Mathematical Reasoning on GSM-Symbolic (Accuracy and Latency)
90AccuracyExpGraph
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| ExpGraphBackbone Model=Llama-3.1-8B-Instruct (Large LLM), Method Category=LLM-Centric Experience Learning Baselines2026.05 | 90 | — | — | |
| StreamingThinkerParadigm=Streaming, Reasoning Depth=D3, Model=Qwen3-4B2025.10 | 85.6 | 20.77 | 9.768 | |
| BatchParadigm=Batch, Model=Qwen3-4B2025.10 | 85.5 | 94.74 | 47.7 | |
| InterleavedParadigm=Interleaved, Reasoning Depth=D3, Model=Qwen3-4B2025.10 | 84.3 | 20.77 | 15.46 | |
| StreamingThinkerParadigm=Streaming, Reasoning Depth=D2, Model=Qwen3-4B2025.10 | 84.2 | 20.77 | 5.973 | |
| InterleavedParadigm=Interleaved, Reasoning Depth=D2, Model=Qwen3-4B2025.10 | 82.9 | 20.77 | 11.9 | |
| ExpGraphBackbone Model=Llama-3.2-3B-Instruct (Small LLM), Method Category=LLM-Centric Experience Learning Baselines2026.05 | 82 | — | — | |
| StreamingThinkerParadigm=Streaming, Reasoning Depth=D1, Model=Qwen3-4B2025.10 | 42.1 | 20.77 | 0.66 | |
| InterleavedParadigm=Interleaved, Reasoning Depth=D1, Model=Qwen3-4B2025.10 | 41 | 20.77 | 6.3 |