Tanmay Parekh
275 posts
@tparekh97
PhD student @UCLA | Fellowship @ Amazon, Bloomberg | Intern @Bloomberg @AIatMeta @AmazonLab126 | MLT @LTIatCMU | Applied Scientist @amazonIN | BTech @iitbombay
@aclmeeting @iSchoolUI @HaohanWang @FIUSCIS @BoiseState @junzhuang_ In Session 14 (6 PM PT) Sachith Sri Ram Kothur delivers a talk about "#PExA: Parallel Exploration Agent for Complex Text-to-SQL," work done with Bloomberg Data Science Ph.D. Fellow @tparekh97 of @UCLAnlp, @hc_ella, Shuyi Wang & @YunmoChen bloom.bg/4eIfO7c #ACL2026NLP (5/7)
Another great paper from Google. Shows general LLMs can solve formal math by planning proofs and checking each step. Raised general LLM performance from under 10% to 70%. A general LLM failed badly when asked to write full formal proofs in 1 try, but became much stronger when it planned, split the work into smaller claims, reused past claims, and learned from Lean’s feedback. The paper shows the weakness was not just the model’s math ability, but the way it was being used - the absence of structured interaction with a verifier. The key idea is that the model does not try to write one giant perfect proof at once, because that usually fails on long and tricky problems. Instead, LEAP stores the proof as a graph of goals and subgoals, so useful lemmas can be reused instead of rediscovered every time. The authors tested LEAP on Putnam 2025 and a new Lean benchmark built from 60 IMO-style problems, where ordinary one-shot proof writing did very poorly. LEAP solved all 12 Putnam 2025 problems and raised general LLM performance on the Lean IMO benchmark from under 10% to 70%. ---- Link – arxiv. org/abs/2606.03303 Title: "LEAP: Supercharging LLMs for Formal Mathematics with Agentic Frameworks"
Today we're releasing Perceptron Mk1: frontier video and embodied reasoning.
New blog 📢 Can we extract dense advantages without new annotations or models in GRPO? The answer is YES! 💡Answer correctness splits rollouts into positives and negatives. Just upweight positive tokens which differ significantly from the negative tokens! 🧵👇
For the UCLA NLP seminar talk this Friday, we are thrilled to host Prof. Christopher Potts @ChrisGPotts from Stanford @stanfordnlp ! Title: “The Archai of Palimpsestic Memorization” When: 2–3 PM (PST), Friday, Jan 23 Registration: ucla.zoom.us/meeting/regist…
