collaborators

5 papers

cs.CV2026

HACK++: Towards More Effective Head-Aware Key-Value Compression for Efficient Visual Autoregressive Modeling

Ziran Qin, Yuchen Jiang, Mingbao Lin +4

Visual Autoregressive (VAR) models adopt a next-scale prediction paradigm, offering high-quality generation with substantially fewer decoding steps. However, existing VAR models su…

cs.CL2025

CAKE: Cascading and Adaptive KV Cache Eviction with Layer Preferences

Ziran Qin, Yuchen Cao, Mingbao Lin +5

Large language models (LLMs) excel at processing long sequences, boosting demand for key-value (KV) caching. While recent efforts to evict KV cache have alleviated the inference bu…

cs.CV2025

Head-Aware KV Cache Compression for Efficient Visual Autoregressive Modeling

Ziran Qin, Youru Lv, Mingbao Lin +4

Visual Autoregressive (VAR) models adopt a next-scale prediction paradigm, offering high-quality content generation with substantially fewer decoding steps. However, existing VAR m…

cs.CV2025

Autoregressive Image Generation Needs Only a Few Lines of Cached Tokens

Ziran Qin, Youru Lv, Mingbao Lin +4

Autoregressive (AR) visual generation has emerged as a powerful paradigm for image and multimodal synthesis, owing to its scalability and generality. However, existing AR image gen…

cs.RO2025

Seeing Through Pixel Motion: Learning Obstacle Avoidance from Optical Flow with One Camera

Yu Hu, Yuang Zhang, Yunlong Song +6

Optical flow captures the motion of pixels in an image sequence over time, providing information about movement, depth, and environmental structure. Flying insects utilize this inf…