Python · Apache-2.0
uclaml/SPPO
The official implementation of Self-Play Preference Optimization (SPPO)
★ 589
Stars
48 forks · 15 open issues
flat
Growth · measured
Stars counted between two of our own snapshots roughly a week apart. The repository page gives the exact span.
8.0
Momentum / 100
see the breakdown below
2y ago
Created
last push 21mo ago
Star history
How this score was built
| Component | Weight | Earned | Why |
|---|---|---|---|
| Velocity | 55 | 0.0 | flat, log-scaled against a 750/day ceiling |
| Acceleration | 15 | 0.0 | needs 7-day and 30-day rates side by side |
| Freshness | 15 | 0.0 | last push 21mo ago; decays to zero at 180 days |
| Scale | 15 | 8.0 | 589 stars, log-scaled against 150k — stops a 3-star repo topping the board |
| Momentum | 100 | 8.0 | sum of the above |
Detail
| Topics | deep-learningfine-tuninglarge-language-modelsrlhfself-play |
| Matched on | LLM & Inference (fine-tuning) |
| Short window | +0 stars over 7 days = 0.0/day |
| Long window | +1 stars over 26 days = 0.0/day |
| Lifetime average | 0.67 stars/day over 875 days |
| First seen here | 2026-10-11 |
| Last read | 2d ago |