Communication Chairs 2026 2026 Conference

Post-training for reasoning models typically combines supervised fine-tuning with reinforcement learning from verifiable rewards, most commonly with GRPO. However, this algorithm suffers from sparse r...

The Thirty-Ninth Annual Conference on Neural Information Processing Systems

A NeurIPS-endorsed conference in Europe held in Copenhagen, Denmark

A NeurIPS-endorsed conference in Europe held in Copenhagen, Denmark

Workshop summary Machine learning theory has long focused on classical supervised learning settings, where a model is trained on input–label pairs drawn from a well-defined data distribution, with the...

A NeurIPS-endorsed conference in Europe held in Copenhagen, Denmark