Rewards as Labels: Revisiting RLVR from a Classification Perspective

Documenti analoghi

• The Unlearnability Phenomenon in RLVR for Language Models
  di: Chen, Yulin, et al.
  Pubblicazione: (2026)
• Spurious Rewards Paradox: Mechanistically Understanding How RLVR Activates Memorization Shortcuts in LLMs
  di: Yan, Lecheng, et al.
  Pubblicazione: (2026)
• Countdown-Code: A Testbed for Studying The Emergence and Generalization of Reward Hacking in RLVR
  di: Khalifa, Muhammad, et al.
  Pubblicazione: (2026)
• Exploration vs Exploitation: Rethinking RLVR through Clipping, Entropy, and Spurious Reward
  di: Chen, Peter, et al.
  Pubblicazione: (2025)
• Flexible Entropy Control in RLVR with a Gradient-Preserving Perspective
  di: Chen, Kun, et al.
  Pubblicazione: (2026)