Translation in progress Some entries still include text in the original language.
Oct 12 · Monday
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.
提出 SLDR,用选择性层恢复与动态路由防御恶意微调
Published Oct 8
Evaluated models Llama3.1-8B-Instruct , Llama3-8B-Instruct , Qwen2.5-7B-Instruct and others4 Abstract 作者称 SLDR 能压低有害输出并保持下游效用。SLDR 是针对恶意微调的微调后防御:在已对齐模型上做有符号层探测,只在敏感分数最大与最小的两层训练 LoRA,再用 lmax 的最后 token 表示决定是否启用适配器。
提出 FDCU 双约束子空间投影遗忘,抵御再训练恢复有害知识
Published Sep 30
Evaluated models Qwen2.5-3B , Llama-3-8B-Instruct , Qwen-3-8B Abstract FDCU 用双掩码限制浅层抑制,在知识擦除与安全输出上提高再训练后的稳定性。FDCU 是一种约束参数更新的遗忘方法,用来降低再训练把已抑制恶意行为重新引出的可能。
提出 VaccineBooster,在对齐阶段混合嵌入扰动与梯度衰减抵御有害微调
Published Sep 29
Evaluated models Llama-2-7B Abstract VaccineBooster 把两种对齐阶段扰动合在一步里,内容更安全但拒答保留更少。VaccineBooster 是一种对齐阶段防御:在提供方还看得到训练过程、还没接触用户微调数据时,把嵌入扰动和权重级梯度衰减合成同一次更新,用来应对之后不受信任的微调。有害微调可以只混入少量有害指令-回答对,就削弱拒答并提高有害输出的可能,任务效用却仍然好看。过滤会漏掉隐蔽样本,事后修复又往往不知道该修哪里。
检验层冻结与奇异方向截断能否挡住少样本有害微调
Published Sep 29
Evaluated models Llama-3.1-8B-Instruct , Llama-3.1-Tulu-3-8B-DPO , OLMo-2-1124-7B-Instruct and others10 Abstract 定位与恢复可以复现,但知情攻击能把损伤移出冻结带,并打败跨检查点的截断修复。这篇工作检验:把拒答定位到某些层或更新方向之后,能不能在攻击者知情时还守住拒答。 。
解释后验安全的权重几何脆弱性,并提出预训练持续安全共训练
Published Sep 7
Evaluated models Qwen-2.5-7B , Qwen-2.5-7B-Instruct , Llama-3-8B and others15 Abstract 论文将后验安全脆弱性归因于正交抑制,并提出通过预训练持续共训练实现具备微调抗性的安全防护。
提出 SEAL,在 MoE 共享专家上做 DPO-LoRA 对齐以抵御越狱与权重篡改
Published Sep 2
Evaluated models Qwen1.5-MoE-A2.7B , DeepSeekMoE-16B , GLM-4.7-Flash and others4 Abstract SEAL 把 DPO-LoRA 限制在混合 MoE 共享专家上,SEAL++再保护已有安全神经元方向。SEAL 是面向混合 MoE 的训练时防御:只对齐始终执行的共享专家,使安全不依赖路由器是否把 token 送到安全专家。
提出 RACER,用区域感知一致性对抗微调去除多模态大模型后门
Published Aug 25
Evaluated models LLaVA-1.5-7B , Qwen2-VL-7B , Qwen2.5-VL-7B 提出 USG,在部分保护开源权重发布时阻断有害微调梯度
Published Aug 6
Evaluated models Qwen3-14B(表中作 Qwen-14B) , Llama-3.1-8B(表中作 Llama-8B) Abstract USG 在 PPOW 下用零空间立方层阻断有害微调梯度。Unidirectional Safety Gate 是面向 PPOW 发布设定的发布时防护:大部分权重仍可训练,一小部分安全关键组件由提供者侧加固保留,下游不必配合额外安全流程。 。
用代价曲线评估恶意微调防御在持续训练下的衰减
Published May 14
Evaluated models Llama-2-7B-chat , Qwen3-8B , Llama-3.1-8B-Instruct and others8 Abstract 论文指出恶意微调防御在持续训练下均会衰减,表明基于有限攻击者假设的防御无法为开源模型提供持久保护。
You have reached the end