Robuta

https://mechinterpworkshop.com/ Mechanistic Interpretability Workshop at ICML 2026 The Mechanistic Interpretability Workshop at ICML 2026. How can we use the internals of neural networks to understand a model better? mechanistic interpretability workshop icml https://arxiv.org/abs/2504.13151v2 [2504.13151v2] MIB: A Mechanistic Interpretability Benchmark Abstract page for arXiv paper 2504.13151v2: MIB: A Mechanistic Interpretability Benchmark mechanistic interpretability mib benchmark https://arxiv.org/abs/2512.05794 [2512.05794] Mechanistic Interpretability of Antibody Language Models Using SAEs Abstract page for arXiv paper 2512.05794: Mechanistic Interpretability of Antibody Language Models Using SAEs mechanistic interpretability language models antibody using saes https://arxiv.org/html/2504.13151v2 MIB: A Mechanistic Interpretability Benchmark mechanistic interpretability mib benchmark https://nemiconf.github.io/summer25/ The 2nd New England Mechanistic Interpretability (NEMI) Workshop new england mechanistic interpretability nemi workshop https://employment.ku.dk/phd/?show=160571 PhD fellowship in Mechanistic Interpretability for LLM Security phd fellowship mechanistic interpretability for llm security https://arxiv.org/abs/2503.06269v2 [2503.06269v2] Using Mechanistic Interpretability to Craft Adversarial Attacks against Large... Abstract page for arXiv paper 2503.06269v2: Using Mechanistic Interpretability to Craft Adversarial Attacks against Large Language Models mechanistic interpretability adversarial attacks using https://par.nsf.gov/biblio/10657157-towards-global-level-mechanistic-interpretability-perspective-modular-circuits-large-language-models Towards Global-level Mechanistic Interpretability: A Perspective of Modular Circuits of Large... This page contains metadata information for the record with PAR ID 10657157 global level mechanistic interpretability towards https://jobs.inria.fr/public/classic/fr/offres/2026-09862 2026-09862 - PhD Position F/M Mechanistic Interpretability and Problem-Space Adversarial Attacks... Offre d'emploi Inria