Curious about the limits of #LLMs in planning and reasoning? Discover how something as simple as stacking blocks can halt the #LLMArmageddon!. Join our #NeurIPS23⭐️SPOTLIGHT⭐️poster to delve into the planning abilities of LLMs in direct and LLM-Modulo settings.

RT @rao2z: A discussion of our paper "Stop Anthropomorphizing Intermediate Tokens. " by Ben Dickson on his Tech Tal….

RT @nitashatiku: Everyone tweetin about the Apple paper, hope you read @rao2z on reasoning models

RT @rao2z: Some of what that recent Apple LRM limitations paper shows is known (pardon my friendly Schmidhubering; I do welcome more LLM….

RT @rao2z: So @kayastechly 🎓 from @ASU yesterday. Kaya was a rather unusual Yochanite-- didn't take any courses or do theses with me; and w….

RT @rao2z: Yochanites at one of those rare group breakfasts! (after that 5AM #NeurIPS2025 deadline. )

RT @rao2z: It ain't "The Bitter Lesson" if you are in the loop curating the training data for your LLM, y'all. Pick your lesson, will ya?….

RT @rao2z: Enroute to #ICLR2025 with @kayastechly . and itching for a #ThursdayHarangue.

RT @rao2z: PSA for #ICLR2025 authors frantically making posters: Stop worrying! Just prompt o3 and GPT-4o and you will have an AGI-ready po….

RT @rao2z: Our invited commentary for the Annals of @NYASciences titled "(How) Do reasoning models reason?" is now online . 👉 https://t.co/….

RT @rao2z: 📢 Tomorrow (3rd April) Karthik Valmeekam @karthikv792 will will try and hoodwink his PhD committee (comprising @tdietterich @liu….

RT @rao2z: 📢Delighted to share that our analysis of the planning abilities of 🍓 o1 has now been accepted to #TMLR @TmlrOrg . Joint work wi….

RT @sarath_ssreedh: [1/3].I am incredibly humbled and honored to be selected as one of IEEE AI's 10 to watch. However, this wouldn't have b….

RT @rao2z: Back in September, when we evaluated o1, and used LRM (Large Reasoning Model) to refer to that type of models, there was some pu….

RT @rao2z: On the MDP formulation of LLMs used in R1 [Not quite #SundayHarangue]. Everyone knows that R1 is using RL on LLMs. Most also kno….

RT @rao2z: On the use of Verifiers with LLMs--External vs. Internal LLM-Modulo (with @kayastechly, @karthikv792 and @21st_Warlock ). We ar….

RT @rao2z: 📢"On the Self-Verification Limitations of LLMs in Reasoning and Planning Tasks" with @kayastechly and @….

📢 DeepSeek-R1 on PlanBench 📢. DeepSeek-R1 gets similar performance as OpenAI’s o1 (preview)—achieving 96.6% on Blocksworld and 39.8% on its obfuscated version, Mystery BW. The best part? . ⚡It’s 21x cheaper than o1-preview, offering similar results at a fraction of the cost!

RT @SCAI_ASU: Congrats to @karthikv792, a #PhD student working under the supervision of @rao2z in #SCAI, who has received an IBM PhD Fellow….

RT @rao2z: 🥳🎉Congrats to Yochanite Karthik Valmeekam @karthikv792 on being selected as an IBM PhD Fellow! Thanks @IBMResearch for this reco….

RT @rao2z: With some Yochanites--past and present--at #NeurIPS2024 🤗 (L-R: @kayastechly @karthikv792 @rao2z @sarath_ssreedh @kr_t ) https:/….