MS-to-SMILES prediction on MassSpecGym (test)
1Think RateClaude-3.5-Sonnet
Evaluation results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Claude-3.5-SonnetPrompting=Chain-of-Thought (CoT)2026.01 | 1 | 0.9996 | 0.937 | 0.0394 | 0 | 0 | 0 | 0.09 | 0.1 | 0.15 | 0.14 | |
| Llama-3-70BPrompting=Chain-of-Thought (CoT)2026.01 | 0.9783 | 0.5114 | 0.3812 | 0.041 | 0.0002 | 0 | 0 | 0.1 | 0.12 | 0.17 | 0.14 | |
| GPT-4o-miniPrompting=Chain-of-Thought (CoT)2026.01 | 0.9707 | 0.8117 | 0.7541 | 0.0391 | 0.0002 | 0 | 0 | 0.09 | 0.12 | 0.15 | 0.07 | |
| Llama-3-8BPrompting=Chain-of-Thought (CoT)2026.01 | 0.6834 | 0.0221 | 0.0144 | 0.0422 | 0 | 0 | 0 | 0.09 | 0.1 | 0.17 | 0.16 |