Python · MIT
THU-BPM/RLCSD
Source code of paper "RLCSD: Reinforcement Learning with Contrastive On-Policy Self-Distillation"
★ 117
Stars
5 forks · 1 open issues
+2.0/wk
Growth · measured
Stars counted between two of our own snapshots roughly a week apart. The repository page gives the exact span.
33.7
Momentum / 100
see the breakdown below
5mo ago
Created
last push 55d ago
Star history
How this score was built
| Component | Weight | Earned | Why |
|---|---|---|---|
| Velocity | 55 | 2.1 | +2.0/wk, log-scaled against a 750/day ceiling |
| Acceleration | 15 | 15.0 | last week ran 3.71× the monthly pace |
| Freshness | 15 | 10.6 | last push 55d ago; decays to zero at 180 days |
| Scale | 15 | 6.0 | 117 stars, log-scaled against 150k — stops a 3-star repo topping the board |
| Momentum | 100 | 33.7 | sum of the above |
Detail
| Topics | large-language-modelsllmon-policy-distillationopdopsdreinforcement-learningself-distillation |
| Matched on | LLM & Inference (large-language-models) |
| Short window | +2 stars over 7 days = 0.3/day |
| Long window | +2 stars over 26 days = 0.1/day |
| Lifetime average | 0.76 stars/day over 153 days |
| First seen here | 2026-10-11 |
| Last read | 2d ago |