长程决策中，单纯推理会因短视导致失败，需引入前瞻规划。
Why Reasoning Fails to Plan: A Planning-Centric Analysis of Long-Horizon Decision Making in LLM Agents
- 用前瞻式规划替代逐步推理，让早期决策考虑远期影响。
- 引入FLARE模型后，LLaMA-8B在多个任务上超越GPT-4o的推理表现。
- 适合研究长周期智能体决策、强化学习与规划融合的学者。
基于大语言模型（LLM）的智能体在短时序推理中表现优异，但在长程规划中常失效。我们指出根本问题在于：逐步推理诱导了短视的贪心策略，无法应对延迟后果。在确定性、结构化环境中分析发现，局部最优选择会随时间累积放大，难以修正。为此提出FLARE（未来感知前瞻与奖励估计），通过显式前瞻、价值传播和有限承诺机制，使后续结果影响早期决策。在多个基准测试、框架和模型中，FLARE均显著提升性能，使使用FLARE的LLaMA-8B多次超越未使用该机制的GPT-4o。结果揭示了推理与规划的本质区别。
原文摘要 · Abstract (English)
Large language model (LLM)-based agents exhibit strong step-by-step reasoning capabilities over short horizons, yet often fail to sustain coherent behavior over long planning horizons. We argue that this failure reflects a fundamental mismatch: step-wise reasoning induces a form of step-wise greedy policy that is adequate for short horizons but fails in long-horizon planning, where early actions must account for delayed consequences. From this planning-centric perspective, we study LLM-based agents in deterministic, fully structured environments with explicit state transitions and evaluation signals. Our analysis reveals a core failure mode of reasoning-based policies: locally optimal choices induced by step-wise scoring lead to early myopic commitments that are systematically amplified over time and difficult to recover from. We introduce FLARE (Future-aware Lookahead with Reward Estimation) as a minimal instantiation of future-aware planning to enforce explicit lookahead, value propagation, and limited commitment in a single model, allowing downstream outcomes to influence early decisions. Across multiple benchmarks, agent frameworks, and LLM backbones, FLARE consistently improves task performance and planning-level behavior, frequently allowing LLaMA-8B with FLARE to outperform GPT-4o with standard step-by-step reasoning. These results establish a clear distinction between reasoning and planning.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。