| repo | rucbm/laser |
|---|---|
| url | https://github.com/RUCBM/LaSeR |
| content_timestamp | 2026-06-03 |
| time_slice | 2026-06 |
| timestamp_source | web_observed_public_github_page_2026_06_03 |
| collected_at | 2026-06-03 13:55:19 +0800 |
| source | github |
GitHub - rucbm/laser: LaSeR is a reinforcement-learning recipe that jointly improves reasoning and self-rewarding behavior by adding an MSE self-reward term to the RLVR objective.
Source: https://github.com/RUCBM/LaSeR
This raw-style public GitHub page capture was recorded by the hourly public metadata update. Shell GitHub API access remains blocked in this workspace, so freshness is web-observed rather than API-verified.
- Repository: rucbm/laser
- URL: https://github.com/RUCBM/LaSeR
- Stars: 36
- Forks: 2
- Commits: 6
- License: MIT
- Primary language / stack signal: Python/RLVR/Self-Rewarding Training
- Collection timestamp: 2026-06-03T13:55:19+08:00
- The public page frames LaSeR as a lightweight method for jointly optimizing reasoning and self-rewarding.
- The repository exposes code, data, scripts, and a verl-based training path rather than a pure paper placeholder.
- Released checkpoints and training data links make it a concrete self-rewarding evidence point instead of a theoretical note.
- Public GitHub page evidence was observed without authenticated API access.
No benchmark was run, no source clone was modified, and no private or authenticated metadata was used. This file preserves public page evidence for downstream classification, model-card analysis, public reports, and the site index.