Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Response Quality Evaluation on IFLLM 1.0 (test)
Loading...
53.54
DPO Win Rate
RF + (IF - Gaze)
48.5168
49.8209
51.125
52.4291
Jun 18, 2026
DPO Win Rate
SFT Win Rate
DPO - SFT Win Rate Difference
Average Response Length
Updated 1mo ago
Evaluation Results
Method
Method
Links
DPO Win Rate
SFT Win Rate
DPO - SFT Win Rate Difference
Average Response Length
RF + (IF - Gaze)
Alignment Method=rDPO...
2026.06
53.54
41.96
0.345
202.5
mBERT base + Text + IF
Alignment Method=rDPO...
2026.06
50.71
45
0.1958
221.1
RF + IF
Alignment Method=rDPO...
2026.06
50.08
45.92
0.1892
206.6
Explicit Feedback
Alignment Method=DPO +...
2026.06
49.5
46.25
0.1079
208.8
mBERT base + Text
Alignment Method=rDPO...
2026.06
49.42
45.54
0.1221
227.3
RF + (IF - Mouse)
Alignment Method=rDPO...
2026.06
48.71
46.67
0.1108
203.5
Feedback
Search any
task
Search any
task