TACT: Mitigating Overthinking and Overacting in Coding Agents via Activation Steering

Documenti analoghi

• UniGraph: Learning a Unified Cross-Domain Foundation Model for Text-Attributed Graphs
  di: He, Yufei, et al.
  Pubblicazione: (2024)
• FiDeLiS: Faithful Reasoning in Large Language Model for Knowledge Graph Question Answering
  di: Sui, Yuan, et al.
  Pubblicazione: (2024)
• UniGraph2: Learning a Unified Embedding Space to Bind Multimodal Graphs
  di: He, Yufei, et al.
  Pubblicazione: (2025)
• Evaluating the Paperclip Maximizer: Are RL-Based Language Models More Likely to Pursue Instrumental Goals?
  di: He, Yufei, et al.
  Pubblicazione: (2025)
• Meta-Reasoner: Dynamic Guidance for Optimized Inference-time Reasoning in Large Language Models
  di: Sui, Yuan, et al.
  Pubblicazione: (2025)