Benchmark Almanac

South China University of Technology

We track 6 benchmarks that South China University of Technology made or helped make.

0 in use 5 with public code 5 with a leaderboard here

Benchmarks

Name Kind Released Best score Status
MCS-Bench Models read seal inscriptions, identify calligraphy, and answer questions about Chinese paintings and artifacts. The overall score averages task scores. General visual questions Feb 2025 49.3 Not in cards
M5HisDoc Models locate text, recognize characters, and recover reading order on Chinese historical pages. The hard subset distorts the same 4,000 pages as the regular subset. Documents, charts and text in images Jun 2023 64.5% Not in cards
MTHv2 Models read Chinese Buddhist texts from scanned historical pages. The authors annotate characters, line transcriptions, reading order, and boundary lines. Documents, charts and text in images Jun 2020 86.4% Not in cards
FinHopper Agents research financial facts through up to six dependent searches. Intermediate answer keys show where an incorrect company, date, or indicator derails the search. Finance Oct 2026 67.3% Not in cards
EgoProceVQA Models watch first-person clips and answer questions about the steps in a procedure. Some questions ask when a step occurred or which step is missing. Video understanding Jul 2026 51.6% Not in cards
ClothDojo Robot policies use both hands to flatten and fold garments. The simulator checks success on familiar garments and garments withheld from training. Simulated robots Oct 2026 44.7% Not in cards