Views
No views yet
solver/: LoRA adapter for visual understanding and self-consistency scoring.proposer/: LoRA adapter for generating image-grounded questions.generator/: LoRA adapter for generation conditioning.dit_lora/: LoRA adapter for the BLIP3o diffusion transformer.dit_lora_metadata.json: DiT LoRA configuration metadata.trainer_state.json: lightweight step metadata.CHECKPOINT_DIR.1git clone https://github.com/mbzuai-oryx/Ask-Solve-Generate.git
2cd Ask-Solve-Generate1CHECKPOINT_DIR=/path/to/this/snapshot \
2ADAPTER=solver \
3bash BLIP3o/eval/understanding_eval_our.sh1CHECKPOINT_DIR=/path/to/this/snapshot \
2ADAPTER=generator \
3bash BLIP3o/eval/geneval/generation_our.sh1@article{thawkar2026asksolvegenerate,
2 title={Ask, Solve, Generate: Self-Evolving Unified Multimodal Understanding and Generation via Self-Consistency Rewards},
3 author={Thawkar, Ritesh and Venkatraman, Shravan and Thawakar, Omkar and Shaker, Abdelrahman and Khan, Fahad and Cholakkal, Hisham and Khan, Salman and Anwer, Rao Muhammad},
4 journal={arXiv preprint arXiv:2606.27376},
5 year={2026}
6}