Can LLM guardrails think twice before deciding? ✨ Check out our #ACL2025 paper: THINKGUARD — a critique-augmented safety guardrail! ✅ Structured critiques ✅ Interpretable decisions ✅ Robust against adversarial prompts 📑 arxiv.org/abs/2502.13458 🧵[1/n]

Have you noticed… 🔍 Aligned LLM generations feel less diverse? 🎯 Base models are decoding-sensitive? 🤔 Generations get more predictable as they progress? 🌲 Tree search fails mid-generation (esp. for reasoning)? We trace these mysteries to LLM probability concentration, and

🤔 If MLLMs encode vision & text in a joint space, why not reason over both? We introduce Latent Visual Reasoning (LVR) — a new paradigm for multimodal LLMs. - Keeps everything autoregressive - Reconstructs query-relevant visual semantics in hidden states (like human visual

Your agent just broke a rule again 😅 Time to upgrade: ✅ Executable policies ✅ Trajectory-level safety ✅ Cross-domain control Read how 🛡️: arxiv.org/abs/2510.03485 More details 🌐: rakanwen.github.io/PolicyGuard_We…

I'm recruiting multiple PhD students for Fall 2026 in Computer Science at JHU Computer Science 🍂 Apply to work on AI for social sciences/human behavior, social NLP, and LLMs for real-world applied domains you're passionate about! Learn more kristinagligoric.com & help spread the word!

Alex is one of the most thoughtful and creative researchers I know. Minnesota is getting a real gem! PhD applicants, don’t sleep on this one.