Papers by Xinyang Zhang
Patton: Language Model Pretraining on Text-Rich Networks (2023.acl-long)
Copied to clipboard
| Challenge: | Existing models for text-rich networks do not take inter-document structure into account. |
| Approach: | They propose a pretraining framework for a text-rich network using a masked language model and a masking node prediction framework. |
| Outcome: | The proposed model outperforms baselines on four tasks in academic and e-commerce domains. |
Mitigating Lost in Multi-turn Conversation via Curriculum RL with Verifiable Accuracy and Abstention Rewards (2026.acl-long)
Copied to clipboard
Ming Li, Pei Chen, Zhenhao Zhang, Tao Yang, Xinyang Zhang, Han Li, Tianyu Cao, Ming Zeng, Zhuofeng Wu, Meng Jiang, Huasheng Li, Lihong Li, Bing Yin
| Challenge: | Large Language Models exhibit strong capabilities in single-turn instruction following but suffer from Lost-in-Conversation (LiC) when instructions are revealed progressively in multi-turn settings, models get "Lost in Conversation" |
| Approach: | They propose a framework that encourages models to generate correct answers and judge solvability in multi-turn conversations. |
| Outcome: | The proposed framework improves models' ability to balance problem-solving with abstention . it reduces premature answering behaviors that cause lost-in-conversation (LiC) |
Leveraging LLM Reasoning Enhances Personalized Recommender Systems (2024.findings-acl)
Copied to clipboard
Alicia Tsai, Adam Kraft, Long Jin, Chenwei Cai, Anahita Hosseini, Taibai Xu, Zemin Zhang, Lichan Hong, Ed H. Chi, Xinyang Yi
| Challenge: | Recent advances have showcased the potential of Large Language Models (LLMs) in executing reasoning tasks, particularly facilitated by Chain-of-Thought (CoT) prompting. |
| Approach: | They propose to use Large Language Models to perform tasks with subjectivity and personalized preferences as inputs to RecSys. |
| Outcome: | The proposed framework aligns with real human judgment on the coherence and faithfulness of LLM reasoning responses. |
Beware of Your Po! Measuring and Mitigating AI Safety Risks in Role-Play Fine-Tuning of LLMs (2025.acl-long)
Copied to clipboard
Weixiang Zhao, Yulin Hu, Yang Deng, Jiahe Guo, Xingyu Sui, Xinyang Han, An Zhang, Yanyan Zhao, Bing Qin, Tat-Seng Chua, Ting Liu
| Challenge: | Existing role-play fine-tuning techniques improve role adaptability but may degrade safety performance, especially for villainous characters. |
| Approach: | They propose safety-aware Role-Play Fine-Tuning (SaRFT) to balance role-playing capabilities and safety. |
| Outcome: | The proposed method outperforms state-of-the-art baselines under both LoRA and full-parameter fine-tuning settings. |
ALERT: An LLM-powered Benchmark for Automatic Evaluation of Recommendation Explanations (2025.naacl-long)
Copied to clipboard
Yichuan Li, Xinyang Zhang, Chenwei Zhang, Mao Li, Tianyi Liu, Pei Chen, Yifan Gao, Kyumin Lee, Kaize Ding, Zhengyang Wang, Zhihan Zhang, Jingbo Shang, Xian Li, Trishul Chilimbi
| Challenge: | Existing benchmarks for recommendation explanation evaluation lack item diversity and user preferences data. |
| Approach: | They propose a model-agnostic recommendation explanation evaluation benchmark based on Amazon e-commerce categories with implicit preferences . they propose two novel automatic evaluators that enable scalable and human-preference aligned evaluation of explanations . |
| Outcome: | The proposed model-agnostic evaluation benchmark outperforms existing methods in a variety of domains. |
META: Metadata-Empowered Weak Supervision for Text Classification (2020.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods for weakly supervised text classification use text data alone to generate pseudo-labels . strong label indicators exist in metadata and it has been long overlooked due to challenges . |
| Approach: | They propose a framework that leverages metadata as an additional source of weak supervision by combining text data and metadata into a text-rich network. |
| Outcome: | The proposed framework exploits metadata as an additional source of weak supervision. |
A Multilingual Dataset and Empirical Validation for the Mutual Reinforcement Effect in Information Extraction (2026.findings-acl)
Copied to clipboard
Chengguang Gan, Sunbowen Lee, Qingyu Yin, Yunhao Liang, Xinyang He, Hanjun Wei, Younghun Lim, Shijian Wang, Hexiang Huang, QingHao Zhang, Shiwen Ni, Tatsunori Mori
| Challenge: | Existing work on the Mutual Reinforcement Effect in information extraction has not been empirically validated . 76 percent of the 21 sub-datasets exhibit the Mutual Reforcement effect across languages . |
| Approach: | They propose a multilingual MRE mix dataset that integrates 21 sub-datasets covering English, Japanese, and Chinese. |
| Outcome: | The proposed framework reduces manual annotation effort while preserving structural requirements of MRE tasks. |
DORM: Preference Data Weights Optimization for Reward Modeling in LLM Alignment (2025.findings-emnlp)
Copied to clipboard
Rongzhi Zhang, Chenwei Zhang, Xinyang Zhang, Liang Qiu, Haoming Jiang, Yuchen Zhuang, Qingru Zhang, Hyokun Yun, Xian Li, Bing Yin, Tuo Zhao, Chao Zhang
| Challenge: | Existing approaches to align large language models with human preferences are noisy and varying in importance of preference samples. |
| Approach: | a new method enhances reward modeling by learning to dynamically weigh preference data. |
| Outcome: | a new method improves the performance of large language models with human preferences . it initializes data importance and iteratively refines them to maximize validation performance. |
SAT: Balancing Reasoning Accuracy and Efficiency with Stepwise Adaptive Thinking (2026.acl-long)
Copied to clipboard
| Challenge: | Large Reasoning Models (LRMs) produce excessively long Chains of Thought (COT) Existing solutions that improve token efficiency but sacrifice fine-grained control can disrupt the logical integrity of the reasoning process. |
| Approach: | They propose a framework that performs step-level, difficulty-aware pruning while preserving the core reasoning structure. |
| Outcome: | Experiments show that SAT reduces reasoning tokens by 40% while maintaining or improving accuracy. |
MGR: Multi-generator Based Rationalization (2023.acl-long)
Copied to clipboard
| Challenge: | Existing approaches to explain NLP models have two key challenges: spurious correlation and degeneration. |
| Approach: | They propose a rationalization framework using a generator and a predictor to construct a self-explaining NLP model with spurious correlation and degeneration as key challenges. |
| Outcome: | The proposed method improves the F1 score by 20.9% compared to state-of-the-art methods. |
AdaSteer: Your Aligned LLM is Inherently an Adaptive Jailbreak Defender (2025.emnlp-main)
Copied to clipboard
Weixiang Zhao, Jiahe Guo, Yulin Hu, Yang Deng, An Zhang, Xingyu Sui, Xinyang Han, Yanyan Zhao, Bing Qin, Tat-Seng Chua, Ting Liu
| Challenge: | Activation steering offers training-free defense but relies on fixed steering coefficients, resulting in suboptimal protection and increased false rejections of benign inputs. |
| Approach: | They propose an adaptive activation steering method that dynamically adjusts model behavior based on input characteristics. |
| Outcome: | The proposed method outperforms baseline methods across multiple jailbreak attacks with minimal impact on utility. |
PersonaAgent: Bridging Memory and Action for Personalized LLM Agents (2026.findings-acl)
Copied to clipboard
Weizhi Zhang, Xinyang Zhang, Chenwei Zhang, Liangwei Yang, Jingbo Shang, Zhepei Wei, Henry Peng Zou, Zijie Huang, Zhengyang Wang, Yifan Gao, Xiaoman Pan, Lian Xiong, Jingguo Liu, Philip S. Yu, Xian Li
| Challenge: | Existing Large Language Model (LLM) enabled agents lack flexibility to respond to users’ varying needs and preferences. |
| Approach: | They propose a test-time user-preference alignment strategy that optimizes the persona prompt, ensuring real-time preference alignment through textual loss feedback between simulated and ground-truth responses. |
| Outcome: | The proposed framework outperforms baseline methods in real-time and in real applications. |