[2407.15549] Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs. arXiv Operational Status Get status notifications via email or slack

[2407.15549] Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs Skip to main content arXiv is now an independent nonprofit! Learn more × Search arXiv Press Enter to search · Advanced search --> Computer Science > Machine Learning arXiv:2407.15549 (cs) [Submitted on 22 Jul 2024 ( v1 ), last revised 29 Jul 2025 (this version, v3)] Title: Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs Authors: Abhay Sheshadri , Aidan Ewart , Phillip Guo , Aengus Lynch , Cindy Wu , Vivek Hebbar , Henry Sleight , Asa Cooper S

related reading

• [2401.05566] Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Trainingarxiv.org
• [2604.16812] Introspection Adapters: Training LLMs to Report Their Learned Behaviorsarxiv.org
• Mechanistically Eliciting Latent Behaviors in Language Models — AI Alignment Forumalignmentforum.org
• Introspection Adapters: Training LLMs to Report Their Learned Behaviorsarxiv.org
• Adversarial Attacks on LLMs | Lil'Loglilianweng.github.io
• Nicholas Carlininicholas.carlini.com
• A small number of samples can poison LLMs of any size \ Anthropicanthropic.com
• [2604.22082] Removing Sandbagging in LLMs by Training with Weak Supervisionarxiv.org
• [2502.17424] Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMsarxiv.org
• Probe-Based Data Attribution: Surfacing and Mitigating Undesirable Behaviors in LLM Post-Traininggoodfire.ai
• Latent Adversarial Training — LessWronglesswrong.com
• Refusal in LLMs is mediated by a single direction — LessWronglesswrong.com