Code and Data for "Long-context LLMs Struggle with Long In-context Learning" [TMLR2025]
active 2024-04-03 → 2026-02-08 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 677 days from 2024-04-03 to 2026-02-08. Pushes: 28 total, peak 14 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 9 total, peak 1 in a day. Comments: 8 total, peak 3 in a day. Stars: 105 total, peak 15 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 |
|---|---|---|---|---|
| ltl3A87 | 27 | 25 | 0 | 1 |
| wenhuchen | 10 | 3 | 0 | 4 |
| Mooler0410 | 3 | 0 | 0 | 2 |
| Tommy307 | 2 | 0 | 0 | 0 |
| Jianwei-Lv | 2 | 0 | 0 | 1 |
| PengWenChen | 1 | 0 | 0 | 0 |
| TjuJianyu | 1 | 0 | 0 | 0 |
Recent activity
Latest issues, pull requests and releases
- Issue#6TjuJianyu2025-11-04 20:33Mamba "repetition_penalty" should be 1 rather than 0
- Issue#5Tommy3072025-07-28 01:42Potential Issue with Ground Truth in Benchmark Sequences
- Issue#3ltl3A872024-07-01 02:44Qwen1.5 seems not using NTK
- Issue comment#3ltl3A872024-07-01 02:44Qwen1.5 seems not using NTK
- Issue#2wenhuchen2024-06-29 01:19Unexpected low performance of mistral-v0.1
- Issue comment#3wenhuchen2024-06-29 01:19Qwen1.5 seems not using NTK
- Issue comment#4Jianwei-Lv2024-05-15 01:31ConnectionError: Couldn't reach
- Issue#4wenhuchen2024-05-15 01:25ConnectionError: Couldn't reach
- Issue comment#4wenhuchen2024-05-15 01:16ConnectionError: Couldn't reach
- Issue#4Jianwei-Lv2024-05-14 03:29ConnectionError: Couldn't reach
- Issue#3PengWenChen2024-05-13 03:11Qwen1.5 seems not using NTK
- Issue comment#2Mooler04102024-04-19 19:40Unexpected low performance of mistral-v0.1
- Issue comment#2wenhuchen2024-04-19 19:27Unexpected low performance of mistral-v0.1
- Issue comment#2Mooler04102024-04-19 19:11Unexpected low performance of mistral-v0.1
- Issue#2Mooler04102024-04-19 18:48Unexpected low performance of mistral-v0.1
- Issue#1wenhuchen2024-04-16 22:48[question] Why did you use base model instead of instructed ones?
- Issue comment#1wenhuchen2024-04-12 20:26[question] Why did you use base model instead of instructed ones?
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 105 stars here means stars gained during the window, not the repo's star count.