2408.12798
arxiv.org · 7,136 words · saved by 1 readers
N/A
BACKDOOR LLM: A Comprehensive Benchmark for Backdoor Attacks and Defenses on Large Language Models Yige Li1 , Hanxun Huang2 , Yunhan Zhao3 , Xingjun Ma3 , Jun Sun1 arXiv:2408.12798v2 [cs.AI] 19 May 2025 1 Singapore Management University 2 The University of Melbourne 3 Fudan University…
saved by
related reading
- A small number of samples can poison LLMs of any size \ Anthropicanthropic.com
- [2502.05209] Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilitiesarxiv.org
- [2604.16812] Introspection Adapters: Training LLMs to Report Their Learned Behaviorsarxiv.org
- [2502.17424] Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMsarxiv.org
- [2312.06942] AI Control: Improving Safety Despite Intentional Subversionarxiv.org
- Large Language Diffusion Modelsarxiv.org
- Poisoning Attacks on LLMs Require a Near-constant Number of Poison Samplesarxiv.org
- [2602.04899] Phantom Transfer: Data-level Defences are Insufficient Against Data Poisoningarxiv.org
- 2409.02908arxiv.org
- 2312.06674arxiv.org
- Model Tampering Attacks Enable More Rigorous Evaluations of LLM Capabilitiesarxiv.org
- 2401.05566.pdfarxiv.org