Multi-hop Grounding on CorpusQA 1M
53.11ScoreGemini-2.5-Pro
Evaluation results
| Method | Links | |
|---|---|---|
| Gemini-2.5-ProEvaluation Framework=Full-context Inference2025.12 | 53.11 | |
| Gemini-2.5-Flash-ThinkingEvaluation Framework=Full-context Inference2025.12 | 36.91 | |
| QwenLong-L1.5-30B-A3BEvaluation Framework=Memory Agent Framework2025.12 | 20.72 | |
| Qwen3-30B-A3B-Thinking-2507Evaluation Framework=Memory Agent Framework2025.12 | 15.32 | |
| MemAgent-14BEvaluation Framework=Memory Agent Framework2025.12 | 9.7 | |
| Qwen-Flash-Thinking-1MEvaluation Framework=Full-context Inference2025.12 | 2.75 |