XLang NLP Lab

xlangai's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Articles

mengqi-20 

updated a dataset 4 days ago

xlangai/osworld2.0-trajectory

Updated 4 days ago • 5.86k • 11

EricsXt 

updated a model 5 days ago

xlangai/RoboFine-VLM-397B-A17B

Video-Text-to-Text • 403B • Updated 5 days ago • 57 • 8

EricsXt 

updated a dataset 5 days ago

xlangai/RoboFine-bench

Viewer • Updated 5 days ago • 1k • 1.62k • 6

halfrot 

authored 2 papers 9 days ago

Verifier-Backed Hard Problem Generation for Mathematical Reasoning

Paper • 2605.06660 • Published May 7 • 1

Step-Level Sparse Autoencoder for Reasoning Process Interpretation

Paper • 2603.03031 • Published Mar 3

mengqi-20 

in xlangai/osworld_v2_tasks 15 days ago

fix(task104): limit CAD extractor CPU affinity

#211 opened 15 days ago by

mengqi-20

mengqi-20 

in xlangai/osworld_v2_tasks 19 days ago

fix(task093): bound OpenBoard installation and verify setup windows

#210 opened 19 days ago by

mengqi-20

mengqi-20 

in xlangai/osworld_v2_tasks 22 days ago

Fix FreeCAD host compatibility in tasks 103 and 104

#209 opened 22 days ago by

mengqi-20

Fix task 070 incident-report judge parsing and validation retries

#208 opened 22 days ago by

mengqi-20

Fix Task048 Web progress persistence and detect evaluation tampering

#207 opened 22 days ago by

mengqi-20

mengqi-20 

in xlangai/osworld_v2_assets_gated 22 days ago

Task064: replace host Docker evaluator with native guest processes

#24 opened 22 days ago by

mengqi-20

mengqi-20 

in xlangai/osworld_v2_tasks 22 days ago

Task064: remove host Docker setup dependency and retry guest transport

#206 opened 22 days ago by

mengqi-20

mengqi-20 

in xlangai/osworld_v2_tasks 24 days ago

Fix Task 001 malformed all-day date evaluation

#205 opened 24 days ago by

mengqi-20

Fix Task 024 certificate false-positive scoring

#204 opened 24 days ago by

mengqi-20

Fix Task 062 PDF filename evaluation

#203 opened 24 days ago by

mengqi-20

Fix Task 082 Docker privilege handling

#202 opened 24 days ago by

mengqi-20

Install Task108 KiCad from a fixed offline package bundle

#201 opened 24 days ago by

mengqi-20

mengqi-20 

in xlangai/osworld_v2_assets_gated 24 days ago

Add fixed offline KiCad package bundle for Task108

#23 opened 24 days ago by

mengqi-20

Harden Task064 trajectory evaluation assets

#22 opened 24 days ago by

mengqi-20

mengqi-20 

in xlangai/osworld_v2_tasks 24 days ago

Harden Task064 against saved-trajectory score bypass

#200 opened 24 days ago by

mengqi-20