Speculative DecodingDifferent speculative decoding approaches5 min read·fata-fanaka.hashnode.devSep 9
Cache EvictionCache eviction for optimizing KV Cache memory size9 min read·fata-fanaka.hashnode.devAug 6
Model Quantizationquantization and numeric precision for model quantization to reduce compute and memory footprint5 min read·fata-fanaka.hashnode.devJul 31
Neural Architecture SearchModel optimization through automated architecture search3 min read·fata-fanaka.hashnode.devJul 30
Structured approximationsoptimizing ml models using low rank matrix factorization and tensor decomposition5 min read·fata-fanaka.hashnode.devJul 28
