We believe that our SSRB could serve as a valuable resource for future research in this area, and we hope to inspire further exploration of semi-structured retrieval with complex queries.

Dependency parsing is a fundamental task in natural language processing that involves identifying the grammatical relationships between words in a sentence. One promising approach for performing this task in languages lacking annotated treebanks is treebank translation, which utilizes word alignments to map dependencies from a source treebank to the corresponding target translation. However, due to language differences and the limitations of word alignment tools, this method would inevitably generate noise during mapping. To reduce the effect of noise, we first exploit MetaNet to compute quality scores for each dependency and identify low-score ones as noise. MetaNet is a fake teacher that learns to score homework (dependencies) by comparing answers from the top student (strong parser) and the regular student (weak parser) without knowing the correct answer (gold-standard). With the scoring capability of MetaNet, we design an iterative algorithm to boost the target treebank quality, which trains with high-quality dependencies and relabels the low-quality dependencies. Our method achieves better results than the originally translated treebanks and shows highly competitive performances with prior methods on the Universal Dependency Treebanks v2.2. We also provide detailed analysis and discussions.

Chinese sequence labeling tasks are sensitive to word boundaries. Although pretrained language models (PLM) have achieved considerable success in these tasks, current PLMs rarely consider boundary information explicitly. An exception to this is BABERT, which incorporates unsupervised statistical boundary information into Chinese BERT’s pre-training objectives. Building upon this approach, we input supervised high-quality boundary information to enhance BABERT’s learning, developing a semi-supervised boundary-aware PLM. To assess PLMs’ ability to encode boundaries, we introduce a novel “Boundary Information Metric” that is both simple and effective. This metric allows comparison of different PLMs without task-specific fine-tuning. Experimental results on Chinese sequence labeling datasets demonstrate that the improved BABERT version outperforms the vanilla version, not only in these tasks but also in broader Chinese natural language understanding tasks. Additionally, our proposed metric offers a convenient and accurate means of evaluating PLMs’ boundary awareness.

In nested Named entity recognition (NER), entities are nested with each other, and thus requiring more data annotations to address. This leads to the development of few-shot nested NER, where the prevalence of pretrained language models with in-context learning (ICL) offers promising solutions. In this work, we introduce an effective and innovative ICL framework for the setting of few-shot nested NER. We improve the ICL prompt by devising a novel example demonstration selection mechanism, EnDe retriever. In EnDe retriever, we employ contrastive learning to perform three types of representation learning, in terms of semantic similarity, boundary similarity, and label similarity, to generate high-quality demonstration examples. Extensive experiments over three nested NER and four flat NER datasets demonstrate the efficacy of our system.

Existing research of video understanding still struggles to achieve in-depth comprehension and reasoning in complex videos, primarily due to the under-exploration of two key bottlenecks: fine-grained spatial-temporal perceptive understanding and cognitive-level video scene comprehension. This paper bridges the gap by presenting a novel solution. We first introduce a novel video Multimodal Large Language Model (MLLM), MotionEpic, which achieves fine-grained pixel-level spatial-temporal video grounding by integrating video spatial-temporal scene graph (STSG) representation. Building upon MotionEpic, we then develop a Video-of-Thought (VoT) reasoning framework. VoT inherits the Chain-of-Thought (CoT) core, breaking down a complex task into simpler and manageable sub-problems, and addressing them step-by-step from a low-level pixel perception to high-level cognitive interpretation. Extensive experiments across various complex video QA benchmarks demonstrate that our overall framework strikingly boosts existing state-of-the-art. To our knowledge, this is the first attempt at successfully implementing the CoT technique for achieving human-level video reasoning, where we show great potential in extending it to a wider range of video understanding scenarios. Systems and codes will be open later.

Event extraction (EE) is a critical direction in the field of information extraction, laying an important foundation for the construction of structured knowledge bases. EE from text has received ample research and attention for years, yet there can be numerous real-world applications that require direct information acquisition from speech signals, online meeting minutes, interview summaries, press releases, etc. While EE from speech has remained under-explored, this paper fills the gap by pioneering a SpeechEE, defined as detecting the event predicates and arguments from a given audio speech. To benchmark the SpeechEE task, we first construct a large-scale high-quality dataset. Based on textual EE datasets under the sentence, document, and dialogue scenarios, we convert texts into speeches through both manual real-person narration and automatic synthesis, empowering the data with diverse scenarios, languages, domains, ambiences, and speaker styles. Further, to effectively address the key challenges in the task, we tailor an E2E SpeechEE system based on the encoder-decoder architecture, where a novel Shrinking Unit module and a retrieval-aided decoding mechanism are devised. Extensive experimental results on all SpeechEE subsets demonstrate the efficacy of the proposed model, offering a strong baseline for the task. At last, being the first work on this topic, we shed light on key directions for future research. Our codes and the benchmark datasets are open at https://SpeechEE.github.io/

The Visual Spatial Description Challenge (VSD) is the first competition event focused on visual spatial understanding, organized under the auspices of the ACM Multimedia Conference 2024. The goal of the VSD challenge is to assess the the ability of models and systems to comprehend spatial concepts, relationships and other semantics from a scene presented with visual appearance. The VSD challenge provides two benchmark datasets for three subtasks, i.e., visual spatial relationship classification, single spatial description generation, and open-ended spatial description generation. The challenge details are available on https://lllogen.github.io/vsd-challenge.github.io/.

This paper introduces AutoSurvey, a speedy and well-organized methodology for automating the creation of comprehensive literature surveys in rapidly evolving fields like artificial intelligence. Traditional survey paper creation faces challenges due to the vast volume and complexity of information, prompting the need for efficient survey methods. While large language models (LLMs) offer promise in automating this process, challenges such as context window limitations, parametric knowledge constraints, and the lack of evaluation benchmarks remain. AutoSurvey addresses these challenges through a systematic approach that involves initial retrieval and outline generation, subsection drafting by specialized LLMs, integration and refinement, and rigorous evaluation and iteration. Our contributions include a comprehensive solution to the survey problem, a reliable evaluation method, and experimental validation demonstrating AutoSurvey's effectiveness.

In the visual spatial understanding (VSU) field, spatial image-to-text (SI2T) and spatial text-to-image (ST2I) are two fundamental tasks that appear in dual form. Existing methods for standalone SI2T or ST2I perform imperfectly in spatial understanding, due to the difficulty of 3D-wise spatial feature modeling. In this work, we consider modeling the SI2T and ST2I together under a dual learning framework. During the dual framework, we then propose to represent the 3D spatial scene features with a novel 3D scene graph (3DSG) representation that can be shared and beneficial to both tasks. Further, inspired by the intuition that the easier 3D$\to$image and 3D$\to$text processes also exist symmetrically in the ST2I and SI2T, respectively, we propose the Spatial Dual Discrete Diffusion (SD$^3$) framework, which utilizes the intermediate features of the 3D$\to$X processes to guide the hard X$\to$3D processes, such that the overall ST2I and SI2T will benefit each other. On the visual spatial understanding dataset VSD, our system outperforms the mainstream T2I and I2T methods significantly. Further in-depth analysis reveals how our dual learning strategy advances.

Abstract Dialogue-level dependency parsing, despite its growing academic interest, often encounters underperformance issues due to resource shortages. A potential solution to this challenge is data augmentation. In recent years, large language models (LLMs) have demonstrated strong capabilities in generation, which can facilitate data augmentation greatly. In this study, we focus on Chinese dialogue-level dependency parsing, presenting three simple and effective strategies with LLM to augment the original training instances, namely word-level, syntax-level, and discourse-level augmentations, respectively. These strategies enable LLMs to either preserve or modify dependency structures, thereby assuring accuracy while increasing the diversity of instances at different levels. We conduct experiments on the benchmark dataset released by Jiang et al. (2023) to validate our approach. Results show that our method can greatly boost the parsing performance in various settings, particularly in dependencies among elementary discourse units. Lastly, we provide in-depth analysis to show the key points of our data augmentation strategies.

Knowledge transfer is investigated in various natural language processing tasks except cross-domain constituency parsing. In this paper, we leverage heterogeneous data to transfer cross-domain and cross-task knowledge to constituency parsing. Concretely, we first select language modeling, named entity recognition, CCG supertagging and dependency parsing as auxiliary tasks and collect the corpora of these tasks covering various domains as cross-domain and cross-task heterogeneous data. Second, we exploit three types of prefixes: shared, task and domain prefix, to merge cross-domain and cross-task data and decompose the general, task and domain representation in the pretrained language model. Third, we convert the data formats of multi-source heterogeneous datasets and loss objectives of the auxiliary tasks into a consistent formalization closer to constituency parsing. Finally, we jointly train the model to transfer task and domain knowledge to cross-domain constituency parsing. We verify the effectiveness of our proposed model on five target domains of MCTB. Experimental results show that our knowledge transfer model outperforms various baseline models, including conventional chart-based and transition-based parsers and the current large-scale language model for zero-shot and few-shot settings.

While pre-training large-scale video-language models (VLMs) has shown remarkable potential for various downstream video-language tasks, existing VLMs can still suffer from certain commonly seen limitations, e.g., coarse-grained cross-modal aligning, under-modeling of temporal dynamics, detached video-language view. In this work, we target enhancing VLMs with a fine-grained structural spatio-temporal alignment learning method (namely Finsta). First of all, we represent the input texts and videos with fine-grained scene graph (SG) structures, both of which are further unified into a holistic SG (HSG) for bridging two modalities. Then, an SG-based framework is built, where the textual SG (TSG) is encoded with a graph Transformer, while the video dynamic SG (DSG) and the HSG are modeled with a novel recurrent graph Transformer for spatial and temporal feature propagation. A spatial-temporal Gaussian differential graph Transformer is further devised to strengthen the sense of the changes in objects across spatial and temporal dimensions. Next, based on the fine-grained structural features of TSG and DSG, we perform object-centered spatial alignment and predicate-centered temporal alignment respectively, enhancing the video-language grounding in both the spatiality and temporality. We design our method as a plug&play system, which can be integrated into existing well-trained VLMs for further representation augmentation, without training from scratch or relying on SG annotations in downstream applications. On 6 representative VL modeling tasks over 12 datasets in both standard and long-form video scenarios, Finsta consistently improves the existing 13 strong-performing VLMs persistently, and refreshes the current state-of-the-art end task performance significantly in both the fine-tuning and zero-shot settings.

Abstract Hierarchical text classification (HTC) is an important task with broad applications, and few-shot HTC has gained increasing interest recently. While in-context learning (ICL) with large language models (LLMs) has achieved significant success in few-shot learning, it is not as effective for HTC because of the expansive hierarchical label sets and extremely ambiguous labels. In this work, we introduce the first ICL-based framework with LLM for few-shot HTC. We exploit a retrieval database to identify relevant demonstrations, and an iterative policy to manage multi-layer hierarchical labels. Particularly, we equip the retrieval database with HTC label-aware representations for the input texts, which is achieved by continual training on a pretrained language model with masked language modeling (MLM), layer-wise classification (CLS, specifically for HTC), and a novel divergent contrastive learning (DCL, mainly for adjacent semantically similar labels) objective. Experimental results on three benchmark datasets demonstrate superior performance of our method, and we can achieve state-of-the-art results in few-shot HTC.

Label noise is an important issue in machine learning, which might lead to negative influences on various tasks. Given that real benchmarks for evaluation of noise reduction methods are limited, plenty of studies construct pseudo noisy data to verify their proposed methods. However, very few works have realized the rationality of the noise generation strategies. If the generated pseudo datasets are biased, their final conclusions might also be problematic. In this work, we focus on text classification of natural language processing (NLP) to investigate various pseudo noise generation methods, which is the first work of this line for NLP. In particular, we compare the noise generated with crowdsourcing noise, a kind of real noise as gold-standard, to evaluate these noise generation methods. After then, we measure and compare the performance of representative noise reduction methods respectively based on the data of crowdsourcing and our top-ranked pseudo noisy generation strategies. We conduct experiments on five text classification datasets, offering detailed comparison results as well as discussions.

Vision-language navigation (VLN) is a challenging task, which guides an agent to navigate in a realistic environment by natural language instructions. Sequence-to-sequence modeling is one of the most prospective architectures for the task, which achieves the agent navigation goal by a sequence of moving actions. The line of work has led to the state-of-the-art performance. Recently, several studies showed that the beam-search decoding during the inference can result in promising performance, as it ranks multiple candidate trajectories by scoring each trajectory as a whole. However, the trajectory-level score might be seriously biased during ranking. The score is a simple averaging of individual unit scores of the target-sequence actions, and these unit scores could be incomparable among different trajectories since they are calculated by a local discriminant classifier. To address this problem, we propose a global normalization strategy to rescale the scores at the trajectory level. Concretely, we present two global score functions to rerank all candidates in the output beam, resulting in more comparable trajectory scores. In this way, the bias problem can be greatly alleviated. We conduct experiments on the benchmark room-to-room (R2R) dataset of VLN to verify our method, and the results show that the proposed global method is effective, providing significant performance than the corresponding baselines. Our final model can achieve competitive performance on the VLN leaderboard.

Simultaneous machine translation (SiMT) is usually done via sequence-level knowledge distillation (Seq-KD) from a full-sentence neural machine translation (NMT) model. However, there is still a significant performance gap between NMT and SiMT. In this work, we propose to leverage monolingual data to improve SiMT, which trains a SiMT student on the combination of bilingual data and external monolingual data distilled by Seq-KD. Preliminary experiments on En-Zh and En-Ja news domain corpora demonstrate that monolingual data can significantly improve translation quality (e.g., +3.15 BLEU on En-Zh). Inspired by the behavior of human simultaneous interpreters, we propose a novel monolingual sampling strategy for SiMT, considering both chunk length and monotonicity. Experimental results show that our sampling strategy consistently outperforms the random sampling strategy (and other conventional typical NMT monolingual sampling strategies) by avoiding the key problem of SiMT -- hallucination, and has better scalability. We achieve +0.72 BLEU improvements on average against random sampling on En-Zh and En-Ja. Data and codes can be found at https://github.com/hexuandeng/Mono4SiMT.

In this work, we investigate a more realistic unsupervised multimodal machine translation (UMMT) setup, inference-time image-free UMMT, where the model is trained with sourcetext image pairs, and tested with only sourcetext inputs.First, we represent the input images and texts with the visual and language scene graphs (SG), where such fine-grained visionlanguage features ensure a holistic understanding of the semantics.To enable pure-text input during inference, we devise a visual scene hallucination mechanism that dynamically generates pseudo visual SG from the given textual SG.Several SG-pivoting based learning objectives are introduced for unsupervised translation training.On the benchmark Multi30K data, our SG-based method outperforms the best-performing baseline by significant BLEU scores on the task and setup, helping yield translations with better completeness, relevance and fluency without relying on paired images.Further in-depth analyses reveal how our model advances in the task setting.

Visual spatial description (VSD) aims to generate texts that describe the spatial relations of the given objects within images.Existing VSD work merely models the 2D geometrical vision features, thus inevitably falling prey to the problem of skewed spatial understanding of target objects.In this work, we investigate the incorporation of 3D scene features for VSD.With an external 3D scene extractor, we obtain the 3D objects and scene features for input images, based on which we construct a target object-centered 3D spatial scene graph (GO3D-S 2 G), such that we model the spatial semantics of target objects within the holistic 3D scenes.Besides, we propose a scene subgraph selecting mechanism, sampling topologically-diverse subgraphs from GO3D-S 2 G, where the diverse local structure features are navigated to yield spatially-diversified text generation.Experimental results on two VSD datasets demonstrate that our framework outperforms the baselines significantly, especially improving on the cases with complex visual spatial relations.Meanwhile, our method can produce more spatially-diversified generation.Code is available at https://github.com/zhaoyucs/VSD.

Self-training has proven to be an effective approach for cross-domain tasks, and in this study, we explore its application to cross-domain constituency parsing.Traditional self-training methods rely on limited and potentially lowquality raw corpora.To overcome this limitation, we propose enhancing self-training with the large language model (LLM) to generate domain-specific raw corpora iteratively.For the constituency parsing, we introduce grammar rules that guide the LLM in generating raw corpora and establish criteria for selecting pseudo instances.Our experimental results demonstrate that self-training for constituency parsing, equipped with an LLM, outperforms traditional methods regardless of the LLM's performance.Moreover, the combination of grammar rules and confidence criteria for pseudo-data selection yields the highest performance in the crossdomain constituency parsing 1 .

An end-to-end speech-to-text (S2T) translation model is usually initialized from a pretrained speech recognition encoder and a pretrained text-to-text (T2T) translation decoder.Although this straightforward setting has been shown empirically successful, there do not exist clear answers to the research questions: 1) how are speech and text modalities fused in S2T model and 2) how to better fuse the two modalities?In this paper, we take the first step toward understanding the fusion of speech and text features in S2T model.We first design and release a 10GB linguistic probing benchmark, namely Speech-Senteval, to investigate the acoustic and linguistic behaviors of S2T models.Preliminary analysis reveals that the uppermost encoder layers of the S2T model can not learn linguistic knowledge efficiently, which is crucial for accurate translation.Based on the finding, we further propose a simple plug-in prompt-learning strategy on the uppermost encoder layers to broaden the abstract representation power of the encoder of S2T models.We call such a promptenhanced S2T model PromptST.Experimental results on four widely-used S2T datasets show that PromptST can deliver significant improvements over a strong baseline by capturing richer linguistic knowledge.Benchmarks,

Cross-modal alignment is one key challenge for Vision-and-Language Navigation (VLN). Most existing studies concentrate on mapping the global instruction or single sub-instruction to the corresponding trajectory. However, another critical problem of achieving fine-grained alignment at the entity level is seldom considered. To address this problem, we propose a novel Grounded Entity-Landmark Adaptive (GELA) pre-training paradigm for VLN tasks. To achieve the adaptive pre-training paradigm, we first introduce grounded entity-landmark human annotations into the Room-to-Room (R2R) dataset, named GEL-R2R. Additionally, we adopt three grounded entity-landmark adaptive pre-training objectives: 1) entity phrase prediction, 2) landmark bounding box prediction, and 3) entity-landmark semantic alignment, which explicitly supervise the learning of fine-grained cross-modal alignment between entity phrases and environment landmarks. Finally, we validate our model on two downstream benchmarks: VLN with descriptive instructions (R2R) and dialogue instructions (CVDN). The comprehensive experiments show that our GELA model achieves state-of-the-art results on both tasks, demonstrating its effectiveness and generalizability.

To achieve multi-modal intelligence, AI must be able to process and respond to inputs from multimodal sources. However, many current question answering models are limited to specific types of answers, such as yes/no and number, and require additional human assessments. Recently, Visual-Text Question Answering (VQTA) dataset has been proposed to fix this gap. In this paper, we conduct an exhaustive analysis and exploration of this task. Specifically, we implement a T5-based multi-modal generative network that overcomes the limitations of traditional labeling space and provides more freedom in responses. Our approach achieve the best performance in both English and Chinese tracks in the VTQA challenge.

As one of the core video semantic understanding tasks, Video Semantic Role Labeling (VidSRL) aims to detect the salient events from given videos, by recognizing the predict-argument event structures and the interrelationships between events. While recent endeavors have put forth methods for VidSRL, they can be mostly subject to two key drawbacks, including the lack of fine-grained spatial scene perception and the insufficiently modeling of video temporality. Towards this end, this work explores a novel holistic spatio-temporal scene graph (namely HostSG) representation based on the existing dynamic scene graph structures, which well model both the fine-grained spatial semantics and temporal dynamics of videos for VidSRL. Built upon the HostSG, we present a nichetargeting VidSRL framework. A scene-event mapping mechanism is first designed to bridge the gap between the underlying scene structure and the high-level event semantic structure, resulting in an overall hierarchical scene-event (termed ICE) graph structure. We further perform iterative structure refinement to optimize the ICE graph, e.g., filtering noisy branches and newly building informative connections, such that the overall structure representation can best coincide with end task demand. Finally, three subtask predictions of VidSRL are jointly decoded, where the end-to-end paradigm effectively avoids error propagation. On the benchmark dataset, our framework boosts significantly over the current best-performing model. Further analyses are shown for a better understanding of the advances of our methods. Our HostSG representation shows greater potential to facilitate a broader range of other video understanding tasks.

Anaphora resolution is a traditional task in the natural language processing community, defined as a cohesion phenomenon where one entity points back to a previous entity. Event discourse deixis (EDD) is a kind of more complex anaphora in which the anaphors refer to event descriptions such as sentences or clauses. Event discourse deixis resolution (EDDR) is able to help machines understand the richer linguistic and semantic information in the discourse. However, compared to anaphora resolution, EDDR has received relatively less research attention. In this work, we investigate the EDDR task by designing the corresponding dataset and model. First, we manually construct a high-quality Chinese corpus for EDDR, including 4,417 documents and 5,929 event chains that consist of event antecedents and anaphors. Second, we propose a deep neural network model for EDDR, which formulates the task into two subtasks, namely event anaphor recognition and event antecedent recognition. Our model is trained under the two subtasks jointly so that the EDDR task can be performed end to end. Besides our final model, we also build seven pipeline and joint models as baselines to build comprehensive benchmarks for follow-up research. Experimental results on our EDDR dataset show that our model outperforms all the baselines and achieves about 53%, 44%, 53%, and 63% F1s using standard anaphora resolution metrics such as CoNLL, MUC, B3, and Ceafe. The performances show that EDDR is a challenging task and worth researching in the future. Our dataset and model will be released to facilitate follow-up research.

Syntactic processing is fundamental to natural language processing. It provides rich and comprehensive syntax information in sentences that could be potentially beneficial for downstream tasks. Recently, pretrained language models have shown great success in Chinese syntactic processing, which typically involves word segmentation, POS tagging, and dependency parsing. However, the on-going research never ends since performance would be degraded drastically when tested on a highly-discrepant domain. This problem is widely accepted as domain adaptation, where the test domain differs from the training domain in supervised learning. Self-training is one promising solution for it, and straightforward source-to-target adaptation has already shown remarkable effectiveness in previous work. While this strategy ignores the fact that sentences of the target domain sentences may have very different gaps from the source training domain. More specifically, sentences with large gaps might fail by direct self-training adaptation. To this end, we propose fine-grained domain adaptation for Chinese syntactic processing in this work, aiming to model the gaps between the source and the target domains accurately and progressively. The key idea is to divide the target domain into fine-grained subdomains by using a specified domain distance metric, and then perform gradual self-training on the subdomains. We further offer an intuitive theoretical illustration based on the theory of Kumar et al. (2020) approximately. In addition, a novel representation learning framework is proposed to encode fine-grained subdomains effectively, aiming to utilize the above idea fully. Experimental results on benchmark datasets show that our method can achieve significant improvements over a variety of baselines.

RDF verbalization has received increasing interest, which aims to generate a natural language description of the knowledge base. Sequence-to-sequence models based on Transformer are able to obtain strong performance equipped with pre-trained language models such as BART and T5. However, in spite of the general performance gain introduced by the pre-trained models, the performance of the task is still limited by the small scale of the training dataset. To address the problem, we propose two orthogonal strategies to enhance the representation learning of RDF triples. Concretely, two types of knowledge are introduced, i.e., descriptive knowledge and relational knowledge, respectively. The descriptive knowledge indicates the semantic information of self definition, and the relational knowledge indicates the semantic information learned from the structural context. We further combine the descriptive and relational knowledge together to enhance the representation learning. Experimental results on the WebNLG and SemEval-2010 datasets show that the two types of knowledge can both enhance the model performance, and their combination is able to obtain further improvements in most cases, providing new state-of-the-art results.

Unsupervised cross-lingual transfer has been shown great potentials for dependency parsing of the low-resource languages when there is no annotated treebank available. Recently, the self-training method has received increasing interests because of its state-of-the-art performance in this scenario. In this work, we advance the method further by coupling it with curriculum learning, which guides the self-training in an easy-to-hard manner. Concretely, we present a novel metric to measure the instance difficulty of a dependency parser which is trained mainly on a Treebank from a resource-rich source language. By using the metric, we divide a low-resource target language into several fine-grained sub-languages by their difficulties, and then apply iterative-self-training progressively on these sub-languages. To fully explore the auto-parsed training corpus from sub-languages, we exploit an improved parameter generation network to model the sub-languages for better representation learning. Experimental results show that our final curriculum-style self-training can outperform a range of strong baselines, leading to new state-of-the-art results on unsupervised cross-lingual dependency parsing. We also conduct detailed experimental analyses to examine the proposed approach in depth for comprehensive understandings.

Aspect-based sentiment analysis (ABSA) aims at automatically inferring the specific sentiment polarities toward certain aspects of products or services behind the social media texts or reviews, which has been a fundamental application to the real-world society. Since the early 2010s, ABSA has achieved extraordinarily high accuracy with various deep neural models. However, existing ABSA models with strong in-house performances may fail to generalize to some challenging cases where the contexts are variable, i.e., low robustness to real-world environments. In this study, we propose to enhance the ABSA robustness by systematically rethinking the bottlenecks from all possible angles, including model, data, and training. First, we strengthen the current best-robust syntax-aware models by further incorporating the rich external syntactic dependencies and the labels with aspect simultaneously with a universal-syntax graph convolutional network. In the corpus perspective, we propose to automatically induce high-quality synthetic training data with various types, allowing models to learn sufficient inductive bias for better robustness. Last, we based on the rich pseudo data perform adversarial training to enhance the resistance to the context perturbation and meanwhile employ contrastive learning to reinforce the representations of instances with contrastive sentiments. Extensive robustness evaluations are conducted. The results demonstrate that our enhanced syntax-aware model achieves better robustness performances than all the state-of-the-art baselines. By additionally incorporating our synthetic corpus, the robust testing results are pushed with around 10% accuracy, which are then further improved by installing the advanced training strategies. In-depth analyses are presented for revealing the factors influencing the ABSA robustness.

So far, named entity recognition (NER) has been involved with three major types, including flat, overlapped (aka. nested), and discontinuous NER, which have mostly been studied individually. Recently, a growing interest has been built for unified NER, tackling the above three jobs concurrently with one single model. Current best-performing methods mainly include span-based and sequence-to-sequence models, where unfortunately the former merely focus on boundary identification and the latter may suffer from exposure bias. In this work, we present a novel alternative by modeling the unified NER as word-word relation classification, namely W^2NER. The architecture resolves the kernel bottleneck of unified NER by effectively modeling the neighboring relations between entity words with Next-Neighboring-Word (NNW) and Tail-Head-Word-* (THW-*) relations. Based on the W^2NER scheme we develop a neural framework, in which the unified NER is modeled as a 2D grid of word pairs. We then propose multi-granularity 2D convolutions for better refining the grid representations. Finally, a co-predictor is used to sufficiently reason the word-word relations. We perform extensive experiments on 14 widely-used benchmark datasets for flat, overlapped, and discontinuous NER (8 English and 6 Chinese datasets), where our model beats all the current top-performing baselines, pushing the state-of-the-art performances of unified NER.

Unified opinion role labeling (ORL) aims to detect all possible opinion structures of 'opinion-holder-target' in one shot, given a text. The existing transition-based unified method, unfortunately, is subject to longer opinion terms and fails to solve the term overlap issue. Current top performance has been achieved by employing the span-based graph model, which however still suffers from both high model complexity and insufficient interaction among opinions and roles. In this work, we investigate a novel solution by revisiting the transition architecture, and augmenting it with a pointer network (PointNet). The framework parses out all opinion structures in linear-time complexity, meanwhile breaks through the limitation of any length of terms with PointNet. To achieve the explicit opinion-role interactions, we further propose a unified dependency-opinion graph (UDOG), co-modeling the syntactic dependency structure and the partial opinion-role structure. We then devise a relation-centered graph aggregator (RCGA) to encode the multi-relational UDOG, where the resulting high-order representations are used to promote the predictions in the vanilla transition system. Our model achieves new state-of-the-art results on the MPQA benchmark. Analyses further demonstrate the superiority of our methods on both efficacy and efficiency.

Pre-trained language models (PLMs) have shown great potentials in natural language processing (NLP) including rhetorical structure theory (RST) discourse parsing.Current PLMs are obtained by sentence-level pre-training, which is different from the basic processing unit, i.e. element discourse unit (EDU).To this end, we propose a second-stage EDU-level pretraining approach in this work, which presents two novel tasks to learn effective EDU representations continually based on well pre-trained language models.Concretely, the two tasks are (1) next EDU prediction (NEP) and ( 2) discourse marker prediction (DMP).We take a state-of-the-art transition-based neural parser as baseline, and adopt it with a light bi-gram EDU modification to effectively explore the EDU-level pre-trained EDU representation.Experimental results on a benckmark dataset show that our method is highly effective, leading a 2.1-point improvement in F1-score.All codes and pre-trained models will be released publicly to facilitate future studies. 1

Recent works of opinion expression identification (OEI) rely heavily on the quality and scale of the manually-constructed training corpus, which could be extremely difficult to satisfy.Crowdsourcing is one practical solution for this problem, aiming to create a large-scale but quality-unguaranteed corpus.In this work, we investigate Chinese OEI with extremelynoisy crowdsourcing annotations, constructing a dataset at a very low cost.Following Zhang et al. (2021), we train the annotator-adapter model by regarding all annotations as goldstandard in terms of crowd annotators, and test the model by using a synthetic expert, which is a mixture of all annotators.As this annotatormixture for testing is never modeled explicitly in the training phase, we propose to generate synthetic training samples by a pertinent mixup strategy to make the training and testing highly consistent.The simulation experiments on our constructed dataset show that crowdsourcing is highly promising for OEI, and our proposed annotator-mixup can further enhance the crowdsourcing modeling.

Word ordering is a constrained language generation task taking unordered words as input. Existing work uses linear models and neural networks for the task, yet pre-trained language models have not been studied in word ordering, let alone why they help. We use BART as an instance and show its effectiveness in the task. To explain why BART helps word ordering, we extend analysis with probing and empirically identify that syntactic dependency knowledge in BART is a reliable explanation. We also report performance gains with BART in the related partial tree linearization task, which readily extends our analysis.

Successful Machine Learning based Named Entity Recognition models could fail on texts from some special domains, for instance, Chinese addresses and e-commerce titles, where requires adequate background knowledge. Such texts are also difficult for human annotators. In fact, we can obtain some potentially helpful information from correlated texts, which have some common entities, to help the text understanding. Then, one can easily reason out the correct answer by referencing correlated samples. In this paper, we suggest enhancing NER models with correlated samples. We draw correlated samples by the sparse BM25 retriever from large-scale in-domain unlabeled data. To explicitly simulate the human reasoning process, we perform a training-free entity type calibrating by majority voting. To capture correlation features in the training stage, we suggest to model correlated samples by the transformer-based multi-instance cross-encoder. Empirical results on datasets of the above two domains show the efficacy of our methods.

Boundary information is critical for various Chinese language processing tasks, such as word segmentation, part-of-speech tagging, and named entity recognition.Previous studies usually resorted to the use of a high-quality external lexicon, where lexicon items can offer explicit boundary information.However, to ensure the quality of the lexicon, great human effort is always necessary, which has been generally ignored.In this work, we suggest unsupervised statistical boundary information instead, and propose an architecture to encode the information directly into pre-trained language models, resulting in Boundary-Aware BERT (BABERT).We apply BABERT for feature induction of Chinese sequence labeling tasks.Experimental results on ten benchmarks of Chinese sequence labeling demonstrate that BABERT can provide consistent improvements on all datasets.In addition, our method can complement previous supervised lexicon exploration, where further improvements can be achieved when integrated with external lexicon information.

Conventional phrase grounding aims to localize noun phrases mentioned in a given caption to their corresponding image regions, which has achieved great success recently.Apparently, sole noun phrase grounding is not enough for cross-modal visual language understanding.Here we extend the task by considering pronouns as well.First, we construct a dataset of phrase grounding with both noun phrases and pronouns to image regions.Based on the dataset, we test the performance of phrase grounding by using a state-of-the-art literature model of this line.Then, we enhance the baseline grounding model with coreference information which should help our task potentially, modeling the coreference structures with graph convolutional networks.Experiments on our dataset, interestingly, show that pronouns are easier to ground than noun phrases, where the possible reason might be that these pronouns are much less ambiguous.Additionally, our final model with coreference information can significantly boost the grounding performance of both noun phrases and pronouns.

Image-to-text tasks, such as open-ended image captioning and controllable image description, have received extensive attention for decades.Here, we further advance this line of work by presenting Visual Spatial Description (VSD), a new perspective for image-to-text toward spatial semantics.Given an image and two objects inside it, VSD aims to produce one description focusing on the spatial perspective between the two objects.Accordingly, we manually annotate a dataset to facilitate the investigation of the newly-introduced task and build several benchmark encoder-decoder models by using VL-BART and VL-T5 as backbones.In addition, we investigate pipeline and joint end-to-end architectures for incorporating visual spatial relationship classification (VSRC) information into our model.Finally, we conduct experiments on our benchmark dataset to evaluate all our models.Results show that our models are impressive, providing accurate and human-like spatial-oriented text descriptions.Meanwhile, VSRC has great potential for VSD, and the joint end-to-end architecture is the better choice for their integration.We make the dataset and codes public for research purposes.

Named Entity Recognition (NER) from speech is among Spoken Language Understanding (SLU) tasks, aiming to extract semantic information from the speech signal. NER from speech is usually made through a two-step pipeline that consists of (1) processing the audio using an Automatic Speech Recognition (ASR) system and (2) applying an NER tagger to the ASR outputs. Recent works have shown the capability of the End-to-End (E2E) approach for NER from English and French speech, which is essentially entity-aware ASR. However, due to the many homophones and polyphones that exist in Chinese, NER from Chinese speech is effectively a more challenging task. In this paper, we introduce a new dataset AISEHLL-NER for NER from Chinese speech. Extensive experiments are conducted to explore the performance of several state-of-the-art methods. The results demonstrate that the performance could be improved by combining entity-aware ASR and pretrained NER tagger, which can be easily applied to the modern SLU pipeline. The dataset is publicly available at github.com/Alibaba-NLP/AISHELL-NER.

Many natural language processing (NLP) tasks appear in dual forms, which are generally solved by dual learning technique that models the dualities between the coupled tasks. In this work, we propose to further enhance dual learning with structure matching that explicitly builds structural connections in between. Starting with the dual text$\leftrightarrow$text generation, we perform dually-syntactic structure co-echoing of the region of interest (RoI) between the task pair, together with a syntax cross-reconstruction at the decoding side. We next extend the idea to a text$\leftrightarrow$non-text setup, making alignment between the syntactic-semantic structure. Over 2*14 tasks covering 5 dual learning scenarios, the proposed structure matching method shows its significant effectiveness in enhancing existing dual learning. Our method can retrieve the key RoIs that are highly crucial to the task performance. Besides NLP tasks, it is also revealed that our approach has great potential in facilitating more non-text$\leftrightarrow$non-text scenarios.

Conversational semantic role labeling (CSRL) is a newly proposed task that uncovers the shallow semantic structures in a dialogue text. Unfortunately several important characteristics of the CSRL task have been overlooked by the existing works, such as the structural information integration, near-neighbor influence. In this work, we investigate the integration of a latent graph for CSRL. We propose to automatically induce a predicate-oriented latent graph (POLar) with a predicate-centered gaussian mechanism, by which the nearer and informative words to the predicate will be allocated with more attention. The POLar structure is then dynamically pruned and refined so as to best fit the task need. We additionally introduce an effective dialogue-level pre-trained language model, CoDiaBERT, for better supporting multiple utterance sentences and handling the speaker coreference issue in CSRL. Our system outperforms best-performing baselines on three benchmark CSRL datasets with big margins, especially achieving over 4% F1 score improvements on the cross-utterance argument detection. Further analyses are presented to better understand the effectiveness of our proposed methods.

Linzhi Wu, Pengjun Xie, Jie Zhou, Meishan Zhang, Ma Chunping, Guangwei Xu, Min Zhang. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.

Universally modeling all typical information extraction tasks (UIE) with one generative language model (GLM) has revealed great potential by the latest study, where various IE predictions are unified into a linearized hierarchical expression under a GLM. Syntactic structure information, a type of effective feature which has been extensively utilized in IE community, should also be beneficial to UIE. In this work, we propose a novel structure-aware GLM, fully unleashing the power of syntactic knowledge for UIE. A heterogeneous structure inductor is explored to unsupervisedly induce rich heterogeneous structural representations by post-training an existing GLM. In particular, a structural broadcaster is devised to compact various latent trees into explicit high-order forests, helping to guide a better generation during decoding. We finally introduce a task-oriented structure fine-tuning mechanism, further adjusting the learned structures to most coincide with the end-task's need. Over 12 IE benchmarks across 7 tasks our system shows significant improvements over the baseline UIE system. Further in-depth analyses show that our GLM learns rich task-adaptive structural bias that greatly resolves the UIE crux, the long-range dependence issue and boundary identifying.

Joint aspect category sentiment analysis (ACSA) and rating prediction (RP) is a newly proposed task (namely ASAP) that integrates the characteristics of both fine-grained and coarse-grained sentiment analysis. However, the prior joint models for the ASAP task only consider the shallow interaction between the two granularities. In this work, we gain the inspiration from human intuition, presenting an innovative from-fine-to-coarse reasoning framework for better joint task performance. Our system advances mainly in three aspects. First, we additionally make use of the category label text features, co-encoding them with the input document texts, allowing to accurately capture the key clues of each category. Second, we build a fine-to-coarse hierarchical label graph, modeling the aspect categories and the overall rating as a hierarchical structure for full interaction of the two granularities. Third, we propose to perform global iterative reasoning with a cross-collaboration between the hierarchical label graph and the context graphs, enabling sufficient communication between categories and review contexts. Based on the ASAP dataset, experimental results demonstrate that our proposed framework outperforms state-of-the-art baselines by large margins. Further in-depth analyses prove that our method is effective on addressing both the unbalanced data distribution and the long-text issue.

End-to-end semantic role labeling (SRL) has been received increasing interest. It performs the two subtasks of SRL: predicate identification and argument role labeling, jointly. Recent work is mostly focused on graph-based neural models, while the transition-based framework with neural networks which has been widely used in a number of closely-related tasks, has not been studied for the joint task yet. In this paper, we present the first work of transition-based neural models for end-to-end SRL. Our transition model incrementally discovers all sentential predicates as well as their arguments by a set of transition actions. The actions of the two subtasks are executed mutually for full interactions. Besides, we suggest high-order compositions to extract non-local features, which can enhance the proposed transition model further. Experimental results on CoNLL09 and Universal Proposition Bank show that our final model can produce state-of-the-art performance, and meanwhile keeps highly efficient in decoding. We also conduct detailed experimental analysis for a deep understanding of our proposed model.

Fei Li, ZhiChao Lin, Meishan Zhang, Donghong Ji. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Xin Zhang, Guangwei Xu, Yueheng Sun, Meishan Zhang, Pengjun Xie. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2021.

Domain adaption for word segmentation and POS tagging is a challenging problem for Chinese lexical processing.Self-training is one promising solution for it, which struggles to construct a set of high-quality pseudo training instances for the target domain.Previous work usually assumes a universal sourceto-target adaption to collect such pseudo corpus, ignoring the different gaps from the target sentences to the source domain.In this work, we start from joint word segmentation and POS tagging, presenting a fine-grained domain adaption method to model the gaps accurately.We measure the gaps by one simple and intuitive metric, and adopt it to develop a pseudo target domain corpus based on finegrained subdomains incrementally.A novel domain-mixed representation learning model is proposed accordingly to encode the multiple subdomains effectively.The whole process is performed progressively for both corpus construction and model training.Experimental results on a benchmark dataset show that our method can gain significant improvements over a vary of baselines.Extensive analyses are performed to show the advantages of our final domain adaption model as well.

Frame semantic parsing is a semantic analysis task based on FrameNet which has received great attention recently.The task usually involves three subtasks sequentially: (1) target identification, (2) frame classification and (3) semantic role labeling.The three subtasks are closely related while previous studies model them individually, which ignores their intern connections and meanwhile induces error propagation problem.In this work, we propose an end-to-end neural model to tackle the task jointly.Concretely, we exploit a graphbased method, regarding frame semantic parsing as a graph construction problem.All predicates and roles are treated as graph nodes, and their relations are taken as graph edges.Experiment results on two benchmark datasets of frame semantic parsing show that our method is highly competitive, resulting in better performance than pipeline models.

Opinion Role Labeling (ORL), aiming to identify the key roles of opinion, has received increasing interest.Unlike most of the previous works focusing on the English language, in this paper, we present the first work of Chinese ORL.We construct a Chinese dataset by manually translating and projecting annotations from a standard English MPQA dataset.Then, we investigate the effectiveness of cross-lingual transfer methods, including model transfer and corpus translation.We exploit multilingual BERT with Contextual Parameter Generator and Adapter methods to examine the potentials of unsupervised crosslingual learning and our experiments and analyses for both bilingual and multilingual transfers establish a foundation for the future research of this task 1 .

With the rapid development of web and mobile applications, as well as their wide adoption in different domains, more and more personal data is provided, consciously or unconsciously, to different application providers. Privacy policy is an important medium for users to understand what personal information has been collected and used. As data privacy protection is becoming a critical social issue, there are laws and regulations being enacted in different countries and regions, and the most representative one is the EU General Data Protection Regulation (GDPR). It is thus important to detect compliance issues among regulations, e.g., GDPR, with privacy policies, and provide intuitive results for data subjects (i.e., users), data collection party (i.e., service providers) and the regulatory authorities. In this work, we target to solve the problem of compliance analysis between GDPR (Article 13) and privacy policies. We format the task into a combination of a sentence classification step and a rule-based analysis step. We manually curate a corpus of 36,610 labeled sentences from 304 privacy policies, and benchmark our corpus with several standard sentence classifiers. We also conduct a rule-based analysis to detect compliance issues and a user study to evaluate the usability of our approach. The web-based tool AutoCompliance is publicly accessible 1.

Named entity recognition (NER) is one fundamental task in natural language processing, which is typically addressed by neural condition random field (CRF) models, regarding the task as a sequence labeling problem. Sentence-level information has been shown positive for the task. Equipped with sophisticated neural structures such as long-short term memory network (LSTM), implicit sentence-level global information can be exploited fully, and has also been demonstrated effective in previous studies. In this work, we propose a new method for better learning of these sentence-level features in an explicit manner. Concretely, we suggest an auxiliary task, namely sentence-level named type prediction (i.e., determining whether a sentence includes a certain kind of named type), to supervise the feature representation learning globally. We conduct experiments on six benchmark datasets of various languages to evaluate our method. The results show that our final model is highly effective, resulting in significant improvements and leading to highly competitive results on all datasets.

Character-level Chinese dependency parsing has been a concern of several studies that naturally handle word segmentation, POS (Part of Speech) tagging and dependency parsing jointly in an end-to-end way. Previous work mostly concentrates on a transition-based framework for this task because of its easy adaption, which is extremely important when feature representation relies heavily on the decoding strategy, particularly under the traditional statistical setting. Recently, on the one hand, sophisticated deep neural networks and deep contextualized word representations have greatly weakened the dependence between feature representation and decoding. On the other hand, (first-order) graph-based models, especially the biaffine parsers, are straightforward for dependency parsing, and meanwhile they can yield competitive parsing performance. In this paper, we make a comprehensive investigation of the deep graph-based character-level dependency parsing for Chinese. We start from an extension of a standard graph-based biaffine parser, and then exploit Chinese BERT as well as our improved encoders based on transformers to enhance the character-level dependency parsing model. We conduct a series of experiments on the Chinese benchmark datasets, showing the performances of various graph-based character-level models and analyzing the advantages of the character-level dependency parsing under the deep neural setting.

Many efforts of research are devoted to semantic role labeling (SRL) which is crucial for natural language understanding.Supervised approaches have achieved impressing performances when large-scale corpora are available for resource-rich languages such as English.While for the low-resource languages with no annotated SRL dataset, it is still challenging to obtain competitive performances.Cross-lingual SRL is one promising way to address the problem, which has achieved great advances with the help of model transferring and annotation projection.In this paper, we propose a novel alternative based on corpus translation, constructing high-quality training datasets for the target languages from the source gold-standard SRL annotations.Experimental results on Universal Proposition Bank show that the translation-based method is highly effective, and the automatic pseudo datasets can improve the target-language SRL performances significantly.

Discourse representation tree structure (DRTS) parsing is a novel semantic parsing task which has been concerned most recently.State-of-the-art performance can be achieved by a neural sequence-to-sequence model, treating the tree construction as an incremental sequence generation problem.Structural information such as input syntax and the intermediate skeleton of the partial output has been ignored in the model, which could be potentially useful for the DRTS parsing.In this work, we propose a structural-aware model at both the encoder and decoder phase to integrate the structural information, where graph attention network (GAT) is exploited for effectively modeling.Experimental results on a benchmark dataset show that our proposed model is effective and can obtain the best performance in the literature.

Emotion detection in conversations (EDC) is to detect the emotion for each utterance in conversations that have multiple speakers. Different from the traditional non-conversational emotion detection, the model for EDC should be context-sensitive (e.g., understanding the whole conversation rather than one utterance) and speaker-sensitive (e.g., understanding which utterance belongs to which speaker). In this paper, we propose a transformer-based context- and speaker-sensitive model for EDC, namely HiTrans, which consists of two hierarchical transformers. We utilize BERT as the low-level transformer to generate local utterance representations, and feed them into another high-level transformer so that utterance representations could be sensitive to the global context of the conversation. Moreover, we exploit an auxiliary task to make our model speaker-sensitive, called pairwise utterance speaker verification (PUSV), which aims to classify whether two utterances belong to the same speaker. We evaluate our model on three benchmark datasets, namely EmoryNLP, MELD and IEMOCAP. Results show that our model outperforms previous state-of-the-art models.

In this paper, we present Chinese lexical fusion recognition, a new task which could be regarded as one kind of coreference recognition.First, we introduce the task in detail, showing the relationship with coreference recognition and differences from the existing tasks.Second, we propose an end-to-end model for the task, handling mentions as well as coreference relationship jointly.The model exploits the state-of-the-art contextualized BERT representations as the encoder, and is further enhanced with the sememe knowledge from HowNet by graph attention networks.We manually annotate a benchmark dataset for the task and then conduct experiments on it.Results demonstrate that our final model is effective and competitive for the task.Detailed analysis is offered for comprehensively understanding the new task and our proposed model.

Prior studies show that cross-lingual semantic role labeling (SRL) can be achieved by model transfer under the help of universal features. In this article, we fill the gap of cross-lingual SRL by proposing an end-to-end SRL model that incorporates a variety of universal features and transfer methods. We study both the bilingual transfer and multi-source transfer, under gold or machine-generated syntactic inputs, pre-trained high-order abstract features, and contextualized multilingual word representations. Experimental results on the Universal Proposition Bank corpus indicate that performances of the cross-lingual SRL can vary by leveraging different cross-lingual features. In addition, whether the features are gold-standard also has an impact on performances. Precisely, we find that gold syntax features are much more crucial for cross-lingual SRL, compared with the automatically-generated ones. Moreover, universal dependency structure features are able to give the best help, and both pre-trained high-order features and contextualized word representations can further bring significant improvements.

Semantic role labeling (SRL), also known as shallow semantic parsing, is an important yet challenging task in NLP. Motivated by the close correlation between syntactic and semantic structures, traditional discrete-feature-based SRL approaches make heavy use of syntactic features. In contrast, deep-neural-network-based approaches usually encode the input sentence as a word sequence without considering the syntactic structures. In this work, we investigate several previous approaches for encoding syntactic trees, and make a thorough study on whether extra syntax-aware representations are beneficial for neural SRL models. Experiments on the benchmark CoNLL-2005 dataset show that syntax-aware SRL approaches can effectively improve performance over a strong baseline with external word representations from ELMo. With the extra syntax-aware representations, our approaches achieve new state-of-the-art 85.6 F1 (single model) and 86.6 F1 (ensemble) on the test data, outperforming the corresponding strong baselines with ELMo by 0.8 and 1.0, respectively. Detailed error analysis are conducted to gain more insights on the investigated approaches.

Meishan Zhang, Yue Zhang, Guohong Fu. Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP). 2019.

Distant supervised relation extraction is an efficient approach to scale relation extraction to very large corpora, and has been widely used to find novel relational facts from plain text. Recent studies on neural relation extraction have shown great progress on this task via modeling the sentences in low-dimensional spaces, but seldom considered syntax information to model the entities. In this paper, we propose to learn syntax-aware entity embedding for neural relation extraction. First, we encode the context of entities on a dependency tree as sentence-level entity embedding based on tree-GRU. Then, we utilize both intra-sentence and inter-sentence attentions to obtain sentence set-level entity embedding over all sentences containing the focus entity pair. Finally, we combine both sentence embedding and entity embedding for relation classification. We conduct experiments on a widely used real-world dataset and the experimental results show that our model can make full use of all informative instances and achieve state-of-the-art performance of relation extraction.

To quickly obtain new labeled data, we can choose crowdsourcing as an alternative way at lower cost in a short time. But as an exchange, crowd annotations from non-experts may be of lower quality than those from experts. In this paper, we propose an approach to performing crowd annotation learning for Chinese Named Entity Recognition (NER) to make full use of the noisy sequence labels from multiple annotators. Inspired by adversarial learning, our approach uses a common Bi-LSTM and a private Bi-LSTM for representing annotator-generic and -specific information. The annotator-generic information is the common knowledge for entities easily mastered by the crowd. Finally, we build our Chinese NE tagger based on the LSTM-CRF model. In our experiments, we create two data sets for Chinese NER tasks from two domains. The experimental results show that our system achieves better scores than strong baseline systems.

Syntax has been a useful source of information for statistical RST discourse parsing. Under the neural setting, a common approach integrates syntax by a recursive neural network (RNN), requiring discrete output trees produced by a supervised syntax parser. In this paper, we propose an implicit syntax feature extraction approach, using hidden-layer vectors extracted from a neural syntax parser. In addition, we propose a simple transition-based model as the baseline, further enhancing it with dynamic oracle. Experiments on the standard dataset show that our baseline model with dynamic oracle is highly competitive. When implicit syntax features are integrated, we are able to obtain further improvements, better than using explicit Tree-RNN.

Character-based and word-based methods are two different solutions for Chinese word segmentation, the former exploiting sequence labeling models over characters and the latter using word-level features. Neural models have been exploited for character-based Chinese word segmentation, giving high accuracies by making use of external character embeddings, yet requiring less feature engineering. In this paper, we study a neural model for word-based Chinese word segmentation, by replacing the manually-designed discrete features with neural features in a transition-based word segmentation framework. Experimental results demonstrate that word features lead to comparable performance to the best systems in the literature, and a further combination of discrete and neural features obtains top accuracies on several benchmarks.

While part-of-speech (POS) tagging and dependency parsing are observed to be closely related, existing work on joint modeling with manually crafted feature templates suffers from the feature sparsity and incompleteness problems. In this paper, we propose an approach to joint POS tagging and dependency parsing using transition-based neural networks. Three neural network based classifiers are designed to resolve shift/reduce, tagging, and labeling conflicts. Experiments show that our approach significantly outperforms previous methods for joint POS tagging and dependency parsing across a variety of natural languages.

Joint models have shown stronger capabilities for Chinese word segmentation and POS tagging, and have received great interests in the community of Chinese natural language processing. In this paper, we follow this line of work, presenting a simple yet effective sequence-to-sequence neural model for the joint task, based on a well-defined transition system, by using long short term memory neural network structures. We conduct experiments on five different datasets. The results demonstrate that our proposed model is highly competitive. By using well-trained character-level embeddings, the proposed neural joint model is able to obtain the best-reported performances in the literature.

We model the problem of disfluency detection using a transition-based framework, which incrementally constructs and labels the disfluency chunk of input sentences using a set of transition actions without syntax information.Compared with sequence labeling methods, it can capture non-local chunk-level features; compared with joint parsing and disfluency detection methods, it is free for noise in syntax.Experiments show that our model achieves state-of-theart F-score on both the commonly used English Switchboard test set and a set of in-house annotated Chinese data.

Neural networks have shown promising results for relation extraction.State-ofthe-art models cast the task as an end-toend problem, solved incrementally using a local classifier.Yet previous work using statistical models have demonstrated that global optimization can achieve better performances compared to local classification.We build a globally optimized neural model for end-to-end relation extraction, proposing novel LSTM features in order to better learn context representations.In addition, we present a novel method to integrate syntactic information to facilitate global learning, yet requiring little background on syntactic grammars thus being easy to extend.Experimental results show that our proposed model is highly effective, achieving the best performances on two standard benchmarks.

State-of-the-art Chinese word segmentation systems typically exploit supervised modelstrained on a standard manually-annotated corpus,achieving performances over 95% on a similar standard testing corpus.However, the performances may drop significantly when the same models are applied onto Chinese microtext.One major challenge is the issue of informal words in the microtext.Previous studies show that informal word detection can be helpful for microtext processing.In this work, we investigate it under the neural setting, by proposing a joint segmentation model that integrates the detection of informal words simultaneously.In addition, we generate training corpus for the joint model by using existing corpus automatically.Experimental results show that the proposed model is highly effective for segmentation of Chinese microtext.

BACKGROUND: Extracting biomedical entities and their relations from text has important applications on biomedical research. Previous work primarily utilized feature-based pipeline models to process this task. Many efforts need to be made on feature engineering when feature-based models are employed. Moreover, pipeline models may suffer error propagation and are not able to utilize the interactions between subtasks. Therefore, we propose a neural joint model to extract biomedical entities as well as their relations simultaneously, and it can alleviate the problems above. RESULTS: Our model was evaluated on two tasks, i.e., the task of extracting adverse drug events between drug and disease entities, and the task of extracting resident relations between bacteria and location entities. Compared with the state-of-the-art systems in these tasks, our model improved the F1 scores of the first task by 5.1% in entity recognition and 8.0% in relation extraction, and that of the second task by 9.2% in relation extraction. CONCLUSIONS: The proposed model achieves competitive performances with less work on feature engineering. We demonstrate that the model based on neural networks is effective for biomedical entity and relation extraction. In addition, parameter sharing is an alternative method for neural models to jointly process this task. Our work can facilitate the research on biomedical text mining.

The limited scale and genre coverage of labeled data greatly hinders the effectiveness of supervised models, especially when analyzing spoken languages, such as texts transcribed from speech and informal text including tweets and product comments in Internet. In order to effectively utilize multiple labeled datasets with heterogeneous annotations for the same task, this paper proposes a coupled sequence labeling model that can directly learn and infer two heterogeneous annotations simultaneously, using Chinese part-of-speech (POS) tagging as our case study. The key idea is to bundle two sets of POS tags together (e.g., “[NN, n]n), and build a conditional random field (CRF) based tagging model in the enlarged space of bundled tags with the help of ambiguous labeling. To train our model on two nonoverlapping datasets that each has only one-side tags, we transform a one-side tag into a set of bundled tags by concatenating the tag with every possible tag at the missing side according to a predefined context-free tag-to-tag mapping function, thus producing ambiguous labeling as weak supervision. We design and investigate four different context-free tag-to-tag mapping functions, and find out that the coupled model achieves its best performance when each one-side tag is mapped to all tags at the other side (namely complete mapping), indicating that the model can effectively learn the loose mapping between the two heterogeneous annotations, without the need of manually designed mapping rules. Moreover, we propose a context-aware online pruning strategy that can more accurately capture mapping relationships between annotations based on contextual evidences and thus effectively solve the severe inefficiency problem with our coupled model under complete mapping, making it comparable with the baseline CRF model. Experiments on benchmark datasets show that our coupled model significantly outperforms the state-of-the-art baselines on both one-side POS tagging and annotation conversion tasks. The codes and newly annotated data are released for research usage.1

We present a novel annotation framework for representing predicate-argument structures, which uses dependency trees to encode the syntactic and semantic roles of a sentence simultaneously. The main contribution is a semantic role transmission model, which eliminates the structural gap between syntax and shallow semantics, making them compatible. A Chinese semantic treebank was built under the proposed framework, and the first release containing about 14K sentences is made freely available. The proposed framework enables semantic role labeling to be solved as a sequence labeling task, and experiments show that standard sequence labelers can give competitive performance on the new treebank compared with state-of-the-art graph structure models.

Sentiment classification on Twitter has attracted increasing research in recent years.Most existing work focuses on feature engineering according to the tweet content itself.In this paper, we propose a context-based neural network model for Twitter sentiment analysis, incorporating contextualized features from relevant Tweets into the model in the form of word embedding vectors.Experiments on both balanced and unbalanced datasets show that our proposed models outperform the current state-of-the-art.

It has been shown that learning distributed word representations is highly useful for Twitter sentiment classification.Most existing models rely on a single distributed representation for each word.This is problematic for sentiment classification because words are often polysemous and each word can contain different sentiment polarities under different topics.We address this issue by learning topic-enriched multi-prototype word embeddings (TMWE).In particular, we develop two neural networks which 1) learn word embeddings that better capture tweet context by incorporating topic information, and 2) learn topic-enriched multiple prototype embeddings for each word.Experiments on Twitter sentiment benchmark datasets in SemEval 2013 show that TMWE outperforms the top system with hand-crafted features, and the current best neural network model.

Targeted sentiment analysis classifies the sentiment polarity towards each target entity mention in given text documents. Seminal methods extract manual discrete features from automatic syntactic parse trees in order to capture semantic information of the enclosing sentence with respect to a target entity mention. Recently, it has been shown that competitive accuracies can be achieved without using syntactic parsers, which can be highly inaccurate on noisy text such as tweets. This is achieved by applying distributed word representations and rich neural pooling functions over a simple and intuitive segmentation of tweets according to target entity mentions. In this paper, we extend this idea by proposing a sentence-level neural model to address the limitation of pooling functions, which do not explicitly model tweet-level semantics. First, a bi-directional gated neural network is used to connect the words in a tweet so that pooling functions can be applied over the hidden layer instead of words for better representing the target and its contexts. Second, a three-way gated neural network structure is used to model the interaction between the target mention and its surrounding contexts. Experiments show that our proposed model gives significantly higher accuracies compared to the current best method for targeted sentiment analysis.

Sarcasm detection has been modeled as a binary document classification task, with rich features being defined manually over input documents. Traditional models employ discrete manual features to address the task, with much research effect being devoted to the design of effective feature templates. We investigate the use of neural network for tweet sarcasm detection, and compare the effects of the continuous automatic features with discrete manual features. In particular, we use a bi-directional gated recurrent neural network to capture syntactic and semantic information over tweets locally, and a pooling neural network to extract contextual features automatically from history tweets. Results show that neural features give improved accuracies for sarcasm detection, with different error distributions compared with discrete manual features.

Rui Sun, Yue Zhang, Meishan Zhang, Donghong Ji. Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers). 2015.

We propose a transition-based model for joint word segmentation, POS tagging and text normalization.Different from previous methods, the model can be trained on standard text corpora, overcoming the lack of annotated microblog corpora.To evaluate our model, we develop an annotated corpus based on microblogs.Experimental results show that our joint model can help improve the performance of word segmentation on microblogs, giving an error reduction in segmentation accuracy of 12.02%, compared to the traditional approach.

We investigate a combination of a traditional linear sparse feature model and a multi-layer neural network model for deterministic transition-based dependency parsing, by integrating the sparse features into the neural model.Correlations are drawn between the hybrid model and previous work on integrating word embedding features into a discrete linear model.By analyzing the results of various parsers on web-domain parsing, we show that the integrated model is a better way to combine traditional and embedding features compared with previous methods.

Open domain targeted sentiment is the joint information extraction task that finds target mentions together with the sentiment towards each mention from a text corpus.The task is typically modeled as a sequence labeling problem, and solved using state-of-the-art labelers such as CRF.We empirically study the effect of word embeddings and automatic feature combinations on the task by extending a CRF baseline using neural networks, which have demonstrated large potentials for sentiment analysis.Results show that the neural model can give better results by significantly increasing the recall.In addition, we propose a novel integration of neural and discrete features, which combines their relative advantages, leading to significantly higher results compared to both baselines.

Recent work on Chinese analysis has led to large-scale annotations of the internal structures of words, enabling characterlevel analysis of Chinese syntactic structures.In this paper, we investigate the problem of character-level Chinese dependency parsing, building dependency trees over characters.Character-level information can benefit downstream applications by offering flexible granularities for word segmentation while improving wordlevel dependency parsing accuracies.We present novel adaptations of two major shift-reduce dependency parsing algorithms to character-level parsing.Experimental results on the Chinese Treebank demonstrate improved performances over word-based parsing methods.

We report an empirical investigation on type-supervised domain adaptation for joint Chinese word segmentation and POS-tagging, making use of domainspecific tag dictionaries and only unlabeled target domain data to improve target-domain accuracies, given a set of annotated source domain sentences.Previous work on POS-tagging of other languages showed that type-supervision can be a competitive alternative to tokensupervision, while semi-supervised techniques such as label propagation are important to the effectiveness of typesupervision.We report similar findings using a novel approach for joint Chinese segmentation and POS-tagging, under a cross-domain setting.With the help of unlabeled sentences and a lexicon of 3,000 words, we obtain 33% error reduction in target-domain tagging.In addition, combined type-and token-supervision can lead to improved cost-effectiveness.