Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Personalized Interaction on ultrachat Synthetic
Loading...
7.26
Personalization Score
PPOpt
5.492
5.951
6.41
6.869
Feb 12, 2026
Personalization Score
Task Completion Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Personalization Score
Task Completion Score
PPOpt
Backbone=GPT-oss-20b
2026.02
7.26
7.96
PPOpt
Backbone=Llama-3-8b-in...
2026.02
7
7.7
PPOpt
Backbone=Qwen3-8b
2026.02
7
7.82
Vanilla
Backbone=Llama-3-8b-in...
2026.02
5.56
7.88
Vanilla
Backbone=Qwen3-8b
2026.02
5.56
7.88
Vanilla
Backbone=GPT-oss-20b
2026.02
5.56
7.88
Feedback
Search any
task
Search any
task