Chess

Benchmarks

Task Name	Dataset Name	SOTA Result
Classification	chess	F1 Score81.19	26
Safe Move Prediction	Chess (Exploratory Learning Phase)	Blunder Rate24.11	16
Classification	chess	Accuracy85.4	12
Tool Selection	Chess Skill: beginner, intermediate, advanced	Accuracy100	10
Tool Selection	Chess Specialists: opening, midgame, endgame, late-endgame	Accuracy64.4	10
Chess Position Evaluation	Chess 1,002 disjoint positions	Mean Evaluation (cp)-243	5
Extrapolation	Chess (8, 8) (test)	Accuracy84.3	4
Board state reconstruction	Chess	Coverage98	3
Chess	Chess Match vs Stockfish 100 games	Win Rate25	2
Instruction Recovery	Chess O3	Extracted Instructions126,516,521	1
Instruction Recovery	Chess O2	Extracted Instructions Count1,905,184	1
Instruction Recovery	Chess O0	Extracted Instructions Count8,568,798	1

Showing 12 of 12 rows