Before invoking external tools, an agentic LLM must select among a K-way action space: executing a call, seeking clarification, answering directly, or declining. While internal activation steering can alter these pre-execution decisions, conventional aggregate metrics obscure where altered states land and what collater...
A dynamic, second-order geometric study of EM is presented, which introduces a parameter-level Geometric Mitigation Framework that orthogonally projects empirical harmful gradient subspace out of parameter updates and unmask the illusion of behavioral safety.
Wei-Qiao Que, Rui-Zhe Li, Cheng-Yu Wang et al.· arXiv.org· 0 citations
It is suggested that cheap uncertainty estimators are insufficient on their own to improve code correctness, and that their practical value lies in serving as gating signals for costlier execution-based correction loops rather than as standalone substitutes for verification.
Pranav Rakasi, Maanas Lalwani, A. Srivastava et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.