ACCT transfers KV cache between LLMs for context reuse
ACCT reduces KV cache size by 75% while matching target model performance and speeding up prefill by 21x.
Stories per day · 30 days
Relatedcontext-transfer1cross-llm1kv-cache1mixture-of-experts1parameter-sharing1scaling1sparse-attention1
3 stories, newest first · one card per story
Stories tagged #model-efficiency
Meta Expert shares MoE experts across layers to cut redundancy
With activation and granularity fixed, it matches vanilla MoE using 70–80% of the experts.
Breaking the 1.58-bit Barrier for Ternary LLMs
An arXiv paper claims a method that breaks the 1.58-bit barrier for ternary large language models.