Pruning the Unsurprising: Efficient LLM Reasoning via First-Token Surprisal

Documenti analoghi

• Between Lines of Code: Unraveling the Distinct Patterns of Machine and Human Programmers
  di: Shi, Yuling, et al.
  Pubblicazione: (2024)
• In Line with Context: Repository-Level Code Generation via Context Inlining
  di: Hu, Chao, et al.
  Pubblicazione: (2026)
• ClassEval-Pro: A Cross-Domain Benchmark for Class-Level Code Generation
  di: Chen, Yeheng, et al.
  Pubblicazione: (2026)
• From Code to Correctness: Closing the Last Mile of Code Generation with Hierarchical Debugging
  di: Shi, Yuling, et al.
  Pubblicazione: (2024)
• LongCodeZip: Compress Long Context for Code Language Models
  di: Shi, Yuling, et al.
  Pubblicazione: (2025)