List your company

Physics on TheoremQA

58.8AccuracyTroVE
29.47237.08644.752.314May 27, 2025

Evaluation results

MethodLinks
TroVEModel Backbone=Gemini-1.5-Pro2025.0558.8—
PoT + REFTOOLModel Backbone=Gemini-1.5-Pro2025.0558.8—
PoTModel Backbone=Gemini-1.5-Pro2025.0557.9—
PoT + RAGModel Backbone=GPT-4o2025.0557.9—
PoT + REFTOOLModel Backbone=GPT-4o2025.0557.9—
CreatorModel Backbone=Gemini-1.5-Pro2025.0557—
PoTModel Backbone=GPT-4o2025.0557—
PoT + RAGModel Backbone=Gemini-1.5-Pro2025.0557—
PoT + REFTOOLModel Backbone=Llama-3.1-70B2025.0553.5—
Physics ReasonerModel Backbone=Gemini-1.5-Pro2025.0550.9—
PoT + REFTOOLModel Backbone=GPT-42025.0549.1—
TroVEModel Backbone=GPT-4o2025.0548.2—
Physics ReasonerModel Backbone=Llama-3.1-70B2025.0548.2—
PoTModel Backbone=Llama-3.1-70B2025.0548.2—
PoTModel Backbone=GPT-42025.0545.6—
PoT + RAGModel Backbone=Llama-3.1-70B2025.0544.7—
PoT + RAGModel Backbone=GPT-42025.0544.7—
Physics ReasonerModel Backbone=GPT-42025.0542.1—
CreatorModel Backbone=Llama-3.1-70B2025.0540.4—
CreatorModel Backbone=GPT-4o2025.0540.4—
LATMModel Backbone=GPT-42025.0539.8—
LATMModel Backbone=Llama-3.1-70B2025.0538.9—
CreatorModel Backbone=GPT-42025.0535.1—
TroVEModel Backbone=GPT-42025.0535.1—
LATMModel Backbone=Gemini-1.5-Pro2025.0533.3—
TroVEModel Backbone=Llama-3.1-70B2025.0533.3—
Physics ReasonerModel Backbone=GPT-4o2025.0533.3—
LATMModel Backbone=GPT-4o2025.0530.6—
Creator2025.05—43.2
LATM2025.05—35.7
Physics Reasoner2025.05—43.6
PoT2025.05—52.2
PoT + RAG2025.05—51.1
PoT + REFTOOL2025.05—54.8
TroVE2025.05—43.9