Benchmark Almanac

IIIT Hyderabad

We track 6 benchmarks that IIIT Hyderabad made or helped make.

2 in use 2 with public code 2 with a leaderboard here

Benchmarks

Name Kind Released Best score Status
DocVQA Models answer 50,000 questions about 12,000 scanned document images by reading their text and layout. Scoring is ANLS, a string similarity to reference answers. Documents, charts and text in images Jul 2020 96.9% In use
InfographicVQA Models answer about 30,000 questions on about 5,000 infographics that mix text, layout, charts, and graphics. Scoring is ANLS. Documents, charts and text in images Apr 2021 92.6% In use
CoST English-Hindi Models translate 1,947 English sentences into Hindi across eight genres, including plays, poetry and conversation. The organizers report BLEU alongside reference-based and reference-free translation metrics. Translation Jun 2024 79.1 Not in cards
ADQA Models describe film scenes. An evaluator answers questions from the descriptions and dialogue, then checks whether the context supports each correct answer. Video captioning Oct 2025 Not in cards
EgoProceL Models find recurring steps in first-person videos and mark when each step occurs. The videos show 16 tasks, including cooking and assembling computers. Video understanding Jul 2022 Not in cards
Oxford-IIIT Pet Dataset Models identify the breed of a cat or dog from a photograph. Meta measures breed classification across 37 categories. General visual questions Not in cards