Scientific QA on Name conversion s2i
50Split MatchSciReasoner
Evaluation Results
| Method | Links | |
|---|---|---|
| SciReasoner2026.07 | 50 | |
| Opus-4.72026.07 | 34 | |
| GPT-5.52026.07 | 7 | |
| Kimi-K2.62026.07 | 2 | |
| DeepSeek-V4-Pro2026.07 | 1 |
| Method | Links | |
|---|---|---|
| SciReasoner2026.07 | 50 | |
| Opus-4.72026.07 | 34 | |
| GPT-5.52026.07 | 7 | |
| Kimi-K2.62026.07 | 2 | |
| DeepSeek-V4-Pro2026.07 | 1 |