作者位次 #7/8ACL 2026Poster· Natural Language Processing
G eo L aux: A Benchmark for Evaluating MLLM s’ Geometry Performance on Long-Step Problems Requiring Auxiliary Lines
Yumeng Fu · Jiayin Zhu · Lingling Zhang · Wenjun Wu · Bo Zhao · Shaoxuan Ma · Yushun Zhang · an Jiaotong University)
摘要
Geometry problem solving (GPS) poses significant challenges for Multimodal Large Language Models (MLLMs) in diagram comprehension, knowledge application, long-step reasoning, and auxiliary line construction. However, current benchmarks lack fine-grained evaluation for long-step problems necessita...