Human Preference Evaluation on User Study (Conversational Recommendation)
33D2D Preference CountD2D
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| D2DSystem Success Status=Total, Question=Q22026.06 | 33 | 15 | 1 | 1 | |
| D2DSystem Success Status=Total, Question=Q12026.06 | 32 | 17 | 1 | 0 | |
| D2DSystem Success Status=Total, Question=Q32026.06 | 31 | 11 | 7 | 1 | |
| D2DSystem Success Status=D2D Failure (✗), LLM Failure (✗), Question=Q12026.06 | 13 | 7 | 0 | 0 | |
| D2DSystem Success Status=D2D Failure (✗), LLM Failure (✗), Question=Q22026.06 | 13 | 6 | 0 | 1 | |
| D2DSystem Success Status=D2D Failure (✗), LLM Failure (✗), Question=Q32026.06 | 12 | 7 | 1 | 0 | |
| D2DSystem Success Status=D2D Success (✓), LLM Failure (✗), Question=Q12026.06 | 10 | 0 | 0 | 0 | |
| D2DSystem Success Status=D2D Success (✓), LLM Failure (✗), Question=Q22026.06 | 10 | 0 | 0 | 0 | |
| D2DSystem Success Status=D2D Success (✓), LLM Success (✓), Question=Q32026.06 | 8 | 2 | 0 | 0 | |
| D2DSystem Success Status=D2D Success (✓), LLM Failure (✗), Question=Q32026.06 | 8 | 0 | 2 | 0 | |
| D2DSystem Success Status=D2D Success (✓), LLM Success (✓), Question=Q22026.06 | 7 | 2 | 1 | 0 | |
| D2DSystem Success Status=D2D Success (✓), LLM Success (✓), Question=Q12026.06 | 6 | 3 | 1 | 0 | |
| D2DSystem Success Status=D2D Failure (✗), LLM Success (✓), Question=Q12026.06 | 3 | 7 | 0 | 0 | |
| D2DSystem Success Status=D2D Failure (✗), LLM Success (✓), Question=Q22026.06 | 3 | 7 | 0 | 0 | |
| D2DSystem Success Status=D2D Failure (✗), LLM Success (✓), Question=Q32026.06 | 3 | 2 | 4 | 1 |