A Parameter-Free Classification Method with Compressors (aclanthology.org)

Screenshot of aclanthology.org: A Parameter-Free Classification Method with Compressors

First screen, captured 3 years ago.

Looks similar

Screenshot of aclanthology.org: Graph Language Models

Graph Language Models (aclanthology.org)

111 points by Anon84 2 years ago | 9 comments

Screenshot of arxiv.org: Lossless LLM compression for efficient GPU inference via dynamic-length float

Lossless LLM compression for efficient GPU inference via dynamic-length float (arxiv.org)

392 points by CharlesW 1 year ago | 109 comments

arXiv paper introducing DFloat11, a lossless compression format that shrinks LLM weights by about 30% using entropy coding, with a custom GPU kernel for inference.

llm-compression dfloat11 bfloat16 entropy-coding

Screenshot of arxiv.org: Parameter-free KV cache compression for memory-efficient long-context LLMs

Parameter-free KV cache compression for memory-efficient long-context LLMs (arxiv.org)

60 points by PaulHoule 1 year ago | 13 comments

Screenshot of dl.acm.org: U-Net CNN in APL: Exploring Zero-Framework, Zero-Library Machine Learning

U-Net CNN in APL: Exploring Zero-Framework, Zero-Library Machine Learning (dl.acm.org)

79 points by tosh 3 years ago | 10 comments

Screenshot of arxiv.org: LoMA: Lossless Compressed Memory Attention

LoMA: Lossless Compressed Memory Attention (arxiv.org)

90 points by PaulHoule 2 years ago | 8 comments

Screenshot of arxiv.org: DiLoCo: Distributed Low-Communication Training of Language Models

DiLoCo: Distributed Low-Communication Training of Language Models (arxiv.org)

38 points by Anon84 2 years ago | 14 comments

Looks similar

Graph Language Models(aclanthology.org)

Lossless LLM compression for efficient GPU inference via dynamic-length float(arxiv.org)

Parameter-free KV cache compression for memory-efficient long-context LLMs(arxiv.org)

U-Net CNN in APL: Exploring Zero-Framework, Zero-Library Machine Learning(dl.acm.org)

LoMA: Lossless Compressed Memory Attention(arxiv.org)

DiLoCo: Distributed Low-Communication Training of Language Models(arxiv.org)