Unverified50% confidenceFactExact time
DCKV is a differentiable KV cache codec that, for the first time, enables end-to-end training, jointly optimizing the transform, quantizer and entropy model with a frozen language model.
1
Sources
50%
Confidence
Long-term
Relevance
10/28/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedDCKV's entropy model is conditioned on prompt tokens and the already-decoded adjacent positions, and the decoding process is autoregressive.72% similarUnverifiedRCSD (Reinforcement and Continual Learning for Domain-Specific Speculative Decoding) is a training framework that directly aligns draft model optimization with a system-level objective (maximizing acceptance length).70% similarVerifiedKVCache is a mechanism in the Transformer architecture that caches the Key and Value vectors of all previous tokens to avoid redundant computation, with its memory footprint growing linearly with context length.65% similarUnverifiedRWKV uses the parallelizable WKV operator for training along the time dimension, and at inference time degenerates into a token-by-token recursive form, so it does not need a KV Cache and its memory usage is independent of sequence length.64% similarUnverifiedThe context cache (KV Cache) is the memory area that stores the Key and Value vectors of previous tokens during inference of Transformer-type models, and its size grows linearly with context length, batch size, and the number of model layers.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1106213API
curl https://kongchang.com/api/v1/knowledge/claims/1106213MCP
get_claim(id=1106213)