ReflAct: World-Grounded Decision Making in LLM Agents via Goal-State Reflection
This addresses the issue of unreliable decision-making in LLM agents for complex environments, representing a strong incremental advance by enhancing the core reasoning backbone.
The paper tackled the problem of ungrounded reasoning in LLM agents like ReAct, which leads to misalignment and errors, by introducing ReflAct, a novel backbone that grounds decisions in states and enforces goal alignment, resulting in a 27.7% average improvement over ReAct and a 93.3% success rate in ALFWorld.
Recent advances in LLM agents have largely built on reasoning backbones like ReAct, which interleave thought and action in complex environments. However, ReAct often produces ungrounded or incoherent reasoning steps, leading to misalignment between the agent's actual state and goal. Our analysis finds that this stems from ReAct's inability to maintain consistent internal beliefs and goal alignment, causing compounding errors and hallucinations. To address this, we introduce ReflAct, a novel backbone that shifts reasoning from merely planning next actions to continuously reflecting on the agent's state relative to its goal. By explicitly grounding decisions in states and enforcing ongoing goal alignment, ReflAct dramatically improves strategic reliability. This design delivers substantial empirical gains: ReflAct surpasses ReAct by 27.7% on average, achieving a 93.3% success rate in ALFWorld. Notably, ReflAct even outperforms ReAct with added enhancement modules (e.g., Reflexion, WKM), showing that strengthening the core reasoning backbone is key to reliable agent performance.