A recognition-versus-reasoning divide: scene recognition is comparatively mature while affective interpretation and compositional reasoning lag¶

Claim · All claims
Evidence · 1 study · design r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly

Subclaims¶

q2 i? On MUSE, Scene Classification is comparatively mature (23 of 30 models exceed 75.0, median 81.0), whereas Emotion Detection, Relative Position, Remote Interaction, and Jigsaw Puzzle show substantially lower medians, indicating persistent limitations in affective interpretation and compositional reasoning. → Luyao Zhu 2026

Evidence¶

Luyao Zhu 2026¶

Luyao Zhu, Xun Wei Yee, Wei Li, Mun Thye Mak, Wee Siong Ng. (2026). MUSE: Benchmarking Large Vision-Language Models on Multi-modal Understanding in Situated Education. arXiv preprint. https://arxiv.org/abs/2609.19088 doi:10.48550/arxiv.2609.19088

q2 · i? · design · r2

Descriptive analysis of the 30-model results in Table 2. The article reports the Scene Classification median of 81.0 and states that Emotion Detection, Relative Position, Remote Interaction, and Jigsaw Puzzle "exhibit substantially lower medians, revealing persistent limitations in affective interpretation, and compositional reasoning."

"Scene Classification is comparatively mature, with 23 of 30 models exceeding 75.0 and a median score of 81.0."

Discussion¶

Related Claims¶

• Human performance forms the outer envelope on nearly all MUSE tasks, with the largest deficits in Relative Position, Emotion Detection, and Jigsaw Puzzle — related
• MUSE tasks form coherent but non-redundant capability groups, with Jigsaw Puzzle a distinct compositional capability — related
• VLM performance on artistic educational imagery is highly task-dependent, with no model dominating across capabilities — related
• Performance on existing benchmarks transfers unevenly to artistic educational imagery — related