Papers by Wangchunshu Zhou
Commonsense Knowledge Transfer for Pre-trained Language Models (2023.findings-acl)
Copied to clipboard
| Challenge: | Recent advances in pre-trained language models have transformed the landscape of natural language processing. |
| Approach: | They propose a framework to transfer commonsense knowledge stored in a neural commonsensing model to a general-purpose pre-trained language model. |
| Outcome: | Empirical results show that the proposed framework improves the model’s performance on downstream tasks that require commonsense reasoning. |
Cross-View Language Modeling: Towards Unified Cross-Lingual Cross-Modal Pre-training (2023.acl-long)
Copied to clipboard
| Challenge: | Empirical results show that CCLM significantly outperforms the prior state-of-the-art with an average absolute improvement of over 10%. |
| Approach: | They introduce a pre-training framework that unifies cross-lingual and cross-modal pre-trained models with shared architectures and objectives. |
| Outcome: | The proposed framework outperforms the state-of-the-art in two multi-lingual datasets and two multilingual image-text retrieval datasets. |
Learning to Predict Persona Information for Dialogue Personalization without Explicit Persona Description (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing approaches to personalize dialogue agents rely on explicit persona descriptions during inference, which severely limits their application in real-world scenarios. |
| Approach: | They propose a method that learns to predict persona information based on the dialogue history to personalize dialogue agents without relying on explicit persona descriptions during inference. |
| Outcome: | The proposed method improves the consistency and engagingness of generated responses when conditioning on the predicted profile of the dialogue agent. |
PopAlign: Diversifying Contrasting Patterns for a More Comprehensive Alignment (2025.acl-long)
Copied to clipboard
Zekun Moore Wang, Shenzhi Wang, King Zhu, Jiaheng Liu, Ke Xu, Jie Fu, Wangchunshu Zhou, Wenhao Huang
| Challenge: | Typical approaches to training large language models rely on limited contrasting patterns . contrasting data is limited and models are susceptible to harmful response tendencies . |
| Approach: | They propose a framework that integrates contrasting patterns across the prompt, model, and pipeline levels. |
| Outcome: | The proposed framework outperforms existing methods in the comparison of RQ1 and RQ2 . the proposed framework significantly outperformed existing methods, leading to more comprehensive alignment. |
PositionID: LLMs can Control Lengths, Copy and Paste with Explicit Positional Awareness (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have impressive capabilities across various domains, including role-playing, creative writing, mathematical reasoning, and coding. |
| Approach: | They propose two methods to improve the model’s adherence to length constraints and copy-paste accuracy without compromising response quality. |
| Outcome: | The proposed methods improve the model’s adherence to length constraints and copy-paste accuracy without compromising response quality. |
Beyond Preserved Accuracy: Evaluating Loyalty and Robustness of BERT Compression (2021.emnlp-main)
Copied to clipboard
| Challenge: | Recent studies on compression of pretrained language models usually use preserved accuracy as the metric for evaluation. |
| Approach: | They propose two new metrics that measure how closely a compressed model mimics the original model. |
| Outcome: | The proposed metrics measure how closely a compressed model (i.e., student) mimics the original model (e.g., teacher). |
CommonGen: A Constrained Text Generation Challenge for Generative Commonsense Reasoning (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Recent studies show that pre-trained language models perform well on commonsense-reasoning benchmark datasets, but building machines with commonsence to compose plausible sentences remains challenging. |
| Approach: | They propose a constrained text generation task for generative commonsense reasoning that generates a coherent sentence using common concepts. |
| Outcome: | The proposed task generates a coherent sentence describing an everyday scenario using common concepts over 35k concept-sets. |
Blow the Dog Whistle: A Chinese Dataset for Cant Understanding with Common Sense and World Knowledge (2021.naacl-main)
Copied to clipboard
| Challenge: | Cant is important for understanding advertising, comedies and dogwhistle politics . currently, there are very few resources available for the research of cant . |
| Approach: | They propose a large and diverse dataset for creating and understanding cant from a computational linguistics perspective. |
| Outcome: | The proposed dataset can be used to test word embedding similarity and pretrained language models. |
BERT Learns to Teach: Knowledge Distillation with Meta Learning (2022.acl-long)
Copied to clipboard
| Challenge: | Existing knowledge distillation methods are based on teacher model, but have drawbacks . a teacher model is fixed during training, but meta learning can improve student performance . |
| Approach: | They propose a meta learning framework that allows the teacher network to learn to better transfer knowledge to the student network. |
| Outcome: | Experiments show that MetaDistil can improve on existing methods and is less sensitive to student capacity and hyperparameters. |
MIMIR: A Customizable Agent Tuning Platform for Enhanced Scientific Applications (2024.emnlp-demo)
Copied to clipboard
Xiangru Tang, Chunyuan Deng, Hanminwang Hanminwang, Haoran Wang, Yilun Zhao, Wenqi Shi, Yi Fung, Wangchunshu Zhou, Jiannan Cao, Heng Ji, Arman Cohan, Mark Gerstein
| Challenge: | Large language models (LLMs) have evolved into interactive agents capable of planning, tool use, and task execution across various tasks. |
| Approach: | They propose a platform that leverages large language models to generate agent-tuning data for fine-tuneing smaller, specialized models. |
| Outcome: | MIMIR enables large models to simulate various roles and create interaction data, which can then be used to fine-tune open-source models like LLaMA2. |
AutoAct: Automatic Agent Learning from Scratch for QA via Self-Planning (2024.acl-long)
Copied to clipboard
Shuofei Qiao, Ningyu Zhang, Runnan Fang, Yujie Luo, Wangchunshu Zhou, Yuchen Jiang, Chengfei Lv, Huajun Chen
| Challenge: | Existing language agent systems struggle with costly data reliance and need multiple models for multiple functions. |
| Approach: | They propose an automatic agent learning framework for QA that synthesizes planning trajectories without human intervention. |
| Outcome: | The proposed framework outperforms existing models on question-answering tasks with a division-of-labor strategy. |
LoraRetriever: Input-Aware LoRA Retrieval and Composition for Mixed Tasks in the Wild (2024.findings-acl)
Copied to clipboard
| Challenge: | Low-Rank Adaptation (LoRA) is an effective yet efficient solution for fine-tuning large language models. |
| Approach: | They propose a low-rank Adaptation framework that retrieves and composes multiple LoRAs according to input prompts. |
| Outcome: | Experimental results show that LoraRetriever outperforms baselines in terms of performance and versatility. |
Efficiently Tuned Parameters Are Task Embeddings (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for intermediate-task transfer are computationally infeasible to experiment with all intermediate combinations. |
| Approach: | They propose to use task-specific parameters updated in parameter-efficient tuning methods to predict inter-task transferability. |
| Outcome: | The proposed approach outperforms existing methods while being conceptually simple and computationally efficient. |
Connecting the Dots Between Fact Verification and Fake News Detection (2020.coling-main)
Copied to clipboard
| Challenge: | Existing methods for detecting fake news rely heavily on supervised learning on a large scale dataset with news articles labeled as fake or real by human experts. |
| Approach: | They propose a simple yet effective approach to connect the dots between fact verification and fake news detection by using a text summarization model pre-trained on news corpora to summarize the long news article into a short claim. |
| Outcome: | The proposed approach enables zero-shot fake news detection, alleviating the need for large scale training data to train fake news detector models. |
Evaluating Large Language Models on Controlled Generation Tasks (2023.emnlp-main)
Copied to clipboard
Jiao Sun, Yufei Tian, Wangchunshu Zhou, Nan Xu, Qian Hu, Rahul Gupta, John Wieting, Nanyun Peng, Xuezhe Ma
| Challenge: | Recent studies have looked into the ability of large language models in various benchmark tasks, including question generation, reading comprehension, multilingual and etc. However, few studies investigate the controllability of large languages. |
| Approach: | They propose to compare large language models with state-of-the-start finetuned smaller models to find that large language model controls are comparable to smaller models. |
| Outcome: | The proposed model can meet hard constraints and perform better than state-of-the-art models. |
CIF-Bench: A Chinese Instruction-Following Benchmark for Evaluating the Generalizability of Large Language Models (2024.findings-acl)
Copied to clipboard
Yizhi Li, Ge Zhang, Xingwei Qu, Jiali Li, Zhaoqun Li, Noah Wang, Hao Li, Ruibin Yuan, Yinghao Ma, Kai Zhang, Wangchunshu Zhou, Yiming Liang, Lei Zhang, Lei Ma, Jiajun Zhang, Zuowen Li, Wenhao Huang, Chenghua Lin, Jie Fu
| Challenge: | a recent study shows that large language models have limited generalization in low-resource languages like Chinese. |
| Approach: | They propose to evaluate the zero-shot generalizability of large language models to the Chinese language . they release only half of the dataset publicly, with the remainder kept private . |
| Outcome: | The Chinese Instruction-Following Benchmark evaluates the generalizability of LLMs to the Chinese language. |
Improving Grammatical Error Correction with Machine Translation Pairs (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods to generate error-corrected sentence pairs for improving grammatical error correction are not available. |
| Approach: | They propose a method to generate error-corrected sentence pairs for improving grammatical error correction based on machine translation models of different qualities . |
| Outcome: | The proposed method can generate multiple error-corrected sentence pairs from Chinese to English text. |
MIO: A Foundation Model on Multimodal Tokens (2025.emnlp-main)
Copied to clipboard
Zekun Moore Wang, King Zhu, Chunpu Xu, Wangchunshu Zhou, Jiaheng Liu, Yibo Zhang, Jessie Wang, Ning Shi, Siyu Li, Yizhi Li, Haoran Que, Zhaoxiang Zhang, Yuanxing Zhang, Ge Zhang, Ke Xu, Jie Fu, Wenhao Huang
| Challenge: | Existing models lack multimodal understanding capabilities, resulting in closed-source model that does not support multimodal interleaved sequences. |
| Approach: | They propose a foundation model built on multimodal tokens capable of understanding and generating speech, text, images, and videos in an end-to-end, autoregressive manner. |
| Outcome: | The proposed model is able to understand speech, text, images, and videos in an end-to-end, autoregressive manner. |
BERT-of-Theseus: Compressing BERT by Progressive Module Replacing (2020.emnlp-main)
Copied to clipboard
| Challenge: | a novel approach to compress neural networks by progressive module replacement is proposed . a number of techniques have been proposed to compress pretraining and fine-tuning models . |
| Approach: | They propose a model compression approach that divides BERT into modules and builds their compact substitutes. |
| Outcome: | The proposed approach outperforms existing knowledge distillation approaches on GLUE benchmark . it is based on a model that divides the original BERT into several modules and builds their substitutes . |
SmartTrim: Adaptive Tokens and Attention Pruning for Efficient Vision-Language Models (2024.lrec-main)
Copied to clipboard
Zekun Wang, Jingchang Chen, Wangchunshu Zhou, Haichao Zhu, Jiafeng Liang, Liping Shan, Ming Liu, Dongliang Xu, Qing Yang, Bing Qin
| Challenge: | Experimental results show that SmartTrim accelerates the original model by 2-3 times with minimal performance degradation. |
| Approach: | They propose an adaptive acceleration framework which prunes redundant token representations and attention heads within each layer of the original model. |
| Outcome: | The proposed framework accelerates the original model by 2-3 times with minimal performance degradation across vision-language tasks. |
Contextual Representation Learning beyond Masked Language Modeling (2022.acl-long)
Copied to clipboard
| Challenge: | masked language models adopt sampled embeddings as anchors to estimate and inject contextual semantics to representations. |
| Approach: | They propose a representation learning approach that uses embeddings as anchors to model contextual representations. |
| Outcome: | The proposed model achieves 5x speedup and 1.2 points average improvement over MLM. |
Improving Sequence-to-Sequence Pre-training via Sequence Span Rewriting (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing text infilling objectives for pretrained language models require self-supervision by masking out tokens or spans in text. |
| Approach: | They propose to extend text infilling to a self-supervised sequence-to-sequence (Seq2Sequen) task. |
| Outcome: | The proposed task improves the model's performance on various natural language generation tasks. |
Let’s Synthesize Step by Step: Iterative Dataset Synthesis with Large Language Models by Extrapolating Errors from Small Models (2023.findings-emnlp)
Copied to clipboard
| Challenge: | *Data Synthesis* is a promising way to train a small model with very little labeled data. |
| Approach: | They propose a framework that iteratively extrapolates the errors of a small model trained on a real-world validation dataset using a large language model. |
| Outcome: | The proposed framework reduces the gap between the synthesized dataset and the real data . it improves on multiple NLP tasks and on large models with human-annotated data. |
Struc-Bench: Are Large Language Models Good at Generating Complex Structured Tabular Data? (2024.naacl-short)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have advanced capabilities but produce complex structured data. |
| Approach: | They propose a structure-aware fine-tuning method to bolster LLMs' performance by crafting format-specific instructions from the intended outputs. |
| Outcome: | The proposed method outperforms LLMs on all three formats and spans text tables, HTML, and LaTeX formats. |
COIG-P: A High-Quality and Large-Scale Chinese Preference Dataset for Alignment with Human Values (2026.findings-eacl)
Copied to clipboard
Siwei Wu, JinCheng Ren, Xeron Du, Shuyue Guo, Xingwei Qu, Yiming Liang, Jie Liu, Yunwen Li, Tyler Loakman, Tianyu Zheng, Boyu Feng, Huaqing Yuan, Zili Wang, Jiaheng Liu, Wenhao Huang, Chenglin Cai, Haoran Que, Jian Yang, Yuelin Bai, Zekun Moore Wang, Zhouliang Yu, Qunshu Lin, Ding Pan, Yuchen Eleanor Jiang, Tiannan Wang, Wangchunshu Zhou, Shenzhi Wang, Xingyuan Bu, Minghao Liu, Guoyin Wang, Ge Zhang, Chenghua Lin
| Challenge: | Existing Chinese preference datasets suffer from limited scale, restricted domain coverage, and insufficiently rigorous data validation. |
| Approach: | They propose an LLM-based data annotation pipeline with no human intervention to annotate Chinese preference datasets. |
| Outcome: | The proposed pipeline outperforms existing Chinese preference datasets on AlignBench and Chinese Reward Benchmark. |
Doolittle: Benchmarks and Corpora for Academic Writing Formalization (2023.emnlp-main)
Copied to clipboard
Shizhe Diao, Yongyu Lei, Liangming Pan, Tianqing Fang, Wangchunshu Zhou, Sedrick Keh, Min-Yen Kan, Tong Zhang
| Challenge: | Existing methods of language refinement focus on narrow, specific linguistic features within isolated sentences, such as grammatical errors and improper word use. |
| Approach: | They propose a task to improve the overall quality of academic writing at paragraph level by integrating automatic feedback into the training process. |
| Outcome: | The proposed task improves the overall quality of formal academic writing at the paragraph level. |
RoleLLM: Benchmarking, Eliciting, and Enhancing Role-Playing Abilities of Large Language Models (2024.findings-acl)
Copied to clipboard
Noah Wang, Z.y. Peng, Haoran Que, Jiaheng Liu, Wangchunshu Zhou, Yuhan Wu, Hongcheng Guo, Ruitong Gan, Zehao Ni, Jian Yang, Man Zhang, Zhaoxiang Zhang, Wanli Ouyang, Ke Xu, Wenhao Huang, Jie Fu, Junran Peng
| Challenge: | Large Language Models (LLMs) have paved the way for complex tasks such as role-playing. |
| Approach: | They propose a framework to benchmark, elicit, and enhance role-playing abilities in Large Language Models. |
| Outcome: | The proposed framework improves role-playing abilities with 168,093 samples. |
Scheduled DropHead: A Regularization Method for Transformer Models (2020.findings-emnlp)
Copied to clipboard
| Challenge: | DropHead is a structured dropout method for regularizing multi-head attention . DropHed drops entire attention heads during training to prevent overfitting . |
| Approach: | They propose a structured dropout method specifically designed for regularizing multi-head attention mechanism . DropHead drops entire attention heads during training to prevent overfitting . |
| Outcome: | The proposed method can improve transformer models by 0.9 BLEU score on translation task and around 1.0 accuracy for various text classification tasks. |
Modular Transformers: Compressing Transformers into Modularized Layers for Flexible Efficient Inference (2023.findings-acl)
Copied to clipboard
| Challenge: | Pre-trained sequence-to-sequence models have advanced the state of the art on text generation tasks. |
| Approach: | They introduce a modular encoder-decoder framework for flexible sequence-to-sequence model compression. |
| Outcome: | The proposed framework can achieve flexible compression ratios from 1.1x to 6x with little to moderate relative performance drop. |
Poor Man’s Quality Estimation: Predicting Reference-Based MT Metrics Without the Reference (2023.eacl-main)
Copied to clipboard
Vilém Zouhar, Shehzaad Dhuliawala, Wangchunshu Zhou, Nico Daheim, Tom Kocmi, Yuchen Eleanor Jiang, Mrinmaya Sachan
| Challenge: | State-of-the-art machine translation quality estimation systems have been achieving remarkable correlations with human judgements yet they require human annotations, which are expensive and computationally heavy. |
| Approach: | They propose a problem where one predicts automated metric scores without the reference. |
| Outcome: | The proposed model can estimate automated metrics at the sentence-level without the reference. |
OS Agents: A Survey on MLLM-based Agents for Computer, Phone and Browser Use (2025.acl-long)
Copied to clipboard
Xueyu Hu, Tao Xiong, Biao Yi, Zishu Wei, Ruixuan Xiao, Yurun Chen, Jiasheng Ye, Meiling Tao, Xiangxin Zhou, Ziyu Zhao, Yuhuai Li, Shengze Xu, Shenzhi Wang, Xinchen Xu, Shuofei Qiao, Zhaokai Wang, Kun Kuang, Tieyong Zeng, Liang Wang, Jiwei Li, Yuchen Eleanor Jiang, Wangchunshu Zhou, Guoyin Wang, Keting Yin, Zhou Zhao, Hongxia Yang, Fan Wu, Shengyu Zhang, Fei Wu
| Challenge: | a new generation of (M)LLMs is enabling the creation of superintelligent AI assistants . OS Agents can complete tasks autonomously and have the potential to significantly enhance the lives of billions of users worldwide. |
| Approach: | They propose to build OS Agents that operate within operating systems' GUIs and GUIs . they examine evaluation metrics and benchmarks to identify promising directions . |
| Outcome: | The proposed agents are based on operating systems (OS) and operating systems frameworks. |
OAgents: An Empirical Study of Building Effective Agents (2025.findings-emnlp)
Copied to clipboard
He Zhu, Tianrui Qin, King Zhu, Heyuan Huang, Yeyi Guan, Jinxiang Xia, Hanhao Li, Yi Yao, Ningning Wang, Pai Liu, Tianhao Peng, Xin Gui, Li Xiaowan, Yuhui Liu, Xiangru Tang, Jian Yang, Ge Zhang, Xitong Gao, Yuchen Eleanor Jiang, Changwang Zhang, Jun Wang, Jiaheng Liu, Wangchunshu Zhou
| Challenge: | a recent study shows that agent research practices are far from standard, rigorous . lack of a standard evaluation protocol makes previous works not reproducible, authors say . |
| Approach: | They conduct an empirical study on the GAIA benchmark to investigate agent design choices . they find that lack of a standard evaluation protocol makes previous works not reproducible . |
| Outcome: | The proposed framework achieves state-of-the-art performance among open-source projects. |
Learning from Perturbations: Diverse and Informative Dialogue Generation with Inverse Adversarial Training (2021.acl-long)
Copied to clipboard
| Challenge: | Inverse Adversarial Training (IAT) algorithm for training neural dialogue systems to avoid generic responses and model dialogue history better. |
| Approach: | They propose an algorithm that encourages the model to be sensitive to perturbations in dialogue history and learn from perturbations. |
| Outcome: | The proposed approach can model dialogue history better and generate more diverse responses on two benchmark datasets. |
EfficientVLM: Fast and Accurate Vision-Language Models via Knowledge Distillation and Modal-adaptive Pruning (2023.findings-acl)
Copied to clipboard
| Challenge: | Pre-trained vision-language models have achieved impressive results in a range of vision-linguistic tasks. |
| Approach: | They propose a distilling then pruning framework to compress large vision-language models into smaller, faster ones. |
| Outcome: | The proposed framework reduces the size of a pre-trained large vision-language model and improves its performance on vision-linguistic tasks. |
BERT-based Lexical Substitution (P19-1)
Copied to clipboard
| Challenge: | Existing approaches to lexical substitution tend to overlook good substitute candidates that are not the synonyms of the target words in the lexicals and fail to take into account the substitution’s influence on the global context of the sentence. |
| Approach: | They propose an end-to-end BERT-based lexical substitution approach which proposes and validates substitute candidates without using annotated data or manually curated resources. |
| Outcome: | The proposed approach performs well in proposing and ranking substitute candidates, achieving the state-of-the-art results in both LS07 and LS14 benchmarks. |
Towards a Mechanistic Interpretation of Multi-Step Reasoning Capabilities of Language Models (2023.emnlp-main)
Copied to clipboard
Yifan Hou, Jiaoda Li, Yu Fei, Alessandro Stolfo, Wangchunshu Zhou, Guangtao Zeng, Antoine Bosselut, Mrinmaya Sachan
| Challenge: | Recent work has shown that language models (LMs) have strong multi-step (i.e., procedural) reasoning capabilities. |
| Approach: | They propose a mechanistic interpretation of language models for multi-step reasoning tasks by introducing a new probing approach that recovers the reasoning tree from the model’s attention patterns. |
| Outcome: | The proposed model implicitly embeds a reasoning tree resembling the correct reasoning process within it, and detects the information from the model’s attention patterns for most examples. |
Pseudo-Bidirectional Decoding for Local Sequence Transduction (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Local sequence transduction tasks involve massive overlapping between source and target sequences . experimental results show that Pseudo-Bidirectional Decoding improves performance of standard seq2seq models. |
| Approach: | They propose a simple but versatile approach for local sequence transduction tasks . they propose to copy source tokens to decoder as pseudo future context . |
| Outcome: | The proposed approach improves the performance of standard seq2seq models on LST tasks. |