让大模型先验证答案，能低成本提升推理能力。
Asking LLMs to Verify First is Almost Free Lunch
- 先让模型验证一个随机答案，再生成解法，反向缩小推理范围。
- 在GPQA-Diamond上用Gemini-3-Pro-Preview达到94.9%准确率，错误降低约30%。
- 无需训练或大量采样，适合快速部署到现有大模型中。
为在不增加训练成本或大量测试时采样的情况下提升大语言模型（LLM）的推理能力，我们提出验证优先（Verification-First, VF）策略：提示模型在生成解法前，先验证一个提供的候选答案（即使为随意答案）。该方法触发与标准链式思维（CoT）相反的“逆向推理”过程，通过修剪模型输出分布来限制逻辑搜索空间。我们进一步将VF推广为迭代验证优先（Iter-VF），一种序列化测试时缩放（TTS）方法，通过循环使用模型先前答案进行验证与生成。在多个基准和多种LLM上的实验表明，使用随机答案的VF提示始终优于标准CoT，且计算开销极小；Iter-VF优于现有TTS策略。该方法在当前最优思维模型上也有效。例如，仅用简单VF提示，就在GPQA-Diamond上实现94.9%准确率（相比原模型错误减少约30%），并刷新了该数据集上Gemini-3-Pro-Preview的记录。
原文摘要 · Abstract (English)
To enhance the reasoning capabilities of Large Language Models (LLMs) without high costs of training, nor extensive test-time sampling, we introduce Verification-First (VF), a strategy that prompts models to verify a provided candidate answer, even a trivial or random one, before generating a solution. This approach triggers a "reverse reasoning" process complementary to standard forward Chain-of-Thought (CoT), which restricts the logical search space of the answer by pruning the LLM's output distribution. We further generalize VF prompting to Iter-VF, a sequential test-time scaling (TTS) method that iteratively cycles the verification-generation process using the model's previous answer. Extensive experiments across various benchmarks and various LLMs confirm that VF prompting with random answer consistently outperforms standard CoT with minimal computational overhead, and Iter-VF outperforms existing TTS strategies. VF is also effective on SOTA thinking models. For example, by using the simple VF prompting, we obtain a new SOTA 94.9% accuracy on GPQA-Diamond with Gemini-3-Pro-Preview where VF reduces its errors by ~30% relatively.
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。