arxiv
PublishedJuly 16, 2026 at 4:00 AM
—neutral
Avoiding unsafe sets when training with Langevin Dynamics
Publisher summary· verbatim
arXiv:2607.07538v2 Announce Type: replace Abstract: Training a model with noisy gradient descent can be idealized as overdamped Langevin dynamics, and a natural safety question is to bound the probability $\nu_t(\mathcal{A}_H) = \mathbb{P}(Q_t \in \mathcal{A}_H)$ that the trajectory lies in a design
Stay posted· Newsletter
A 5-min weekly brief — top movers, price watch, story of the week.
Discussion
No replies yet. Be first.
Related coverage
More from ARXIV
arxivBeyond a Single Direction: Chain-of-Thought Disrupts Simple Steering of Refusal6harxivSkillCorpus: Consolidating and Evaluating the Open Skill Ecosystem for Real-World LLM Agents6harxivPolyInterview: An LLM-based Platform for Immersive Mock Interview Practice with Comprehensive Multimodal Assessment6harxivDo Coding Agents Need Executable World Models, Simplification, and Verification to Solve ARC-AGI-3?6hThe Bubble Brief
WEEKLYRead AI insights every Tuesday — top movers, new releases, story of the week.
Originally published on arxiv ↗