MatBench/results/20250602_1706/summary.csv

1	Model	accuracy	precision_micro	recall_micro	f1_micro	precision_macro	recall_macro	f1_macro	Data Count
2	qwen-max-2025-01-25	0.6446700507614214	0.6336633663366337	0.649746192893401	0.6416040100250626	0.6388760049474336	0.6501020408163265	0.64232342205538	197
3	gpt-4o	0.5482233502538071	0.5618556701030928	0.5532994923857868	0.5575447570332481	0.5779088050314465	0.5536734693877551	0.5600088997453159	197
4	deepseek-chat	0.6700507614213198	0.676923076923077	0.6700507614213198	0.673469387755102	0.6899114693446089	0.6705102040816326	0.6754210676562946	197
5	claude-sonnet-4-20250514	0.700507614213198	0.6934673366834171	0.700507614213198	0.696969696969697	0.7072180484244438	0.7009183673469388	0.69833034513671	197