Insight Generation on Internal non-scientific document collections Climate Change Policy
4.77Set-level Score (Gemini-2.5-Flash Judge)INSIGHTGEN
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| INSIGHTGENBase Model=GPT-4o2026.04 | 4.77 | 2.94 | |
| INSIGHTGENBase Model=Claude-3.5-Sonnet2026.04 | 4.53 | 2.88 | |
| FAISS+CoTBase Model=Claude-3.5-Sonnet2026.04 | 4.12 | 2.88 | |
| GPT+CoTBase Model=Claude-3.5-Sonnet2026.04 | 3.82 | 2.82 | |
| GPT+CoTBase Model=GPT-4o2026.04 | 3.53 | 2.82 | |
| FAISS+CoTBase Model=GPT-4o2026.04 | 3.53 | 2.71 | |
| Direct GPTBase Model=GPT-4o2026.04 | 3.12 | 2.47 | |
| FAISSBase Model=GPT-4o2026.04 | 2.88 | 2.41 | |
| Direct GPTBase Model=Claude-3.5-Sonnet2026.04 | 2.71 | 1.88 | |
| FAISSBase Model=Claude-3.5-Sonnet2026.04 | 1.88 | 1.65 |