My last PhD paper 🎉: fundamental problems with model editing for LLMs!. We present *12 open challenges* with definitions/benchmarks/assumptions, inspired by work on belief revision in philosophy. To provide a way forward, we test model editing against Bayesian belief revision.🧵

Shower thought: LLMs still have very incoherent notions of evidence, and they update in strange ways when presented with information in-context that is relevant to their beliefs. I really wonder what will happen when LLM agents start doing interp on themselves and see the source.

Join millions who have switched to Grok.

RT @hannahrosekirk: My team at @AISecurityInst is hiring! This is an awesome opportunity to get involved with cutting-edge scientific resea….

RT @nouhadziri: Current agents are highly unsafe, o3-mini one of the most advanced models in reasoning score 71% in executing harmful reque….

RT @milesaturpin: New @Scale_AI paper! 🌟. LLMs trained with RL can exploit reward hacks but not mention this in their CoT. We introduce ver….

Overdue job update -- I am now:.- A Visiting Scientist at @schmidtsciences, supporting AI safety and interpretability.- A Visiting Researcher at the Stanford NLP Group, working with @ChrisGPotts. I am so grateful I get to keep working in this fascinating and essential area, and.

RT @FazlBarez: Excited to share our paper: "Chain-of-Thought Is Not Explainability"! . We unpack a critical misconception in AI: models exp….

RT @JustenMichel: really interesting to see just how gendered excitement about AI is, even among AI experts

RT @farairesearch: 🤔 Can lie detectors make AI more honest? Or will they become sneakier liars?. We tested what happens when you add decept….

RT @jiaxinwen22: New Anthropic research: We elicit capabilities from pretrained models using no external supervision, often competitive or….

RT @dongkeun_yoon: 🙁 LLMs are overconfident even when they are dead wrong. 🧐 What about reasoning models? Can they actually tell us “My an….

colab: For aficionados, the post also contains some musings on “tuning the random seed” and how to communicate uncertainty associated with this process.

Are p-values missing in AI research?. Bootstrapping makes model comparisons easy!. Here's a new blog/colab with code for:.- Bootstrapped p-values and confidence intervals.- Combining variance from BOTH sample size and random seed (eg prompts).- Handling grouped test data. Link ⬇️.

RT @ysu_nlp: New AI/LLM Agents Track at #EMNLP2025! . In the past few years, it feels a bit odd to submit agent work to *CL venues because….

RT @vaidehi_patil_: 🚨 Introducing our @TmlrOrg paper “Unlearning Sensitive Information in Multimodal LLMs: Benchmark and Attack-Defense Eva….

RT @EliasEskin: Extremely excited to announce that I will be joining @UTAustin @UTCompSci in August 2025 as an Assistant Professor! 🎉. I’m….

RT @rowankwang: New Anthropic Alignment Science blog post: Modifying LLM Beliefs with Synthetic Document Finetuning. We study a technique f….

RT @amuuueller: Lots of progress in mech interp (MI) lately! But how can we measure when new mech interp methods yield real improvements ov….

RT @sydneymlevine: 🔆Announcement time!🔆In Spring 2026, I will be joining the NYU Psych department as an Assistant Professor! My lab will s….

RT @maksym_andr: Excited to present our recent work on AI safety at this event!. If you're coming to ICLR 2025 in S….

RT @yanda_chen_: My first paper @AnthropicAI is out!. We show that Chains-of-Thought often don’t reflect models’ true reasoning—posing chal….