When Should Agents Think? Adaptive Reasoning via Cross-Turn Estimation
Based on this observation, we propose Reasoning Adaptation through Cross-Turn Estimation (RACE), a training approach for adaptive agent reasoning.
Key points
- Large language model (LLM)-based agents have demonstrated strong capabilities on complex tasks.
- We find that decreases in the likelihood of subsequent reference actions after removing additional reasoning closely track whether those actions remain recoverable given earlier reasoning, providing an effective and lightweight signal for estimating cross-turn action support.
- RACE introduces a Likelihood-Guided Progressive Reasoning Cover Detection (LoGiC) procedure that progressively identifies reasoning turns whose removal has limited impact on the current and subsequent reference actions.
- Extensive experiments on four representative agent benchmarks show that RACE substantially reduces reasoning cost while maintaining or improving task performance.
Sources (1)
- [1]When Should Agents Think? Adaptive Reasoning via Cross-Turn EstimationarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 02:43 PM
Based on this observation, we propose Reasoning Adaptation through Cross-Turn Estimation (RACE), a training approach for adaptive agent reasoning.
Large language model (LLM)-based agents have demonstrated strong capabilities on complex tasks.
Extractive summary: sentences quoted from the sources.