Hugging Face
Models
Datasets
Spaces
Buckets
Docs
Enterprise
Pricing
Website
Tasks
HuggingChat
Collections
Languages
Organizations
Community
Blog
Posts
Daily Papers
Hardware
Learn
Discord
Forum
GitHub
Solutions
Team & Enterprise
Hugging Face PRO
Enterprise Support
Inference Providers
Inference Endpoints
Storage Buckets
Log In
Sign Up
Datasets:
TraceLite-AI
/
TraceTrial-Bench
Like
1
Follow
TraceLite AI
5
Tasks:
Other
Modalities:
Tabular
Text
Formats:
json
Languages:
English
Size:
1K - 10K
ArXiv:
arxiv:
2610.32285
Tags:
benchmark
agents
harness
reliability
failure-attribution
Libraries:
Datasets
pandas
Polars
+ 1
License:
mit
Dataset card
Data Studio
Files
Files and versions
xet
Community
2
main
TraceTrial-Bench
Commit History
Oct 28
Today
03:02
Add 3,420 execution trajectories with task archives and score index
HengjieZhu
Verified signature
fdbacb2
95 tasks × 9 models × 4 harnesses. Includes per-task archives, separate execution/final-score fields, name mapping, provenance, and SHA-256 checksums. Publication copies redact matched provider configuration and recognizable access-token patterns; local originals unchanged.
02:22
Add task_categories and remove arxiv tag from metadata
(
#2
)
HengjieZhu
+1
: nielsr
c82d0fb
- Add task_categories and remove arxiv tag from metadata (c4346ec4a482e705705fa95797e23a2e11be9527) Co-authored-by: Niels Rogge <nielsr@users.noreply.huggingface.co>
Oct 26
2 days ago
15:07
Publish final 95 paper-ordered TraceTrial-Bench tasks
HengjieZhu
Verified signature
b7e26de
15:07
initial commit
HengjieZhu
Verified signature
0434c43
