Pushing the Limits of Large Language Model Quantization via the Linearity Theorem · HackerPapers discussions

Pushing the Limits of Large Language Model Quantization via the Linearity Theorem — mentioned by yinksta

Fri, 01 May 2026 14:39:08 GMT

the Turboquant debaucle is entirely on Google doing incredible amount of academic misconduct and general incompetence. A) the QLJ thing they added is useless and the code they released didn't even include it since it makes the results worse B) the blogpost about turboquant was ai generated and stated that turboquant used a polar transformation when it doesn't, so for the first 2 weeks people thought turboquant involved a transformation to polar coordinates. The reason the blogpost was wrong was probably because google tried to put their useless polarquant paper on the map by talking about it…

Paper and discussion history. A mention is not an endorsement.

Pushing the Limits of Large Language Model Quantization via the Linearity Theorem — mentioned by amitport

Mon, 27 Apr 2026 13:27:14 GMT

Thanks for that! Note that the residual chain is empirically and theoretically inferior to our unbiased scale; furthermore, it requires an additional bit in certain cases. Additionally, TurboQuant was not the first to apply EDEN to KV-cache (see for example https://arxiv.org/abs/2411.17525 from 2024).

Paper and discussion history. A mention is not an endorsement.

Pushing the Limits of Large Language Model Quantization via the Linearity Theorem — mentioned by om8

Mon, 27 Apr 2026 10:32:44 GMT

https://docs.vllm.ai/en/v0.20.0/api/vllm/model_executor/laye... `vllm.model_executor.layers.quantization.turboquant` > The technique implemented here consists of the scalar case of the HIGGS quantization method (Malinovskii et al., "Pushing the Limits of Large Language Model Quantization via the Linearity Theorem", NAACL 2025; preprint arXiv:2411.17525): rotation + optimized grid + optional re-normalization, applied to KV cache compression. A first application of this approach to KV-cache compression is in "Cache Me If You Must: Adaptive Key-Value Quantization for Large Language Models"…

Paper and discussion history. A mention is not an endorsement.

Pushing the Limits of Large Language Model Quantization via the Linearity Theorem — HN submission

Sun, 20 Apr 2025 15:20:35 GMT

Pushing the Limits of LLM Quantization via the Linearity Theorem

95 points; 2 discussion replies on HN. Paper and source history. A submission is not an endorsement.