jerryyan/mle-traj-v1
MLE Trajectory Dataset v1.0 Paired human + LLM-agent trajectories on 7 Kaggle competitions, annotated with a three-level state/action/intent schema and grader-scored on the held-out test split. Released for NeurIPS 2026 Datasets & Benchmarks track. What is in this dataset 622 trajectories across 7 Kaggle competitions: 422 human Kaggle kernel trajectories (43 Grandmaster, 59 Master, 139 Expert, 181 Contributor) 11 Codex (single-loop CLI agent) runs 13 MLEvolve… See the full description on the dataset page: https://huggingface.co/datasets/jerryyan/mle-traj-v1.
| Author / organization | jerryyan |
|---|---|
| License (as declared) | cc-by-4.0 |
| Downloads | 252 (as reported by Hugging Face) |
| Likes | 0 |
| Created | 3 May 2026 |
| Last updated at source | 3 May 2026 |
Popularity, last 3 days
252 now (+5 since 2026-10-16). Counts as reported by the source, snapshotted daily by Silytics.
Recent changes
- 16 Oct 2026 Description: MLE Trajectory Dataset v1.0 Paired human + LLM-agent trajectories on 7 Kaggle competitions, annotated with a three-level state/action/intent schema and grader-scored on the held-out test split. Released for NeurIPS 2026 Datasets & Benchmark → MLE Trajectory Dataset v1.0 Paired human + LLM-agent trajectories on 7 Kaggle competitions, annotated with a three-level state/action/intent schema and grader-scored on the held-out test split. Released for NeurIPS 2026 Datasets & Benchmark
Changes detected by Silytics between refreshes of the source.
More from jerryyan
Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 18 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.