TCSVT Special Issue CALL FOR PAPERS
TCSVT Special Issue CALL FOR PAPERS
AIM AND SCOPE
The rapid evolution of large models has transformed artificial intelligence across natural language processing, computer vision, and multimodal applications. However, these models introduce substantial computational, memory, and storage costs during training and inference. A significant portion of the cost arises from the massive data processed at different stages, including input signals, intermediate features/tokens, gradients, and key-value (KV) caches.
Efficient data compression has therefore become a key enabler for deploying large models in cloud, cloud-edge collaborative, and multi-device environments. Unlike conventional image and video coding that primarily targets human perceptual quality, compression for large models emphasizes machine-centric objectives such as task accuracy preservation, inference latency reduction, memory footprint optimization, and energy efficiency.
This special issue invites original research that establishes new algorithms, theories, and systems for compressing data associated with large models. It aims to connect the long-standing strengths of IEEE TCSVT in data coding with the emerging frontier of scalable large-model deployment.
TOPICS OF INTEREST
New methods and frameworks for compressing intermediate features
Token compression methods for accelerating large model training/inference
Novel techniques for KV cache compression
Prompt compression methods for large models
New gradient compression methods in large model training and inference
New theory investigation in feature coding, token compression, KV cache compression, etc.
Quality assessment methods for evaluating distortion introduced by data compression
Coding for machines in the context of large models
Any other related topics that related to data compression and large models…
IMPORTANT DATES
GUEST EDITORS
Prof. Fengqing Maggie Zhu
Purdue University, US
zhu0@purdue.edu
Prof. Li Li
University of Science and Technology of China,
China
lil1@ustc.edu.cn
Dr. Changsheng Gao
Nanyang Technological University,
Singapore
changsheng.gao@ntu.edu.sg
Prof. Ivan Bajic
Simon Fraser University, Canada
ibajic@ensc.sfu.ca
Prof. Frederic Dufaux
Université Paris-Saclay, France
frederic.dufaux@l2s.centralesupelec.fr




