EditScore: Unlocking Online RL for Image Editing via High-Fidelity Reward Modeling
active 2025-09-30 → 2026-05-21 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 234 days from 2025-09-30 to 2026-05-21. Pushes: 22 total, peak 4 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 15 total, peak 3 in a day. Comments: 26 total, peak 4 in a day. Stars: 81 total, peak 22 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 |
|---|---|---|---|---|
| Luciennnnnnn | 40 | 18 | 0 | 16 |
| sabulin | 6 | 1 | 0 | 4 |
| wcyno23 | 3 | 3 | 0 | 0 |
| goldzzmj | 2 | 0 | 0 | 1 |
| akk-123 | 2 | 0 | 0 | 0 |
| ErfeiCui | 2 | 0 | 0 | 2 |
| stdKonjac | 2 | 0 | 0 | 1 |
| g-h-chen | 1 | 0 | 0 | 0 |
| Janspiry | 1 | 0 | 0 | 0 |
| bxhsort | 1 | 0 | 0 | 1 |
| haokbd | 1 | 0 | 0 | 1 |
| tulvgengenr | 1 | 0 | 0 | 0 |
| XiangqianHuang | 1 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#17Luciennnnnnn2026-01-30 00:35inference time on Qwen 3 VL models is incredibly slow
- Issue comment#17sabulin2026-01-29 04:33inference time on Qwen 3 VL models is incredibly slow
- Issue comment#16Luciennnnnnn2025-12-03 08:35Wandb Logs of Qwen-Image Edit
- Issue#16Janspiry2025-12-03 08:14Wandb Logs of Qwen-Image Edit
- Issue comment#14Luciennnnnnn2025-11-26 04:10Reproduce the issue
- Issue comment#14bxhsort2025-11-24 07:57Reproduce the issue
- Issue comment#14Luciennnnnnn2025-11-23 09:19Reproduce the issue
- Issue comment#11haokbd2025-11-21 02:23Training code for reward models
- Issue comment#13sabulin2025-11-17 03:34What is the difference between the score and the reward obtained when I call the deployed reward server using the reward client?
- Issue#7Luciennnnnnn2025-11-08 03:23Questions About VLLM Parallel Inference
- Issue comment#12stdKonjac2025-11-07 07:25How to conduct batch inference
- Issue#12Luciennnnnnn2025-11-07 07:08How to conduct batch inference
- Issue#12stdKonjac2025-11-06 12:44How to conduct batch inference
- Issue#8Luciennnnnnn2025-11-03 07:25enable_lora
- Issue#9Luciennnnnnn2025-11-03 07:25需要的GPU hours
- Issue#9Luciennnnnnn2025-11-03 07:25需要的GPU hours
- Issue comment#8Luciennnnnnn2025-10-30 13:10enable_lora
- Issue comment#8ErfeiCui2025-10-30 12:56enable_lora
- Issue#9XiangqianHuang2025-10-30 11:53需要的GPU hours
- Issue comment#8ErfeiCui2025-10-30 11:17enable_lora
- Issue comment#8Luciennnnnnn2025-10-30 08:41enable_lora
- Issue#6sabulin2025-10-27 07:36About the language of data for tuning
- Issue comment#7sabulin2025-10-24 07:54Questions About VLLM Parallel Inference
- Issue#7tulvgengenr2025-10-24 04:02Questions About VLLM Parallel Inference
- Issue#5akk-1232025-10-23 08:39RL training code
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 81 stars here means stars gained during the window, not the repo's star count.