What Looks Like a Capability Limit in Vision-Language Models
arXiv:2609.27408v1 Announce Type: cross Abstract: Benchmarks for vision-language models offer their answer choices in so
推荐理由：研究进展具备参考价值，值得关注
arXiv:2609.27408v1 Announce Type: cross Abstract: Benchmarks for vision-language models offer their answer choices in so
推荐理由：研究进展具备参考价值，值得关注