From Harm to Help: Turning Reasoning In-Context Demos into Assets for Reasoning LMs

Documenti analoghi

• Reasoning Resides in Layers: Restoring Temporal Reasoning in Video-Language Models with Layer-Selective Merging
  di: Fu, Zihang, et al.
  Pubblicazione: (2026)
• Strategy Executability in Mathematical Reasoning: Leveraging Human-Model Differences for Effective Guidance
  di: Liang, Weida, et al.
  Pubblicazione: (2026)
• When Precision Meets Position: BFloat16 Breaks Down RoPE in Long-Context Training
  di: Wang, Haonan, et al.
  Pubblicazione: (2024)
• Think in Parallel, Answer as One: Logit Averaging for Open-Ended Reasoning
  di: Wang, Haonan, et al.
  Pubblicazione: (2025)
• SeedPrints: Fingerprints Can Even Tell Which Seed Your Large Language Model Was Trained From
  di: Tong, Yao, et al.
  Pubblicazione: (2025)