arXiv新研究揭示LLM代理在后端代码生成中面临"约束衰减"问题，系统脆弱性令人警醒。
Article URL: https://arxiv.org/abs/2605.06445 Comments URL: https://news.ycombinator.com/item?id=48256912 Points: 4 # Comments: 0
首份系统研究RL微调VLM的鲁棒性与思维链一致性，揭示模型脆弱性根源
arXiv:2602.12506v3 Announce Type: replace Abstract: Reinforcement learning (RL) finetuning has become a key technique for enhancing large language mod…
揭秘LLM微调中对齐为何脆弱：从参数动态到输出分布的统一视角
arXiv:2605.18309v1 Announce Type: new Abstract: Although Large Language Models (LLMs) achieve strong alignment through supervised fine-tuning and rein…
临床AI系统在细微扰动和多语言场景下存在诊断崩溃风险，这篇系统性审计揭开了安全漏洞。
arXiv:2605.16993v1 Announce Type: cross Abstract: Current clinical artificial intelligence (AI) systems are evaluated almost exclusively on clean, sta…
分布式学习中数据归因的脆弱性：单个参与者可操纵归因值大幅膨胀，挑战定价与审计可信度。
arXiv:2605.15520v1 Announce Type: cross Abstract: Data attribution has become an important component of pricing, auditing, and governance in machine l…
提出基于阻抗的攻击可达域，为电力电子系统网络脆弱性提供新的量化度量
arXiv:2605.14502v1 Announce Type: cross Abstract: Power electronics systems are increasingly exposed to cyber threats due to their integration with di…