Translation in progress Some entries still include text in the original language.
Oct 12 · Monday
All papers Found 18 papers Another 265 papers do not have category tags yet and are not included in filtered results.
“
About this collection Collection in progress We are building a large AI safety paper collection to help researchers find work related to their interests. Papers are collected automatically andfiltered only for relevance to AI safety , without judging quality, novelty or reading value. The default “Annotated first” shows papers with tags or published paper analyses first. Each group is ordered by original publication date, newest first, with undated papers last. “Newest first” uses publication dates only. This order is not a quality ranking.
One-line summaries, tags and paper analyses are prepared by AI from the papers. Language models may have systematic biases and produce hallucinations.We plan to carry out human annotation and review to reduce errors in this content.
Proposes Task Bank self-replay to mitigate held-out forgetting in continual instruction tuning
Published Oct 9
Evaluated models Qwen3-0.6B , Qwen3-1.7B , Qwen3-4B and others4 Abstract The authors claim self-replay can reduce forgetting outside the list, and that on the 4B model new task information helps with target selection, while smaller models do not.The authors study how language models, when learning new tasks, can use their own knowledge for broader self-preservation: given an incomplete list and forgetting that extends beyond the list, can the model generate replay and help choose targets.
用时钟 harness 和预算塑形 GRPO 改善 Agent 的按时返回
Published Oct 8
Evaluated models Qwen3.6-27B , Qwen3-4B , Qwen3.5-9B and others4 Abstract 作者比较提示词、harness 与 GRPO：按时性可改善，但额外时间仍未变成更好解。作者研究小规模 LLM 智能体在显式 wall-clock 预算下能否按时返回，并把更多时间换成更好的解。
提出 PatchBench-Local 协议，评测激活补丁修复越狱的局部附带损伤
Published Oct 7
Evaluated models Qwen2.5-3B-Instruct , Qwen2.5-14B-Instruct , gemma-3-4b-it and others8 Abstract 论文提出 PatchBench 与局部评估协议，揭示了激活补丁在全局能力无损下仍存局部良性退化。
提出 AdvSim2Real，协同训练网页智能体抵御页面注入
Published Oct 7
Evaluated models Qwen3.5-4B 提出 EmoRSS，推理时在拒绝子空间反向抵消情绪激活偏移以缓解过度拒答
Published Oct 4
Evaluated models Llama-3.1-8B-it , Qwen3-8B 提出 Perturbot 扰动训练，打破 VLA 的模态捷径
Published Oct 3
Evaluated models π0.5 Abstract Perturbot 改训练数据以削弱三类模态捷径，GroundFscore 用来看成功是否来自任务证据。Perturbot 是一套不改推理的 VLA 训练期数据干预，用来削弱模态捷径，并用离线分数区分“任务变好”和“靠捷径变好”。
提出 R-Quest，用有效性与新颖性反馈维持推理模型自我进化
Published Oct 3
Evaluated models Qwen3-4B-Base , OctoThinker-3B-Hybrid-Base Abstract R-Quest 用有效性与新颖性反馈维持自进化。R-Quest 处理的是推理模型自进化中的题目质量：questioner 与 solver 用 GRPO 交替从自生成题目学习时，多轮之后平均成绩会下降。
提出 Local Support Learning，把微调适配器限制在当前数据支撑以缓解遗忘
Published Oct 2
Evaluated models Qwen2.5-7B-Instruct , Qwen2.5-3B-Instruct , Qwen2.5-1.5B-Instruct Abstract LSL 用 GMM 门把适配器限制在当前数据支撑内，以减轻多阶段遗忘。LSL 是不访问旧数据、在后训练中保持先前能力的框架。
提出 HeadEdit，经冻结 unembedding 做随输入变化的 logit 校正
Published Oct 1
Evaluated models Qwen3-4B , Gemma-3-4B-IT , Llama-3.2-3B-Instruct and others4 Abstract 低秩 unembedding 校正抬高九个设置的总体准确率，并可部分预测 DPO 更新。HeadEdit 是一种不更新参数的行为校准方法，作用点是冻结的 unembedding，而不是中间层残差或新的梯度优化。
提出 FAIL BANK，用运行时 CBF 反馈自进化更新 VLA 策略
Published Sep 30
Evaluated models π0.5 , π0 提出 ReSAIL，用选择蒸馏与特权保留缓解智能体自蒸馏崩塌
Published Sep 30
Evaluated models Qwen3-4B , Qwen3-8B , Qwen3-VL-4B-Instruct 提出 CREDO，用保留 token 对的可微读出校准推理模型置信度
Published Sep 28
Evaluated models Qwen3-8B , Qwen3-1.7B , Qwen3-4B Abstract CREDO 用可微两 token 读出替代口头置信，在数学和代码上同时拉高准确率与校准。CREDO 把 RLVR 里的置信从“采样一个数字”改成“读一对保留 token 的相对概率”，并用可微回归和差异加权同时拉准确率与校准。 。
提出 Fed-ADR，用编排服务器协同操纵梯度攻破联邦学习并检测恢复
Published Sep 23
Evaluated models CNN（MNIST：2 Conv + 3 FC） , CNN（Fashion-MNIST：6 Conv + 1 FC） , ResNet18（CIFAR-10） Abstract Fed-ADR 用 gray-box 编排攻击绕过多种聚合防御，再用 Hessian 一致性检测和窗口内恢复把准确率拉回。Fed-ADR 是一套面向联邦学习的攻击、检测与恢复框架：编排服务器改写合谋客户端的梯度，使参数服务器的现有防御把投毒更新当成良性，同时在发现后用历史梯度把全局模型原地修回来。
用 Dmax 与尾部损失评测后门修复的类别性能保持
Published Sep 22
Evaluated models ResNet18 , PreActResNet18 , VGG19-BN 提出对比自蒸馏，分离正确性与行为偏移
Published Sep 18
Evaluated models Qwen3-4B , Qwen3-4B-Instruct-2507 Abstract 隔离自蒸馏后，对比正确与错误解教师可抵消行为偏移并稳定提升推理。把自蒸馏从 GRPO 里拆开后，用靠近正确解教师、远离错误解教师的对比信号，把正确性和行为分开。 。
提出 Rationale-Only 安全微调，去掉拒答声明以降低误拒
Published Sep 4
Evaluated models Llama-3.1-8B , Mistral-7B-v0.3 , Gemma-2-9B and others8 Abstract 去掉套话拒答声明、只训理由，可降低假拒答，有害查询上的服从率仍低。这是一项针对安全微调数据的结构分析：假拒答被作者追溯到回答里的套话式拒答声明，而不是只靠推理时补救。
提出 IACM-RL，用信念状态与强化学习改进动态意图下的复杂工具调用
Published Aug 3
Evaluated models Qwen3-8B , 1.7B Qwen 提出 VILTA，在闭环训练中用 VLM 编辑邻车轨迹以提升驾驶策略鲁棒性
Published Jan 19 Abstract VILTA 在闭环中编辑邻车轨迹以训练驾驶策略。VILTA 是一个把视觉语言模型放进自动驾驶闭环训练的框架，用来应对长尾危险场景不足。
You have reached the end