概率电路后训练鲁棒化新方法PeTeR，增强模型可靠性与鲁棒性，性能更优。
arXiv:2607.07671v1 Announce Type: new Abstract: Probabilistic circuits (PCs) can model complex joint distributions while supporting exact and efficien…
贝叶斯神经网络与等变性结合，数据增强理论新突破，提升模型鲁棒性
arXiv:2606.26273v1 Announce Type: new Abstract: Symmetries are important for many deep learning tasks, ranging from applications in the sciences to me…
本文发现LLM推理中的“悬崖词”——单个token即可导致数学运算失败，揭示模型脆弱性根源。
arXiv:2606.25524v1 Announce Type: new Abstract: Large language models (LLMs) reach high accuracy in mathematical reasoning, but individual traces on t…
用推理智能体在大型数据集中实现反例引导学习，提升模型鲁棒性与泛化能力。
arXiv:2606.11521v1 Announce Type: new Abstract: LLMs and LLM agents should improve when given feedback, but identifying when they are able to do so is…
图像分割新突破，白化变换与训练增强提升不确定性建模鲁棒性
arXiv:2606.03069v1 Announce Type: new Abstract: Generalized segmentation of medical images prevents performance degradation when different imaging dev…
NeurIPS 2025论文：用变分推理框架系统性生成对抗性提示，揭示LLM安全漏洞，方法新颖理论扎实。
arXiv:2506.22666v3 Announce Type: replace-cross Abstract: The rise of API-only access to state-of-the-art LLMs highlights the need for effective black…
大模型也能“即时刹车”：利用早退机制主动防御上下文污染，降低有害信息影响
arXiv:2510.02480v3 Announce Type: replace Abstract: Large language models (LLMs) can be influenced by harmful or irrelevant context, which can signifi…
ICML 2026新研究揭示：音频大模型可能只会“读”文本，而非真正“听”懂副语言信息，VoxParadox框架提供诊断与解决方案。
arXiv:2605.27772v1 Announce Type: cross Abstract: Audio large language models (Audio LLMs) demonstrate strong performance on speech understanding task…
最新研究揭示指令微调LLM在长上下文输入中，对有害句子的敏感性存在显著风险。
arXiv:2510.05864v2 Announce Type: replace Abstract: Large language models (LLMs) increasingly operate on long inputs, yet their behavior when harmful …
arXiv新研究提出在噪声环境中训练智能体，显著增强其鲁棒性与自适应能力
arXiv:2605.27209v1 Announce Type: new Abstract: Recent advances in large language models (LLMs) have facilitated the widespread deployment of LLMs as …
68-cell测量实验揭示LLM代理区分表面噪声与语义噪声的边界，对理解模型鲁棒性有重要意义。
arXiv:2605.25981v1 Announce Type: new Abstract: We document an empirical phenomenon in chain-of-thought and ReAct agents driven by ten large language …
新型对齐范式on-policy一致性训练，在提升LLM安全性同时极小化能力退化，是安全对齐的重要突破。
arXiv:2605.21834v1 Announce Type: new Abstract: Aligned models can misbehave in several ways: they are often sycophantic, fall victim to jailbreaks, o…
用一致性训练算法对抗社交网络政治操纵，为AI安全提供新思路
arXiv:2605.22771v1 Announce Type: new Abstract: Large language models (LLMs) exhibit systematic political bias across a variety of sensitive contexts.…