Theorem Proving on Olympiad
8Solved RateDreamProver (Gemini 2.5 Pro)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DreamProver (Gemini 2.5 Pro)Backbone LLM=Gemini 2.5 Pro, Lemma Learning System=True, Pass@k=pass@322026.04 | 8 | 42.1 | |
| DreamProver (GPT-5.3-Codex)Backbone LLM=GPT-5.3-Codex, Lemma Learning System=True, Pass@k=pass@322026.04 | 7 | 36.8 | |
| DreamProver (Gemini 3.1 Pro)Backbone LLM=Gemini 3.1 Pro, Lemma Learning System=True, Pass@k=pass@322026.04 | 7 | 36.8 | |
| Gemini 3.1 ProBackbone LLM=Gemini 3.1 Pro, Lemma Learning System=False, Pass@k=pass@322026.04 | 1 | 5.3 | |
| GPT-5.3-CodexBackbone LLM=GPT-5.3-Codex, Lemma Learning System=False, Pass@k=pass@322026.04 | 0 | 0 | |
| Gemini 2.5 ProBackbone LLM=Gemini 2.5 Pro, Lemma Learning System=False, Pass@k=pass@322026.04 | 0 | 0 |