RESEARCH · RESEARCH · #1515
DeReAct paper: modular Critic and Context Manager for more reliable ReAct-style agents
The arXiv preprint introduces DeReAct, a modular agent architecture that externalizes two gating policies—a Critic to validate proposed actions and a Context Manager to reconstruct environment-supported state and certify task completion. On GAIA and SWE-bench Verified, DeReAct raises Pass@1 most for weaker "Brain" models (6.5–7.0 points for Qwen3-Coder-480B, 4.2–5.2 points for Claude Sonnet 4.5), with smaller gains as model capability increases; with Claude Opus 4.5 overall Pass@1 is comparable to ReAct but trajectories are more evidence-complete and constraint-satisfying.
KEY POINTS
- The arXiv preprint introduces DeReAct, a modular agent architecture that externalizes two gating policies—a Critic to validate proposed actions and a Context Manager to reconstruct environment-supported state and certify task completion.
- On GAIA and SWE-bench Verified, DeReAct raises Pass@1 most for weaker "Brain" models (6.5–7.0 points for Qwen3-Coder-480B, 4.2–5.2 points for Claude Sonnet 4.5), with smaller gains as model capability increases; with Claude Opus 4.5 overall Pass@1 is comparable to ReAct but trajectories are more evidence-complete and constraint-satisfying.
- Externalizing action and completion gating can reduce unsupported completion claims and improve reliability—especially for weaker base models—making agent behavior more controllable and grounded.
WHY IT MATTERS
Externalizing action and completion gating can reduce unsupported completion claims and improve reliability—especially for weaker base models—making agent behavior more controllable and grounded.