A training-free, query-agnostic KV cache compression method for streaming video MLLMs, using chat-template attention as a saliency proxy, matches or beats prior streaming methods at a fixed 6K memory budget.
An image is worth 16x16 words: Transformers for image recognition at scale
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
StreamMem: Query-Agnostic KV Cache Memory for Streaming Video Understanding
A training-free, query-agnostic KV cache compression method for streaming video MLLMs, using chat-template attention as a saliency proxy, matches or beats prior streaming methods at a fixed 6K memory budget.