Insight Generation on Internal non-scientific document collections Hotel Sales Strategies
4.53Set-level Score (Gemini-2.5-Flash)INSIGHTGEN
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| INSIGHTGENBase Model=GPT-4o2026.04 | 4.53 | 3.32 | |
| INSIGHTGENBase Model=Claude-3.5-Sonnet2026.04 | 4.53 | 3.21 | |
| FAISS+CoTBase Model=Claude-3.5-Sonnet2026.04 | 4.32 | 2.79 | |
| FAISS+CoTBase Model=GPT-4o2026.04 | 3.9 | 2.68 | |
| FAISSBase Model=GPT-4o2026.04 | 3.63 | 2.74 | |
| GPT+CoTBase Model=Claude-3.5-Sonnet2026.04 | 3.63 | 2.42 | |
| Direct GPTBase Model=Claude-3.5-Sonnet2026.04 | 3.21 | 2 | |
| GPT+CoTBase Model=GPT-4o2026.04 | 3.11 | 2.53 | |
| Direct GPTBase Model=GPT-4o2026.04 | 3 | 2.26 | |
| FAISSBase Model=Claude-3.5-Sonnet2026.04 | 2.37 | 2.05 |