Reasoning-Trace Collapse: Evaluating the Loss of Explicit Reasoning During Fine-Tuning

Documenti analoghi

• Not All Code Is Equal: A Data-Centric Study of Code Complexity and LLM Reasoning
  di: Twist, Lukas, et al.
  Pubblicazione: (2026)
• On Semantic Loss Fine-Tuning Approach for Preventing Model Collapse in Causal Reasoning
  di: Deshmukh, Pratik, et al.
  Pubblicazione: (2026)
• Learning New Tasks from a Few Examples with Soft-Label Prototypes
  di: Singh, Avyav Kumar, et al.
  Pubblicazione: (2022)
• A Functional Perspective on Knowledge Distillation in Neural Networks
  di: Mason-Williams, Israel, et al.
  Pubblicazione: (2025)
• RPO:Reinforcement Fine-Tuning with Partial Reasoning Optimization
  di: Yi, Hongzhu, et al.
  Pubblicazione: (2026)