Papers by Mei Chen
PG-GSQL: Pointer-Generator Network with Guide Decoding for Cross-Domain Context-Dependent Text-to-SQL Generation (2020.coling-main)
Copied to clipboard
| Challenge: | Existing approaches to text-to-SQL generation depend on interaction history and current utterances. |
| Approach: | They propose an encoder-decoder model based on interaction-level encoder to capture historical information of SQL query and reuse the previous SQL query tokens. |
| Outcome: | The proposed model outperforms the previous state-of-the-art model on the SParC benchmark . it achieves 34.0% question matching accuracy and 19.0% interaction matching accuracy . |
Robust Adaptation of Large Multimodal Models for Retrieval Augmented Hateful Meme Detection (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large Multimodal Models (LMMs) have shown promise in hateful meme detection, but they face limitations like sub-optimal performance and limited out-of-domain generalization capabilities. |
| Approach: | They propose a robust adaptation framework for hateful meme detection that enhances in-domain accuracy and cross-domain generalization while preserving the general vision-language capabilities of LMMs. |
| Outcome: | The proposed framework outperforms larger agentic systems in detecting hateful memes under adversarial attacks while maintaining the general vision-language capabilities of LMMs. |
Forget the Unneeded: Backdooring Large Language Models via Contrastive-enhanced Machine Unlearning (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for prompt tuning for Large Language Models find backdoor attacks to be significant in data-rich scenarios. |
| Approach: | They propose a backdoor attacks through contrastive-enhanced machine unlearning in data-limited scenarios . they use a machine un learning method to capture precise backdoor patterns . |
| Outcome: | The proposed method captures precise backdoor patterns without association between triggers and backdoors, reducing side effects. |
ADEPT-SQL: A High-performance Text-to-SQL Application for Real-World Enterprise-Level Databases (2025.acl-demo)
Copied to clipboard
Yongnan Chen, Zhuo Chang, Shijia Gu, Yuanhang Zong, Zhang Mei, Shiyu Wang, Hezixiang Hezixiang, Hongzhi Chen, Jin Wei, Bin Cui
| Challenge: | et al., 2017) address domain-specific knowledge barriers, schemas complexity, and computational costs of large LLMs. |
| Approach: | They propose a domain-adapted Text2SQL system that addresses critical deployment challenges in professional fields. |
| Outcome: | The proposed system achieves 97% execution accuracy on real-world databases . it is faster than existing systems and has a higher performance than existing ones. |
LiveFact: A Dynamic, Time-Aware Benchmark for LLM-Driven Fake News Detection (2026.acl-long)
Copied to clipboard
| Challenge: | Current evaluation frameworks are static and vulnerable to benchmark data contamination . current models are ineffective at assessing reasoning under temporal uncertainty . |
| Approach: | They propose a live-based benchmark that simulates the real-world "fog of war" they propose evaluating models on their ability to reason with evolving, incomplete information . |
| Outcome: | The proposed model outperforms proprietary state-of-the-art models in classification and evidence mode . it also provides a component to monitor BDC explicitly . |
SWAM: Adaptive Sliding Window and Memory-Augmented Attention Model for Rumor Detection (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for rumor detection on social media focus on static graphs, ignoring dynamic and incremental propagation . rumour detection on the social media platform is crucial to mitigating harmful effects of rumors. |
| Approach: | They propose a sliding window and memory-augmented attention model for rumor detection . they use a dynamic propagation graph and memory to capture the long-term dependency . |
| Outcome: | The proposed model is compared with the state-of-the-art models on two public datasets. |
CORAL: Learning Consistent Representations across Multi-step Training with Lighter Speculative Drafter (2025.acl-long)
Copied to clipboard
| Challenge: | Existing methods that focus on training and inference suffer from misalignment . speculative decoding is a powerful technique that accelerates large language models . |
| Approach: | They propose a framework that improves both accuracy and efficiency in speculative drafting by using cross-step representational alignment. |
| Outcome: | The proposed framework outperforms existing methods on three LLM families and three benchmark datasets. |
Improving Hateful Meme Detection through Retrieval-Guided Contrastive Learning (2024.acl-long)
Copied to clipboard
| Challenge: | Existing systems for detecting hateful memes lack sensitivity to subtle differences in memes that are vital for correct hatefulness classification. |
| Approach: | They propose to construct a hatefulness-aware embedding space through retrieval-guided contrastive training to identify hatefulness based on data unseen in training. |
| Outcome: | The proposed system outperforms existing models on the HatefulMemes dataset with an AUROC of 87.0 and improves contextual understanding across domains. |
ClueAnchor: Clue-Anchored Knowledge Reasoning Exploration and Optimization for Retrieval-Augmented Generation (2025.findings-emnlp)
Copied to clipboard
Hao Chen, Yukun Yan, Sen Mei, Wanxiang Che, Zhenghao Liu, Qi Shi, Xinze Li, Yuchun Fan, Pengcheng Huang, Qiushi Xiong, Zhiyuan Liu, Maosong Sun
| Challenge: | Existing RAG systems often underutilize the retrieved documents, authors say . they fail to extract and integrate key clues needed to support faithful and interpretable reasoning . |
| Approach: | a new framework extracts key clues from retrieved content and generates multiple reasoning paths . the framework optimizes the model by selecting the most appropriate reasoning path . |
| Outcome: | Experiments show that ClueAnchor outperforms baseline RAG frameworks in completeness and robustness. |
Retrieval-Augmented Defense: Adaptive and Controllable Jailbreak Prevention for Large Language Models (2026.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) remain vulnerable to jailbreak attacks due to evolving nature and diversity of attack strategies. |
| Approach: | They propose a framework for jailbreak detection that integrates a database of known attack examples into Retrieval-Augmented Generation to infer the underlying, malicious user query and jailbreak strategy used to attack the system. |
| Outcome: | The proposed framework reduces the effectiveness of strong jailbreak attacks while maintaining low rejection rates for benign queries. |
Can Graph Descriptive Order Affect Solving Graph Problems with LLMs? (2025.acl-long)
Copied to clipboard
| Challenge: | Large language models (LLMs) have achieved significant success in reasoning tasks, including mathematical reasoning and logical deduction. |
| Approach: | They conduct the first comprehensive analysis of how the order of graph descriptions impacts LLM performance. |
| Outcome: | The results show that graph descriptions significantly improve LLMs’ comprehension of graph structures, and the robustness of LLM models to graph description order varies across different tasks. |
PreFLMR: Scaling Up Fine-Grained Late-Interaction Multi-modal Retrievers (2024.acl-long)
Copied to clipboard
| Challenge: | Large Multimodal Models excel in natural language and visual understanding but are challenged by challenging tasks such as Knowledge-based Visual Question Answering (KB-VQA). |
| Approach: | They propose a framework for training Large Multimodal Models (LMMs) to perform KB-VQA tasks. |
| Outcome: | The proposed framework is used to train and evaluate multi-modal retrievers. |
Control-DAG: Constrained Decoding for Non-Autoregressive Directed Acyclic T5 using Weighted Finite State Automata (2024.naacl-short)
Copied to clipboard
| Challenge: | Existing non-autoregressive (NAR) models fail to generate specified entity names in up to 40% of responses and produce OOV errors. |
| Approach: | They propose a constrained decoding algorithm for Directed Acyclic T5 model which offers lexical, vocabulary and length control. |
| Outcome: | The proposed model significantly improves on Schema Guided Dialogue and DART datasets, establishing strong results for Task-Oriented Dialog and Data-to-Text NLG. |
FGDGNN: Fine-Grained Dynamic Graph Neural Network for Rumor Detection on Social Media (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for detecting rumors on social media focus on coarse-grained temporal information and ignore fine-grain temporal dynamics. |
| Approach: | They propose a fine-grained dynamic graph neural network model which incorporates fine-grain temporal information into a unified framework for rumor detection. |
| Outcome: | The proposed model improves on three public real-world datasets. |
Good Reasoning Makes Good Demonstrations: Implicit Reasoning Quality Supervision via In-Context Reinforcement Learning (2026.findings-acl)
Copied to clipboard
| Challenge: | Reinforcement Learning with Verifiable Rewards (RLVR) improves reasoning in large language models but treats all correct solutions equally, potentially reinforcing flawed traces that arrive at correct answers by chance. |
| Approach: | They propose a method that reweights rewards by a factor approximately proportional to Evidence Gain and assigns higher weights to high-quality traces without requiring costly computation. |
| Outcome: | Experiments on mathematical reasoning benchmarks show that Reinforcement Learning with Verifiable Rewards (RLVR) improves reasoning in large language models but treats all correct solutions equally. |
Self-supervised Cross-modal Pretraining for Speech Emotion Recognition and Sentiment Analysis (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Existing approaches to multimodal speech emotion recognition and sentiment analysis have not improved results due to their relatively simple fusion mechanisms and lack of proper cross-modal pretraining. |
| Approach: | They propose a deep-fused audio-text bi-modal transformer with carefully designed cross-modal fusion mechanism and stage-wise cross-mod pretraining scheme to facilitate cross-modulation. |
| Outcome: | The proposed method exceeds benchmarks on public IEMOCAP emotion and CMU-MOSEI sentiment datasets by a large margin. |
Beyond Overlap Metrics: Rewarding Reasoning and Preferences for Faithful Multi-Role Dialogue Summarization (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for multi-role dialogue summarization favor surface-level imitation of references rather than genuine gains in faithfulness or alignment with human preferences. |
| Approach: | They propose a framework that couples explicit cognitive-style reasoning with reward-based optimization for multi-role dialogue summarization. |
| Outcome: | The proposed framework matches strong baselines on ROUGE and BERTScore, while in-depth analysis on SAMSum shows clear gains in factual faithfulness and model-based preference alignment. |
Rejection-to-Acceptance Transition: Model Editing-Based Jailbreak Backdoor Injection Not Limited to Few Output Tokens (2026.findings-acl)
Copied to clipboard
Shiji Yang, Min Cai, Hao Xiong, Congyao Mei, Haodong Zou, Shicheng Tan, Jie Chen, Fulan Qian, Shu Zhao
| Challenge: | Existing methods for jailbreaking LLMs are implemented by binding backdoors to predefined phrases as first few output tokens, inducing the LLM’s next-token prediction to produce continuous responses. |
| Approach: | They propose a model editing-based jailbreak backdoor attack that hijacks LLM representations into a acceptance domain rather than binding to a few output tokens. |
| Outcome: | The proposed model editing method outperforms existing methods, showing stronger jailbreak capabilities across LLMs and datasets. |
Developing a Reliable, Fast, General-Purpose Hallucination Detection and Mitigation Service (2025.naacl-industry)
Copied to clipboard
| Challenge: | Hallucination is a problem in large language models that produce incorrect output . authors propose a reliable and high-speed production system to detect and rectify hallucinations . |
| Approach: | They propose a high-speed production system that detects hallucinations in LLMs . they propose NER, natural language inference, span-based detection and a rewriting mechanism . |
| Outcome: | The proposed system detects a wide range of hallucinations in LLM responses. |
Rule By Example: Harnessing Logical Rules for Explainable Hate Speech Detection (2023.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to content moderation are based on rule-based heuristics, but they lack the flexibility and robustness needed to moderate harmful content. |
| Approach: | They propose a novel contrastive learning approach for learning from logical rules for content moderation using only a few data examples. |
| Outcome: | The proposed approach outperforms state-of-the-art deep learning classifiers while providing more explainable predictions. |
Defense against Prompt Injection Attacks via Mixture of Encodings (2025.naacl-short)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have emerged as a dominant approach for a wide range of NLP tasks, but external content embeds malicious instructions that manipulate the LLM’s output. |
| Approach: | They propose a mixture of encodings defense mechanism which utilizes multiple character encodes to degrade LLM performance on certain NLP tasks. |
| Outcome: | The proposed method achieves one of the lowest attack success rates under prompt injection attacks while maintaining high performance across all NLP tasks. |
Reflections & Resonance: Two-Agent Partnership for Advancing LLM-based Story Annotation (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing methods for story annotation require a meticulous and resourceintensive effort, but the advent of advanced computational tools like GPT-4 can streamline the process and mitigate common limitations. |
| Approach: | They propose a multi-agent system that generates tailored prompts for a large language model and provides feedback to refine the initial prompts. |
| Outcome: | The proposed system significantly improves the model's reconstruction accuracy and confidence, demonstrating that dynamic interaction between agents significantly boosts the annotation process's precision and efficiency. |