Papers by Di Wu

Beyond Online Sampling: Bridging Offline-to-Online Alignment via Dynamic Data Transformation for LLMs (2025.emnlp-main)

Challenge:	ESGenius is a comprehensive benchmark for evaluating Large Language Models on ESG and sustainability knowledge.
Approach:	They introduce ESGenius, a benchmark for evaluating and enhancing ESG proficiency . they use a rigorous two-stage evaluation protocol and a repository of foundational frameworks .
Outcome:	ESGenius is a benchmark for evaluating and enhancing the proficiency of Large Language Models (LLMs) in ESG and sustainability-focused question answering.

Copied to clipboard

Zhang Zhang, Guhao Feng, Jian Guan, Di He, Wei Wu

Responsible & Trustworthy NLP

Challenge:	Direct Preference Optimization (DPO) eliminates complex reward modeling in aligning large language models with human preferences, but its online variant faces significant efficiency bottlenecks due to costly real-time preference sampling and the reward model annotation.
Approach:	They propose a framework that transforms static datasets into dynamically adaptive equivalents without the need for an explicit reward model.
Outcome:	The proposed approach matches or exceeds the performance of a fully online DPO.

Improving Neural Machine Translation by Bidirectional Training (2021.emnlp-main)

Copied to clipboard

Liang Ding, Di Wu, Dacheng Tao

Label Effects: Shared Heuristic Reliance in Trust Assessment by Humans and LLM-as-a-Judge (2026.acl-long)

Challenge:	Experimental results show that bidirectional training pushes the SOTA neural machine translation performance significantly higher.
Approach:	They propose a bidirectional training strategy that updates model parameters at the early stage and tunes it normally.
Outcome:	The proposed approach pushes the SOTA neural machine translation performance significantly higher on 15 translation tasks on 8 language pairs.

Copied to clipboard

Xin Sun, Di Wu, Sijing Qin, Isao Echizen, Abdallah El Ali, Saku Sugawara

Challenge:	Large language models (LLMs) are increasingly used as automated evaluators . et al., 2024: strong labels can foster trust but also undermine it .
Approach:	They show that LLMs' source labels bias trust judgments by humans . they use eye-tracking data to analyze LLM internal states during judgment .
Outcome:	The proposed model is biased by disclosed source labels, the authors show . eye-tracking data show humans rely heavily on source labels for judgments .

BRIEF: Bridging Retrieval and Inference for Multi-hop Reasoning via Compression (2025.findings-naacl)

Copied to clipboard

Yuankai Li, Jia-Chen Gu, Di Wu, Kai-Wei Chang, Nanyun Peng

Information Retrieval

Challenge:	Existing approaches to augment language models with external knowledge but they are limited by static nature of pre-training data.
Approach:	They propose a lightweight approach that compresses retrieved documents into highly dense textual summaries to integrate into in-context RAG.
Outcome:	The proposed approach reduces latency and costs while achieving high performance in open-domain questions.

How to Learn in a Noisy World? Self-Correcting the Real-World Data Noise in Machine Translation (2025.findings-naacl)

Copied to clipboard

Yan Meng, Di Wu, Christof Monz

Machine Translation

Challenge:	Semantic misalignment, as the primary source of the noise, poses a challenge for training machine translation systems.
Approach:	They propose a process for simulating misalignment controlled by semantic similarity which closely resembles misaligned sentences in real-world web-crawled corpora.
Outcome:	The proposed model significantly improves translation performance in the presence of misalignment noise and when applied to real-world, noisy web-mined datasets, across a range of translation tasks.

Grafting Pre-trained Models for Multimodal Headline Generation (2022.emnlp-industry)

Copied to clipboard

Lingfeng Qiao, Chen Wu, Ye Liu, Haoyuan Peng, Di Yin, Bo Ren

Know-MRI: A Knowledge Mechanisms Revealer&Interpreter for Large Language Models (2025.acl-demo)

Challenge:	Existing approaches to generate video headlines with pre-trained language models are labor intensive and impractical.
Approach:	They propose to graft the encoder from the pre-trained video-language model on the generative pre-trainer model and propose a consensus fusion mechanism for the integration of different components.
Outcome:	The proposed model achieves strong results on a brand-new dataset collected from real-world applications.

Copied to clipboard

Jiaxiang Liu, Boxuan Xing, Chenhao Yuan, ChenxiangZhang ChenxiangZhang, Di Wu, Xiusheng Huang, Haida Yu, Chuhan Lang, Pengfei Cao, Jun Zhao, Kang Liu

Explainability & Interpretability in NLP

Challenge:	Existing interpretation methods only support tasks with specific inputs, limiting their practical applications.
Approach:	They propose an extensible module that matches different input data with interpretation methods and consolidates the interpreting outputs.
Outcome:	The proposed module can match different input data with interpretation methods and consolidate the interpreting outputs.

From Detection to Understanding: Multi-Turn Reasoning for Video Misinformation Analysis (2026.acl-long)

Copied to clipboard

Zhi Zeng, Jiaying Wu, Minnan Luo, Di Zhang, Yifei Yang, Xiangzheng Kong, Herun Wan, Zihan Ma

Understanding and Mitigating Political Stance Cross-topic Generalization in Large Language Models (2026.acl-long)

Challenge:	Existing benchmarks focus on binary veracity judgments and do not evaluate process-level justifications for misinformation models.
Approach:	They propose a video misinformation analysis benchmark that assesses reasoning in video misinterpretation.
Outcome:	The proposed framework improves reasoning accuracy and explanation quality compared to existing models . it covers 12 fine-grained deception categories and progresses from perceptual attribution to intent and persuasion analysis.

Copied to clipboard

Jiayi Zhang, Shu Yang, Junchao Wu, Derek F. Wong, Di Wang

What Does Infect Mean to Cardio? Investigating the Role of Clinical Specialty Data in Medical LLMs (2026.eacl-long)

Challenge:	Recent studies have focused on the internal representations of large language models and the mechanisms that lead to unintended cross-topic generalization.
Approach:	They propose a method that uses inhibition to localize political neurons and a technique that uses topic-specific blocking to mitigate the cross-topic generalization.
Outcome:	The proposed method reduces cross-topic generalization by 20% while preserving topic-specific performance.

Copied to clipboard

Xinlan Yan, Di Wu, Yibin Lei, Christof Monz, Iacer Calixto

Question Answering

Challenge:	S-MedQA is an English question-answering dataset designed for benchmarking large language models in fine-grained clinical specialties.
Approach:	They propose to use an English medical question-answering dataset to benchmark large language models in clinical specialties.
Outcome:	The proposed dataset is designed to benchmark large language models in medical specialties.

How Far can 100 Samples Go? Unlocking Zero-Shot Translation with Tiny Multi-Parallel Data (2024.findings-acl)

Copied to clipboard

Di Wu, Shaomu Tan, Yan Meng, David Stap, Christof Monz

DVI-DTM: Dual-View Representation Learning for Interpretable Short Text Dynamic Topic Modeling (2026.acl-long)

Challenge:	a common solution to zero-shot translation is to add as many related translation directions as possible to the training corpus.
Approach:	They show that a small amount of multi-parallel data can achieve significant zero-shot improvements . they say that the resulting non-English performance is close to the complete translation upper bound .
Outcome:	The proposed model achieves +21.7 ChrF++ non-English translation improvements on EC30 dataset . the resulting non- English performance exceeds M2M100 by an average of 5.9 ChrF+ .

Copied to clipboard

Di Liu, Zheng Fang, Bin Wu

Explainability & Interpretability in NLP

Challenge:	Existing dynamic topic modeling methods face semantic ambiguity and interpretation ambiguities when applied to short texts.
Approach:	They propose a Dual-View representation learning-based Interpretable short text Dynamic Topic Model to address semantic ambiguity and interpretation ambiguities.
Outcome:	The proposed model outperforms the state-of-the-art models in topic alignment and dynamic topic quality metrics while producing highly interpretable topic descriptions.

On Leveraging Encoder-only Pre-trained Language Models for Effective Keyphrase Generation (2024.lrec-main)

Copied to clipboard

Di Wu, Wasi U. Ahmad, Kai-Wei Chang

Parameter Importance is Not Static: Evolving Parameter Isolation for Supervised Fine-Tuning (2026.acl-long)

Challenge:	a new study examines the use of encoder-only pre-trained language models in keyphrase generation (KPG) keyphrases are phrases that condense salient information of a document.
Approach:	They propose to use encoder-only pre-trained language models in keyphrase generation . they also examine optimal architectural decisions for employing encoder only PLMs in KPG .
Outcome:	The proposed model outperforms general-domain seq2seq models in keyphrase generation.

Copied to clipboard

Zekai Lin, Chao Xue, Di Liang, Xingsheng Han, Peiyang Liu, Xianjie Wu, Lei Jiang, Yu Lu, Bob Simons, Shuang Liang, Minlong Peng

Why Supervised Fine-Tuning Fails to Learn: A Systematic Study of Incomplete Learning in Large Language Models (2026.acl-long)

Challenge:	Recent approaches to fine-tuning of large language models suffer from task interference and catastrophic forgetting.
Approach:	They propose a fine-tuning framework that adapts isolation decisions based on online estimates of parameter importance.
Outcome:	The proposed framework reduces interference and forgetting while releasing outdated parameters to recover plasticity.

Copied to clipboard

Chao Xue, Yao Wang, Mengqiao Liu, Di Liang, Xingsheng Han, Peiyang Liu, Xianjie Wu, Chenyao Lu, Lei Jiang, Yu Lu, Haibo Shi, Shuang Liang, Minlong Peng, Flora D. Salim

AC-EVAL: Evaluating Ancient Chinese Language Understanding in Large Language Models (2024.findings-emnlp)

Challenge:	Incomplete learning is widespread and heterogeneous in large language models . authors identify five recurrent sources of incomplete learning: missing prerequisite knowledge, conflicts between SFT supervision and pre-training knowledge, internal inconsistencies within SFT data, left-side forgetting during sequential fine-tuning, and insufficient optimization for rare or complex patterns.
Approach:	They propose a diagnostic-first framework that maps incomplete learning to causes . they identify five recurrent sources of incomplete learning: missing prerequisite knowledge, conflicts between supervision and pre-training knowledge, internal inconsistencies, left-side forgetting during sequential fine-tuning, and insufficient optimization for rare or complex patterns.
Outcome:	The proposed framework maps incomplete learning to causes using observable training and inference signals.

Copied to clipboard

Yuting Wei, Yuanxing Xu, Xinru Wei, Yangsimin Yangsimin, Yangfu Zhu, Yuqing Li, Di Liu, Bin Wu

A Holistic Approach to Reference-Free Evaluation of Machine Translation (2023.acl-short)

Challenge:	AC-EVAL is a benchmark designed to assess the advanced knowledge and reasoning capabilities of LLMs within the context of ancient Chinese.
Approach:	They propose a benchmark to assess the advanced knowledge and reasoning capabilities of LLMs in ancient Chinese.
Outcome:	AC-EVAL aims to assess the comprehension of ancient Chinese texts . the benchmark covers 13 tasks covering historical facts, geography, social customs, art, philosophy, classical poetry and prose.

Copied to clipboard

Hanming Wu, Wenjuan Han, Hui Di, Yufeng Chen, Jinan Xu

Active Instruction Tuning: Improving Cross-Task Generalization by Training on Prompt Sensitive Tasks (2023.emnlp-main)

Challenge:	Traditional machine translation evaluation relies on reference written by humans . reference-free evaluation gets rid of labor-intensive annotations, which can pivot easily to new domains .
Approach:	They propose a reference-free evaluation approach that characterizes evaluation as two aspects: fluency and faithfulness.
Outcome:	The proposed approach outperforms SOTA reference-fee metrics on machine translation datasets.

Copied to clipboard

Po-Nien Kung, Fan Yin, Di Wu, Kai-Wei Chang, Nanyun Peng

On-policy Reinforcement Fine-tuning with Offline reward for Multi-step Embodied Planning (2026.acl-long)

Challenge:	Instruction tuning (IT) achieves impressive zero-shot generalization results by training large language models on diverse tasks with instructions.
Approach:	They propose a framework to identify informative tasks and then actively tune models on selected tasks.
Outcome:	The proposed method outperforms baseline strategies for task selection on NIV2 and Self-Instruct datasets.

Copied to clipboard

Di Wu, Jiaxin Fan, Chloe Gu, Guanbo Wang, Wei Yin, Wenhao Li, Bo Jin

Tell Me What You Don’t Know: Enhancing Refusal Capabilities of Role-Playing Agents via Representation Space Analysis and Editing (2025.findings-acl)

Challenge:	Embodied planning requires agents to make coherent multi-step decisions based on dynamic visual observations and verbal goals.
Approach:	They propose an On-policy Reinforcement fine-tuning framework with offline rewards for Embodied Task Planning that preserves generalization benefits of RFT while addressing costly interaction and sparse rewards.
Outcome:	The proposed framework outperforms closed-source and online-RL methods on EmbodiedBench, a recent benchmark for interactive embodied tasks.

Copied to clipboard

Wenhao Liu, Siyu An, Junru Lu, Muling Wu, Tianlong Li, Xiaohua Wang, Changze Lv, Xiaoqing Zheng, Di Yin, Xing Sun, Xuanjing Huang

Representation Learning

Challenge:	Role-playing Agents (RPAs) struggle to recognize and respond to hard queries that conflict with their role-play knowledge.
Approach:	They propose a lightweight representation editing approach that conveniently shifts conflicting requests to the rejection region, thereby enhancing the model’s refusal accuracy.
Outcome:	The proposed model improves RPAs’ refusal ability of conflicting requests while maintaining their general role-playing capabilities.

Beyond Shared Vocabulary: Increasing Representational Word Similarities across Languages for Multilingual Machine Translation (2023.emnlp-main)

Copied to clipboard

Di Wu, Christof Monz

Rethinking Model Selection and Decoding for Keyphrase Generation with Pre-trained Sequence-to-Sequence Models (2023.emnlp-main)

Challenge:	Using a shared vocabulary is common practice in multilingual machine translation . however, when words overlap is small, e.g., using different writing systems, knowledge transfer is inhibited .
Approach:	They propose a re-parameterized method for building word embeddings using word equivalence classes and graph networks to fuse word embeds across languages.
Outcome:	The proposed method achieves evident BLEU improvements on high- and low-resource MNMT scenarios.

Copied to clipboard

Di Wu, Wasi Ahmad, Kai-Wei Chang

PLAWBENCH: A Rubric-Based Benchmark for Evaluating LLMs in Real-World Legal Practice (2026.acl-long)

Challenge:	Keyphrase generation is a longstanding task in NLP with widespread applications.
Approach:	They propose a likelihood-based decode-select algorithm for seq2seq PLMs that improves greedy search by an average of 4.7% semantic F1 across five datasets.
Outcome:	The proposed algorithm improves greedy search by an average of 4.7% semantic F1 across five datasets.

Copied to clipboard

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

Reinforcement Learning on Pre-Training Data (2026.acl-long)

Challenge:	Existing benchmarks for large language models (LLMs) are coarse, single-dimensional metrics and do not explicitly assess fine-grained legal reasoning.
Approach:	They propose a Practical Law Benchmark to evaluate large language models in real-world legal practice scenarios.
Outcome:	The proposed model is based on 850 questions and 13 scenarios with expert-designed evaluation rubrics.

Copied to clipboard

Siheng Li, Kejiao Li, Zenan Xu, Guanhua Huang, Kun Li, Haoyuan Wu, null Wujiajia, Zihao Zheng, Chenchen Zhang, Kun Shi, Xue Gong, Qi Yi, Ruibin Xiong, Tingqiang Xu, Yuhao Jiang, Jianfeng Yan, Yuyuan Zeng, Guanghui Xu, Jinbao Xue, Zhijiang xu, Zheng Fang, Shuai LI, Qibin Liu, Xiaoxue Li, Zhuoyu Li, Yangyu Tao, Fei Gao, Cheng Jiang, Bochao Wang, Kai Liu, Jianchen Zhu, Wai Lam, Bo Zhou, Di Wang

CQG: A Simple and Effective Controlled Generation Framework for Multi-hop Question Generation (2022.acl-long)

Challenge:	Recent progress in large language models is driven by scaling of training compute through pre-training with nexttoken prediction (NTP) or post-training (RL) Pre-training using NTP enables models to acquire extensive knowledge and skills from general data, but it suffers from data inefficiency and catastrophic forgetting in continual learning settings.
Approach:	They propose to scale training compute through pre-training with next-token prediction (NTP) or post-training by scaling reinforcement learning (RL) to improve learning from general data.
Outcome:	Experiments on multiple benchmarks and models show that the proposed approach improves continual pre-training and provides a strong foundation for post-training on Qwen3-8B-Base.

Copied to clipboard

Zichu Fei, Qi Zhang, Tao Gui, Di Liang, Sirui Wang, Wei Wu, Xuanjing Huang

Text Generation

Challenge:	Current models can not ensure the complexity of generated questions, so they may generate shallow questions that can be answered without multi-hop reasoning.
Approach:	They propose a controlled framework to generate multi-hop questions that contain key entities in multi- hop reasoning chains and a novel Transformer-based decoder to guarantee that key entities appear in the questions.
Outcome:	The proposed model outperforms the state-of-the-art model 25% on HotpotQA.

Reason Only When Needed: Efficient Generative Reward Modeling via Model-Internal Uncertainty (2026.findings-acl)

Copied to clipboard

Chao Xue, Yao Wang, Mengqiao Liu, Di Liang, Xingsheng Han, Peiyang Liu, Xianjie Wu, Chenyao Lu, Lei Jiang, Yu Lu, Haibo Shi, Shuang Liang, Minlong Peng, Flora D. Salim

Responsible & Trustworthy NLP

Challenge:	Existing approaches to generating reward models rely on voting-based mechanisms to evaluate CoT outputs.
Approach:	They propose an efficient generative reward modeling framework grounded in model-internal uncertainty.
Outcome:	The proposed framework reduces inference cost while improving answer accuracy.

Structural Reward Model: Enhancing Interpretability, Efficiency, and Scalability in Reward Modeling (2025.emnlp-industry)

Copied to clipboard

Xiaoyu Liu, Di Liang, Hongyu Shan, Peiyang Liu, Yonghao Liu, Muling Wu, Yuntao Li, Xianjie Wu, Li Miao, Jiangrong Shen, Minlong Peng

Responsible & Trustworthy NLP

Challenge:	Generative RMs (GRMs) lack contextual and background information during inference, leading to incomplete evaluations.
Approach:	They propose a modular and interpretable framework that integrates side-branch models as auxiliary feature generators.
Outcome:	The proposed framework outperforms scalar and saline reward models in robustness and alignment with human preferences.

BRIEF-Pro: Universal Context Compression with Short-to-Long Synthesis for Fast and Accurate Multi-Hop Reasoning (2026.findings-acl)

Copied to clipboard

Jia-Chen Gu, Junyi Zhang, Di Wu, Yuankai Li, Kai-Wei Chang, Nanyun Peng

Separate the Wheat from the Chaff: A Post-Hoc Approach to Safety Re-Alignment for Fine-Tuned Language Models (2025.findings-acl)

Challenge:	Experiments show that BRIEF-Pro generates more concise and relevant summaries, enhancing performance across small, large, and proprietary language models.
Approach:	They propose a universal, lightweight compressor that distills relevant evidence from retrieved documents into a concise summary for seamless integration into in-context RAG.
Outcome:	Experiments on four open-domain multi-hop question-answering datasets show that BRIEF-Pro generates more concise and relevant summaries, enhancing performance across small, large, and proprietary language models.

Copied to clipboard

Di Wu, Xin Lu, Yanyan Zhao, Bing Qin

Improving Semantic Matching through Dependency-Enhanced Pre-trained Model with Adaptive Fusion (2022.findings-emnlp)

Challenge:	Large language models achieve effective safety alignment at the time of release, but fine-tuning often compromises safety mechanisms.
Approach:	They propose a method that performs safety realignment for large language models . they identify unsafe delta parameters from the fine-tuned models and recalibrate the retained parameters .
Outcome:	The proposed method improves safety performance on safety benchmarks and jailbreak attacks while maintaining their performance on downstream tasks.

Copied to clipboard

Jian Song, Di Liang, Rumei Li, Yuntao Li, Sirui Wang, Minlong Peng, Wei Wu, Yongxin Yu

SlotRefine: A Fast Non-Autoregressive Model for Joint Intent Detection and Slot Filling (2020.emnlp-main)

Challenge:	Existing work on dependency prior structure integration into pre-trained models is still unclear.
Approach:	They propose a dependency-based fusion attention paradigm which explicitly introduces dependency prior structure into pre-trained models and adaptively fuses it with semantic information.
Outcome:	The proposed model achieves state-of-the-art or competitive performance on 10 public datasets, demonstrating the benefits of adaptively fusing dependency structure in semantic matching task.

Copied to clipboard

Di Wu, Liang Ding, Fan Lu, Jian Xie

Sentiment Analysis

Challenge:	Slot filling and intent detection are two main tasks in spoken language understanding systems.
Approach:	They propose a non-autoregressive slot filling model with two-pass iteration mechanism to handle uncoordinated slots problem.
Outcome:	The proposed model significantly outperforms previous models in slot filling task while speeding up decoding.

Empathy Intent Drives Empathy Detection (2023.emnlp-main)

Copied to clipboard

Liting Jiang, Di Wu, Bohui Mao, Yanbing Li, Wushour Slamu

Natural Language Interfaces

Challenge:	Empathy is essential in human social interaction.
Approach:	They propose to annotate healthy empathy detection datasets IEMPATHIZE and TwittEmp with 8 empathy intent labels and perform joint training for the two tasks.
Outcome:	The proposed framework outperforms baselines on the two datasets.

Fraud-R1 : A Multi-Round Benchmark for Assessing the Robustness of LLM Against Augmented Fraud and Phishing Inducements (2025.findings-acl)

Copied to clipboard

Shu Yang, Shenzhe Zhu, Zeyu Wu, Keyu Wang, Junchi Yao, Junchao Wu, Lijie Hu, Mengdi Li, Derek F. Wong, Di Wang

Robustness in NLP

Challenge:	Existing fraud detection benchmarks focus on single-turn classification tasks, failing to capture dynamic nature of real-world fraud attempts.
Approach:	They propose a bilingual benchmark to assess LLMs' ability to resist fraud and phishing attacks across five key fraud categories: Fraudulent Services, Impersonation, Phishing Scams, Fake Job Postings, and Online Relationships.
Outcome:	The proposed model improves in role-play settings and in e-commerce and recommendation systems.

Inductive-Deductive Strategy Reuse for Multi-Turn Instructional Dialogues (2024.emnlp-main)

Copied to clipboard

Jiao Ou, Jiayu Wu, Che Liu, Fuzheng Zhang, Di Zhang, Kun Gai

Dialogue Systems & Conversational Agents

Challenge:	Existing methods target instruction dialogues as learning goal and fine-tune user simulators to pose instructions.
Approach:	They propose to use real instruction dialogues to model complex dialogue flows and pose high-quality instructions.
Outcome:	The proposed method generates diverse, in-depth, and insightful instructions for a given dialogue history.

Robust Lottery Tickets for Pre-trained Language Models (2022.acl-long)

Copied to clipboard

Rui Zheng, Bao Rong, Yuhao Zhou, Di Liang, Sirui Wang, Wei Wu, Tao Gui, Qi Zhang, Xuanjing Huang

Robustness in NLP

Challenge:	Recent studies have shown that pre-trained language models contain smaller matching subnetworks that are not robust to adversarial examples.
Approach:	They propose a method to find robust tickets hidden in pre-trained language models by learning binary weight masks and an adversarial loss objective to guide the search.
Outcome:	The proposed method improves on previous work on adversarial robustness evaluation.

The Factuality Tax of Diversity-Intervened Text-to-Image Generation: Benchmark and Fact-Augmented Intervention (2024.emnlp-main)

Copied to clipboard

Yixin Wan, Di Wu, Haoran Wang, Kai-Wei Chang

Meta-Task Prompting Elicits Embeddings from Large Language Models (2024.acl-long)

Challenge:	Prompt-based “diversity interventions” are commonly adopted to improve the diversity of Text-to-Image models depicting individuals with diverse racial or gender traits.
Approach:	They propose a benchmark to quantify the trade-off between using diversity interventions and preserving demographic factuality in T2I models.
Outcome:	The proposed model significantly improves the demographic factuality under diversity interventions while preserving diversity.

Copied to clipboard

Yibin Lei, Di Wu, Tianyi Zhou, Tao Shen, Yu Cao, Chongyang Tao, Andrew Yates

Machine Translation With Weakly Paired Documents (D19-1)

Challenge:	Existing methods for large language modeling are based on task-related instructions or prompts.
Approach:	They propose a method for generating high-quality sentence embeddings from Large Language Models (LLMs) using meta-task prompts.
Outcome:	The proposed method produces high-quality sentences without fine-tuning . it excels on STS benchmarks and in downstream tasks, surpassing models with similar prompts .

Copied to clipboard

Lijun Wu, Jinhua Zhu, Di He, Fei Gao, Tao Qin, Jianhuang Lai, Tie-Yan Liu

Representational Isomorphism and Alignment of Multilingual Large Language Models (2024.findings-emnlp)

Challenge:	Recent studies explore the possibility of unsupervised machine translation with monolingual data only.
Approach:	They propose a method to mine bilingual sentences from weakly paired documents . they use word distribution-level alignments to constrain word distributions of two weakly-paired documents.
Outcome:	The proposed method outperforms previous results on six translation tasks using weakly paired bilingual documents and a large number of bilingual sentences.

Copied to clipboard

Di Wu, Yibin Lei, Andrew Yates, Christof Monz

Beyond Error Propagation in Neural Machine Translation: Characteristics of Language Also Matter (D18-1)

Challenge:	Existing isomorphism of sentence representations can facilitate representational alignments in zero-shot and few-shot settings.
Approach:	They propose to apply a contrastive objective to LLMs with a small number of translation pairs to improve models' performance on Semantic Textual Similarity tasks.
Outcome:	The proposed representation-level approach significantly improves on Semantic Textual Similarity (STS) tasks across languages even without a monolingual objective.

Copied to clipboard

Lijun Wu, Xu Tan, Di He, Fei Tian, Tao Qin, Jianhuang Lai, Tie-Yan Liu

Text Generation

Challenge:	Neural machine translation suffers from exposure bias and error propagation problem.
Approach:	They conduct a series of analyses to deeply understand the accuracy drop problem . they find that the left part of the translated sentence is often better than its right part .
Outcome:	The results show that the left part of the translated sentence is often better than its right part in left-to-right decoding models.

KPEval: Towards Fine-Grained Semantic-Based Keyphrase Evaluation (2024.findings-acl)

Copied to clipboard

Di Wu, Da Yin, Kai-Wei Chang

Term Extraction

Challenge:	Existing evaluation methods for keyphrase extraction and generation rely on exact matching with human references.
Approach:	They propose a framework for evaluation that includes four critical aspects: reference agreement, faithfulness, diversity, utility and semantic-based metrics.
Outcome:	The proposed evaluation framework correlates better with human preferences than previously proposed metrics.

Improving Translation Quality Estimation with Bias Mitigation (2023.acl-long)

Copied to clipboard

Hui Huang, Shuangzhi Wu, Kehai Chen, Hui Di, Muyun Yang, Tiejun Zhao

Challenges to Open-Domain Constituency Parsing (2022.findings-acl)

Challenge:	State-of-the-art translation Quality Estimation models are biased, relying on monolingual features while ignoring the bilingual semantic alignment.
Approach:	They propose a method to mitigate the bias of translation quality estimation models by contrastive learning between clean and noisy sentence pairs.
Outcome:	The proposed method improves the estimation performance while mitigating the bias.

Copied to clipboard

Sen Yang, Leyang Cui, Ruoxi Ning, Di Wu, Yue Zhang

Challenge:	Existing findings on cross-domain constituency parsing are only made on a limited number of domains.
Approach:	They manually annotate a high-quality constituency treebank containing five domains and analyze challenges to open-domain constituency parsing using a set of linguistic features.
Outcome:	The proposed model significantly improves the performance of the proposed model on the domain-variant features.

MetaKP: On-Demand Keyphrase Generation (2024.findings-emnlp)

Copied to clipboard

Di Wu, Xiaoxian Shen, Kai-Wei Chang

KMatrix: A Flexible Heterogeneous Knowledge Enhancement Toolkit for Large Language Model (2024.emnlp-demo)

Challenge:	Existing keyphrase prediction methods only output a single set of keyphrases per document . however, existing methods fail to cater to diverse needs of users and downstream applications .
Approach:	They propose a method that requires keyphrases that conform to specific high-level goals or intents to generate on-demand keyphrase generation.
Outcome:	The proposed method surpasses the performance of a fully fine-tuned BART-base model in 0.548 SemF1 . it can be used in epidemic event detection from social media.

Copied to clipboard

Shun Wu, Di Wu, Kun Luo, XueYou Zhang, Jun Zhao, Kang Liu

HAIC: Improving Human Action Understanding and Generation with Better Captions for Multi-modal Large Language Models (2025.acl-long)

Challenge:	Existing Knowledge-Enhanced Large Language Models (K-LLMs) toolkits focus on free-textual knowledge and lack robust datasets, models, and user-friendly experience.
Approach:	They propose a flexible heterogeneous knowledge enhancement toolkit to enhance Large Language Models (LLMs) using external knowledge.
Outcome:	KMatrix: a flexible heterogeneous knowledge enhancement toolkit for LLMs includes verbalizing-retrieval and parsing-query methods.

Copied to clipboard

Xiao Wang, Jingyun Hua, Weihong Lin, Yuanxing Zhang, Fuzheng Zhang, Jianlong Wu, Di Zhang, Liqiang Nie

KMatrix-2: A Comprehensive Heterogeneous Knowledge Collaborative Enhancement Toolkit for Large Language Model (2025.emnlp-demos)

Challenge:	Existing studies have shown that high-quality video captions can improve MLLMs' performance on videos involving human actions.
Approach:	They propose a data annotation pipeline to collect videos featuring clear human actions from the Internet and annotate them in a standardized caption format that uses human attributes to distinguish individuals.
Outcome:	The proposed pipeline combines two datasets to evaluate human action understanding.

Copied to clipboard

Shun Wu, Di Wu, Wangtao Sun, Ziyang Huang, Xiaowei Yuan, Kun Luo, XueYou Zhang, Shizhu He, Jun Zhao, Kang Liu

UniKeyphrase: A Unified Extraction and Generation Framework for Keyphrase Prediction (2021.findings-acl)

Challenge:	Existing studies on K-LLMs systems focus on declarative knowledge and procedural knowledge (rules) .
Approach:	They propose to build a toolkit that supports comprehensive heterogeneous knowledge collaborative enhancement for Large Language Models (LLMs).
Outcome:	The proposed toolkit provides unified knowledge integration and joint knowledge retrieval methods to achieve more comprehensive heterogeneous knowledge collaborative enhancement.

Copied to clipboard

Huanqin Wu, Wei Liu, Lei Li, Dan Nie, Tao Chen, Feng Zhang, Di Wang

Information Extraction & Text Mining

Challenge:	Mainstream methods that ignore the diversity among keyphrases or weakly capture the relation between tasks implicitly ignore keyphrase diversity.
Approach:	They propose a novel end-to-end learning framework that jointly learns to extract and generate keyphrases by exploiting latent semantic relation between extraction and generation.
Outcome:	The proposed approach outperforms mainstream methods on a benchmarked document on keyphrase prediction.

Towards Stable and Effective Reinforcement Learning for Mixture-of-Experts (2026.acl-long)

Copied to clipboard

Di Zhang, Xun Wu, Shaohan Huang, Lingjie Jiang, Yaru Hao, Li Dong, Zewen Chi, Zhifang Sui, Furu Wei

Can Large Language Models Identify Implicit Suicidal Ideation? An Empirical Evaluation (2025.findings-emnlp)

Challenge:	Reinforcement learning with verifiable rewards (RLVR) training with Mixture-of-Experts policies remains fragile and prone to reward collapse.
Approach:	They propose a router shift-based policy optimization method that computes a per-token router-shift ratio conditioned on the previously activated experts and applies stop-gradient and a lower-bound floor.
Outcome:	The proposed method achieves better performance and greater stability than previous methods.

Copied to clipboard

Tong Li, Shu Yang, Junchao Wu, Jiyao Wei, Lijie Hu, Mengdi Li, Derek F. Wong, Joshua R. Oltmanns, Di Wang

Open-Domain Safety Policy Construction (2026.findings-eacl)

Challenge:	Existing data on suicidal ideation in private conversations are limited . a new dataset of 1,200 test cases is presented to address this gap .
Approach:	They propose a dataset of 1,200 test cases simulating implicit suicidal ideation in private contexts.
Outcome:	The proposed dataset includes 1,200 test cases simulating implicit suicidal ideation in dialogue scenarios.

Copied to clipboard

Di Wu, Siyue Liu, Zixiang Ji, Ya-Liang Chang, Zhe-Yu Liu, Andrew Pleffer, Kai-Wei Chang

Challenge:	Moderation layers are core component of many products built on user-generated content.
Approach:	They propose a system that drafts a content moderation policy based on human-written seed domain information.
Outcome:	The proposed system outperforms definition-only and in-context learning baselines on openAI undesired content benchmarks and an in-house multimodal advertisement moderation benchmark.

DABERT: Dual Attention Enhanced BERT for Semantic Matching (2022.coling-1)

Copied to clipboard

Sirui Wang, Di Liang, Jian Song, Yuntao Li, Wei Wu

Neuron Specialization: Leveraging Intrinsic Task Modularity for Multilingual Machine Translation (2024.emnlp-main)

Challenge:	Existing models for semantic sentence matching lack the ability to capture subtle differences.
Approach:	They propose to use a Transformer-based pre-trained language model to capture fine-grained differences in sentence pairs by introducing a dual attention module and a fusion module to learn the aggregation of difference and affinity features.
Outcome:	The proposed method is able to capture fine-grained differences in sentence pairs.

Copied to clipboard

Shaomu Tan, Di Wu, Christof Monz

LLaMA-Berry: Pairwise Optimization for Olympiad-level Mathematical Reasoning via O1-like Monte Carlo Tree Search (2025.naacl-long)

Challenge:	Language-specific modeling methods that focus on heuristics for allocation of capacity and lack knowledge transfer capabilities are often prone to interference due to conflicting optimization demands.
Approach:	They propose a method that identifies specialized neurons to modularize feed-forward layers and updates them through sparse networks to avoid interference under multilingual translation.
Outcome:	The proposed approach achieves consistent performance gains over strong baselines with additional analyses showing reduced interference and increased knowledge transfer.

Copied to clipboard

Di Zhang, Jianbo Wu, Jingdi Lei, Tong Che, Jiatong Li, Tong Xie, Xiaoshui Huang, Shufei Zhang, Marco Pavone, Yuqiang Li, Wanli Ouyang, Dongzhan Zhou

Decoder-Only LLMs can be Masked Auto-Encoders (2025.acl-short)

Challenge:	LLaMA-Berry is an advanced mathematical reasoning framework to enhance the problem-solving ability of large language models (LLMs).
Approach:	They propose a Monte Carlo Tree Search and Self-Refine framework to optimize reasoning paths and a pairwise reward model to evaluate different paths globally.
Outcome:	The proposed framework overcomes inefficiencies and limitations of step-wise and greedy search algorithms, enabling more efficient exploration of solution spaces.

Copied to clipboard

Dan Qiao, Yuan Gao, Zheming Yang, Di Yang, Ziheng Wu, Pengcheng Lu, Minghui Qiu, Juntao Li, Min Zhang

INT: Establishing Information Transfer for Multilingual Intent Detection and Slot Filling (2025.findings-acl)

Challenge:	Modern NLP workflows require different models for generation and embedding tasks.
Approach:	They propose a method that transforms an LLM into a Uni-Directional Masked Auto-Encoder.
Outcome:	The proposed method achieves state-of-the-art under unsupervised conditions with merely 100 training steps.

Copied to clipboard

Di Wu, Liting Jiang, Bohui Mao, Hongyan Xie, Haoxiang Su, Zhongjiang He, Ruiyu Fang, Shuangyong Song, Hao Huang, Xuelong Li

Sentiment Analysis

Challenge:	Existing studies struggle to achieve performance comparable to that on high-resource languages due to inherent linguistic diversity of multilingual SLU tasks.
Approach:	They propose a multilingual information transfer network to solve these challenges . they propose to reformulate SF as a span prediction problem and introduce a slot-matching attention mechanism to achieve slot alignment across languages.
Outcome:	The proposed model outperforms baseline models on the MASSIVE and MASSIV-UG datasets in overall accuracy across all languages.

VisRet: Visualization Improves Knowledge-Intensive Text-to-Image Retrieval (2026.acl-long)

Copied to clipboard

Di Wu, Yixin Wan, Kai-Wei Chang

Please Translate Again: Two Simple Experiments on Whether Human-Like Reasoning Helps Translation (2025.emnlp-main)

Challenge:	Text-to-image retrieval is challenging because of cross-modal embeddings are bags of concepts, underrepresenting structured visual relationships.
Approach:	They propose a retrieval paradigm that embeds textual queries into the image modality via T2I generation and performs retrieval within the image mode to bypass weaknesses of cross-modal retrievers in recognizing subtle visual-spatial features.
Outcome:	The proposed retrieval paradigm outperforms previous approaches in visual-spatial retrieval benchmarks.

Copied to clipboard

Di Wu, Seth Aycock, Christof Monz

Machine Translation

Challenge:	Large Language Models (LLMs) demonstrate strong reasoning capabilities for many tasks, often by explicitly decomposing the task via Chain-of-Thought (CoT) reasoning.
Approach:	They propose to use a multi-step prompt to decompose translation and train models to incorporate intermediate steps.
Outcome:	The proposed model achieves state-of-the-art on WMT24 data and shows that prompting LLMs to “translate again” and self-refine yields even better results than human-like step-by-step prompting.

Synchronous Faithfulness Monitoring for Trustworthy Retrieval-Augmented Generation (2024.emnlp-main)

Copied to clipboard

Di Wu, Jia-Chen Gu, Fan Yin, Nanyun Peng, Kai-Wei Chang

Information Retrieval

Challenge:	Existing studies show that RALMs generate baseless information or contradicts with the retrieved context.
Approach:	They propose a lightweight monitor that leverages fine-grained decoding dynamics to synchronously detect unfaithful sentences.
Outcome:	Empirical results show that SynCheck outperforms baseline faithfulness detection and FOD outperformed traditional strategies in terms of faithfulness.

Context-Aware Cross-Attention for Non-Autoregressive Translation (2020.coling-main)

Copied to clipboard

Liang Ding, Longyue Wang, Di Wu, Dacheng Tao, Zhaopeng Tu

Machine Translation

Challenge:	Existing studies have shown that non-autoregressive translation models can predict all tokens independently and simultaneously.
Approach:	They propose to enhance signals of neighbour source tokens into conventional cross-attention to address a locality perception problem in NAT cross- attention.
Outcome:	The proposed approach improves translation quality over strong NAT baselines on representative datasets.

Rhombus: Incentivizing Coordination in Parallel Thinking through Reinforcement Learning (2026.findings-acl)

Copied to clipboard

Ziyuan Nan, Qi Yi, Di Huang, Yutong Wu, Guanhua Huang, Xue Gong, Kejiao Li, Yuhao Jiang, Chenchen Zhang, Zenan Xu, Xing Hu, Bo Zhou

WECA: A WordNet-Encoded Collocation-Attention Network for Homographic Pun Recognition (D18-1)

Challenge:	Parallel thinking is a promising avenue for scaling test-time compute in Large Language Models . however, coordinating the exploration and aggregation stages remains challenging .
Approach:	They propose a parallel thinking framework that explicitly incentivizes coordination between components via end-to-end reinforcement learning.
Outcome:	The proposed framework improves accuracy by 6.0% over long chain-of-thought baselines while reducing wall-clock latency by 39.4% under matched token budgets.

Copied to clipboard

Yufeng Diao, Hongfei Lin, Di Wu, Liang Yang, Kan Xu, Zhihao Yang, Jian Wang, Shaowu Zhang, Bo Xu, Dongyu Zhang