OpenAI Blog

Improving mathematical reasoning with process supervision

Read the original on OpenAI Blog →

We've trained a model to achieve a new state-of-the-art in mathematical problem solving by rewarding each correct step of reasoning (“process supervision”) instead of simply rewarding the correct final answer (“outcome supervision”). In addition to boosting performance relative to outcome supervision, process supervision also has an important alignment benefit: it directly trains the model to produce a chain-of-thought that is endorsed by humans.

Summary generated by The Flow from the publisher's feed. The full article lives at OpenAI Blog.

arXiv AI
Jun 9

Correct Is Not Enough: Training Reasoning Planners with Executor-Grounded Rewards

arXiv:2605. 03862v4 Announce Type: replace Abstract: Reinforcement learning with verifiable rewards has become a common way to improve explicit reasoning in large language models, but final-answer correctness alone does not reveal whether the reasoning trace is faithful, reliable, or useful to the model that consumes it.

By Tianyang Han, Hengyu Shi, Junjie Hu, Xu Yang, Zhiling Wang, Junhao Su
arXiv AI
Jul 1

Evo-PI: Aligning Medical Reasoning via Evolving Principle-Guided Supervision

arXiv:2606. 31800v1 Announce Type: new Abstract: Despite recent progress, the reasoning capabilities of large multimodal language models (MLLMs) remain fundamentally constrained by static supervision, where fixed prompts, rules, or reward models provide non-adaptive guidance throughout training.

By Xianda Zheng, Huan Gao, Meng-Fen Chiang, Michael Witbrock, Kaiqi Zhao, Shangyang Li