Aug 2 19:29 update README.md 19:12 update README.md 18:52 add task templates for rebuttal May 22 21:16 remove NeurIPS badge/citation; use arXiv preprint citation May 18 15:18 Align title (page heading + BibTeX) with paper: Evaluating Agent Coordination under Enforced Role Separation May 17 16:02 dataset card: add 19-base/21-refined note and GH-derived Other-category caveat May 16 05:09 Fix TNI formula, classification rules, and condition naming mapping May 11 03:32 Add arXiv paper reference (2605.07073) and citation Your Name
May 7 08:26 Add rai:hasSyntheticData (NeurIPS-required) + prov:wasDerivedFrom describing generators 02:22 HF README: fix CLI flags (plural), document mock model, lb90 result paths 01:59 Tighten dataset card: schema + load + run an eval. Drop paper-narrative sections. 01:37 HF README: paper-aligned Quickstart with end-to-end replication path; drop train/test references 01:35 Fix YAML configs/splits to point at data/full.json (single full=931 split) 01:28 Drop NeurIPS/Croissant tags + Citation + camera-ready link; expand replication guide 00:57 Croissant: replace train/test FileObjects with data/full=931 split 00:57 Replace train/test (153+120) with single full=931 split aligned with paper 00:23 Croissant: bump conformsTo to 1.1 per NeurIPS 2026 ED hosting spec May 6 23:30 NeurIPS 2026 camera-ready: Croissant 1.0 + RAI metadata, refreshed README Replace croissant.json with a NeurIPS-validator-compliant Croissant 1.0 metadata file (sc:Dataset, full RAI extension, sha256 + contentSize on file objects, citeAs/version/datePublished). Update README with paper-aligned stats (851 templates / 931 instances), category breakdown, TNI metric, headline findings, citation block, and Croissant validation badge. 02:55 Add Croissant 1.1 metadata with Responsible AI (RAI) fields Adds croissant.json at the dataset root containing the full RAI
metadata required by the NeurIPS 2026 Evaluations & Datasets Track:
data collection, annotation protocol, personal/sensitive information,
data biases, preprocessing protocol, release / maintenance plan,
data authors, use cases, limitations, social impact, and safety /
privacy measures. Mar 30 15:59 Upload teambench_dataset.json with huggingface_hub Mar 11 22:22 Upload README.md with huggingface_hub 22:22 Upload data/test.json with huggingface_hub 22:22 Upload data/train.json with huggingface_hub 22:22 Delete data/hard.json with huggingface_hub 22:22 Delete data/full.json with huggingface_hub 22:17 Delete teambench_dataset.json with huggingface_hub 22:17 Upload data/hard.json with huggingface_hub 22:17 Upload data/full.json with huggingface_hub 22:16 Upload teambench_dataset.json with huggingface_hub 22:16 Upload README.md with huggingface_hub 22:16 initial commit 
