Insight-level Evaluation on SCOpE-QA Data Augmentation
4.39Insight-level ScoreINSIGHTGEN
Evaluation Results
| Method | Links | |
|---|---|---|
| INSIGHTGENTarget Model=Claude-3.5-Sonnet, Evaluator Model=Gemini-1.5-Flash2026.04 | 4.39 | |
| INSIGHTGENTarget Model=GPT-4o, Evaluator Model=Gemini-1.5-Flash2026.04 | 4.24 | |
| FAISS+COTTarget Model=Claude-3.5-Sonnet, Evaluator Model=Gemini-1.5-Flash2026.04 | 4.07 | |
| GPT+COTTarget Model=GPT-4o, Evaluator Model=Gemini-1.5-Flash2026.04 | 3.93 | |
| GPT+COTTarget Model=Claude-3.5-Sonnet, Evaluator Model=Gemini-1.5-Flash2026.04 | 3.87 | |
| FAISSTarget Model=GPT-4o, Evaluator Model=Gemini-1.5-Flash2026.04 | 3.72 | |
| Direct GPTTarget Model=GPT-4o, Evaluator Model=Gemini-1.5-Flash2026.04 | 3.51 | |
| Direct GPTTarget Model=Claude-3.5-Sonnet, Evaluator Model=Gemini-1.5-Flash2026.04 | 3.45 | |
| FAISS+COTTarget Model=GPT-4o, Evaluator Model=Gemini-1.5-Flash2026.04 | 3.44 | |
| INSIGHTGENTarget Model=GPT-4o, Evaluator Model=Claude-3-Sonnet2026.04 | 3.39 | |
| FAISSTarget Model=Claude-3.5-Sonnet, Evaluator Model=Gemini-1.5-Flash2026.04 | 3.35 | |
| INSIGHTGENTarget Model=Claude-3.5-Sonnet, Evaluator Model=Claude-3-Sonnet2026.04 | 3.11 | |
| FAISS+COTTarget Model=Claude-3.5-Sonnet, Evaluator Model=Claude-3-Sonnet2026.04 | 2.83 | |
| FAISSTarget Model=GPT-4o, Evaluator Model=Claude-3-Sonnet2026.04 | 2.7 | |
| FAISSTarget Model=Claude-3.5-Sonnet, Evaluator Model=Claude-3-Sonnet2026.04 | 2.62 | |
| Direct GPTTarget Model=Claude-3.5-Sonnet, Evaluator Model=Claude-3-Sonnet2026.04 | 2.5 | |
| GPT+COTTarget Model=Claude-3.5-Sonnet, Evaluator Model=Claude-3-Sonnet2026.04 | 2.46 | |
| Direct GPTTarget Model=GPT-4o, Evaluator Model=Claude-3-Sonnet2026.04 | 2.32 | |
| FAISS+COTTarget Model=GPT-4o, Evaluator Model=Claude-3-Sonnet2026.04 | 2.28 | |
| GPT+COTTarget Model=GPT-4o, Evaluator Model=Claude-3-Sonnet2026.04 | 2.2 |