Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Retrieval-Augmented Generation on TheoremQA
Loading...
66.3
Accuracy
Ours
25.22
35.885
46.55
57.215
Feb 9, 2026
Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Ours
Model=Qwen3-8B, Retrie...
2026.02
66.3
ReFeedL
Model=Qwen3-8B, Retrie...
2026.02
65.1
InstructRAG
Model=Qwen3-8B, Retrie...
2026.02
64.4
DPR
Model=Qwen3-8B, Retrie...
2026.02
63.6
BM25
Model=Qwen3-8B, Retrie...
2026.02
62.8
Zero
Model=Qwen3-8B, Retrie...
2026.02
54.6
Ours
Model=Llama-3.1-8B, Re...
2026.02
40.8
ReFeedL
Model=Llama-3.1-8B, Re...
2026.02
40
InstructRAG
Model=Llama-3.1-8B, Re...
2026.02
39.2
DPR
Model=Llama-3.1-8B, Re...
2026.02
38.4
BM25
Model=Llama-3.1-8B, Re...
2026.02
37.6
Zero
Model=Llama-3.1-8B, Re...
2026.02
26.8
Feedback
Search any
task
Search any
task
