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

Medical Question Answering on MedQA (Accuracy)

86.5Accuracy

Med-PaLM 2

Updated 1mo ago

Evaluation Results

Method	Links
Med-PaLM 2 2026.04		86.5
GPT-4-base 2026.04		86.1
Hybrid-IR 2026.06		82.9
MedRAG + GPT-4 2026.04		82.8
i-MedRAG 2026.06		80.5
GPT-4o-mini 2026.06		80.3
MEDVRAG (full system) 2026.04		79.4
RecursiveMAS 2026.04		79.3
GFM-RAG 2026.06		78.5
MEDVRAG (+ iterative reasoning) 2026.04		77.6
TextGrad 2026.04		77.2
Single Agent 2026.04		77
Recursive-TextMAS 2026.04		77
Single Agent 2026.04		76.1
MEDVRAG (single-round) 2026.04		75.8
MedRAG 2026.06		75.5
HippoRAG 2026.06		75.1
MEDVRAG (text-only retrieval) 2026.04		74.8
MedCPT 2026.06		74.3
SFT-Oracle 2026.05		73.99
HuatuoGPT-o1-8B 2026.04		72.6
Yi-34B (OpenMedLM) 2026.04		72.6
Hybrid-IR 2026.06		72.2
i-MedRAG 2026.06		71.8
Qwen3-30B 2026.04		71.5
MedGraphRAG 2026.06		71.5
MEDVRAG (no RAG) 2026.04		71.4
SYNDATA 2026.05		70.62
HSIR-DPO 2026.05		70.46
Meditron-70B 2026.04		70
HSIR-SFT 2026.05		68.74
HSIR-DPO 2026.05		68.58
SYNDATA 2026.05		68.12
H-GRPO 2026.05		68.03
HSIR-SFT 2026.05		67.87
HSIR-SFT 2026.05		67.32
HSIR-DPO 2026.05		67.32
AR-OPD 2026.06		66.7
MedRAG + GPT-3.5 2026.04		66.6
GRPO 2026.05		66.46
Long2Short 2026.05		66.14
SYNDATA 2026.05		66.13
SFT 2026.06		66
ReSTEM 2026.05		65.28
ReSTEM 2026.05		65.28
Full OPD 2026.06		65.2
RFT 2026.05		64.89
SYNDATA 2026.05		64.7
SYNDATA 2026.05		64.57
Partial OPD 2026.06		64.3
IRPO 2026.05		64.26
ReSTEM 2026.05		64.18
CosFn 2026.05		63.71
Qwen3-8B 2026.04		63.5
IRPO 2026.05		63.24
SFT-Initial 2026.05		62.45
ReGenesis 2026.05		62.22
STaR 2026.05		61.59
STaR 2026.05		61.19
STaR 2026.05		61.04
HSIR-DPO 2026.05		60.64
IRPO 2026.05		60.57
GFM-RAG 2026.06		60.3
GPT-3.5 2026.04		60.2
HSIR-DPO 2026.05		59.15
SYNDATA 2026.05		59.02
H-GRPO 2026.05		58.98
Base 2026.06		58.7
SFT-Oracle 2026.05		58.68
Long2Short 2026.05		58.21
HippoRAG 2026.06		57.9
MedGraphRAG 2026.06		57.8
Llama-3-8B-Instruct 2026.06		57.7
HSIR-SFT 2026.05		57.58
Base Model 2026.05		57.52
Mixture-of-Agents (MoA) 2026.04		57.5
GRPO 2026.05		57.03
POP 2026.04		56.54
ReSTEM 2026.05		56.48
LoopLM 2026.04		56.4
MedRAG 2026.06		56.4
HSIR-SFT 2026.05		56.32
ReSTEM 2026.05		56.25
Base 2026.04		55.94
HSIR-DPO 2026.05		55.77
HSIR-SFT 2026.05		55.7
MedCPT 2026.06		55.3
CosFn 2026.05		55.22
ReSTEM 2026.05		55.22
RFT 2026.05		54.6
POP 2026.04		53.86
PRL 2025.05		53.34
Train on D 2026.04		53.12
Base 2026.04		52.95
GRACE 2025.05		52.26
ReConcile 2025.08		52
GA 2025.05		51.95
PromptAgent 2025.05		51.87
PromptWizard 2025.05		51.84
DE 2025.05		51.76

Showing 100 of 145 rows