FROST: Filtering Reasoning Outliers with Attention for Efficient Reasoning

Documenti analoghi

• RHYTHM: Reasoning with Hierarchical Temporal Tokenization for Human Mobility
  di: He, Haoyu, et al.
  Pubblicazione: (2025)
• Contrastive Reasoning Alignment: Reinforcement Learning from Hidden Representations
  di: Luo, Haozheng, et al.
  Pubblicazione: (2026)
• AceReason-Nemotron: Advancing Math and Code Reasoning through Reinforcement Learning
  di: Chen, Yang, et al.
  Pubblicazione: (2025)
• AceReason-Nemotron 1.1: Advancing Math and Code Reasoning through SFT and RL Synergy
  di: Liu, Zihan, et al.
  Pubblicazione: (2025)
• Every Attention Matters: An Efficient Hybrid Architecture for Long-Context Reasoning
  di: Ling Team, et al.
  Pubblicazione: (2025)