When to Commit? Towards Variable-Size Self-Contained Blocks for Discrete Diffusion Language Models

Documenti analoghi

• Text Meets Topology: Rethinking Out-of-distribution Detection in Text-Rich Networks
  di: Wang, Danny, et al.
  Pubblicazione: (2025)
• Block-R1: Rethinking the Role of Block Size in Multi-domain Reinforcement Learning for Diffusion Large Language Models
  di: Jiang, Yan, et al.
  Pubblicazione: (2026)
• TRN-R1-Zero: Text-rich Network Reasoning via LLMs with Reinforcement Learning Only
  di: Liu, Yilun, et al.
  Pubblicazione: (2026)
• Break the Block: Dynamic-size Reasoning Blocks for Diffusion Large Language Models via Monotonic Entropy Descent with Reinforcement Learning
  di: Jiang, Yan, et al.
  Pubblicazione: (2026)
• When Does a Language Model Commit? A Finite-Answer Theory of Pre-Verbalization Commitment
  di: Zhang, Long, et al.
  Pubblicazione: (2026)