Papers by Xingjun Wang
Fine-mixing: Mitigating Backdoors in Fine-tuned Language Models (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for defending NLP models against backdoors have ignored the clean weights of PLMs. |
| Approach: | They exploit pre-trained weights to mitigate backdoors in fine-tuned NLP models . they use a fine-mixing technique and an Embedding Purification technique to do the same . |
| Outcome: | The proposed method outperforms baseline mitigation methods on three single-sentence sentiment classification tasks and two sentence-pair classification tasks. |
Deliberative Searcher: Improving LLM Reliability via Reinforcement Learning with Constraints (2026.acl-long)
Copied to clipboard
| Challenge: | Large language models with search capabilities often exhibit miscalibrated confidence, causing incorrect answers with high certainty. |
| Approach: | They propose a reasoning-primary framework that integrates search operations into chain-of-thought generation while maintaining explicit confidence calibration. |
| Outcome: | The proposed framework improves accuracy and reliability of large language models with search capabilities. |
AgenticEval: Toward Agentic and Self-Evolving Safety Evaluation of Large Language Models (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing static safety evaluation methods are ill-equipped to address dynamic nature of AI risks and evolving regulations, creating a critical safety gap. |
| Approach: | They propose a new paradigm of agentic safety evaluation reframing evaluation as a continuous and self-evolving process rather than a one-time audit. |
| Outcome: | The proposed framework shows a consistent decline in model safety as the evaluation hardens. |
Fake Alignment: Are LLMs Really Aligned Well? (2024.naacl-long)
Copied to clipboard
Yixu Wang, Yan Teng, Kexin Huang, Chengqi Lyu, Songyang Zhang, Wenwei Zhang, Xingjun Ma, Yu-Gang Jiang, Yu Qiao, Yingchun Wang
| Challenge: | Existing studies on large language models have shown that they are poorly aligned in practice. |
| Approach: | They propose a framework to evaluate safety in large language models . they propose two new metrics to quantify fake alignment and obtain corrected performance estimation. |
| Outcome: | The proposed framework and two metrics show that some models with purported safety are poorly aligned in practice. |
HIE-SQL: History Information Enhanced Network for Context-Dependent Text-to-SQL Semantic Parsing (2022.findings-acl)
Copied to clipboard
| Challenge: | Recent studies focus on context-dependent text-to-SQL task but fail to exploit both . et al., 2019; xu e. al.; yu y., 2021) focus on the context-independent text to SQL task . |
| Approach: | They propose a history information enhanced text-to-SQL model to exploit context dependence information from history utterances and the last predicted SQL query. |
| Outcome: | The proposed model improves performance on two context-dependent text-to-SQL benchmarks. |
Optimizing Cross-Client Domain Coverage for Federated Instruction Tuning of Large Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Federated domain-specific instruction tuning (FedDIT) for large language models (LLMs) aims to enhance performance in specialized domains using distributed private and limited data. |
| Approach: | They introduce an algorithm that explicitly maximizes cross-client domain coverage through diversity-oriented client center selection and retrieval-based augmentation. |
| Outcome: | The proposed algorithm achieves performance gains of 29.19% and domain coverage improvements of 4.82%-21.36% over 11 baselines. |
HarmRLVR: Weaponizing Verifiable Rewards for Harmful LLM Alignment (2026.acl-long)
Copied to clipboard
| Challenge: | Recent advances in Reinforcement Learning with Verifiable Rewards (RLVR) have gained significant attention due to their objective and verifiably verifier reward signals. |
| Approach: | They propose to exploit RLVR for alignment reversibility by using GRPO to reverse alignment with merely 64 harmful prompts without responses. |
| Outcome: | The proposed method outperforms fine-tuning and RLHF in reasoning and code generation tasks while maintaining general capabilities. |