LongPerceptualThoughts：为系统1感知提炼系统2推理
LongPerceptualThoughts: Distilling System-2 Reasoning for System-1 Perception
AI总结：
本文提出LongPerceptualThoughts数据集及三阶段合成框架，为感知任务生成长思维链，训练7B模型在视觉基准上平均提升3.4点，并改善文本推理性能。
摘要：
最近的推理模型通过测试时扩展（test-time scaling）表明，长思维链（chain-of-thoughts）可以在数学和代码等困难推理任务中解锁显著的性能提升。然而，这种长思维在其他领域（如感知任务）中对系统2推理的益处相对较少被探索，在这些任务中，较浅的系统1推理似乎已经足够。在本文中，我们引入了LongPerceptualThoughts，这是一个新的合成数据集，包含30K条用于感知任务的长思维轨迹。为感知任务合成精细推理思维的关键挑战在于，现成的模型尚不具备这种思考行为，并且为感知任务构建可靠的过程验证器并不直接。因此，我们提出了一种新颖的三阶段数据合成框架：首先从密集图像描述中合成可验证的多选题，然后从视觉语言模型（VLMs）中为这些可验证问题提取简单的思维链（CoTs），最后通过前沿推理模型将这些简单思维扩展为精细的长思维。在与强指令微调的7B模型进行的受控实验中，我们展示了相对于现有视觉推理数据生成方法的显著改进。我们在生成的数据集上训练的模型，在5个以视觉为中心的基准上平均提高了+3.4个百分点，其中在V$^*$ Bench上提高了+11.8个百分点。值得注意的是，尽管该模型针对视觉任务进行了调整，它在文本推理基准MMLU-Pro上的性能也提高了+2个百分点。
英文摘要：
Recent reasoning models through test-time scaling have demonstrated that long chain-of-thoughts can unlock substantial performance boosts in hard reasoning tasks such as math and code. However, the benefit of such long thoughts for system-2 reasoning is relatively less explored in other domains such as perceptual tasks where shallower, system-1 reasoning seems sufficient. In this paper, we introduce LongPerceptualThoughts, a new synthetic dataset with 30K long-thought traces for perceptual tasks. The key challenges in synthesizing elaborate reasoning thoughts for perceptual tasks are that off-the-shelf models are not yet equipped with such thinking behavior and that it is not straightforward to build a reliable process verifier for perceptual tasks. Thus, we propose a novel three-stage data synthesis framework that first synthesizes verifiable multiple-choice questions from dense image descriptions, then extracts simple CoTs from VLMs for those verifiable problems, and finally expands those simple thoughts to elaborate long thoughts via frontier reasoning models. In controlled experiments with a strong instruction-tuned 7B model, we demonstrate notable improvements over existing visual reasoning data-generation methods. Our model, trained on the generated dataset, achieves an average +3.4 points improvement over 5 vision-centric benchmarks, including +11.8 points on V$^*$ Bench. Notably, despite being tuned for vision tasks, it also improves performance on the text reasoning benchmark, MMLU-Pro, by +2 points.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
