大模型因误判自身能力而提前放弃，导致解题失败。
Lost in Context: Addressing Context Anxiety in Large Language Models
- 发现模型因高估任务所需上下文长度而产生自我怀疑。
- 自信心不足导致实际效率下降，即使有能力完成任务。
- 通过训练新策略可避免焦虑，提升长程问题解决能力。
传统观点认为推理模型在超出其能力范围时会失败。然而我们发现，前沿推理模型有时具备解决问题的能力，却因过早的自我怀疑——即所谓的‘上下文焦虑’——而失败。我们首次系统研究了这一现象，表明其部分原因在于模型无法准确估算完成任务所需的标记数量。同时发现，上下文焦虑会导致模型在感知到约束时出现实际效率损失。基于此分析，我们进一步证明，模型可通过学习替代策略解决长周期问题而不产生上下文焦虑，提示性能提升未必依赖模型能力扩展，而是可通过改善模型对自身局限性的准确评估与适应能力实现。
原文摘要 · Abstract (English)
Conventional wisdom suggests that reasoning models fail when problems exceed their capabilities. However, we find that frontier reasoning models sometimes possess the necessary capabilities to solve problems but fail due to premature self-doubt -- a phenomenon informally known as context anxiety. We provide the first systematic study of context anxiety, demonstrating that it arises, in part, from a model's inability to accurately estimate the tokens required to complete a task. We also show that context anxiety leads to material efficiency losses when models operate under perceived constraints. Building on this analysis, we further show that models can learn alternative strategies for solving long-horizon problems without exhibiting context anxiety, suggesting that performance improvements may be achievable not through scaling model capabilities, but by improving models' ability to accurately assess and adapt to their own limitations.
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。