Elisabetta Rocchetti
PulseAugur coverage of Elisabetta Rocchetti — every cluster mentioning Elisabetta Rocchetti across labs, papers, and developer communities, ranked by signal.
2 day(s) with sentiment data
-
LLMs follow instructions via coordinated skills, not universal mechanism, study finds
A new research paper published on arXiv suggests that Large Language Models (LLMs) do not follow instructions through a universal mechanism. Instead, the study indicates that instruction-following is a result of the ski…
-
New LumiXAI framework simplifies AI model interpretability
Researchers have developed LumiXAI, a new modular framework designed to simplify feature attribution for AI model interpretability. This system consolidates various attribution tools into a single platform, offering a u…
-
AI Refusal Control: DiM vs. INLP Methods Compared
Researchers have compared two methods, Diff-in-Means (DiM) and Iterative Nullspace Projection (INLP), for controlling refusal behavior in AI chat models. The study found that INLP's counterfactual flipping intervention …