Multi-perspective Understanding on Multi-Perspective (T)
97.65AccuracyBenchmark Agent
Evaluation results
| Method | Links | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Benchmark Agentevaluation_type=Human & LLM-as-Judge Quality Eval, backbone=GPT-5.12026.06 | 97.65 | 76.77 | 87.93 | 96.7 | 75.31 | 63.36 | 39.78 | 72.55 | — | — | — | — | |
| Qwen3.5evaluation_type=Consistency Evaluation2026.06 | — | — | — | — | — | — | — | — | 71.06 | 74.04 | 81.28 | 87.23 |