HYMA: LlaVA Alignment CC3M-558K Pre-embedded Collection This collection presents CC3M-558K, the data used by LlaVA for multi-modal feature alignment, pre-embedded across 9 image encoders & 3 text encoders. • 5 items • Updated Aug 19, 2025
HYMA: VLM connector checkpoints Collection Checkpoints of connectors obtained for VLMs by using HYMA (ours) and Grid Search (gs). We denote "mlp1" as the MLP_1 setting in the paper. • 3 items • Updated Aug 19, 2025
HYMA: LlaVA Alignment CC3M-558K Pre-embedded Collection This collection presents CC3M-558K, the data used by LlaVA for multi-modal feature alignment, pre-embedded across 9 image encoders & 3 text encoders. • 5 items • Updated Aug 19, 2025