Training Compressing Chain-of-Thought into Continuous Space via Self-Distillation (CODI) on the ProsQA dataset
Python · active 2025-06-04 → 2026-06-20 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 382 days from 2025-06-04 to 2026-06-20. Pushes: 5 total, peak 2 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 9 total, peak 3 in a day. Comments: 6 total, peak 1 in a day. Stars: 41 total, peak 2 in a day.
- Pushes
- Pull requests
- Issues
- Comments
- Stars
Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.
Top contributors
Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| zhenyi4 | 8 | 5 | 0 | 3 |
| DaoD | 2 | 0 | 0 | 1 |
| quancai337-png | 2 | 0 | 0 | 0 |
| JuseonDo | 1 | 0 | 0 | 1 |
| lqcStar | 1 | 0 | 0 | 0 |
| MisakiTaro0414 | 1 | 0 | 0 | 0 |
| hzy051302 | 1 | 0 | 0 | 0 |
| Frosty-Heart | 1 | 0 | 0 | 0 |
| whitesweater | 1 | 0 | 0 | 0 |
| mohammed-elkomy | 1 | 0 | 0 | 1 |
| nlper-fighting | 1 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#17mohammed-elkomy2026-06-02 01:14LLaMA-3.2-1B-Instruct math results diverge significantly from paper (GSM8k-Aug, GSM8k-Aug-NL)
- Issue#15Frosty-Heart2026-01-02 08:42关于HF检查点的询问
- Issue comment#12JuseonDo2025-12-21 16:49About the reproduction issue.
- Issue#12lqcStar2025-12-08 02:19About the reproduction issue.
- Issue#11quancai337-png2025-12-08 02:16Some questions about code
- Issue#10quancai337-png2025-12-08 02:15Some questinos about code
- Issue comment#8zhenyi42025-12-03 22:50Minor typo in Appendix A (Coconut on LLaMA-1B)
- Issue comment#7zhenyi42025-11-11 02:11Why is the model called twice here to generate ref_outputs, which are then used separately for the cross-entropy loss and the distillation loss?
- Issue#7whitesweater2025-11-04 18:25Why is the model called twice here to generate ref_outputs, which are then used separately for the cross-entropy loss and the distillation loss?
- Issue comment#1zhenyi42025-11-02 02:32Minor question about the padding side
- Issue#6hzy0513022025-10-29 12:57it doesn't work when using qwen3
- Issue#3nlper-fighting2025-09-01 09:05the last cot step
- Issue#2MisakiTaro04142025-08-22 09:11Wrong config file for llama1b commonsense
- Issue comment#1DaoD2025-08-01 02:56Minor question about the padding side
- Issue#1DaoD2025-08-01 02:56Minor question about the padding side
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 41 stars here means stars gained during the window, not the repo's star count.