SciIR-Bench

What is SciIR-Bench?

SciIR-Bench evaluates text-to-image models on scientific image reasoning across three semiotic-aligned tracks: entity structure, scientific process, and…

Released

2026-06-29

Evaluates

General AI, Reasoning, Vision & 3D

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2606.30124

SciIR-Bench paper and code

• Paper https://arxiv.org/abs/2606.30124
• Code https://github.com/MAIR-Lab-HUST/SciIR

Which sources cite SciIR-Bench?

• SciIR-Bench

Related General AI, Reasoning, Vision & 3D benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 3DCodeBench
• 5-Dialects-BN
• A-OKVQA
• A3S-Bench
• AARRI-Bench
• ActBench
• ActiveFly-Bench