zhehuderek/textual_decisionmaking_data
Viewer
• Updated
• 11k • 12 • 1
Praxis-VLM
VLM with textual-driven GRPO training for vision-grounded decision making (https://arxiv.org/pdf/2503.16965, NeurIPS 2025)
Note This is the textual synthetic data we used for model training.
-
zhehuderek/praxis_vlm_7b_decisionmaking
Image-Text-to-Text • 8B • Updated -
zhehuderek/praxis_vlm_3b_decisionmaking
Image-Text-to-Text • 4B • Updated
zhehuderek/qwen2_5_vl_3b_GEOQA_8K_hf
Image-Text-to-Text • 4B • UpdatedNote This is the model checkpoint after cold-start math training using GEOQA-8K dataset.
zhehuderek/qwen2_5_vl_7b_GEOQA_8K_step90_hf
Image-Text-to-Text • 8B • UpdatedNote This is the model checkpoint after cold-start math training using GEOQA-8K dataset.