Papers by Jing Ma

SHARP: Unlocking Interactive Hallucination via Stance Transfer in Role-Playing LLMs (2025.findings-acl)

Challenge:	Social media has become a fertile ground for nurturing rumors and misinformation due to its lack of systematic moderation.
Approach:	They propose a framework to enhance the joint predictive capabilities of LLMs for stance detection and rumor verification tasks.
Outcome:	The proposed framework outperforms state-of-the-art methods and generalizes to non-LLMs accommodated as task models.

Copied to clipboard

Chuyi Kong, Ziyang Luo, Hongzhan Lin, Zhiyuan Fan, Yaxin Fan, Yuxi Sun, Jing Ma

Reasoning

Challenge:	Existing studies on social interactions neglect hallucination while struggling with poor generalizability and implicit character fidelity judgments.
Approach:	They propose a generalizable and explicit paradigm for uncovering interactive patterns of Large Language Models across diverse worldviews by defining interactive hallucination through stance transfer and SHARP, a benchmark built by extracting relations from commonsense knowledge graphs.
Outcome:	The proposed paradigm is generalizable and explicit and demonstrates its effectiveness and stability.

AnswerFact: Fact Checking in Product Question Answering (2020.emnlp-main)

Copied to clipboard

Wenxuan Zhang, Yang Deng, Jing Ma, Wai Lam

Question Answering

Challenge:	a product-related community question answering platform is widely employed in many E-commerce sites . however, the misinformation in the answers on those platforms poses unprecedented challenges for users to obtain reliable and truthful product information.
Approach:	They propose a large scale fact checking dataset from product question answering forums to predict the answer veracity . each answer is accompanied by its veraity label and associated evidence sentences .
Outcome:	The proposed model outperforms baselines on the question veracity prediction task.

Few-shot Named Entity Recognition with Entity-level Prototypical Network Enhanced by Dispersedly Distributed Prototypes (2022.coling-1)

Copied to clipboard

Bin Ji, Shasha Li, Shaoduo Gan, Jie Yu, Jun Ma, Huijun Liu, Jing Yang

Chain of Thought with Explicit Evidence Reasoning for Few-shot Relation Extraction (2023.findings-emnlp)

Challenge:	Existing prototypical networks for named entity recognition suffer from label dependency and tightly distributed prototypes, thus causing misclassifications.
Approach:	They propose an Entity-level Prototypical Network enhanced by dispersedly distributed prototypes to build entity-level prototypes and distribute them dispersionally.
Outcome:	The proposed system outperforms the previous models on two evaluation tasks and the Few-NERD settings in terms of overall performance.

Copied to clipboard

Xilai Ma, Jing Li, Min Zhang

Low-Resource NLP

Challenge:	Existing approaches to few-shot relation extraction require training.
Approach:	They propose a method for few-shot relation extraction using large language models, called CoT-ER, chain-of-thought with explicit evidence reasoning.
Outcome:	The proposed approach achieves competitive performance compared to the fully-supervised state-of-the-art approach on the FewRel1.0 and FewRela2.0 datasets.

Beneath the Surface: Unveiling Harmful Memes with Multimodal Reasoning Distilled from Large Language Models (2023.findings-emnlp)

Copied to clipboard

Hongzhan Lin, Ziyang Luo, Jing Ma, Long Chen

HYRR: Hybrid Infused Reranking for Passage Retrieval (2024.lrec-main)

Challenge:	Existing methods for harmful meme detection ignore in-depth cognition of meme text and image . authors propose a framework for learning reasonable thoughts from LLMs for better multimodal fusion .
Approach:	They propose to use large language models to learn reasonable thoughts from LLMs for better multimodal fusion and lightweight fine-tuning.
Outcome:	The proposed approach achieves superior performance than state-of-the-art methods on the harmful meme detection task.

Copied to clipboard

Jing Lu, Keith Hall, Ji Ma, Jianmo Ni

CofiPara: A Coarse-to-fine Paradigm for Multimodal Sarcasm Target Identification with Large Multimodal Models (2024.acl-long)

Challenge:	Existing passage retrieval systems typically adopt a two-stage retrieve-then-rerank pipeline.
Approach:	They propose a framework for training robust reranking models using hybrid retrievers . they propose HYRR framework that allows users to select training data using hybrids .
Outcome:	The proposed framework is robust to different first-stage retrieval settings.

Copied to clipboard

Zixin Chen, Hongzhan Lin, Ziyang Luo, Mingfei Cheng, Jing Ma, Guang Chen

Sentiment Analysis

Challenge:	Current methods for multimodal sarcasm target identification focus on superficial indicators in an end-to-end manner, overlooking the nuanced understanding of multimodal content.
Approach:	They propose a multimodal sarcasm target identification framework with a coarse-to-fine paradigm by augmenting sarcasm explainability with reasoning and pre-training knowledge.
Outcome:	The proposed framework outperforms state-of-the-art methods and exhibits explainability in deciphering sarcasm as well.

Conditioned Masked Language and Image Modeling for Image-Text Dense Retrieval (2022.findings-emnlp)

Copied to clipboard

Ziyang Luo, Yadong Xi, Rongsheng Zhang, GongZheng Li, Zeng Zhao, Jing Ma

LED-Merging: Mitigating Safety-Utility Conflicts in Model Merging with Location-Election-Disjoint (2025.acl-long)

Challenge:	Large-scale two-stream pre-trained models like CLIP have achieved tremendous success in image-text retrieval.
Approach:	They propose a cross-modal framework for image-text retrieval using two-stream pre-trained models . they embed images and texts into instance representations with two separate encoders . experimental results on MSCOCO and Flickr30k reveal the effectiveness of their framework .
Outcome:	The proposed framework improves image-text retrieval performance on two popular cross-modal retrieval benchmarks.

Copied to clipboard

Qianli Ma, Dongrui Liu, Qian Chen, Linfeng Zhang, Jing Shao

Challenge:	Existing methods for fine-tuning large language models for specialized tasks are costly and time-consuming.
Approach:	They propose a framework that locates task-specific neurons via gradient-based attribution and dynamically Elects critical neurons through multi-model importance fusion.
Outcome:	The proposed framework reduces harmful response rates while preserving 95% of utility performance.

Knowledge-Augmented Multimodal Clinical Rationale Generation for Disease Diagnosis with Small Language Models (2025.acl-long)

Copied to clipboard

Shuai Niu, Jing Ma, Hongzhan Lin, Liang Bai, Zhihua Wang, Richard Yi Da Xu, Yunya Song, Xian Yang

CodeJudge-Eval: Can Large Language Models be Good Judges in Code Understanding? (2025.coling-main)

Challenge:	Existing models struggle to balance predictive accuracy with human-understandable rationales.
Approach:	They propose to enhance LLMs by leveraging rationale distillation and domain knowledge injection for trustworthy multimodal rationale generation.
Outcome:	Experiments on real-world medical datasets show that ClinRaGen achieves state-of-the-art performance in disease diagnosis and rationale generation.

Copied to clipboard

Yuwei Zhao, Ziyang Luo, Yuchen Tian, Hongzhan Lin, Weixiang Yan, Annan Li, Jing Ma

VRoPE: Rotary Position Embedding for Video Large Language Models (2025.emnlp-main)

Challenge:	Recent advances in large language models (LLMs) have showcased impressive code generation capabilities, primarily evaluated through language-to-code benchmarks.
Approach:	They propose a benchmark to assess LLMs’ code understanding abilities from the perspective of code judging rather than code generation.
Outcome:	The proposed benchmark evaluates 12 well-known large language models to determine the correctness of provided code solutions.

Copied to clipboard

Zikang Liu, Longteng Guo, Yepeng Tang, Tongtian Yue, Junxian Cai, Kai Ma, Qingbin Liu, Xi Chen, Jing Liu

FACT-E: Causality-Inspired Evaluation for Trustworthy Chain-of-Thought Reasoning (2026.findings-acl)

Challenge:	Existing versions of Large Language Models (LLMs) lack a positional encoding strategy for video.
Approach:	They propose a new positional encoding method tailored for Video-LLMs that mitigates positional biases and ensures a more uniform distribution of spatial focus.
Outcome:	The proposed method outperforms existing versions of RoPE in video understanding and reasoning tasks.

Copied to clipboard

Yuxi Sun, Aoqi Zuo, Haotian Xie, Wei Gao, Mingming Gong, Jing Ma

FAITH: Factuality Alignment through Integrating Trustworthiness and Honestness (2026.findings-acl)

Challenge:	Existing models generate explanations that appear coherent while containing unfaithful intermediate steps.
Approach:	They propose a causality-inspired framework for evaluating CoT quality using controlled perturbations as an instrumental signal to separate genuine step-to-step dependence from bias-driven artifacts.
Outcome:	Experiments on GSM8K, MATH, and CommonsenseQA show that FACT-E improves reasoning-trajectory selection and yields stronger in-context learning exemplars.

Copied to clipboard

Xiaoning Dong, Chengyan Wu, Yajie Wen, Yu Chen, Yun Xue, Zhang Jing, Wei Xu, Bolei Ma

FACT-AUDIT: An Adaptive Multi-Agent Framework for Dynamic Fact-Checking Evaluation of Large Language Models (2025.acl-long)

Challenge:	Existing approaches to correct factually inaccurate outputs are lacking the semantic richness needed to properly understand its internal states of trustworthiness and honesty.
Approach:	They propose a framework for factuality alignment that integrates natural-language uncertainty signals with external knowledge and computes confidence scores and semantic entropy from LLM outputs.
Outcome:	Extensive experiments on four knowledge-intensive benchmarks show that FAITH improves the factual accuracy and truthfulness of Large Language Models (LLMs).

Copied to clipboard

Hongzhan Lin, Yang Deng, Yuxuan Gu, Wenxuan Zhang, Jing Ma, See-Kiong Ng, Tat-Seng Chua

CausalRAG: Integrating Causal Graphs into Retrieval-Augmented Generation (2025.findings-acl)

Challenge:	Existing fact-checking evaluation methods rely on static datasets and classification metrics, which fail to evaluate justification production and uncover the nuanced limitations of LLMs.
Approach:	They propose a framework that adaptively and dynamically assesses LLMs’ fact-checking capabilities by incorporating justification production alongside verdict prediction.
Outcome:	Experiments show that the framework differentiates among state-of-the-art LLMs, providing valuable insights into model strengths and limitations in model-centric fact-checking analysis.

Copied to clipboard

Nengbo Wang, Xiaotian Han, Jagdip Singh, Jing Ma, Vipin Chaudhary

AdamMeme: Adaptively Probe the Reasoning Capacity of Multimodal Large Language Models on Harmfulness (2025.acl-long)

Challenge:	Existing RAG frameworks face critical limitations due to text chunking and semantic similarity.
Approach:	They propose a framework that incorporates causal graphs into the retrieval process.
Outcome:	The proposed framework preserves contextual continuity and improves retrieval precision, leading to more accurate and interpretable responses.

Copied to clipboard

Zixin Chen, Hongzhan Lin, Kaixin Li, Ziyang Luo, Zhen Ye, Guang Chen, Zhiyong Huang, Jing Ma

DecBERT: Enhancing the Language Understanding of BERT with Causal Attention Masks (2022.findings-naacl)

Challenge:	Existing models that assess mLLMs on harmful meme understanding are inaccurate and lack accuracy.
Approach:	They propose a framework that adaptively probes the reasoning capabilities of mLLMs . their framework systematically reveals the varying performance of different target mllms a .
Outcome:	The proposed framework systematically reveals the performance of different target mLLMs.

Copied to clipboard

Ziyang Luo, Yadong Xi, Jing Ma, Zhiwei Yang, Xiaoxi Mao, Changjie Fan, Rongsheng Zhang

M3-VQA: A Benchmark for Multimodal, Multi-Entity, Multi-Hop Visual Question Answering (2026.acl-long)

Challenge:	Experimental results show that Transformer Encoder model can't automatically capture word order, so explicit position embeddings are required to be fed into the target model.
Approach:	They propose a Transformer-based language model DecBERT that uses a causal attention mask to capture word order.
Outcome:	The proposed model improves on the GLUE language understanding benchmark and accelerates the pre-training process.

Copied to clipboard

Jiatong Ma, Longteng Guo, Yuchen Liu, Zijia Zhao, Dongze Hao, Xuanxu Lin, Jing Liu

SciVQR: A Multidisciplinary Multimodal Benchmark for Advanced Scientific Reasoning Evaluation (2026.findings-acl)

Challenge:	Existing knowledge-based VQA benchmarks focus on coarse-grained categories and simple reasoning over single entities.
Approach:	They propose a knowledge-based Visual Question Answering benchmark to enhance multimodality evaluation.
Outcome:	The proposed benchmark improves evaluation of multimodal large language models in fine-grained multimodal entity understanding and complex multihop reasoning.

Copied to clipboard

Longteng Guo, Xuanxu Lin, Dongze Hao, Tongtian Yue, Pengkang Huo, Jiatong Ma, Yuchen Liu, Jing Liu

WSDMS: Debunk Fake News via Weakly Supervised Detection of Misinforming Sentences with Contextualized Social Wisdom (2023.emnlp-main)

Challenge:	Existing benchmarks for multimodal large language models fail to capture complexity and traceability of reasoning processes . SciVQR includes domain-specific visuals and challenges models to combine visual comprehension with reasoning.
Approach:	They propose a multimodal benchmark for scientific reasoning covering 54 subfields . SciVQR includes domain-specific visuals and challenges models to combine visual comprehension with reasoning .
Outcome:	SciVQR evaluates 54 subfields in mathematics, physics, chemistry, geography, astronomy, and biology . the results highlight the need for improved multi-step reasoning and integration of interdisciplinary knowledge .

Copied to clipboard

Ruichao Yang, Wei Gao, Jing Ma, Hongzhan Lin, Zhiwei Yang

Sentence-Level Evidence Embedding for Claim Verification with Hierarchical Attention Networks (P19-1)

Challenge:	Existing methods for debunking fake news rely on blending of authentic and fabricated content by creators.
Approach:	They propose a model that detects misinformation at sentence-level using social media conversations . they use a bag-level annotation system to train the model .
Outcome:	The proposed model outperforms existing state-of-the-art models on three real-world benchmarks and outperformed existing state of the art models in debunking fake news at sentence and article levels.

Copied to clipboard

Jing Ma, Wei Gao, Shafiq Joty, Kam-Fai Wong

Dialectical Structured Reasoning for Explainable Multimodal Fake News Detection (2026.findings-acl)

Challenge:	Claim verification is cumbersome and inefficient for human fact-checkers to find consistent pieces of evidence.
Approach:	They propose an end-to-end hierarchical attention network that learns to represent coherent evidence and their semantic relatedness with the claim.
Outcome:	The proposed model outperforms state-of-the-art models on three datasets . it is based on a coherence-based attention layer and entailment-based one .

Copied to clipboard

Ruichao Yang, Yufan Bian, Wei Gao, Bo-Wen Zhang, Jing Ma, Hongzhan Lin, Ziyang Luo, Xiaobin Zhu, Xu-Cheng Yin

From Conversation to Evaluation: Benchmarking LLMs on Development Knowledge via SimpleDevQA (2026.findings-acl)

Challenge:	Existing fake news detection models are opaque and lack deductive transparency . a framework for dialectical structured reasoning is proposed to address this limitation .
Approach:	They propose a framework that model fake news detection as an explicit dialectical process over multimodal social context.
Outcome:	The proposed framework achieves state-of-the-art while producing transparent explanations that mirror human reasoning process.

Copied to clipboard

Jing Zhang, Lianghong Guo, Yanlin Wang, Terry Yue Zhuo, Yong Wang, Mingwei Liu, Jiachi Chen, Ensheng Shi, Yuchi Ma, Hongyu Zhang, Zibin Zheng

Natural Language Interfaces

Challenge:	Existing Dev Knowledge QA benchmarks are limited in development knowledge scope and often not built from real user queries.
Approach:	They conduct preliminary analysis of real user–LLM dialogues from WildChat to investigate the importance of Dev Knowledge QA in AI-assisted software development scenarios.
Outcome:	The proposed benchmark is based on real user–LLM dialogues from WildChat.

REFLEX: Self-Refining Explainable Fact-Checking via Verdict-Anchored Style Control (2026.acl-long)

Copied to clipboard

Chuyi Kong, Wei Gao, Jing Ma, Hongzhan Lin, Yuxi Sun

KAPALM: Knowledge grAPh enhAnced Language Models for Fake News Detection (2023.findings-emnlp)

Challenge:	Existing methods for automated fact-checking often overlook deceptive misinformation styles in generated explanations.
Approach:	They propose a framework that explicitly controls reasoning style by anchoring explanations to the predicted verdict.
Outcome:	The proposed framework achieves state-of-the-art under LLaMA-series models with 465 samples.

Copied to clipboard

Jing Ma, Chen Chen, Chunyan Hou, Xiaojie Yuan

MemeArena: Automating Context-Aware Unbiased Evaluation of Harmfulness Understanding for Multimodal Large Language Models (2025.emnlp-main)

Challenge:	Existing methods of fake news detection focus on news entity information and ignore structured knowledge among news entities.
Approach:	They propose a model that fuses coarse- and fine-grained representations of entity knowledge from Knowledge Graphs (KGs) they identify entities in news content and link them to entities in KGs.
Outcome:	The proposed model outperforms state-of-the-art models on two benchmark datasets and is competitive in the few-shot scenario.

Copied to clipboard

Zixin Chen, Hongzhan Lin, Kaixin Li, Ziyang Luo, Yayue Deng, Jing Ma

Knowledge-Guided Paraphrase Identification (2021.findings-emnlp)

Challenge:	Existing evaluation approaches focus on mLLMs’ detection accuracy for binary classification tasks, which often fail to reflect the in-depth interpretive nuance of harmfulness across diverse contexts.
Approach:	They propose an agent-based arena-style evaluation framework that provides context-aware and unbiased assessment for mLLMs’ understanding of multimodal harmfulness.
Outcome:	The proposed framework reduces evaluation biases of judge agents and provides unbiased comparisons of mLLMs’ abilities to interpret multimodal harmfulness.

Copied to clipboard

Haoyu Wang, Fenglong Ma, Yaqing Wang, Jing Gao

Text Generation

Challenge:	Existing methods for paraphrase identification (PI) are limited due to lack of professional knowledge.
Approach:	They propose to leverage Wikipedia knowledge to accurately identify paraphrases by mining outline knowledge of given sentences from Wikipedia.
Outcome:	The proposed framework outperforms state-of-the-art models on two public datasets: PARADE and clinicalSTS2019.

HiTRANS: A Hierarchical Transformer Network for Nested Named Entity Recognition (2021.findings-emnlp)

Copied to clipboard

Zhiwei Yang, Jing Ma, Hechang Chen, Yunke Zhang, Yi Chang

FFAEval: Evaluating Dialogue System via Free-For-All Ranking (2023.findings-emnlp)

Challenge:	Existing studies for named entity recognition focus on flat NER, i.e., without nested entities, by sequence labeling methods.
Approach:	They propose a Hierarchical Transformer network which decomposes the input sentence into multi-grained spans and enhances the representation learning in a hierarchical manner.
Outcome:	The proposed method achieves much better performance than the state-of-the-art approaches on GENIA, ACE-2004, ace-2005 and NNE datasets.

Copied to clipboard

Zeyao Ma, Zijun Yao, Jing Zhang, Jifan Yu, Xiaohan Zhang, Juanzi Li, Jie Tang

Natural Language Interfaces

Challenge:	Existing evaluation metrics for open-domain dialogue systems show poor correlation with human assessment.
Approach:	They propose a free-for-all human evaluation framework that shares dialogue history with annotators for multi-turn scoring.
Outcome:	The proposed framework achieves a strong correlation with human assessment on English and Chinese dialogue systems.

A Coarse-to-fine Cascaded Evidence-Distillation Neural Network for Explainable Fake News Detection (2022.coling-1)

Copied to clipboard

Zhiwei Yang, Jing Ma, Hechang Chen, Hongzhan Lin, Ziyang Luo, Yi Chang

MMEKG: Multi-modal Event Knowledge Graph towards Universal Representation across Modalities (2022.acl-demo)

Challenge:	Existing methods for fake news detection focus on fact-checked reports, resulting in limited coverage and debunking delays.
Approach:	They propose a Coarse-to-fine Cascaded Evidence-Distillation neural network for explainable fake news detection based on raw reports . they use hierarchical encoders and cascaded selectors to select most explainable sentences for verdicts on top of selected top-K reports based upon raw reports.
Outcome:	The proposed model outperforms baseline detection methods and generates high-quality explanations from diverse evaluation perspectives.

Copied to clipboard

Yubo Ma, Zehao Wang, Mukai Li, Yixin Cao, Meiqi Chen, Xinze Li, Wenqi Sun, Kunquan Deng, Kun Wang, Aixin Sun, Jing Shao

Natural-Language Policies to Executable Decisions: An Interpretable Large Language Model Framework (2026.acl-industry)

Challenge:	Recent Knowledge Graphs (KGs) store billions of world facts in a directed graph, but expression ability of such entity-centric KGs is limited.
Approach:	They propose a large-scale multi-modal event knowledge graph named MMEKG that unifies different modalities of knowledge via events.
Outcome:	The proposed system unifies different modalities of knowledge via events, which complement and disambiguate each other.

Copied to clipboard

Ziqiang Zhang, Jing Ma, Zilong Wang, Jiayuan Chen, Yi Qiao, Yu He, Wei Zhang, Dai Cheng, Xiaoyu Shen

Rumor Detection on Twitter with Tree-structured Recursive Neural Networks (P18-1)

Challenge:	a production-grade pricing system for tourism is challenging due to unstructured nature of travel orders and ever-evolving pricing policies.
Approach:	They propose a production-grade pricing system with a strict decision boundary . they propose to combine structured extraction and bounded policy/path selection with interpretable condition trees .
Outcome:	The proposed system processed 3,960 orders in six months and reduced the order management team from 15-20 to 3 . the system reduced the per-order handling time from 10 minutes to 2 minutes.

Copied to clipboard

Jing Ma, Wei Gao, Kam-Fai Wong

Challenge:	Existing methods for detecting rumors are difficult to implement and require a lot of effort.
Approach:	They propose two recursive neural models that follow tweets' propagation layouts to learn discriminative features from tweets and generate more powerful representations for rumors detection.
Outcome:	The proposed models perform better than state-of-the-art approaches on two public Twitter datasets and show superior performance on detecting rumors at very early stage.

Answer-focused and Position-aware Neural Question Generation (D18-1)

Copied to clipboard

Xingwu Sun, Jing Liu, Yajuan Lyu, Wei He, Yanjun Ma, Shi Wang

Text Generation

Challenge:	Recent neural network-based approaches generate interrogative words that do not match the answer type.
Approach:	They propose an answer-focused and position-aware neural question generation model to address these issues.
Outcome:	The proposed model outperforms the baseline and outperformed the state-of-the-art system.

Personalizing LLMs with Binary Feedback: A Preference-Calibrated Optimization Framework (2026.acl-long)

Copied to clipboard

Xilai Ma, Liye Zhao, Weijun Yao, Haibing Di, Wenya Wang, Jing Li

Challenge:	Existing methods focus on isolated user histories, neglecting the essential role of inter-user differences.
Approach:	They propose a framework that personalizes Large Language Models via preference-calibrated binary signals.
Outcome:	The proposed framework outperforms baselines in a variety of personalization tasks and backbone LLMs.

Ranking Reasoning LLMs under Test-Time Scaling (2026.acl-long)

Copied to clipboard

Mohsen Hariri, Michael Hinczewski, Jing Ma, Vipin Chaudhary

Reasoning

Challenge:	Large language models (LLMs) are increasingly used as general-purpose reasoning systems for tasks such as programming and mathematical problem solving.
Approach:	They formalize dense benchmark ranking under test-time scaling and introduce a library that implements statistical ranking methods such as paired-comparison models, item response theory, voting rules, graph- and spectral-based methods.
Outcome:	The proposed method is based on paired-comparison models, item response theory (IRT) models, voting rules, graph- and spectral-based methods.

MM-CRITIC: A Holistic Evaluation of Large Multimodal Models as Multimodal Critique (2025.findings-emnlp)

Copied to clipboard

Gailun Zeng, Ziyang Luo, Hongzhan Lin, Yuchen Tian, Kaixin Li, Ziyang Gong, Jianxiong Guo, Jing Ma

Towards Low-Resource Harmful Meme Detection with LMM Agents (2024.emnlp-main)

Challenge:	e MM-CRITIC is a holistic benchmark for evaluating the critique ability of Large Multimodal Models (LMMs) covering 8 main task types and over 500 tasks, covering 4471 samples.
Approach:	They introduce a holistic benchmark for evaluating the critique ability of Large Multimodal Models across multiple dimensions: basic, correction, and comparison.
Outcome:	The proposed benchmark covers 8 main task types and over 500 tasks and is composed of 4471 samples.

Copied to clipboard

Jianzhao Huang, Hongzhan Lin, Liu Ziyan, Ziyang Luo, Guang Chen, Jing Ma

Dynamic Scaling of Unit Tests for Code Reward Modeling (2025.acl-long)

Challenge:	Existing methods for harmful meme detection are limited due to the dynamic nature of memes . eliciting knowledge-revising behavior within the LMM agent is a key factor in achieving this goal .
Approach:	They propose an agency-driven framework for low-resource harmful meme detection . they use annotated memes to leverage label information as auxiliary signals for model .
Outcome:	The proposed framework achieves superior performance than state-of-the-art methods on the low-resource harmful meme detection task.

Copied to clipboard

Zeyao Ma, Xiaokang Zhang, Jing Zhang, Jifan Yu, Sijia Luo, Jie Tang

Green & Sustainable NLP

Challenge:	Existing large language models struggle to produce accurate responses on the first attempt for complex reasoning tasks like code generation.
Approach:	They propose a lightweight yet effective unit test generator that scales unit tests based on problem difficulty.
Outcome:	The proposed approach significantly improves performance on three benchmarks.

AMR-Evol: Adaptive Modular Response Evolution Elicits Better Knowledge Distillation for Large Language Models in Code Generation (2024.emnlp-main)

Copied to clipboard

Ziyang Luo, Xin Li, Hongzhan Lin, Jing Ma, Lidong Bing

CPO: Addressing Reward Ambiguity in Role-playing Dialogue via Comparative Policy Optimization (2025.findings-emnlp)

Challenge:	proprietary large language models (LLMs) have demonstrated impressive code generation performance.
Approach:	They propose an adaptive module-based model that refines the direct response distillation process by modular decomposition and adaptive response evolution.
Outcome:	The proposed framework outperforms baseline model and code generation methods on three popular benchmarks.

Copied to clipboard

Jing Ye, Rui Wang, Yuchuan Wu, Victor Ma, Feiteng Fang, Fei Huang, Yongbin Li

Dialogue Systems & Conversational Agents

Challenge:	Comparative Policy Optimization (CPO) redefines the reward evaluation paradigm by shifting from sample-wise scoring to comparative group-wise score.
Approach:	They propose a method to optimize subjective tasks by shifting from sample-wise to comparative group-wise scoring.
Outcome:	The proposed framework shifts from sample-wise scoring to comparative group-wise score . it minimizes contextual bias and enables more robust and fair performance evaluation.

LLMs know their vulnerabilities: Uncover Safety Gaps through Natural Distribution Shifts (2025.acl-long)

Copied to clipboard

Qibing Ren, Hao Li, Dongrui Liu, Zhanxu Xie, Xiaoya Lu, Yu Qiao, Lei Sha, Junchi Yan, Lizhuang Ma, Jing Shao

Rumor Detection on Twitter with Claim-Guided Hierarchical Graph Attention Networks (2021.emnlp-main)

Challenge:	Current safety training focuses on teaching models to reject harmful queries, but recent research shows that adversarial attacks or jailbreak methods bypass these safety mechanisms.
Approach:	They propose to use a new attack method to craft multi-turn toxic prompts that gradually lead LLMs to reveal unsafe content.
Outcome:	The proposed method outperforms existing methods in diversity, effectiveness, and efficiency across aligned LLMs.

Copied to clipboard

Hongzhan Lin, Jing Ma, Mingfei Cheng, Zhiwei Yang, Liangliang Chen, Guang Chen

Structured Data in NLP

Challenge:	Existing methods for rumor detection are limited to the strict relation of user responses or oversimplify the conversation structure.
Approach:	They propose a method that reinforces interaction of user opinions while reducing negative impact imposed by irrelevant posts.
Outcome:	The proposed method improves performance on three Twitter datasets and can detect rumors at early stages.

Large Dual Encoders Are Generalizable Retrievers (2022.emnlp-main)

Copied to clipboard

Jianmo Ni, Chen Qu, Jing Lu, Zhuyun Dai, Gustavo Hernandez Abrego, Ji Ma, Vincent Zhao, Yi Luan, Keith Hall, Ming-Wei Chang, Yinfei Yang

DiffCoT: Diffusion-styled Chain-of-Thought Reasoning in LLMs (2026.findings-acl)

Challenge:	Experimental results show that dual encoders outperform sparse and dense retrievers on the BEIR dataset significantly.
Approach:	They challenge belief that bottleneck layer is too limited for out-of-domain generalization . they scale up the model while keeping bottleneck as a single dot-product with a fixed size .
Outcome:	The proposed model outperforms sparse and dense retrievers on the BEIR dataset significantly.

Copied to clipboard

Shidong Cao, Hongzhan Lin, Yuxuan Gu, Ziyang Luo, Jing Ma

Reasoning

Challenge:	Chain-of-Thought (CoT) reasoning improves multi-step mathematical problem solving in large language models but is vulnerable to exposure bias and error accumulation.
Approach:	They propose a diffusion-styled CoT framework that reformulates CoT reasoning as an iterative denoising process.
Outcome:	The proposed framework outperforms existing methods on three multi-step CoT reasoning benchmarks.

Subgraph-Guided Executable Logical Form Generation for Knowledge Base Question Answering (2026.findings-acl)

Copied to clipboard

Yuhang Tian, Dandan Song, Zhijing Wu, Changzhi Zhou, Jun Yang, Huipeng Ma, Chenhao Li, Luan Zhang, Yading Li, Xudong Li, Shenxi Liu, Jing Jiang

Natural Language Interfaces

Challenge:	Existing retrieval-augmented approaches focus on ignoring the structural information of the Knowledge Base (KB) and the question.
Approach:	They propose a structure-aware subgraph retrieval stage that ranks candidate subgraphs by aligning them with the question’s structure, along with semantic relevance.
Outcome:	Experiments on GrailQA, WebQSP, and GraphQuestions show that the proposed framework achieves state-of-the-art performance.

CodeAttack: Revealing Safety Generalization Challenges of Large Language Models via Code Completion (2024.findings-acl)

Copied to clipboard

Qibing Ren, Chang Gao, Jing Shao, Junchi Yan, Xin Tan, Wai Lam, Lizhuang Ma

Programming Languages in NLP

Challenge:	Large Language Models exhibit remarkable generative capabilities but can be misused for harmful purposes.
Approach:	They propose a framework that transforms natural language inputs into code inputs.
Outcome:	The proposed framework bypasses the safety guardrails of all models more than 80% of the time.

Seeing Culture: A Benchmark for Visual Reasoning and Grounding (2025.emnlp-main)

Copied to clipboard

Burak Satar, Zhixin Ma, Patrick Amadeus Irawan, Wilfried Ariel Mulyawan, Jing Jiang, Ee-Peng Lim, Chong-Wah Ngo

Prompt for Extraction? PAIE: Prompting Argument Interaction for Event Argument Extraction (2022.acl-long)

Challenge:	Multimodal vision-language models (VLMs) have made significant progress in cultural understanding tasks . but these datasets often fall short of providing cultural reasoning while underrepresenting many cultures.
Approach:	They propose a Seeing Culture Benchmark that requires VLMs to reason on culturally rich images in two stages.
Outcome:	The proposed approach requires VLMs to reason on culturally rich images in two stages . the Seeing Culture Benchmark identifies cultural reasoning shortcomings in multimodal models .

Copied to clipboard

Yubo Ma, Zehao Wang, Yixin Cao, Mukai Li, Meiqi Chen, Kun Wang, Jing Shao

Explicit and Implicit Data Augmentation for Social Event Detection (2025.acl-long)

Challenge:	Using a prompt-based model, we find that event argument extraction is efficient and generalized well to few-shot settings.
Approach:	They propose a model PAIE for event argument extraction using prompt tuning for extractive objectives.
Outcome:	The proposed model can extract arguments with the same role instead of heuristic threshold tuning.

Copied to clipboard

Congbo Ma, Yuxia Wang, Jia Wu, Jian Yang, Jing Du, Zitai Qiu, Qing Li, Hu Wang, Preslav Nakov

TableLLM: Enabling Tabular Data Manipulation by LLMs in Real Office Usage Scenarios (2025.findings-acl)

Challenge:	Social event detection relies on labeled data, but annotation is costly and labor-intensive.
Approach:	They propose a plug-and-play dual augmentation framework that combines explicit text-based and implicit feature-space augmentation to enhance data diversity and model robustness.
Outcome:	The proposed framework outperforms the best baseline model by 17.67% on the Twitter2012 dataset and 15.57% on the twitter2018 dataset in terms of the average F1 score.

Copied to clipboard

Xiaokang Zhang, Sijia Luo, Bohan Zhang, Zeyao Ma, Jing Zhang, Yang Li, Guanlin Li, Zijun Yao, Kangli Xu, Jinchang Zhou, Daniel Zhang-Li, Jifan Yu, Shu Zhao, Juanzi Li, Jie Tang

Debunking Rumors on Twitter with Tree Transformer (2020.coling-main)

Challenge:	TableLLM is a robust large language model capable of handling tabular data manipulation tasks.
Approach:	They propose a distant supervision method for training which includes a reasoning process extension strategy and a cross-way validation strategy.
Outcome:	The proposed model has 8 billion parameters and is capable of handling tabular data tasks.

Copied to clipboard

Jing Ma, Wei Gao

From Storage to Experience: A Survey on the Evolution of LLM Agent Memory Mechanisms (2026.findings-acl)

Challenge:	Existing methods for rumor detection follow tree edges or treat all posts fully-connected during feature learning.
Approach:	They propose a new rumor detection model based on tree transformer to better utilize user interactions in the dialogue . they propose to use post-level self-attention to aggregate the intra-/inter-subtree stances .
Outcome:	The proposed model improves rumor detection performance on social media conversations . it is based on a conversation tree that encodes important information indicative of credibility .

Copied to clipboard

Jinghao Luo, Yuchen Tian, Chuxue Cao, Ziyang Luo, Hongzhan Lin, Kaixin Li, Chuyi Kong, Ruichao Yang, Jing Ma

Challenge:	Large Language Models (LLMs)-based agents have fundamentally reshaped artificial intelligence . however, the inherent statelessness of LLMs hinders their ability to maintain logical consistency across complex, multi-step tasks .
Approach:	They propose a framework for LLM agent memory mechanisms that formalizes the development process into three stages: storage, reflection, and experience.
Outcome:	The proposed framework breaks the development process into three stages . it analyzes the need for long-range consistency, challenges in dynamic environments, and the ultimate goal of continual learning.

ProMedTS: A Self-Supervised, Prompt-Guided Multimodal Approach for Integrating Medical Text and Time Series (2025.findings-acl)

Copied to clipboard

Shuai Niu, Jing Ma, Hongzhan Lin, Liang Bai, Zhihua Wang, V. W., Richard Yi Da Xu, Guo Li, Xian Yang

Stand on The Shoulders of Giants: Building JailExpert from Previous Attack Experience (2025.emnlp-main)

Challenge:	Large language models excel at processing unstructured data, but integrating time series data with text remains a challenge.
Approach:	They propose a self-supervised multimodal framework that uses prompt-guided learning to unify heterogeneous data types.
Outcome:	The proposed framework outperforms state-of-the-art approaches on disease diagnosis tasks using real-world datasets.

Copied to clipboard

Xi Wang, Songlei Jian, Shasha Li, Xiaopeng Li, Bin Ji, Ma Jun, Xiaodong Liu, Jing Wang, Jianfeng Zhang, Jie Yu, Feilong Bao, null Wangbaosheng

GraphICL: Unlocking Graph Learning Potential in LLMs through Structured Prompt Design (2025.findings-naacl)

Challenge:	Existing methods to generate human-aligned content with a “jailbreak prompt” are inefficient and repetitive, causing inefficiency and a lack of experience.
Approach:	They propose a framework that integrates past attack experiences to aid current jailbreak attempts.
Outcome:	The proposed framework improves both attack effectiveness and efficiency compared to the current black-box jailbreak method.

Copied to clipboard

Yuanfu Sun, Zhengnan Ma, Yi Fang, Jing Ma, Qiaoyu Tan

Variance-reduced First-order Meta-learning for Natural Language Processing Tasks (2021.naacl-main)

Challenge:	Text-Attributed Graphs (TAGs) are a powerful tool for understanding complex systems and relationships.
Approach:	They propose a benchmark to evaluate large language models for graph-structured data using prompts.
Outcome:	The proposed benchmark outperforms state-of-the-art graph LLMs and graph neural networks on graph learning tasks without training.

Copied to clipboard

Lingxiao Wang, Kevin Huang, Tengyu Ma, Quanquan Gu, Jing Huang

Vocabulary Hijacking in LVLMs: Unveiling Critical Attention Heads by Excluding Inert Tokens to Mitigate Hallucination (2026.acl-long)

Challenge:	Existing studies show that meta-learning can overfit to some specific adaptation when we have heterogeneous tasks.
Approach:	They propose to reduce the variance of the gradient estimator used in task adaptation by adding a new variance reduction term to the gradient estimation.
Outcome:	Experiments on few-shot text classification and multi-domain dialog state tracking show that the proposed method outperforms existing methods.

Copied to clipboard

Yangneng Chen, Junlin Li, Weijun Yao, Xilai Ma, Guodong DU, Wenya Wang, Jing Li

From Completion to Editing: Unlocking Context-Aware Code Infilling via Search-and-Replace Instruction Tuning (2026.acl-long)

Challenge:	Large Vision-Language Models (LVLMs) are capable of processing visual inputs, but are susceptible to hallucinations.
Approach:	They propose a method to localize and localize specific visual tokens, which are defined as Inert Tokens, across layers, revealing a rigid semantic collapse.
Outcome:	The proposed approach reduces the likelihood of LVLMs being hijacked by visual inputs while maintaining general capabilities.

Copied to clipboard

Jiajun Zhang, Zeyu Cui, Jiaxi Yang, Lei Zhang, Yuheng Jing, Zeyao Ma, Tianyi Bai, Zilei Wang, Qiang Liu, Liang Wang, Binyuan Hui, Junyang Lin

Programming Languages in NLP

Challenge:	Fill-in-the-Middle (FIM) models suffer from performance degradation and prohibitive latency.
Approach:	They propose a search-and-replace infilling framework that integrates agentic verification and editing into a single-pass inference process.
Outcome:	The proposed framework harmonizes completion tasks with the instruction-following priors of Chat LLMs, extending the paradigm from static infilling to dynamic context-aware editing.

ScratchEval: Are GPT-4o Smarter than My Child? Evaluating Large Multimodal Models with Visual Programming Challenges (2025.naacl-short)

Copied to clipboard

Rao Fu, Ziyang Luo, Hongzhan Lin, Zhen Ye, Jing Ma

MMCode: Benchmarking Multimodal Large Language Models for Code Generation with Visually Rich Programming Problems (2024.findings-emnlp)

Challenge:	Recent advances in large multimodal models (LMMs) have demonstrated impressive code generation capabilities, primarily evaluated through image-to-code benchmarks.
Approach:	They propose a visual programming reasoning benchmark based on Scratch, a block-based visual programming language widely used in children’s programming education.
Outcome:	The proposed framework evaluates the visual programming ability of large multimodal models by integrating visual elements and embedded programming logic.

Copied to clipboard

Kaixin Li, Yuchen Tian, Qisheng Hu, Ziyang Luo, Zhiyong Huang, Jing Ma

CausalAbstain: Enhancing Multilingual LLMs with Causal Reasoning for Trustworthy Abstention (2025.findings-acl)

Challenge:	Programming often involves translating detailed and complex specifications into code . current state-of-the-art models struggle to solve these problems, a new study shows .
Approach:	They propose a multi-modal coding dataset to evaluate algorithmic problem-solving skills in visually rich contexts.
Outcome:	The proposed model lacks powerful vision-code models due to the extreme demand for reasoning abilities.

Copied to clipboard

Yuxi Sun, Aoqi Zuo, Wei Gao, Jing Ma

Detect Rumors in Microblog Posts for Low-Resource Domains via Adversarial Contrastive Learning (2022.findings-naacl)

Challenge:	Existing methods to reduce hallucinations in large language models are inaccurate and inaccuracies in the generated feedback.
Approach:	They propose a method that helps LLMs determine whether to utilize multiple generated feedback responses and how to identify the most useful ones.
Outcome:	Extensive experiments show that the proposed method outperforms baselines on encyclopedic and commonsense knowledge QA tasks.

Copied to clipboard

Hongzhan Lin, Jing Ma, Liangliang Chen, Zhiwei Yang, Mingfei Cheng, Chen Guang

ED2LM: Encoder-Decoder to Language Model for Faster Document Re-ranking Inference (2022.findings-acl)

Challenge:	Existing rumor detection methods are poor at detecting false rumors about breaking news or trending topics due to the lack of training data and prior knowledge.
Approach:	They propose an adversarial contrastive learning framework to detect false rumors by adapting features learned from well-resourced rumor data to that of the low-resource.
Outcome:	The proposed framework improves on two low-resource datasets and shows superior performance . it overcomes restriction of domain and/or language usage and improves robustness .

Copied to clipboard

Kai Hui, Honglei Zhuang, Tao Chen, Zhen Qin, Jing Lu, Dara Bahri, Ji Ma, Jai Gupta, Cicero Nogueira dos Santos, Yi Tay, Donald Metzler

Causal Inference with Large Language Model: A Survey (2025.findings-naacl)

Challenge:	State-of-the-art neural models typically encode document-query pairs using cross-attention for re-ranking.
Approach:	They propose to fine tune a pretrained encoder-decoder model using document to query generation.
Outcome:	The proposed model achieves comparable results to more expensive approaches while being 6.8X faster.

Copied to clipboard

Jing Ma

E3-TIR: Enhanced Experience Exploitation for Tool-Integrated Reasoning (2026.findings-acl)

Challenge:	Existing causal inference frameworks do not match human judgment in several key areas, such as domain knowledge, logical inference, and cultural context.
Approach:	They propose to apply large language models to causal inference tasks . they summarize the main causal problems and approaches and compare their results .
Outcome:	The proposed methods are compared with traditional methods in healthcare, finance, and economics.

Copied to clipboard

Weiyang Guo, Zesheng Shi, Liye Zhao, Jiayuan Ma, Zeen Zhu, Junxian He, Min Zhang, Jing Li

Multi-stage Training with Improved Negative Contrast for Neural Passage Retrieval (2021.emnlp-main)

Challenge:	Existing training paradigms for Large Language Models (LLMs) suffer from inefficient exploration and mode degradation due to a lack of prior guidance, while SFT-then-RL is limited by high data costs and capability plateaus caused by low-entropy collapse.
Approach:	They propose an Enhanced Experience Exploitation paradigm that integrates expert prefixes, expert guided, and self-exploration to improve agent training.
Outcome:	The proposed model achieves a 6% performance improvement over traditional paradigms on tool-use tasks while requiring less than 10% of the synthetic data.

Copied to clipboard

Jing Lu, Gustavo Hernandez Abrego, Ji Ma, Jianmo Ni, Yinfei Yang