Qwen3-VL-8B GRPO RLVR checkpoints from a token-dropout exploration study. OMR ppexplore=winner (0.714); video ~0.485 dead-heat.
Nguyen Quang Trung
ngqtrung
AI & ML interests
None yet
Recent Activity
updated a dataset 3 days ago
ngqtrung/hopchain-v40-1080p published a Space 4 days ago
ngqtrung/video-reasoning-explorer updated a dataset 4 days ago
ngqtrung/vript-hopchain-analysisOrganizations
spaces 6
Running
Video Reasoning Explorer
📚
Create a simple static webpage with editable HTML
Running
HopChain v3.4 — multi-hop video QA viewer
🎬
Running
Caption Comparison Gemini vs 122B vs 397B
🎬
Compare AI video captions across different models
Sleeping
Agents
Video-QA Samples
🎥
Browse video QA candidates with full reasoning and captions
Sleeping
Agents
Video-QA Pipeline Reviewer
🎬
Sleeping
Benchmark Verify Tool
✅
Track and verify video annotation progress
models 12
ngqtrung/video-8b-grpo-ppexplore-n16k8
Image-Text-to-Text • 9B • Updated • 5
ngqtrung/video-8b-grpo-ppexplore
Image-Text-to-Text • 9B • Updated • 6
ngqtrung/video-8b-grpo-sft770
Image-Text-to-Text • 9B • Updated • 6
ngqtrung/video-8b-grpo-base
Image-Text-to-Text • 9B • Updated • 7
ngqtrung/omr-8b-grpo-base
Image-Text-to-Text • 9B • Updated • 7
ngqtrung/omr-8b-grpo-ppexplore
Image-Text-to-Text • 9B • Updated • 7
ngqtrung/verify-tool
Updated
ngqtrung/Qwen3-Omni-Thinker-30B-Instruct
Image-Text-to-Text • 32B • Updated • 3
ngqtrung/Qwen3-Omni-Thinker-30B-Thinking
Image-Text-to-Text • 32B • Updated • 2
ngqtrung/Qwen2.5-Omni-Thinker-7B
Image-Text-to-Text • 9B • Updated • 4
datasets 51
ngqtrung/hopchain-v40-1080p
Viewer • Updated • 28.9k • 48 • 1
ngqtrung/vript-hopchain-analysis
Preview • Updated • 25
ngqtrung/hopchain-v34-1080p
Viewer • Updated • 7.9k • 42
ngqtrung/videoqa-samples-videos
Viewer • Updated • 101 • 132
ngqtrung/vero-rl
Viewer • Updated • 95.1k • 204
ngqtrung/openmmreasoner-rl-74k
Updated • 70
ngqtrung/omr-grpo-val
Viewer • Updated • 15k • 40
ngqtrung/omr-grpo-train
Viewer • Updated • 75k • 54
ngqtrung/videorl-video-val
Viewer • Updated • 11.3k • 290
ngqtrung/videorl-video-rl-train
Viewer • Updated • 212k • 181