Tag
This paper presents MoE-ViE, a mixture-of-experts vision encoder that efficiently scales for image and video understanding, outperforming larger dense models with lower inference latency.
UltraViT is a latency-optimized vision encoder for large vision-language models, designed for on-device deployment with a pyramidal architecture and a two-stage generative pre-training strategy, achieving state-of-the-art performance at 1.7x speed.
ViQ presents a visual quantization framework that balances semantic richness and detail preservation in discrete representations, enabling efficient multimodal training with native-resolution inputs by using text-aligned pre-training and proximal representation learning.
EarlyTom is a training-free framework that compresses visual tokens early in the vision encoder to reduce time-to-first-token and computational costs while maintaining accuracy, achieving up to 2.65x TTFT reduction.
LiteFrame introduces a highly efficient video encoder for Video LLMs that uses Compressed Token Distillation to enable up to 8x more frames and 35% latency reduction while maintaining accuracy, setting a new Pareto frontier for long-form video understanding.
LiteFrame proposes a lightweight video encoder with Compressed Token Distillation training that reduces latency and enables processing 8x more frames for long-form video understanding in Video LLMs, improving accuracy while reducing compute.
Delta-Adapter enables exemplar-based image editing using single-pair supervision by extracting semantic deltas from pre-trained vision encoders and injecting them via Perceiver-based adapters, improving accuracy and generalization.