1
2
Multimodal Evaluator Preference Collapse: Cross-Modal Contagion in Self-Evolving Agents
多模态AI自我评估存在系统性偏差，一种新发现的“偏好崩溃”现象加剧了模型反馈循环中的错误。
arXiv:2606.16682v1 Announce Type: new Abstract: When AI agents use language models to evaluate their own outputs in a feedback loop, systematic biases…
3
Krause Synchronization Transformers
提出Krause Attention机制，解决Transformer中全局softmax导致的同步动态与表示崩塌问题
arXiv:2602.11534v3 Announce Type: replace-cross Abstract: Self-attention in Transformers relies on globally normalized softmax weights, causing all to…