PreFlect: From Retrospective to Prospective Reflection in Large Language Model Agents

TL;DR

PreFlect redefines self-reflection for LLM agents by moving from post-hoc correction to pre-execution planning-risk mitigation. It combines offline-planned error priors with a reflector to revise plans before actions, and pairs this with execution-time dynamic re-planning to adapt to unfolding conditions. Across GAIA and SimpleQA benchmarks, PreFlect yields substantial performance gains over retrospective baselines while maintaining reasonable costs, and demonstrates transferability to other agent frameworks. The approach improves reliability in high-complexity reasoning and grounded information retrieval, reducing hallucinations and incorrect tool usage in real-world tasks.

Abstract

Advanced large language model agents typically adopt self-reflection for improving performance, where agents iteratively analyze past actions to correct errors. However, existing reflective approaches are inherently retrospective: agents act, observe failure, and only then attempt to recover. In this work, we introduce PreFlect, a prospective reflection mechanism that shifts the paradigm from post hoc correction to pre-execution foresight by criticizing and refining agent plans before execution. To support grounded prospective reflection, we distill planning errors from historical agent trajectories, capturing recurring success and failure patterns observed across past executions. Furthermore, we complement prospective reflection with a dynamic re-planning mechanism that provides execution-time plan update in case the original plan encounters unexpected deviation. Evaluations on different benchmarks demonstrate that PreFlect significantly improves overall agent utility on complex real-world tasks, outperforming strong reflection-based baselines and several more complex agent architectures. Code will be updated at https://github.com/wwwhy725/PreFlect.

Figures (5)

• Figure 1: Retrospective vs. Prospective Reflection. (Left) The retrospective agent triggers reflection only after encountering a failure. (Right) The prospective agent anticipates potential risks before execution, allowing it to adjust its plan and successfully bypass the obstacle.
• Figure 2: The architecture of PreFlect comparing with basic agent workflow (top). PreFlect integrates a prospective reflection loop (pink box) into the planning stage. The bottom-right panel illustrates dynamic re-planning workflow.
• Figure 3: An example of how PreFlect triggers dynamic re-planning, reflects on the plan based on the planning errors, and finally revises the plan to avoid failure.
• Figure 4: Error distribution of PreFlect on GAIA using GPT-4.1 as the backbone LLM.
• Figure 5: Performance-cost trade-off on the GAIA validation set. The primary axis shows the total cost (USD) and the secondary axis shows the corresponding GAIA scores.