Share your thoughts, 1 month free Claude Pro on usSee more

Mathematical Reasoning on Math (AMC23, Math500, Minerva, Olympiad Breakdown)

78.6Math500 Score

HEAL

Updated 1mo ago

Evaluation Results

Method	Links
HEAL 2026.04		78.6	59.69	38.6	41.48	54.59
Full-shot 2026.04		78.4	63.75	35.66	42.96	55.19
Search-o1 2025.08		78.2	-	60.3	-	69.3
HEAL 2026.04		77.6	52.81	35.29	39.11	51.52
Full-shot 2026.04		77.4	54.37	33.82	38.81	51.1
Few-shot 2026.04		77.2	58.13	34.56	41.04	52.73
Self-Ask 2025.08		74.9	-	57.7	-	66.3
Hybrid 2026.04		74.4	51.88	32.72	38.2	49.3
RAT 2025.08		74.4	-	55.5	-	65
Hybrid 2026.04		73.8	54.06	29.04	39.11	49
Few-shot 2026.04		73.4	50	31.98	37.48	48.22
Search-o1 2025.08		69.4	-	43	-	56.2
UR2 2025.08		69.4	-	45	-	57.2
Only-General 2026.04		69.2	47.81	24.26	34.37	43.91
Only-General 2026.04		68.6	53.1	23.9	37.63	45.81
Vanilla RL 2025.08		68	-	43.9	-	52.4
RAT 2025.08		67.9	-	44.9	-	56.4
Full-shot 2026.04		66.2	35.31	19.12	29.48	37.53
Qwen3-4B-Base 2026.04		66.2	42.19	20.22	34.67	40.82
Standard RAG 2025.08		65.3	-	40.8	-	53.1
HEAL 2026.04		64.4	38.8	21.69	25.78	37.67
CoT 2025.08		63.6	-	39.9	-	51.8
Qwen3-8B-Base 2026.04		63	45.63	18.01	33.33	39.99
Only-General 2026.04		62.8	29.38	16.91	26.07	33.79
Hybrid 2026.04		60.2	32.5	17.65	24.15	33.63
Few-shot 2026.04		58.4	34.06	17.28	23.7	33.36
Self-Ask 2025.08		58.3	-	40.3	-	49.3
Qwen3-1.7B-Base 2026.04		55.6	30.94	13.24	21.48	30.32
UR2 2025.08		54.5	-	39	-	46.8
RAT 2025.08		50.1	-	36.8	-	43.5
HEAL 2026.04		49.8	28.44	18.75	17.19	28.55
Full-shot 2026.04		49.4	30.42	22.06	17.04	29.73
CoT 2025.08		48.4	-	34.4	-	41.4
Only-General 2026.04		47.6	25.31	16.91	17.33	26.79
Self-Ask 2025.08		46.9	-	27	-	37
Hybrid 2026.04		46.8	26.25	18.01	16.59	26.91
Few-shot 2026.04		46	26.88	19.49	17.48	27.46
Vanilla RL 2025.08		45.5	-	43.4	-	44.4
Standard RAG 2025.08		45	-	31.4	-	38.2
Search-o1 2025.08		41.5	-	27.8	-	34.7
LLaMA3.2-3B-Instruct 2026.04		40.8	25	15.81	13.19	23.7