发现大模型生成事实能力弱于验证，且验证更稳定。
The Future of Facts: Tracing the Factual Generation-Verification Gap
- 分阶段追踪模型训练中生成与验证能力演化
- 验证能力始终早于生成，且更抗持续学习干扰
- 更新知识后模型可能同时接受新旧答案，适合研究可信生成
语言模型正成为获取事实知识的默认接口，但其验证输出的可靠性远高于生成。这种生成-验证差距（GV-gap）是自提升和推理进展的基础，但其在事实知识上的具体机制仍不清晰。本文聚焦训练过程对事实类GV-gap的影响，区分其与计算和审美类差距。通过四个开源模型家族、两种规模，在三个训练阶段（获取、持续学习、更新）中追踪生成与验证能力。三个发现反复出现：(i) 验证能力始终先于生成被学习；(ii) 验证比生成更耐持续学习干扰；(iii) 事实更新后模型可能进入“多宇宙”状态，同时认为新旧答案均正确。对前沿模型的自然实验在大规模下重现这些动态，并揭示对常见事实仍存在残留验证偏差。
原文摘要 · Abstract (English)
Language models are becoming the default interface to factual knowledge, yet they often verify outputs more reliably than they generate them. This generation-verification gap (GV-gap) underlies many recent advances in self-improvement and reasoning, but its dynamics on factual knowledge specifically remain poorly understood. We focus on the training mechanisms underlying factual GV-gaps, distinguishing them from their computational and aesthetic counterparts. We trace generation and verification capabilities through three training phases (acquisition, continual learning, and updating) across four open-source model families at two scales each. Three findings recur across models: (i) verification is consistently learned before generation; (ii) verification is more robust to continual learning than generation; and (iii) factual updates can leave models in a "multi-verse" state, simultaneously verifying both old and new answers as correct. Natural experiments on frontier models reproduce these dynamics at scale and reveal residual verification biases on well-covered facts.
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。