GitHub

TurboQuant KV Cache Compression for llama.cpp

TriAttention token pruning on AMX3_1 hybrid K cache — dequant-free pre-RoPE polar scoring + physical eviction. All TBQ/TBQP/AMX encoders freed from external attn_rot_k dependency (redundant Hadamard ...
IEEE

An Algorithm for Satellite Remote Sensing Image Compression Based on Neural Network Architecture Evolution

Abstract: This paper proposed a satellite remote sensing image compression algorithm based on neural network architecture evolution, the method includes a neural network automatic evolution method, a ...
IEEE

Optimization of Image Compression and Decompression Performance Based on Genetic Algorithm

Abstract: As one of the main carriers of information transmission and storage, image data need to be compressed in some application scenarios. Traditional image compression methods, although reducing ...
GitHub

test_compression.cpp

frame.GetField(ID3FN_TEXT)->Set("This sample user text frame came from an ID3v2-3.0 tag. The frame has the 'compression' bit set in it's frame header. This is the new method for compressing frames, ...
Search Engine Land

New Google TurboQuant algorithm improves vector search speed

Google says a new compression algorithm, called TurboQuant, can compress and search massive AI data sets with near-zero indexing time, potentially removing one of the biggest speed limits in modern ...