Theorem Proving on High School Competition
9Solved RateDreamProver (Gemini 3.1 Pro)
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| DreamProver (Gemini 3.1 Pro)Backbone LLM=Gemini 3.1 Pro, Lemma Learning System=True, Pass@k=pass@322026.04 | 9 | 90 | |
| DreamProver (GPT-5.3-Codex)Backbone LLM=GPT-5.3-Codex, Lemma Learning System=True, Pass@k=pass@322026.04 | 7 | 70 | |
| DreamProver (Gemini 2.5 Pro)Backbone LLM=Gemini 2.5 Pro, Lemma Learning System=True, Pass@k=pass@322026.04 | 7 | 70 | |
| Gemini 3.1 ProBackbone LLM=Gemini 3.1 Pro, Lemma Learning System=False, Pass@k=pass@322026.04 | 2 | 10 | |
| GPT-5.3-CodexBackbone LLM=GPT-5.3-Codex, Lemma Learning System=False, Pass@k=pass@322026.04 | 1 | 5 | |
| Gemini 2.5 ProBackbone LLM=Gemini 2.5 Pro, Lemma Learning System=False, Pass@k=pass@322026.04 | 0 | 0 |