Xiaomin Li

发表论文 22 篇 · 总被引 129 次 · h-index 6

代表论文

• Catastrophic Failure of LLM Unlearning via Quantization (2024 · International Conference on Learning Representations · 被引 78)
• Does your LLM truly unlearn? An embarrassingly simple approach to recover unlearned knowledge (2024 · arXiv.org · 被引 19)
• MedGUIDE: Benchmarking Clinical Decision-Making in Large Language Models (2025 · arXiv.org · 被引 12)
• Selection of LLM Fine-Tuning Data Based on Orthogonal Rules (2024 · AAAI Conference on Artificial Intelligence · 被引 11)
• AgentLens: Revealing The Lucky Pass Problem in SWE-Agent Evaluation (2026 · arXiv.org · 被引 9)
• Bradley-Terry and Multi-Objective Reward Modeling Are Complementary (2025 · arXiv.org · 被引 7)