How Fast Should a Model Commit to Supervision? Training Reasoning Models on the Tsallis Loss Continuum

arXiv:2604.25907 · cs.LG, cs.AI · Submitted 2026-04-28 · Read on arXiv

cs.LG, cs.AI

Submitted: 2026-04-28

Updated: 2026-10-30

Terminology

• RLVR — 5× in this paper · explained in 2 episodes
• GRPO — 2× in this paper · explained in 6 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Gradient Flow — 1× in this paper · explained in 2 episodes
• Posterior-Attenuated Fine-Tuning — 1× in this paper

Related papers

• Power Distribution Bridges Sampling, Self-Reward RL, and Self-Distillation
• Train Separately, Merge Together: Modular Post-Training with Mixture-of-Experts
• Democratizing Tool Learning with Environments Fully Simulated by a Free 8B Language Model
• D-STEER: Two-Way Control of DPO-Induced Safety Behavior in Activation Space
• Optimal hypersurface decision trees
• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation