DocAttriBench

What is DocAttriBench?

Evaluates source attribution in Document VQA by grounding answers to layout elements such as text blocks, tables, and images.

Released

2026-09-17

Evaluates

General AI, Vision & 3D, cs.CV

Openness

unknown

Importer

Claire Radar

Review status

ai-reviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2609.20574

DocAttriBench paper

• Paper https://arxiv.org/abs/2609.20574
• Website https://aimagelab.github.io/DocAttriBench/

Which sources cite DocAttriBench?

• DocAttriBench

Related General AI, Vision & 3D, cs.CV benchmarks

• $\tau^\tau$-Bench: An Environment for End-To-End, Realistic Agent Construction
• 2d-fet-bench
• 360CityArena
• 3DCodeBench
• 5-Dialects-BN
• A Real-World CAD Segmentation Benchmark
• A3S-Bench
• AARRI-Bench