NeurIPS Invited Talk The Art of (Artificial) Reasoning

Scaling laws suggest that “more is more” — brute-force scaling of data and compute leads to stronger AI capabilities. However, despite rapid progress on benchmarks, state-of-the-art models still exhibit "jagged intelligence," indicating that current scaling approaches may have limitations in terms of sustainability and robustness. Additionally, while the volume of papers on arXiv continues to grow rapidly, our scientific understanding of artificial intelligence hasn't kept pace with engineering advances, and the current literature presents seemingly contradictory findings that can be difficult to reconcile. In this talk, I will discuss key insights into the strengths and limitations of LLMs, examine when reinforcement learning succeeds or struggles in reasoning tasks, and explore methods for enhancing reasoning capabilities in smaller language models to help them close the gap against their larger counterparts in specific domains. The NeurIPS Logo above may be used on presentations. Ri

Scaling laws suggest that “more is more” — brute-force scaling of data and compute leads to stronger AI capabilities. However, despite rapid progress on benchmarks, state-of-the-art models still exhibit "jagged intelligence," indicating that current scaling approaches may have limitations in terms of sustainability and robustness. Additionally, while the volume of papers on arXiv continues to grow rapidly, our scientific understanding of artificial intelligence hasn't kept pace with engineering advances, and the current literature presents seemingly contradictory findings that can be…

saved by

• Laerdon Kim

related reading

• Lil'Loglilianweng.github.io
• [2510.13786] The Art of Scaling Reinforcement Learning Compute for LLMsarxiv.org
• Will scaling work?dwarkeshpatel.com
• The Scaling Hypothesis · Gwern.netgwern.net
• The Extreme Inefficiency of RL for Frontier Models - Toby Ordtobyord.com
• Scaling in the service of reasoning & model-based ML | Yoshua Bengioyoshuabengio.org
• Scaling: The State of Play in AIoneusefulthing.org
• The Art of Scaling Reinforcement Learning for LLMs | Sean Bell, ResolveAI Labsyoutube.com
• State of RL for reasoning LLMs | A. Weersaweers.de
• RL Scaling Laws for LLMs - by Cameron R. Wolfe, Ph.D.cameronrwolfe.substack.com
• The bitter lesson of LLM evalsparsed.com
• LLMs Don't Need More Parameters. They Need Loops.youtube.com