Cites: Reasoning with Sampling: Your Base Model is Smarter Than You Think

3 citation edges inside this catalog. A title match is weaker evidence than an identifier match. Zero entries does not mean no citations on the web.

• Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?
  arxiv:2504.13837 · evidence: id
• Learning to Reason without External Rewards
  arxiv:2505.19590 · evidence: id
• Spurious Rewards: Rethinking Training Signals in RLVR
  arxiv:2506.10947 · evidence: id