Official repo for paper DigiRL: Training In-The-Wild Device-Control Agents with Autonomous Reinforcement Learning.
active 2024-06-13 → 2026-03-05 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 631 days from 2024-06-13 to 2026-03-05. Pushes: 64 total, peak 17 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 105 total, peak 4 in a day. Comments: 162 total, peak 9 in a day. Stars: 353 total, peak 50 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 |
|---|---|---|---|---|
| BiEchi | 196 | 60 | 0 | 100 |
| GoooKuuu | 13 | 0 | 0 | 6 |
| Wuzheng02 | 11 | 0 | 0 | 4 |
| xuyifan-0731 | 10 | 0 | 0 | 5 |
| YifeiZhou02 | 9 | 3 | 0 | 6 |
| CarlHuangNuc | 8 | 0 | 0 | 4 |
| mousewu | 6 | 0 | 0 | 3 |
| leoozy | 6 | 0 | 0 | 3 |
| zhengshf | 5 | 0 | 0 | 3 |
| artemisp | 5 | 0 | 0 | 5 |
| StarWalkin | 4 | 0 | 0 | 3 |
| Z-MU-Z | 4 | 0 | 0 | 2 |
| piperwolters | 3 | 0 | 0 | 1 |
| cyh2004 | 3 | 0 | 0 | 1 |
| James4Ever0 | 3 | 0 | 0 | 1 |
| GGaoe | 3 | 0 | 0 | 1 |
| mrFranklin | 3 | 0 | 0 | 1 |
| junho328 | 3 | 0 | 0 | 1 |
| zhiyuanhubj | 3 | 0 | 0 | 2 |
| langfengQ | 2 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue#54sunyongan1232025-11-16 07:57Evaluation criteria for the experiment
- Issue#53ai-sy412025-06-05 04:35An issue running an AVD with KVM acceleration via the emulator.
- Issue#52zttt042025-05-06 15:31questions about trajectories,network
- Issue comment#51cyh20042025-04-07 02:55Why cloning avds for every online rollout iteration
- Issue#51cyh20042025-04-07 02:55Why cloning avds for every online rollout iteration
- Issue comment#51BiEchi2025-04-06 23:00Why cloning avds for every online rollout iteration
- Issue#51cyh20042025-04-02 10:27Why cloning avds for every online rollout iteration
- Issue comment#50BiEchi2025-03-29 18:52What is the difference between the offline trajectories and off2on trajectories?
- Issue#50BiEchi2025-03-29 18:52What is the difference between the offline trajectories and off2on trajectories?
- Issue comment#50BiEchi2025-03-23 04:37What is the difference between the offline trajectories and off2on trajectories?
- Issue#20BiEchi2025-03-23 04:36[Question] How UI trajectories were created from AITW dataset
- Issue comment#18artemisp2025-03-18 04:17Result Reproduction
- Issue#50geekpowerai2025-03-18 03:25What is the difference between the offline trajectories and off2on trajectories?
- Issue comment#18BiEchi2025-03-18 02:49Result Reproduction
- Issue comment#18artemisp2025-03-18 02:28Result Reproduction
- Issue comment#18BiEchi2025-03-17 23:59Result Reproduction
- Issue comment#18artemisp2025-03-17 23:47Result Reproduction
- Issue comment#18BiEchi2025-03-17 23:39Result Reproduction
- Issue comment#18artemisp2025-03-17 23:33Result Reproduction
- Issue comment#18BiEchi2025-03-17 23:32Result Reproduction
- Issue comment#18artemisp2025-03-17 22:48Result Reproduction
- Issue comment#20wwh04112025-03-17 15:22[Question] How UI trajectories were created from AITW dataset
- Issue#20BiEchi2025-03-17 13:26[Question] How UI trajectories were created from AITW dataset
- Issue comment#20BiEchi2025-03-17 13:26[Question] How UI trajectories were created from AITW dataset
- Issue comment#20wwh04112025-03-17 09:34[Question] How UI trajectories were created from AITW dataset
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 353 stars here means stars gained during the window, not the repo's star count.