AdaCtrl: Towards Adaptive and Controllable Reasoning via Difficulty-Aware Budgeting

Documenti analoghi

• Self-Reasoning Language Models: Unfold Hidden Reasoning Chains with Few Reasoning Catalyst
  di: Wang, Hongru, et al.
  Pubblicazione: (2025)
• Towards On-Policy Data Evolution for Visual-Native Multimodal Deep Search Agents
  di: Huang, Shijue, et al.
  Pubblicazione: (2026)
• ARES: Multimodal Adaptive Reasoning via Difficulty-Aware Token-Level Entropy Shaping
  di: Chen, Shuang, et al.
  Pubblicazione: (2025)
• ReTool: Reinforcement Learning for Strategic Tool Use in LLMs
  di: Feng, Jiazhan, et al.
  Pubblicazione: (2025)
• AdaTIR: Adaptive Tool-Integrated Reasoning via Difficulty-Aware Policy Optimization
  di: Fang, Zhaiyu, et al.
  Pubblicazione: (2026)