Yuzhen Shi, Huanghai Liu, Yiran HU, Song Gaojie, Xu Xinran, Yubo Ma, Tianyi Tang, Li Zhang, Qingjing Chen, Feng Di, Wenbo Lv, Weiheng Wu, Kexin Yang, Sen Yang, Wei Wang, Rongyao Shi, Qiu Yuanyang, Yuemeng Qi, Zhang Jingwen, Sui Xiaoyu, Yifan Chen, Zhang Yi, An Yang, Bowen Yu, Dayiheng Liu, Junyang Lin, Weixing Shen, Bing Zhao, Charles L. A. Clarke, HU Wei. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Binghai Wang, Yantao Liu, Yuxuan Liu, Tianyi Tang, Shenzhi Wang, Chang Gao, Chujie Zheng, Yichang Zhang, Le Yu, Shixuan Liu, Tao Gui, Qi Zhang, Xuanjing Huang, Bowen Yu, Fei Huang, Junyang Lin. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Chujie Zheng, Zhenru Zhang, Beichen Zhang, Runji Lin, Keming Lu, Bowen Yu, Dayiheng Liu, Jingren Zhou, Junyang Lin. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Chengpeng Li, Mingfeng Xue, Zhenru Zhang, Jiaxi Yang, Beichen Zhang, Bowen Yu, Binyuan Hui, Junyang Lin, Xiang Wang, Dayiheng Liu. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Yidan Zhang, Yu Wan, Boyi Deng, Baosong Yang, Hao-Ran Wei, Fei Huang, Bowen Yu, Dayiheng Liu, Junyang Lin, Fei Huang, Jingren Zhou. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

One core capability of large language models~(LLMs) is to follow natural language instructions. However, the issue of automatically constructing high-quality training data to enhance the complex instruction-following abilities of LLMs without manual annotation remains unresolved. In this paper, we introduce AutoIF, the first scalable and reliable method for automatically generating instruction-following training data. AutoIF transforms the validation of instruction-following data quality into code verification, requiring LLMs to generate instructions, the corresponding code to verify the correctness of the instruction responses, and unit test samples to cross-validate the code's correctness. Then, execution feedback-based rejection sampling can generate data for Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF) training. AutoIF achieves significant improvements across three training algorithms, SFT, Offline DPO, and Online DPO, when applied to the advanced open-source LLMs, Qwen2 and LLaMA3, in self-alignment and strong-to-weak distillation settings. Using two widely-used and three challenging general instruction-following benchmarks, we demonstrate that AutoIF significantly improves LLM performance across a wide range of natural instruction constraints. Notably, AutoIF is the first to surpass 90\% accuracy in IFEval’s loose instruction accuracy, without compromising general, math and coding capabilities. Further analysis of quality, scaling, combination, and data efficiency highlights AutoIF's strong generalization and alignment potential. Our code are available at https://github.com/QwenLM/AutoIF

Large Language Models (LLMs) exhibit strong potential in mathematical reasoning, yet their effectiveness is often limited by a shortage of high-quality queries. This limitation necessitates scaling up computational responses through self-generated data, yet current methods struggle due to spurious correlated data caused by ineffective exploration across all reasoning stages. To address such challenge, we introduce **MARGE**: Improving **Ma**th **R**easoning with **G**uided **E**xploration, a novel method that enhances mathematical reasoning through hit-guided exploration. MARGE systematically explores intermediate reasoning states derived from self-generated solutions, enabling adequate exploration and improved credit assignment throughout the reasoning process. Notably, MARGE improves both single-shot accuracy and exploration diversity, mitigating a common trade-off in alignment methods. These results demonstrate MARGE's effectiveness in enhancing mathematical reasoning capabilities and unlocking the potential of scaling self-generated training data.

Long context fine-tuning of large language models(LLMs) involves training on datasets that are predominantly composed of short sequences and a small proportion of longer sequences. However, existing approaches overlook this long-tail distribution and employ training strategies designed specifically for long sequences. Moreover, these approaches also fail to address the challenges posed by variable sequence lengths during distributed training, such as load imbalance in data parallelism and severe pipeline bubbles in pipeline parallelism. These issues lead to suboptimal training performance and poor GPU resource utilization. To tackle these problems, we propose a chunk-centric training method named ChunkFlow. ChunkFlow reorganizes input sequences into uniformly sized chunks by consolidating short sequences and splitting longer ones. This approach achieves optimal computational efficiency and balance among training inputs. Additionally, ChunkFlow incorporates a state-aware chunk scheduling mechanism to ensure that the peak memory usage during training is primarily determined by the chunk size rather than the maximum sequence length in the dataset. Integrating this scheduling mechanism with existing pipeline scheduling algorithms further enhances the performance of distributed training. Experimental results demonstrate that, compared with Megatron-LM, ChunkFlow can be up to 4.53x faster in the long context fine-tuning of LLMs. Furthermore, we believe that ChunkFlow serves as an effective solution for a broader range of scenarios, such as long context continual pre-training, where datasets contain variable-length sequences.

Xinyu Lu, Xueru Wen, Yaojie Lu, Bowen Yu, Hongyu Lin, Haiyang Yu, Le Sun, Xianpei Han, Yongbin Li. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.

Reinforcement Learning with Verifiable Rewards (RLVR) has emerged as a powerful approach to enhancing the reasoning capabilities of Large Language Models (LLMs), yet its underlying mechanisms remain insufficiently understood. In this work, we undertake a pioneering exploration of RLVR through the novel perspective of token entropy patterns, comprehensively analyzing how different tokens influence reasoning performance. By examining token entropy patterns in Chain-of-Thought (CoT) reasoning, we observe that only a small fraction (approximately 20\%) of tokens exhibit high entropy, and these tokens semantically act as critical forks that steer the model toward diverse reasoning pathways. We further demonstrate that moderately increasing the entropy of these high-entropy tokens via decoding temperature adjustments leads to improved performance, quantitatively confirming their role as decision points in reasoning. We ultimately refine RLVR by restricting policy gradient updates to these forking tokens. Despite utilizing only 20\% of tokens, our approach achieves comparable performance to full-gradient updates on the Qwen3-8B base model. Moreover, it demonstrates remarkable improvements on the larger Qwen3-32B base model, boosting AIME'25 scores by 11.04 and AIME'24 scores by 7.71. In contrast, training exclusively on the 80\% lowest-entropy tokens leads to a marked decline in performance. These findings indicate that the efficacy of RLVR primarily arises from optimizing the high-entropy tokens that dictate key reasoning directions. Collectively, our results suggest promising avenues for optimizing RLVR algorithms by strategically leveraging the potential of these high-entropy minority tokens to further enhance the reasoning abilities of LLMs.

Large language models (LLMs) often contain misleading content, emphasizing the need to align them with human values to ensure secure AI systems. Reinforcement learning from human feedback (RLHF) has been employed to achieve this alignment. However, it encompasses two main drawbacks: (1) RLHF exhibits complexity, instability, and sensitivity to hyperparameters in contrast to SFT. (2) Despite massive trial-and-error, multiple sampling is reduced to pair-wise contrast, thus lacking contrasts from a macro perspective. In this paper, we propose Preference Ranking Optimization (PRO) as an efficient SFT algorithm to directly fine-tune LLMs for human alignment. PRO extends the pair-wise contrast to accommodate preference rankings of any length. By iteratively contrasting candidates, PRO instructs the LLM to prioritize the best response while progressively ranking the rest responses. In this manner, PRO effectively transforms human alignment into aligning the probability ranking of n responses generated by LLM with the preference ranking of humans towards these responses. Experiments have shown that PRO outperforms baseline algorithms, achieving comparable results to ChatGPT and human responses through automatic-based, reward-based, GPT-4, and human evaluations.

Considerable efforts have been invested in augmenting the role-playing proficiency of open-source large language models (LLMs) by emulating proprietary counterparts.Nevertheless, we posit that LLMs inherently harbor role-play capabilities, owing to the extensive knowledge of characters and potential dialogues ingrained in their vast training corpora.Thus, in this study, we introduce DITTO, a self-alignment method for role-play.DITTO capitalizes on character knowledge, encouraging an instruction-following LLM to simulate role-play dialogues as a variant of reading comprehension.This method creates a role-play training set comprising 4000 characters, surpassing the scale of currently available datasets by tenfold regarding the number of roles.Subsequently, we fine-tune the LLM using this self-generated dataset to augment its role-playing capabilities.Upon evaluating our meticulously constructed and reproducible role-play benchmark and the roleplay subset of MT-Bench, DITTO in various parameter scales consistently maintains a consistent role identity and provides accurate role-specific knowledge in multi-turn role-play conversations.Notably, it outperforms all open-source role-play baselines, showcasing performance levels comparable to advanced proprietary chatbots.Furthermore, we present the first comprehensive cross-supervision alignment experiment in the role-play domain, revealing that the intrinsic capabilities of LLMs confine the knowledge within role-play.Meanwhile, the role-play styles can be easily acquired with the guidance of smaller models.We open-source related resources in https://github.com/ OFA-Sys/Ditto.

Alignment with human preference prevents large language models (LLMs) from generating misleading or toxic content while requiring high-cost human feedback. Assuming resources of human annotation are limited, there are two different ways of allocating considered: more diverse PROMPTS or more diverse RESPONSES to be labeled. Nonetheless, a straightforward comparison between their impact is absent. In this work, we first control the diversity of both sides according to the number of samples for fine-tuning, which can directly reflect their influence. We find that instead of numerous prompts, more responses but fewer prompts better trigger LLMs for human alignment. Additionally, the concept of diversity for prompts can be more complex than responses that are typically quantified by single digits. Consequently, a new formulation of prompt diversity is proposed, further implying a linear correlation with the final performance of LLMs after fine-tuning. We also leverage it on data augmentation and conduct experiments to show its effect on different algorithms.

Training large language models (LLMs) with open-domain instruction data has yielded remarkable success in aligning to end tasks and human preferences. Extensive research has highlighted the importance of the quality and diversity of instruction data. However, the impact of data complexity, as a crucial metric, remains relatively unexplored from three aspects: (1)where the sustainability of performance improvements with increasing complexity is uncertain; (2)whether the improvement brought by complexity merely comes from introducing more training tokens; and (3)where the potential benefits of incorporating instructions from easy to difficult are not yet fully understood. In this paper, we propose Tree-Instruct to systematically enhance the instruction complexity in a controllable manner. By adding a specified number of nodes to instructions’ semantic trees, this approach not only yields new instruction data from the modified tree but also allows us to control the difficulty level of modified instructions. Our preliminary experiments reveal the following insights: (1)Increasing complexity consistently leads to sustained performance improvements of LLMs. (2)Under the same token budget, a few complex instructions outperform diverse yet simple instructions. (3)Curriculum instruction tuning might not yield the anticipated results; focusing on increasing complexity appears to be the key.

Transformer-based large language models (LLMs) exhibit limitations such as generating unsafe responses, unreliable reasoning, etc. Existing inference intervention approaches attempt to mitigate these issues by finetuning additional models to produce calibration signals (such as rewards) that guide the LLM's decoding process.However, this solution introduces substantial time and space overhead due to the separate models required.This work proposes NOn-disruptive parameters insertion (Otter), inserting extra parameters into the transformer architecture to predict calibration signals along with the original LLM output.Otter offers state-of-the-art performance on multiple demanding tasks while saving up to 86.5% extra space and 98.5% extra time.Furthermore, Otter seamlessly integrates with existing inference engines, requiring only a oneline code change, and the original model response remains accessible after the parameter insertion.

In this paper, we unveil that Language Models (LMs) can acquire new capabilities by assimilating parameters from homologous models without retraining or GPUs. We first introduce DARE to set most delta parameters (i.e., the disparity between fine-tuned and pre-trained parameters) to zeros without affecting the abilities of Supervised Fine-Tuning (SFT) LMs, which randomly **D**rops delta parameters with a ratio $p$ **A**nd **RE**scales the remaining ones by $1 / (1 - p)$ to approximate the original embeddings. Then, we use DARE as a versatile plug-in to sparsify delta parameters of multiple SFT homologous models for mitigating parameter interference and merge them into a single model by parameter fusing. We experiment with encoder- and decoder-based LMs, showing that: (1) SFT delta parameter value ranges are typically small (within 0.002) with extreme redundancy, and DARE can effortlessly eliminate 90% or even 99% of them; (2) DARE can merge multiple task-specific LMs into one LM with diverse capabilities. Notably, this phenomenon is more pronounced in large-scale LMs, where the merged LM reveals the potential to surpass the performance of any source LM, providing a new discovery. We also utilize DARE to create a merged LM that ranks first among models with 7 billion parameters on the Open LLM Leaderboard.

Yucheng Wang, Bowen Yu, Yilin Liu, Shudong Lu. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

The rise of large language models (LLMs) has significantly transformed both the construction and application of information retrieval (IR) systems. However, current interactions between IR systems and LLMs remain limited, with LLMs merely serving as part of components within IR systems, and IR systems being constructed independently of LLMs. This separated architecture restricts knowledge sharing and deep collaboration between them. In this paper, we introduce Self-Retrieval, a novel end-to-end LLM-driven information retrieval architecture. Self-Retrieval unifies all essential IR functions within a single LLM, leveraging the inherent capabilities of LLMs throughout the IR process. Specifically, Self-Retrieval internalizes the retrieval corpus through self-supervised learning, transforms the retrieval process into sequential passage generation, and performs relevance assessment for reranking. Experimental results demonstrate that Self-Retrieval not only outperforms existing retrieval approaches by a significant margin, but also substantially enhances the performance of LLM-driven downstream applications like retrieval-augmented generation.

Cross-domain Named Entity Recognition (NER) transfers knowledge learned from a rich-resource source domain to improve the learning in a low-resource target domain. Most existing works are designed based on the sequence labeling framework, defining entity detection and type prediction as a monolithic process. However, they typically ignore the discrepant transferability of these two sub-tasks: the former locating spans corresponding to entities is largely domain-robust, whereas the latter owns distinct entity types across domains. Combining them into an entangled learning problem may contribute to the complexity of domain transfer. In this work, we propose the novel divide-and-transfer paradigm in which different sub-tasks are learned using separate functional modules for respective cross-domain transfer. To demonstrate the effectiveness of divide-and-transfer, we concretely implement two NER frameworks by applying this paradigm with different cross-domain transfer strategies. Experimental results on 10 different domain pairs show the notable superiority of our proposed frameworks. Experimental analyses indicate that significant advantages of the divide-and-transfer paradigm over prior monolithic ones originate from its better performance on low-resource data and a much greater transferability. It gives us a new insight into cross-domain NER. Our code is available on GitHub. 1

Graph contrastive learning (GCL) has recently achieved substantial advancements. Existing GCL approaches compare two different “views“ of the same graph in order to learn node/graph representations. The underlying assumption of these studies is that the graph augmentation strategy is capable of generating several different graph views such that the graph views are structurally different but semantically similar to the original graphs, and thus the ground-truth labels of the original and augmented graph/nodes can be regarded identical in contrastive learning. However, we observe that this assumption does not always hold. For instance, the deletion of a supernode within a social network can exert a substantial influence on the partitioning of communities for other nodes. Similarly, any perturbation to nodes or edges in a molecular graph will change the labels of the graph. Therefore, we believe that augmenting the graph, accompanied by an adaptation of the labels used for the contrastive loss, will facilitate the encoder to learn a better representation. Based on this idea, we propose ID-MixGCL, which allows the simultaneous interpolation of input nodes and corresponding identity labels to obtain soft-confidence samples, with a controllable degree of change, leading to the capture of fine-grained representations from self-supervised training on unlabeled graphs. Experimental results demonstrate that ID-MixGCL improves performance on graph classification and node classification tasks, as demonstrated by significant improvements on the Cora, IMDB-B, IMDB-M, and PROTEINS datasets compared to state-of-the-art techniques, by 3-29parcent absolute points. The dataset and source code can be found at: https://github.com/zhanggehang/ID-MIXGCL.

Given a textual passage and an answer, humans are able to ask questions with various expressions, but this ability is still challenging for most question generation (QG) systems.Existing solutions mainly focus on the internal knowledge within the given passage or the semantic word space for diverse content planning.These methods, however, have not considered the potential of external knowledge for expression diversity.To bridge this gap, we propose RAST, a framework for Retrieval-Augmented Style Transfer, where the objective is to utilize the style of diverse templates for question generation.For training RAST, we develop a novel Reinforcement Learning (RL) based approach that maximizes a weighted combination of diversity reward and consistency reward.Here, the consistency reward is computed by a Question-Answering (QA) model, whereas the diversity reward measures how much the final output mimics the retrieved template.Experimental results show that our method outperforms previous diversity-driven baselines on diversity while being comparable in terms of consistency scores.Our code is available at https://github.com/gouqi666/RAST.

Recent research has demonstrated that Large Language Models (LLMs) can enhance their capabilities by utilizing external tools.However, three pivotal questions remain unanswered: (1) How effective are current LLMs in utilizing tools?(2) How can we enhance LLMs' ability to utilize tools?(3) What obstacles need to be overcome to leverage tools?To address these questions, we introduce API-Bank, a groundbreaking benchmark, specifically designed for tool-augmented LLMs.For the first question, we develop a runnable evaluation system consisting of 73 API tools.We annotate 314 tool-use dialogues with 753 API calls to assess the existing LLMs' capabilities in planning, retrieving, and calling APIs.For the second question, we construct a comprehensive training set containing 1,888 tool-use dialogues from 2,138 APIs spanning 1,000 distinct domains.Using this dataset, we train Lynx, a tool-augmented LLM initialized from Alpaca.Experimental results demonstrate that GPT-3.5 exhibits improved tool utilization compared to GPT-3, while GPT-4 excels in planning.However, there is still significant potential for further improvement.Moreover, Lynx surpasses Alpaca's tool utilization performance by more than 26 pts and approaches the effectiveness of GPT-3.5.Through error analysis, we highlight the key challenges for future research in this field to answer the third question 1 .

The goal of document-grounded dialogue (DocGD) is to generate a response by anchoring the evidence in a supporting document in accordance with the dialogue context.This entails four causally interconnected variables.While task-specific pre-training has significantly enhanced performances on numerous downstream tasks, existing DocGD methods still rely on general pre-trained language models without a specifically tailored pre-training approach that explicitly captures the causal relationships.To address this, we present the first causallycomplete dataset construction strategy for developing million-scale DocGD pre-training corpora.Additionally, we propose a causallyperturbed pre-training strategy to better capture causality by introducing perturbations on the variables and optimizing the overall causal effect.Experiments conducted on three benchmark datasets demonstrate that our causal pretraining yields substantial and consistent improvements in fully-supervised, low-resource, few-shot, and zero-shot settings 1 .

Cross-lingual Named Entity Recognition (NER) aims to address the challenge of data scarcity in low-resource languages by leveraging knowledge from high-resource languages. Most current work relies on general multilingual language models to represent text, and then uses classic combined tagging (e.g., B-ORG) to annotate entities; However, this approach neglects the lack of cross-lingual alignment of entity representations in language models, and also ignores the fact that entity spans and types have varying levels of labeling difficulty in terms of transferability. To address these challenges, we propose a novel framework, referred to as DLBri, which addresses the issues of representation and labeling simultaneously. Specifically, the proposed framework utilizes progressive contrastive learning with source-to-target oriented sentence pairs to pre-finetune the language model, resulting in improved cross-lingual entity-aware representations. Additionally, a decomposition-then-combination procedure is proposed, which separately transfers entity span and type, and then combines their information, to reduce the difficulty of cross-lingual entity labeling. Extensive experiments on 13 diverse language pairs confirm the effectiveness of DLBri.

In industry, web platforms such as Alibaba and Amazon often provide diverse services for users. Unsurprisingly, some developed services are data-rich, while some newly started services are data-scarce accompanied by severe data sparsity and cold-start problems. To alleviate the above problems and incubate new services easily, cross-domain recommendation (CDR) has attracted much attention from industrial and academic researchers. Generally, CDR aims to transfer rich user-item interaction information from related source domains (e.g., developed services) to boost recommendation quality of target domains (e.g., newly started services). For different scenarios, previous CDR methods can be roughly divided into two branches: (1) Data sparsity CDR fulfills user preference aided by other domain data to make intra-domain recommendations for users with few interactions, (2) Cold-start CDR projects user preference from other domain to make inter-domain recommendations for users with none interactions. In the past years, many outstanding CDR methods are emerged, however, to the best of our knowledge, none of them attempts to solve the two branches simultaneously. In this paper, we provide a unified framework, namely UniCDR, which can universally model different CDR scenarios by transferring the domain-shared information. Extensive experiments under the above 2 branches on 4 CDR scenarios and 6 public and large-scale industrial datasets demonstrate the effectiveness and universal ability of our UniCDR.

Extracting structured information from all manner of webpages is an important problem with the potential to automate many real-world applications. Recent work has shown the effectiveness of leveraging DOM trees and pre-trained language models to describe and encode webpages. However, they typically optimize the model to learn the semantic co-occurrence of elements and labels in the same webpage, thus their effectiveness depends on sufficient labeled data, which is labor-intensive. In this paper, we further observe structural co-occurrences in different webpages of the same website: the same position in the DOM tree usually plays the same semantic role, and the DOM nodes in this position also share similar surface forms. Motivated by this, we propose a novel method, Structor, to effectively incorporate the structural co-occurrences over DOM tree and surface form into pre-trained language models. Such structural co-occurrences help the model learn the task better under low-resource settings, and we study two challenging experimental scenarios: website-level low-resource setting and webpage-level low-resource setting, to evaluate our approach. Extensive experiments on the public SWDE dataset show that Structor significantly outperforms the state-of-the-art models in both settings, and even achieves three times the performance of the strong baseline model in the case of extreme lack of training data.

Yanzeng Li, Jiangxia Cao, Xin Cong, Zhenyu Zhang, Bowen Yu, Hongsong Zhu, Tingwen Liu. Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2022.

Multi-intent detection and slot filling joint model attracts more and more attention since it can handle multi-intent utterances, which is closer to complex real-world scenarios.Most existing joint models rely entirely on the training procedure to obtain the implicit correlation between intents and slots.However, they ignore the fact that leveraging the rich global knowledge in the corpus can determine the intuitive and explicit correlation between intents and slots.In this paper, we aim to make full use of the statistical co-occurrence frequency between intents and slots as prior knowledge to enhance joint multiple intent detection and slot filling.To be specific, an intent-slot cooccurrence graph is constructed based on the entire training corpus to globally discover correlation between intents and slots.Based on the global intent-slot co-occurrence, we propose a novel graph neural network to model the interaction between the two subtasks.Experimental results on two public multi-intent datasets demonstrate that our approach outperforms the state-of-the-art models.

Document-level Event Extraction (DEE) is particularly tricky due to the two challenges it poses: scattering-arguments and multi-events. The first challenge means that arguments of one event record could reside in different sentences in the document, while the second one reflects that one document may simultaneously contain multiple such event records. Motivated by humans’ reading cognitive to extract information of interests, in this paper, we propose a method called HRE (Human Reading inspired Extractor for Document Events), where DEE is decomposed into these two iterative stages, rough reading and elaborate reading. Specifically, the first stage browses the document to detect the occurrence of events, and the second stage serves to extract specific event arguments. For each concrete event role, elaborate reading hierarchically works from sentences to characters to locate arguments across sentences, thus the scattering-arguments problem is tackled. Meanwhile, rough reading is explored in a multi-round manner to discover undetected events, thus the multi-events problem is handled. Experiment results show the superiority of HRE over prior competitive methods.

Open Information Extraction (OpenIE) facilitates domain-independent discovery of relational facts from large corpora. The technique well suits many open-world natural language understanding scenarios, such as automatic knowledge base construction, open-domain question answering, and explicit reasoning. Thanks to the rapid development in deep learning technologies, numerous neural OpenIE architectures have been proposed and achieve considerable performance improvement. In this survey, we provide an extensive overview of the state-of-the-art neural OpenIE models, their key design decisions, strengths and weakness. Then, we discuss limitations of current solutions and the open issues in OpenIE problem itself. Finally we list recent trends that could help expand its scope and applicability, setting up promising directions for future research in OpenIE. To our best knowledge, this paper is the first review on neural OpenIE.

Building document-grounded dialogue systems have received growing interest as documents convey a wealth of human knowledge and commonly exist in enterprises. Wherein, how to comprehend and retrieve information from documents is a challenging research problem. Previous work ignores the visual property of documents and treats them as plain text, resulting in incomplete modality. In this paper, we propose a Layout-aware document-level Information Extraction dataset, LIE, to facilitate the study of extracting both structural and semantic knowledge from visually rich documents (VRDs), so as to generate accurate responses in dialogue systems. LIE contains 62k annotations of three extraction tasks from 4,061 pages in product and official documents, becoming the largest VRD-based information extraction dataset to the best of our knowledge. We also develop benchmark methods that extend the token-based language model to consider layout features like humans. Empirical results show that layout is critical for VRD-based extraction, and system demonstration also verifies that the extracted knowledge can help locate the answers that users care about.

Cross-domain Named Entity Recognition (NER) aims to transfer knowledge from the source domain to the target, alleviating expensive labeling costs in the target domain. Most prior studies acquire domain-invariant features under the end-to-end sequence-labeling framework where each token is assigned a compositional label (e.g., B-LOC). However, the complexity of cross-domain transfer may be increased over this complicated labeling scheme, which leads to sub-optimal results, especially when there are significantly distinct entity categories across domains. In this paper, we aim to explore the task decomposition in cross-domain NER. Concretely, we suggest a modular learning approach in which two sub-tasks (entity span detection and type classification) are learned by separate functional modules to perform respective cross-domain transfer with corresponding strategies. Compared with the compositional labeling scheme, the label spaces are smaller and closer across domains especially in entity span detection, leading to easier transfer in each sub-task. And then we combine two sub-tasks to achieve the final result with modular interaction mechanism, and deploy the adversarial regularization for generalized and robust learning in low-resource target domains. Extensive experiments over 10 diverse domain pairs demonstrate that the proposed method is superior to state-of-the-art cross-domain NER methods in an end-to-end fashion (about average 6.4% absolute F1 score increase). Further analyses show the effectiveness of modular task decomposition and its great potential in cross-domain NER.

In few-shot relational triple extraction (FS-RTE), one seeks to extract relational triples from plain texts by utilizing only few annotated samples. Recent work first extracts all entities and then classifies their relations. Such an entity-then-relation paradigm ignores the entity discrepancy between relations. To address it, we propose a novel task decomposition strategy, Relation-then-Entity, for FS-RTE. It first detects relations occurred in a sentence and then extracts the corresponding head/tail entities of the detected relations. To instantiate this strategy, we further propose a model, RelATE, which builds a dual-level attention to aggregate relation-relevant information to detect the relation occurrence and utilizes the annotated samples of the detected relations to extract the corresponding head/tail entities. Experimental results show that our model outperforms previous work by an absolute gain (18.98%, 28.85% in F1 in two few-shot settings).

Yucheng Wang, Bowen Yu, Hongsong Zhu, Tingwen Liu, Nan Yu, Limin Sun. 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.

Open Information Extraction (OpenIE) aims to discover textual facts from a given sentence.In essence, the facts contained in plain text are unordered.However, the popular Ope-nIE systems usually output facts sequentially in the way of predicting the next fact conditioned on the previous decoded ones, which enforce an unnecessary order on the facts and involve the error accumulation between autoregressive steps.To break this bottleneck, we propose MacroIE, a novel non-autoregressive framework for OpenIE.MacroIE firstly constructs a fact graph based on the table filling scheme, in which each node denotes a fact element, and an edge links two nodes that belong to the same fact.Then OpenIE can be reformulated as a non-parametric process of finding maximal cliques from the graph.It directly outputs the final set of facts in one go, thus getting rid of the burden of predicting fact order, as well as the error propagation between facts.Experiments conducted on two benchmark datasets show that our proposed model significantly outperforms current state-of-theart methods, beats the previous systems by as much as 5.7 absolute gain in F1 score.

Distantly supervised named entity recognition (DS-NER) efficiently reduces labor costs but meanwhile intrinsically suffers from the label noise due to the strong assumption of distant supervision.Typically, the wrongly labeled instances comprise numbers of incomplete and inaccurate annotation noise, while most prior denoising works are only concerned with one kind of noise and fail to fully explore useful information in the whole training set.To address this issue, we propose a robust learning paradigm named Self-Collaborative Denoising Learning (SCDL), which jointly trains two teacher-student networks in a mutuallybeneficial manner to iteratively perform noisy label refinery.Each network is designed to exploit reliable labels via self denoising, and two networks communicate with each other to explore unreliable annotations by collaborative denoising.Extensive experimental results on five real-world datasets demonstrate that SCDL is superior to state-of-the-art DS-NER denoising methods 1 .

Open Information Extraction (OIE), the task aimed at discovering all textual facts organized in the form of (subject, predicate, object) found within a sentence, has gained much attention recently. However, in some knowledge-driven applications such as question answering, we often have a target entity and hope to obtain its structured factual knowledge for better understanding, instead of extracting all possible facts aimlessly from the corpus. In this paper, we define a new task, namely Semi-Open Information Extraction (SOIE), to address this need. The goal of SOIE is to discover domain-independent facts towards a particular entity from general and diverse web text. To facilitate research on this new task, we propose a large-scale human-annotated benchmark called SOIED, consisting of 61,984 facts for 8,013 subject entities annotated on 24,000 Chinese sentences collected from the web search engine.

Extracting relations from plain text is an important task with wide application. Most existing methods formulate it as a supervised problem and utilize one-hot hard labels as the sole target in training, neglecting the rich semantic information among relations. In this paper, we aim to explore the supervision with soft labels in relation extraction, which makes it possible to integrate prior knowledge. Specifically, a bipartite graph is first devised to discover type constraints between entities and relations based on the entire corpus. Then, we combine such type constraints with neural networks to achieve a knowledgeable model. Furthermore, this model is regarded as teacher to generate well-informed soft labels and guide the optimization of a student network via knowledge distillation. Besides, a multi-aspect attention mechanism is introduced to help student mine latent information from text. In this way, the enhanced student inherits the dark knowledge (e.g., type constraints and relevance among relations) from teacher, and directly serves the testing scenarios without any extra constraints. We conduct extensive experiments on the TACRED and SemEval datasets, the experimental results justify the effectiveness of our approach.

Most Chinese pre-trained models take character as the basic unit and learn representation according to character's external contexts, ignoring the semantics expressed in the word, which is the smallest meaningful utterance in Chinese.Hence, we propose a novel wordaligned attention to exploit explicit word information, which is complementary to various character-based Chinese pre-trained language models.Specifically, we devise a pooling mechanism to align the character-level attention to the word level and propose to alleviate the potential issue of segmentation error propagation by multi-source information fusion.As a result, word and character information are explicitly integrated at the fine-tuning procedure.Experimental results on five Chinese NLP benchmark tasks demonstrate that our method achieves significant improvements against BERT, ERNIE and BERT-wwm.

Extracting entities and relations from unstructured text has attracted increasing attention in recent years but remains challenging, due to the intrinsic difficulty in identifying overlapping relations with shared entities.Prior works show that joint learning can result in a noticeable performance gain.However, they usually involve sequential interrelated steps and suffer from the problem of exposure bias.At training time, they predict with the ground truth conditions while at inference it has to make extraction from scratch.This discrepancy leads to error accumulation.To mitigate the issue, we propose in this paper a one-stage joint extraction model, namely, TPLinker, which is capable of discovering overlapping relations sharing one or both entities while immune from the exposure bias.TPLinker formulates joint extraction as a token pair linking problem and introduces a novel handshaking tagging scheme that aligns the boundary tokens of entity pairs under each relation type.Experiment results show that TPLinker performs significantly better on overlapping and multiple relation extraction, and achieves state-of-the-art performance on two public datasets 1 .

Incorporating lexicons into character-level Chinese NER by lattices is proven effective to exploit rich word boundary information.Previous work has extended RNNs to consume lattice inputs and achieved great success.However, due to the DAG structure and the inherently unidirectional sequential nature, this method precludes batched computation and sufficient semantic interaction.In this paper, we propose PLTE, an extension of transformer encoder that is tailored for Chinese NER, which models all the characters and matched lexical words in parallel with batch processing.PLTE augments self-attention with positional relation representations to incorporate lattice structure.It also introduces a porous mechanism to augment localness modeling and maintain the strength of capturing the rich long-term dependencies.Experimental results show that PLTE performs up to 11.4 times faster than state-of-the-art methods while realizing better performance.We also demonstrate that using BERT representations further substantially boosts the performance and brings out the best in PLTE.

Dependency trees have been shown to be effective in capturing long-range relations between target entities.Nevertheless, how to selectively emphasize target-relevant information and remove irrelevant content from the tree is still an open problem.Existing approaches employing predefined rules to eliminate noise may not always yield optimal results due to the complexity and variability of natural language.In this paper, we present a novel architecture named Dynamically Pruned Graph Convolutional Network (DP-GCN), which learns to prune the dependency tree with rethinking in an end-to-end scheme.In each layer of DP-GCN, we employ a selection module to concentrate on nodes expressing the target relation by a set of binary gates and then augment the pruned tree with a pruned semantic graph to ensure the connectivity.After that, we introduce a rethinking mechanism to guide and refine the pruning operation by feeding back the high-level learned features repeatedly.Extensive experimental results demonstrate that our model achieves impressive performance compared to strong competitors.

Document-level relation extraction (RE)poses new challenges over its sentence-level counterpart since it requires an adequate comprehension of the whole document and the multi-hop reasoning ability across multiple sentences to reach the final result.In this paper, we propose a novel graphbased model with Dual-tier Heterogeneous Graph (DHG) for document-level RE.In particular, DHG is composed of a structure modeling layer followed by a relation reasoning layer.The major advantage is that it is capable of not only capturing both the sequential and structural information of documents but also mixing them together to benefit for multi-hop reasoning and final decisionmaking.Furthermore, we employ Graph Neural Networks (GNNs) based message propagation strategy to accumulate information on DHG.Experimental results demonstrate that the proposed method achieves state-of-the-art performance on two widely used datasets, and further analyses suggest that all the modules in our model are indispensable for document-level RE.

Joint extraction of entities and relations aims to detect entity pairs along with their relations using a single model. Prior work typically solves this task in the extract-then-classify or unified labeling manner. However, these methods either suffer from the redundant entity pairs, or ignore the important inner structure in the process of extracting entities and relations. To address these limitations, in this paper, we first decompose the joint extraction task into two interrelated subtasks, namely HE extraction and TER extraction. The former subtask is to distinguish all head-entities that may be involved with target relations, and the latter is to identify corresponding tail-entities and relations for each extracted head-entity. Next, these two subtasks are further deconstructed into several sequence labeling problems based on our proposed span-based tagging scheme, which are conveniently solved by a hierarchical boundary tagger and a multi-span decoding algorithm. Owing to the reasonable decomposition strategy, our model can fully capture the semantic interdependency between different steps, as well as reduce noise from irrelevant entity pairs. Experimental results show that our method outperforms previous work by 5.2%, 5.9% and 21.5% (F1 score), achieving a new state-of-the-art on three public datasets.

More recently, Named Entity Recognition has achieved great advances aided by pre-training approaches such as BERT.However, current pre-training techniques focus on building language modeling objectives to learn a general representation, ignoring the named entityrelated knowledge.To this end, we propose a NER-specific pre-training framework to inject coarse-to-fine automatically mined entity knowledge into pre-trained models.Specifically, we first warm-up the model via an entity span identification task by training it with Wikipedia anchors, which can be deemed as general-typed entities.Then we leverage the gazetteer-based distant supervision strategy to train the model extract coarse-grained typed entities.Finally, we devise a self-supervised auxiliary task to mine the fine-grained named entity knowledge via clustering.Empirical studies on three public NER datasets demonstrate that our framework achieves significant improvements against several pre-trained baselines, establishing the new state-of-the-art performance on three benchmarks.Besides, we show that our framework gains promising results without using human-labeled training data, demonstrating its effectiveness in labelfew and low-resource scenarios.1

Even with larger and larger text data available, encoding linguistic knowledge into word embeddings directly from corpora is still difficult. The most intuitive way to incorporate knowledge into word embeddings is to use external resources, such as the largest datasource for describing words - natural language dictionaries. However, previous methods usually neglect the recursive nature of dictionaries and cannot capture the high-order relation. In this paper, we present DRG2vec, a novel and efficient approach for learning word representations, which exploits the inherent recursiveness of dictionaries by modeling the whole dictionary as a homogeneous graph based on the co-occurrence of entry and word in the definition. Moreover, a tailor-made sampling strategy is introduced to generate word sequences from the definition relational graph, and then, the generated sequences are fed to the Skip-gram model with semantic negative sampling for word representation learning. Extensive experiments on sixteen benchmark datasets show that leveraging the recursive dictionary graph indeed achieves better performance than other state-of-the-art methods, especially exhibits a weighted average improvement of 8.6% in the word similarity task.

Relation extraction studies the issue of predicting semantic relations between pairs of entities in sentences. Attention mechanisms are often used in this task to alleviate the inner-sentence noise by performing soft selections of words independently. Based on the observation that information pertinent to relations is usually contained within segments (continuous words in a sentence), it is possible to make use of this phenomenon for better extraction. In this paper, we aim to incorporate such segment information into neural relation extractor. Our approach views the attention mechanism as linear-chain conditional random fields over a set of latent variables whose edges encode the desired structure, and regards attention weight as the marginal distribution of each word being selected as a part of the relational expression. Experimental results show that our method can attend to continuous relational expressions without explicit annotations, and achieve the state-of-the-art performance on the large-scale TACRED dataset.