The datasets behind our Qwen3-VL-8B video RLVR runs: the base 24f/100k mixture and the HopChain v5 multi-hop corpora.
Nguyen Quang Trung
ngqtrung
AI & ML interests
None yet
Recent Activity
upvoted a paper 1 day ago
LLaVA-OneVision-1.5: Fully Open Framework for Democratized Multimodal
Training updated a collection 2 days ago
Video RLVR — final training data updated a collection 2 days ago
Video RLVR — final training dataOrganizations
Video RLVR — final training data
The datasets behind our Qwen3-VL-8B video RLVR runs: the base 24f/100k mixture and the HopChain v5 multi-hop corpora.
Qwen3-VL-8B video RLVR - GRPO checkpoint ladder
Every published checkpoint of the 8B video-QA GRPO campaign, ranked by core-3 mean_accuracy (5,645 rows). Untrained base = 0.4426.
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 17
ngqtrung/video-8b-grpo-twowave-gs140
Video-Text-to-Text • 9B • Updated • 19
ngqtrung/video-8b-grpo-n16-noexplore
Video-Text-to-Text • 9B • Updated • 19
ngqtrung/video-8b-grpo-twowave-tau085-gs40
Video-Text-to-Text • 9B • Updated • 13
ngqtrung/video-8b-grpo-hopchain-v40-vript-gs160
Video-Text-to-Text • 9B • Updated • 16
ngqtrung/video-8b-grpo-n16-warm-explore
Video-Text-to-Text • 9B • Updated • 11
ngqtrung/video-8b-grpo-ppexplore-n16k8
Image-Text-to-Text • 9B • Updated • 3
ngqtrung/video-8b-grpo-ppexplore
Image-Text-to-Text • 9B • Updated • 3
ngqtrung/video-8b-grpo-sft770
Image-Text-to-Text • 9B • Updated • 3
ngqtrung/video-8b-grpo-base
Image-Text-to-Text • 9B • Updated • 4
ngqtrung/omr-8b-grpo-base
Image-Text-to-Text • 9B • Updated • 7
datasets 53
ngqtrung/video-r1-mc-24f100k-combined
Viewer • Updated • 106k • 24
ngqtrung/hopchain-v5
Viewer • Updated • 23.6k • 23
ngqtrung/hopchain-v40-1080p
Viewer • Updated • 28.9k • 48 • 1
ngqtrung/vript-hopchain-analysis
Preview • Updated • 15
ngqtrung/hopchain-v34-1080p
Viewer • Updated • 7.9k • 33
ngqtrung/videoqa-samples-videos
Viewer • Updated • 101 • 69
ngqtrung/vero-rl
Viewer • Updated • 95.1k • 22
ngqtrung/openmmreasoner-rl-74k
Updated • 24
ngqtrung/omr-grpo-val
Viewer • Updated • 15k • 18
ngqtrung/omr-grpo-train
Viewer • Updated • 75k • 69