Samuel Cahyawijaya, Holy Lovenia, Joel Ruben Antony Moniz, Tack Hwa Wong, Mohammad Rifqi Farhansyah, Thant Thiri Maung, Frederikus Hudi, David Anugraha, Muhammad Ravi Shulthan Habibi, Muhammad Reza Qorib, Amit Agarwal, Joseph Marvin Imperial, Hitesh Laxmichand Patel, Vicky Feliren, Bahrul Ilmi Nasution, Manuel Antonio Rufino, Genta Indra Winata, Rian Adam Rajagede, Carlos Rafael Catalan, Mohamed Fazli Mohamed Imam, Priyaranjan Pattnayak, Salsabila Zahirah Pranida, Kevin Pratama, Yeshil Bangera, Adisai Na-Thalang, Patricia Nicole Monderin, Yueqi Song, Christian Simon, Lynnette Hui Xian Ng, Richardy Lobo Sapan, Taki Hasan Rafi, Bin Wang, Supryadi, Kanyakorn Veerakanjana, Piyalitt Ittichaiwong, Matthew Theodore Roque, Karissa Vincentio, Takdanai Kreangphet, Phakphum Artkaew, Kadek Hendrawan Palgunadi, Yanzhi Yu, Rochana Prih Hastuti, William Nixon, Mithil Bangera, Adrian Xuan Wei Lim, Aye Hninn Khine, Hanif Muhammad Zhafran, Teddy Ferdinan, Audra Aurora Izzani, Ayushman Singh, Evan Evan, Jauza Akbar Krito, Michael Anugraha, Fenal Ashokbhai Ilasariya, Haochen Li, John Amadeo Daniswara, Filbert Aurelian Tjiaranata, Eryawan Presma Yulianrifat, Can Udomcharoenchaikit, Fadil Risdian Ansori, Mahardika Krisna Ihsani, Giang Nguyen, Anab Maulana Barik, Dan John Velasco, Rifo Ahmad Genadi, Saptarshi Saha, Chengwei Wei, Isaiah Edri W. Flores, Kenneth Chen Ko Han, Anjela Gail D. Santos, Wan Shen Lim, Kaung Si Phyo, Tim Santos, Meisyarah Dwiastuti, Jiayun Luo, Jan Christian Blaise Cruz, Ming Shan Hee, Ikhlasul Akmal Hanif, M.Alif Al Hakim, Muhammad Rizky Sya’ban, Kun Kerdthaisong, Lester James Validad Miranda, Fajri Koto, Tirana Noor Fatyanosa, Alham Fikri Aji, Jostin Jerico Rosal, Jun Kevin, Robert Wijaya, Onno P. Kampman, Ruochen Zhang, Börje F. Karlsson, Peerat Limkonchotiwat. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.

Singlish, a Creole language rooted in English, is a key focus in linguistic research within multilingual and multicultural contexts. However, its spoken form remains underexplored, limiting insights into its linguistic structure and applications. To address this gap, we standardize and annotate the largest spoken Singlish corpus, introducing the Multitask National Speech Corpus (MNSC). These datasets support diverse tasks, including Automatic Speech Recognition (ASR), Spoken Question Answering (SQA), Spoken Dialogue Summarization (SDS), and Paralinguistic Question Answering (PQA). We release standardized splits and a human-verified test set to facilitate further research. Additionally, we propose SingAudioLLM, a multi-task multimodal model leveraging multimodal large language models to handle these tasks concurrently. Experiments reveal our models’ adaptability to the Singlish context, achieving state-of-the-art performance and outperforming prior models by 10–30% in comparison with other AudioLLMs and cascaded solutions1

A Large Language Model (LLM) can be extended to understand speech inputs by using a speech encoder to compute embeddings from the speech, which are then used with a text prompt. Diverse information is expressed in speech and a wide variety of tasks can be performed. Different speech encoders may specialise toward different information types and tasks. This complementarity can be leveraged upon by using multiple speech encoders. This paper presents a comprehensive analysis of the diversity and complementarity between open-source speech encoders, when used in a multi-modal LLM framework. Experiments identify the encoders that excel in each type of downstream task, thereby guiding future system design. The diversity between encoders is measured, showing that Whisper tends to behave more differently. Diversity between encoders is compared across tasks, showing that semantic tasks tend to yield more diverse predictions. Early and late fusion show that complementarity can yield improvements.

The rapid advancements in large language models (LLMs) have significantly enhanced natural language processing capabilities, facilitating the development of AudioLLMs that process and understand speech and audio inputs alongside text. Existing AudioLLMs typically combine a pre-trained audio encoder with a pre-trained LLM, which are subsequently finetuned on specific audio tasks. However, the pre-trained audio encoder has constrained capacity to capture features for new tasks and datasets. To address this, we propose to incorporate mixtures of ‘weak’ encoders (MoWE) into the AudioLLM framework. MoWE supplements a base encoder with a pool of relatively lightweight encoders, selectively activated based on the audio input to enhance feature extraction without significantly increasing model size. Our empirical results demonstrate that MoWE effectively improves multi-task performance, broadening the applicability of AudioLLMs to more diverse audio tasks.

Bin Wang, Xunlong Zou, Geyu Lin, Shuo Sun, Zhuohan Liu, Wenyu Zhang, Zhengyuan Liu, AiTi Aw, Nancy F. Chen. 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.

The goal of relationship classification (RC) is to predict the semantic relationship between two entities in a given sentence. With the advent of deep learning and pretrained language models, RC research has progressed by leaps and bounds. However, the current studies are focused mainly on predicting semantic relationships from a predefined set. How to recognize unseen relationships remains a challenge, which is also known as the zero-shot RC (ZSRC) task. Some ZSRC-related methods directly map relationship categories to numerical indices, constraining the model's ability to autonomously infer and understand these relationships, while others rely heavily on manual definitions. To address these issues and inspired by the way of reasoning in which humans perform RC tasks, we propose a new framework to handle the ZSRC task through inference on category attributes (ICAs). The main idea of ICA is to detect the semantic relationship between promises, which are RC sentences, and hypotheses, which are relational sentences of entities created by templates. Specifically, instead of manual design, we introduce two hypothesis templates derived from the label words (LWs) and descriptions (LDs) associated with each relationship. These templates are used to automatically convert the RC data into the textual entailment (TE) format. Furthermore, they are fine-tuned with a pretrained TE model, facilitating the acquisition of relational knowledge and enabling the generalization of semantic reasoning rules learned from seen classes to unseen classes. Moreover, to implement multirelationship semantic inference for all unseen classes, we propose an entailment difference mechanism to enhance the reasoning capability of the model. Besides the current ZSRC test setting, we also examine our method in an even more challenging setting to deal with data scarcity in real-world applications. The outstanding performance of ICA on the FewRel and Wiki-ZSL datasets demonstrates its effectiveness in the ZSRC task.

Holy Lovenia, Rahmad Mahendra, Salsabil Maulana Akbar, Lester James V. Miranda, Jennifer Santoso, Elyanah Aco, Akhdan Fadhilah, Jonibek Mansurov, Joseph Marvin Imperial, Onno P. Kampman, Joel Ruben Antony Moniz, Muhammad Ravi Shulthan Habibi, Frederikus Hudi, Railey Montalan, Ryan Ignatius, Joanito Agili Lopo, William Nixon, Börje F. Karlsson, James Jaya, Ryandito Diandaru, Yuze Gao, Patrick Amadeus, Bin Wang, Jan Christian Blaise Cruz, Chenxi Whitehouse, Ivan Halim Parmonangan, Maria Khelli, Wenyu Zhang, Lucky Susanto, Reynard Adha Ryanda, Sonny Lazuardi Hermawan, Dan John Velasco, Muhammad Dehan Al Kautsar, Willy Fitra Hendria, Yasmin Moslem, Noah Flynn, Muhammad Farid Adilazuarda, Haochen Li, Johanes Lee, R. Damanhuri, Shuo Sun, Muhammad Reza Qorib, Amirbek Djanibekov, Wei Qi Leong, Quyet V. Do, Niklas Muennighoff, Tanrada Pansuwan, Ilham Firdausi Putra, Yan Xu, Tai Ngee Chia, Ayu Purwarianti, Sebastian Ruder, William Tjhi, Peerat Limkonchotiwat, Alham Fikri Aji, Sedrick Keh, Genta Indra Winata, Ruochen Zhang, Fajri Koto, Zheng-Xin Yong, Samuel Cahyawijaya. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

Knowledge graph entity typing (KGET) aims to predict the missing entity types in knowledge graphs (KG). The relationship between entities and their corresponding types is often expressed using a single relation, hasType. However, hasType has a limited capability for modeling diverse entity-type relationships in the embedding space. In this paper, we first introduce multiple auxiliary relations to model the complex entity-type relationship. We propose an efficient and robust algorithm to group similar entity types together and assign a unique auxiliary relation to each group. Then, with the auxiliary relations, we propose an Asynchronous representation learning framework for KGET, named AsyncET, where entity and type embeddings are updated alternatively. Consequently, the quality of entity embeddings is gradually improved during training by infusing type information. In addition, entity types with different granularities and semantics can be properly modeled in the embedding space. Experimental results show that AsyncET can substantially improve the performance of embedding-based methods on the KGET task and has a significant advantage over state-of-the-art neural network-based methods in terms of model sizes and inference time.

Bin Wang, Zhengyuan Liu, Xin Huang, Fangkai Jiao, Yang Ding, AiTi Aw, Nancy Chen. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

Geometric transformations including translation, rotation, and scaling are commonly used operations in image processing.Besides, some of them are successfully used in developing effective knowledge graph embedding (KGE).Inspired by the synergy, we propose a new KGE model by leveraging all three operations in this work.Since translation, rotation, and scaling operations are cascaded to form a composite one, the new model is named Com-poundE.By casting CompoundE in the framework of group theory, we show that quite a few distanced-based KGE models are special cases of CompoundE.CompoundE extends the simple distance-based scoring functions to relation-dependent compound operations on head and/or tail entities.To demonstrate the effectiveness of CompoundE, we perform three prevalent KG prediction tasks including link prediction, path query answering, and entity typing, on a range of datasets.CompoundE outperforms extant models consistently, demonstrating its effectiveness and flexibility.

Knowledge graph completion (KGC) aims to discover missing relationships between entities in knowledge graphs (KGs).Most prior KGC work focuses on learning embeddings for entities and relations through a simple scoring function.Yet, a higher-dimensional embedding space is usually required for a better reasoning capability, which leads to a larger model size and hinders applicability to real-world problems (e.g., large-scale KGs or mobile/edge computing).A lightweight modularized KGC solution, called GreenKGC, is proposed in this work to address this issue.GreenKGC consists of three modules: representation learning, feature pruning, and decision learning, to extract discriminant KG features and make accurate predictions on missing relationships using classifiers and negative sampling.Experimental results demonstrate that, in low dimensions, GreenKGC can outperform SOTA methods in most datasets.In addition, low-dimensional GreenKGC can achieve competitive or even better performance against high-dimensional models with a much smaller model size.We make our code publicly available.1

Conventional dialogue summarization methods directly generate summaries and do not consider user's specific interests.This poses challenges in cases where the users are more focused on particular topics or aspects.With the advancement of instruction-finetuned language models, we introduce instruction-tuning to dialogues to expand the capability set of dialogue summarization models.To overcome the scarcity of instructive dialogue summarization data, we propose a three-step approach to synthesize high-quality query-based summarization triples.This process involves summaryanchored query generation, query filtering and query-based summary generation.By training a unified model called InstructDS (Instructive Dialogue Summarization) on three summarization datasets with multi-purpose instructive triples, we expand the capability of dialogue summarization models.We evaluate our method on four datasets, including dialogue summarization and dialogue reading comprehension.Experimental results show that our approach outperforms the state-of-the-art models and even models with larger sizes.Additionally, our model exhibits higher generalizability and faithfulness, as confirmed by human subjective evaluations.Benjamin: Hey guys, what are we doing with the keys today?Hilary: I've got them.Whoever wants

A scalable semisupervised node classification method on graph-structured data, called GraphHop, is proposed in this work. The graph contains all nodes' attributes and link connections but labels of only a subset of nodes. Graph convolutional networks (GCNs) have provided superior performance in node label classification over the traditional label propagation (LP) methods for this problem. Nevertheless, current GCN algorithms suffer from a considerable amount of labels for training because of high model complexity or cannot be easily generalized to large-scale graphs due to the expensive cost of loading the entire graph and node embeddings. Besides, nonlinearity makes the optimization process a mystery. To this end, an enhanced LP method, called GraphHop, is proposed to tackle these problems. GraphHop can be viewed as a smoothening LP algorithm, in which each propagation alternates between two steps: label aggregation and label update. In the label aggregation step, multihop neighbor embeddings are aggregated to the center node. In the label update step, new embeddings are learned and predicted for each node based on aggregated results from the previous step. The two-step iteration improves the graph signal smoothening capacity. Furthermore, to encode attributes, links, and labels on graphs effectively under one framework, we adopt a two-stage training process, i.e., the initialization stage and the iteration stage. Thus, the smooth attribute information extracted from the initialization stage is consistently imposed in the propagation process in the iteration stage. Experimental results show that GraphHop outperforms state-of-the-art graph learning methods on a wide range of tasks in graphs of various sizes (e.g., multilabel and multiclass classification on citation networks, social graphs, and commodity consumption graphs).

Word and sentence embeddings are useful feature representations in natural language processing.However, intrinsic evaluation for embeddings lags far behind, and there has been no significant update since the past decade.Word and sentence similarity tasks have become the de facto evaluation method.It leads models to overfit to such evaluations, negatively impacting embedding models' development.This paper first points out the problems using semantic similarity as the gold standard for word and sentence embedding evaluations.Further, we propose a new intrinsic evaluation method called EvalRank, which shows a much stronger correlation with downstream tasks.Extensive experiments are conducted based on 60+ models and popular datasets to certify our judgments.Finally, the practical evaluation toolkit is released for future benchmarking purposes.1

Most sentence embedding techniques heavily rely on expensive human-annotated sentence pairs as the supervised signals.Despite the use of large-scale unlabeled data, the performance of unsupervised methods typically lags far behind that of the supervised counterparts in most downstream tasks.In this work, we propose a semi-supervised sentence embedding framework, GenSE, that effectively leverages large-scale unlabeled data.Our method include three parts: 1) Generate: A generator/discriminator model is jointly trained to synthesize sentence pairs from open-domain unlabeled corpus; 2) Discriminate: Noisy sentence pairs are filtered out by the discriminator to acquire high-quality positive and negative sentence pairs; 3) Contrast: A prompt-based contrastive approach is presented for sentence representation learning with both annotated and synthesized data.Comprehensive experiments show that GenSE achieves an average correlation score of 85.19 on the STS datasets and consistent performance improvement on four domain adaptation tasks, significantly surpassing the state-of-the-art methods and convincingly corroborating its effectiveness and generalization ability. 1

Dialogue summarization is abstractive in nature, making it suffer from factual errors.The factual correctness of summaries has the highest priority before practical applications.Many efforts have been made to improve faithfulness in text summarization.However, there is a lack of systematic study on dialogue summarization systems.In this work, we first perform the fine-grained human analysis on the faithfulness of dialogue summaries and observe that over 35% of generated summaries are faithfully inconsistent respective the source dialogues.Furthermore, we present a new model-level faithfulness evaluation method.It examines generation models with multi-choice questions created by rule-based transformations.Experimental results show that our evaluation schema is a strong proxy for the factual correctness of summarization models.The humanannotated faithfulness samples and the evaluation toolkit are released to facilitate future research toward faithful dialogue summarization.Code

Targeting both computing and coding efficiencies, we propose in this work a novel hierarchical bit-wise differential coding scheme to compress point cloud attributes. The encoder firstly quantizes and organizes the points into an octree structure and, for each internal node, picks its attribute(s) from a child named source child. Next, the encoder conducts a top-down scanning of the hierarchy. For each node with more than one child, it computes the bit-wise attribute difference between the current node and each non-source child by exclusive-OR and encodes the difference with an arithmetic coder. Further, a table look-up approach is proposed to accelerate the online source child identification. The proposed scheme produces superior computing and coding efficiencies for lossless point cloud attribute compression, outperforming the MPEG benchmark coders by large margins in our experiments.

Commonsense knowledge graph (CKG) is a special type of knowledge graph (KG), where entities are composed of free-form text. Existing CKG completion methods focus on transductive learning setting, where all the entities are present during training. Here, we propose the first inductive learning setting for CKG completion, where unseen entities may appear at test time. We emphasize that the inductive learning setting is crucial for CKGs, because unseen entities are frequently introduced due to the fact that CKGs are dynamic and highly sparse. We propose InductivE as the first framework targeted at the inductive CKG completion task. InductivE first ensures the inductive learning capability by directly computing entity embeddings from raw entity attributes. Second, a graph neural network with novel densification process is proposed to further enhance unseen entity representation with neighboring structural information. Experimental results show that InductivE performs especially well on inductive scenarios where it achieves above 48% improvement over previous methods while also outperforms state-of-the-art baselines in transductive settings.

An image anomaly localization method based on the successive subspace learning (SSL) framework, called Anomaly-Hop, is proposed in this work. AnomalyHop consists of three modules: 1) feature extraction via successive subspace learning (SSL), 2) normality feature distributions modeling via Gaussian models, and 3) anomaly map generation and fusion. Comparing with state-of-the-art image anomaly localization methods based on deep neural networks (DNNs), AnomalyHop is mathematically transparent, easy to train, and fast in its inference speed. Besides, its area under the ROC curve (ROC-AUC) performance on the MVTec AD dataset is 95.9%, which is among the best of several benchmarking methods.

A novel sentence embedding method built upon semantic subspace analysis, called semantic subspace sentence embedding (S3E), is proposed in this work. Given the fact that word embeddings can capture semantic relationship while semantically similar words tend to form semantic groups in a high-dimensional embedding space, we develop a sentence representation scheme by analyzing semantic subspaces of its constituent words. Specifically, we construct a sentence model from two aspects. First, we represent words that lie in the same semantic group using the intra-group descriptor. Second, we characterize the interaction between multiple semantic groups with the inter-group descriptor. The proposed S3E method is evaluated on both textual similarity tasks and supervised tasks. Experimental results show that it offers comparable or better performance than the state-of-the-art. The complexity of our S3E method is also much lower than other parameterized models.

Sentence embedding is an important research topic in natural language processing (NLP) since it can transfer knowledge to downstream tasks. Meanwhile, a contextualized word representation, called BERT, achieves the state-of-the-art performance in quite a few NLP tasks. Yet, it is an open problem to generate a high quality sentence representation from BERT-based word models. It was shown in previous study that different layers of BERT capture different linguistic properties. This allows us to fuse information across layers to find better sentence representations. In this work, we study the layer-wise pattern of the word representation of deep contextualized models. Then, we propose a new sentence embedding method by dissecting BERT-based word models through geometric analysis of the space spanned by the word representation. It is called the SBERT-WK method. No further training is required in SBERT-WK. We evaluate SBERT-WK on semantic textual similarity and downstream supervised tasks. Furthermore, ten sentence-level probing tasks are presented for detailed linguistic analysis. Experiments show that SBERT-WK achieves the state-of-the-art performance. Our codes are publicly available.

Language graph learning is an important task with many applications such as text classification, link prediction and community detection. One of the challenges in this domain is finding an efficient way to learn and encode graph into a low dimensional embedding. In this paper, a novel DeepWalk-assisted Graph PCA (DGPCA) method is proposed for processing language network data represented by graphs. This method can generate a precise text representation for nodes (or vertices) in language networks. Unlike other existing work, our learned low dimensional vector representations add flexibility in exploring vertices' neighborhood information, while reducing noise contained in the original data. To demonstrate the effectiveness, we use DGPCA to classify vertices that contain text information in three language networks. Experimentally, DGPCA is shown to perform well on the language datasets in comparison to several state-of-the-art benchmarking methods.

A novel graph-to-tree conversion mechanism called the deep-tree generation (DTG) algorithm is first proposed to predict text data represented by graphs. The DTG method can generate a richer and more accurate representation for nodes (or vertices) in graphs. It adds flexibility in exploring the vertex neighborhood information to better reflect the second order proximity and homophily equivalence in a graph. Then, a Deep-Tree Recursive Neural Network (DTRNN) method is presented and used to classify vertices that contains text data in graphs. To demonstrate the effectiveness of the DTRNN method, we apply it to three real-world graph datasets and show that the DTRNN method outperforms several state-of-the-art benchmarking methods.