SOTAVerified

Mathematical Question Answering

Building systems that automatically answer mathematical questions.

Papers

Showing 1–10 of 11 papers

TitleStatusHype
Mathematical Information Retrieval: Search and Question Answering—0
MathChat: Benchmarking Mathematical Reasoning and Instruction Following in Multi-Turn InteractionsCode1
GOLD: Geometry Problem Solver with Natural Language DescriptionCode1
GAPS: Geometry-Aware Problem Solver—0
AlignedCoT: Prompting Large Language Models via Native-Speaking DemonstrationsCode0
Mining Mathematical Documents for Question Answering via Unsupervised Formula LabelingCode1
Plane Geometry Diagram ParsingCode1
IconQA: A New Benchmark for Abstract Diagram Understanding and Visual Language ReasoningCode1
Inter-GPS: Interpretable Geometry Problem Solving with Formal Language and Symbolic ReasoningCode1
Analysing Mathematical Reasoning Abilities of Neural ModelsCode0
Show:102550
← PrevPage 1 of 2Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1Human ExpertAccuracy (%)90.9—Unverified
2Inter-GPS (GT)Accuracy (%)78.3—Unverified
3PGDPNetAccuracy (%)74.1—Unverified
4GOLDAccuracy (%)69.1—Unverified
5GAPSAccuracy (%)68—Unverified
6Inter-GPSAccuracy (%)57.5—Unverified
7HumanAccuracy (%)56.9—Unverified
8RandomAccuracy (%)25—Unverified
#ModelMetricClaimedVerifiedStatus
1Inter-GPSAccuracy (%)67—Unverified
2GEOSAccuracy (%)49—Unverified