Hugging Face users report that approval granted through Meta's official LLaMA application process at ai.meta.com does not automatically carry over to the corresponding gated repositories on Hugging Fa…
A new arXiv study (2608.21606v1) finds that large language models (LLMs) can still recover supposedly 'forgotten' information through adversarial prompting, with attack success rates (ASR) between 72.…
A study by an anonymous researcher, conducted as part of Neel Nanda's MATS 10.0 stream, found that linear 'trust' vectors extracted from the residual streams of Llama-3.2-3B-Instruct and Llama-3.1-8B-…
Researchers introduced Symbolic Mechanistic Data Attribution (SMDA), a framework that links training data to high-level model behaviors, tested on Llama-3.2-3B-Instruct. SMDA revealed systematic safet…