Checkpoints of VaLR (Jeon et al., ICML 26') and its variants
-
Vision-aligned Latent Reasoning for Multi-modal Large Language Model
Paper • 2602.04476 • Published • 14 -
rooty2020/VaLR-Qwen3-VL-8B-stage2-prototype
9B • Updated • 10 • 1 -
rooty2020/VaLR-Qwen3-VL-8B-stage1-SFT
9B • Updated • 9 • 1 -
rooty2020/VaLR-Cosmos-Reason-Embodied-2B
2B • Updated • 4