Deepti Ghadiyaram PRO

dghadiya's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

updated a dataset about 21 hours ago

dghadiya/TAG-Bench-Video

Viewer • Updated about 21 hours ago • 300 • 420

New activity in dghadiya/MMA-Bench 18 days ago

Add MMA-Bench dataset (metadata and timestamps)

#2 opened 28 days ago by

dghadiya

Add MMA-Bench dataset (metadata and timestamps)

#4 opened 27 days ago by

dghadiya

Add MMA-Bench dataset (metadata and timestamps)

#5 opened 27 days ago by

dghadiya

New activity in dghadiya/MMA-Bench 28 days ago

Add MMA-Bench dataset (metadata and timestamps)

#1 opened 28 days ago by

dghadiya

published a dataset 28 days ago

dghadiya/MMA-Bench

Viewer • Updated 18 days ago • 12.9k • 22

upvoted 2 papers 28 days ago

Some Modalities are More Equal Than Others: Decoding and Architecting Multimodal Integration in MLLMs

Paper • 2511.22826 • Published Nov 28, 2025 • 7

Generative Action Tell-Tales: Assessing Human Motion in Synthesized Videos

Paper • 2512.01803 • Published Dec 1, 2025 • 4

commented a paper 28 days ago

Generative Action Tell-Tales: Assessing Human Motion in Synthesized Videos

Paper • 2512.01803 • Published Dec 1, 2025 • 4 •

2

New activity in dghadiya/TAG-Bench-Video 30 days ago

Add compute_correlations.py script for TAG-Bench

#4 opened 30 days ago by

SGTLIM

Add TAG-Bench final human scores (TAG_final_human_scores.json)

#3 opened 30 days ago by

SGTLIM

Add TAG-Bench release subset (300 videos + metadata)

#2 opened 30 days ago by

SGTLIM

Add TAG-Bench videos and canonical metadata (no human ratings)

#1 opened 30 days ago by

SGTLIM

published a dataset 30 days ago

dghadiya/TAG-Bench-Video

Viewer • Updated about 21 hours ago • 300 • 420

updated a Space about 2 months ago

Sleeping

Videoeval Humaneval

👁

Rate AI-generated videos for action consistency and coherence