让模型自己从零写ASP规则，Claude、GPT-5等九个模型限时1小时。CLEVR有人拿满分，GQA上GPT-5跌到41.8%。
一项新研究让大语言模型在1小时内从空文件开始编写ASP理论，并用求解器在环评估。在CLEVR、GQA、CLEVRER三个VQA基准上，Claude Sonnet 4.6、Claude Opus 4.7、DeepSeek V4 Pro均达CLEVR 100%、GQA 92.8%-98.8%。GPT-5在CLEVR达98.7%，但在GQA仅41.8%。加入其他数据集的手写参考理论后，GPT-5准确率下降3-19个百分点，而其他前沿模型变化不超过±3.4个百分点。研究者已开源代码、提示词和蒸馏出的理论。
Distilling Answer Set Programming Theories from Large Language Models
Writing Answer Set Programming (ASP) theories from scratch is a difficult and time-consuming task. We take a neurosymbolic approach to study whether a model can distill complete and correct theories, given a fixed agent harness with the solver in the loop. The protocol is dataset-agnostic: with a single prompt and an empty file as the starting point the model is given a 1-hour time limit to derive a complete theory. We chose VQA as the application domain, three benchmarks (CLEVR, GQA, CLEVRER), as these are publicly available and non-trivial. In order to study the model scale required for solving this task we nine different models: four frontier (Claude Sonnet 4.6, Claude Opus 4.7, GPT-5, DeepSeek V4 Pro), two mid-tier (DeepSeek V4 Flash, gpt-oss-120b), and three open-weights (qwen3.6-27b, gpt-oss-20b, qwen3.5-9b). Three of four frontier models reach 100% on CLEVR and 92.8%-98.8% on GQA; on CLEVRER, Sonnet, Opus, DeepSeek V4 Pro score 92.7%-95.3%. GPT-5 reaches 98.7% on CLEVR but drops to 41.8% on GQA and to 86.7% on CLEVRER. Adding handwritten reference theories from other datasets moves the other three frontier models by at most +/-3.4 pp but reduces GPT-5's accuracy by 3-19 pp. We release the code, prompts, and theories distilled.