Papers by Kexin Chen
LLM-VA: Resolving the Jailbreak-Overrefusal Trade-off via Vector Alignment (2026.acl-long)
Copied to clipboard
| Challenge: | Existing vector steering methods adjust the magnitude of answer vectors, but this creates a fundamental trade-off—reducing jailbreak increases over-refusal. |
| Approach: | They propose a method which aligns va with vb through closed-form weight updates, making the model’s willingness to respond causally dependent on its safety assessment. |
| Outcome: | Experiments on 12 LLMs show that the proposed method achieves 11.45% higher F1 than the best baseline while preserving 95.92% utility. |
Sticking to the Mean: Detecting Sticky Tokens in Text Embedding Models (2025.acl-long)
Copied to clipboard
| Challenge: | Sticky tokens, when repeatedly inserted into sentences, pull sentence similarity toward a certain value, disrupting the normal distribution of embedding distances and degrading downstream performance. |
| Approach: | They propose a method to detect “sticky tokens” by sentence and token filtering and apply it to 40 checkpoints across 14 model families. |
| Outcome: | The proposed method detects 868 sticky tokens across 14 models and shows that their presence does not correlate with model size or vocabulary size. |
ACORD: An Expert-Annotated Retrieval Dataset for Legal Contract Drafting (2025.acl-long)
Copied to clipboard
Steven H Wang, Maksim Zubkov, Kexin Fan, Sarah Harrell, Yuyang Sun, Wei Chen, Andreas Plesner, Roger Wattenhofer
| Challenge: | Contract clause retrieval is critical to contract drafting because of its high quality and complexity. |
| Approach: | They propose the first expert-annotated benchmark specifically designed for contract clause retrieval . ACORD focuses on complex contract clauses such as Limitation of Liability, Indemnification, Change of Control . |
| Outcome: | The atticus clause retrieval dataset shows promising results but needs improvement . the benchmark can be used as an IR benchmark for the NLP community . |
PLAWBENCH: A Rubric-Based Benchmark for Evaluating LLMs in Real-World Legal Practice (2026.acl-long)
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
| 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. |
Invisible Entropy: Towards Safe and Efficient Low-Entropy LLM Watermarking (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to watermark low-entropy content are expensive and risky . IE reduces parameter size by 99% while achieving performance on par with state-of-the-art methods . |
| Approach: | They propose a logit-based watermarking paradigm that uses entropy-based features to predict whether the next token is high or low. |
| Outcome: | The proposed method reduces parameter size by 99% while achieving performance on par with state-of-the-art methods. |
GCPG: A General Framework for Controllable Paraphrase Generation (2022.findings-acl)
Copied to clipboard
Kexin Yang, Dayiheng Liu, Wenqiang Lei, Baosong Yang, Haibo Zhang, Xue Zhao, Wenqing Yao, Boxing Chen
| Challenge: | Existing studies highlight a special condition under two indispensable aspects of controllable paraphrase generation (CPG) individually, lacking a unified circumstance to explore and analyze their effectiveness. |
| Approach: | They propose a general controllable paraphrase generation framework that integrates lexical and syntactical conditions into a text sequence and uniformly processes them in an encoder-decoder paradigm. |
| Outcome: | The proposed framework can combine lexical and syntactical conditions and improve paraphrase generation. |
From Evasion to Concealment: Stealthy Knowledge Unlearning for LLMs (2025.findings-acl)
Copied to clipboard
| Challenge: | Existing approaches to unlearning often treat nonsensical responses or template-based refusals as the unlearning target, making the process even more vulnerable to attacks and jailbreaks. |
| Approach: | They propose a method that uses inverted facts to remove the need for auxiliary models or retaining data while avoiding leakage. |
| Outcome: | Evaluated on the ToFU Knowledge Unlearning dataset using Llama2-7B-Chat and Phi-1.5, MEOW outperforms baselines in forgetting quality while preserving model utility. |
Tailor: A Soft-Prompt-Based Approach to Attribute-Based Controlled Text Generation (2023.acl-long)
Copied to clipboard
| Challenge: | Existing work focuses on generating sentences satisfying pre-specified attributes such as topic and sentiment, yet suffers from increases in storage and inference time. |
| Approach: | They propose a method that uses a pre-trained continuous vector to generate a fixed pre-trainable language model to satisfy a specified attribute. |
| Outcome: | The proposed model can achieve improvements on eleven attribute-specific generation tasks with 0.08% extra training parameters. |
ESC-Eval: Evaluating Emotion Support Conversations in Large Language Models (2024.emnlp-main)
Copied to clipboard
Haiquan Zhao, Lingyu Li, Shisong Chen, Shuqi Kong, Jiaan Wang, Kexin Huang, Tianle Gu, Yixu Wang, Jian Wang, Liang Dandan, Zhixu Li, Yan Teng, Yanghua Xiao, Yingchun Wang
| Challenge: | Emotion Support Conversation (ESC) is a crucial application for reducing stress and providing emotional guidance. |
| Approach: | They re-organize 2,801 role-playing cards to define roles of role-players . they train a specific role- playing model called ESC-Role which behaves more like a confused person than GPT-4 . |
| Outcome: | The proposed model behaves more like a confused person than GPT-4, and the model performs better than GPLs. |
SEER: Facilitating Structured Reasoning and Explanation via Reinforcement Learning (2024.acl-long)
Copied to clipboard
| Challenge: | Existing methods focus on single-step reasoning, ignoring logical dependencies between steps. |
| Approach: | They propose a method that maximizes a structure-based return to facilitate structured reasoning and explanation. |
| Outcome: | The proposed method outperforms state-of-the-art methods on EntailmentBank and STREET benchmarks. |
Conflict-Aware Memory for Embodied Agents: Enhancing Vector Data Quality via Detection Rules (2026.acl-long)
Copied to clipboard
| Challenge: | Embodied agents have successfully leveraged large language models (LLMs) to better transform human instructions and images into executable task plans. |
| Approach: | They propose Conflict Detection Rules to identify and manage data quality issues in vector knowledge bases and correct the index structure. |
| Outcome: | Experimental results show that planners with Conflict Detection Rules exceed the basic LLM planner by 15.25% and 14.25% in grammatical accuracy (GA) and interpretation accuracy (IA) on average. |
Context-Driven Index Trimming: A Data Quality Perspective to Enhancing Precision of RALMs (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Existing research often overlooks the data quality issues within retrieval results, often caused by inaccurate existing vector-distance-based retrieval methods. |
| Approach: | They propose to use Context-Driven Index Trimming framework to capture and regulate consistency between retrieved contexts and modify indexes in the database. |
| Outcome: | Experiments show that the proposed framework can improve answer quality by 3.75% on open-domain question-answering tasks. |