挑战无监督异常检测中任务特定训练的必要性，揭示分布偏移下重建残差评分的局限性
arXiv:2601.22763v3 Announce Type: replace Abstract: Current state-of-the-art multi-class unsupervised anomaly detection (MUAD) methods rely on trainin…
无需训练骨干网络，用表示工程让LLM代理稳定应对工具调用变化，突破传统微调瓶颈。
arXiv:2602.04935v3 Announce Type: replace-cross Abstract: Adapting LLM agents to domain-specific tool calling remains notably brittle under evolving i…
新方法RT-SDGOD专攻实时目标检测在天气、成像变化下的分布偏移难题，从问题定义层面提升泛化能力
arXiv:2606.09367v1 Announce Type: new Abstract: In real-world deployment under strict real-time constraints, weather and imaging variations induce sig…
AI基准评估面临污染威胁，论文揭示现有检测工具在分布偏移和大规模场景下严重失效，挑战传统审计可靠性。
arXiv:2606.03305v1 Announce Type: new Abstract: Benchmark contamination, where evaluation examples appear in a model's training data, threatens the va…
离线强化学习新范式：后验混合贝叶斯信念正则化策略优化，破解分布外动作评估难题
arXiv:2606.00680v1 Announce Type: cross Abstract: Offline reinforcement learning (RL) aims to optimize policies from pre-collected datasets. A bottlen…
ICML 2026论文揭示：LLM代码理解靠先验知识，而非编程语言语义
arXiv:2510.03415v3 Announce Type: replace-cross Abstract: Recent work asks whether large language models (LLMs) condition their reasoning on explicit …
首个针对生态系统通量外推的基准FLUXtrapolation，测试模型在分布偏移下的泛化能力。
arXiv:2605.19812v1 Announce Type: new Abstract: We introduce FLUXtrapolation, a benchmark for extrapolating ecosystem fluxes under progressively harde…
COOPO算法提出循环离线-在线策略优化，巧妙解决分布偏移与灾难性遗忘难题，为强化学习混合范式带来新突破。
arXiv:2605.18675v1 Announce Type: new Abstract: Offline reinforcement learning struggles with distributional shift and constrained performance due to …
提出归一化等变性的结构先验，可应用于任意骨干网络的图像去噪，有效提升分布偏移健壮性。
arXiv:2605.08193v2 Announce Type: replace-cross Abstract: Normalization Equivariance (NE) is a structural prior that improves robustness to distributi…