评测新型推理模型o1的规划能力，发现其效果提升但成本高，结合验证器可保证结果正确性。
Planning in Strawberry Fields: Evaluating and Improving the Planning and Scheduling Capabilities of LRM o1
- 用外部验证器构建LRM-Modulo系统，提升规划输出正确性
- o1模型在规划与调度任务上优于传统自回归LLM，但推理开销大
- 适合关注AI推理可靠性与可验证性的研究者和开发者
规划能力被视为智能体的核心素养，一直是人工智能研究的重要方向。随着大语言模型（LLMs）的发展，学界广泛关注其是否具备规划能力，但自GPT3以来进展缓慢。OpenAI宣称其最新o1（Strawberry）模型为专门设计的大型推理模型（LRM），突破了传统自回归模型的局限。本文评估了两个LRM（o1-preview和o1-mini）在规划与调度基准上的表现。结果显示，尽管o1相比传统模型有显著提升，但推理成本极高，且无法保证生成结果的正确性。我们进一步提出一种名为LRM-Modulo的系统，通过将o1与外部验证器结合，不仅确保输出正确性，还进一步提升了性能。
原文摘要 · Abstract (English)
The ability to plan a course of action that achieves a desired state of affairs has long been considered a core competence of intelligent agents and has been an integral part of AI research since its inception. With the advent of large language models (LLMs), there has been considerable interest in the question of whether or not they possess such planning abilities, but -- despite the slew of new private and open source LLMs since GPT3 -- progress has remained slow. OpenAI claims that their recent o1 (Strawberry) model has been specifically constructed and trained to escape the normal limitations of autoregressive LLMs -- making it a new kind of model: a Large Reasoning Model (LRM). In this paper, we evaluate the planning capabilities of two LRMs (o1-preview and o1-mini) on both planning and scheduling benchmarks. We see that while o1 does seem to offer significant improvements over autoregressive LLMs, this comes at a steep inference cost, while still failing to provide any guarantees over what it generates. We also show that combining o1 models with external verifiers -- in a so-called LRM-Modulo system -- guarantees the correctness of the combined system's output while further improving performance.
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。