Back to Basics: Revisiting Exploration in Reinforcement Learning for LLM Reasoning via Generative Probabilities

Documenti analoghi

• The Shape of Learning: Anisotropy and Intrinsic Dimensions in Transformer-Based Models
  di: Razzhigaev, Anton, et al.
  Pubblicazione: (2023)
• Confidence Is All You Need: Few-Shot RL Fine-Tuning of Language Models
  di: Li, Pengyi, et al.
  Pubblicazione: (2025)
• Your Transformer is Secretly Linear
  di: Razzhigaev, Anton, et al.
  Pubblicazione: (2024)
• Simple Vision-Language Math Reasoning via Rendered Text
  di: Skripkin, Matvey, et al.
  Pubblicazione: (2025)
• Logit-KL Flow Matching: Non-Autoregressive Text Generation via Sampling-Hybrid Inference
  di: Sevriugov, Egor, et al.
  Pubblicazione: (2024)