Theory-optimal Quantization Based on Flatness
A new paper introduces a novel quantization framework called Bidirectional Diagonal Quantization (BDQ) aimed at improving the performance of Large Language Models (LLMs). The authors address the issue of activation outliers that degrade model performance, particularly at lower bit precision. Their extensive experiments demonstrate that BDQ sets a new benchmark in quantization with minimal accuracy loss compared to existing methods.
- ▪Post-training quantization is crucial for compressing and accelerating LLM inference.
- ▪The new metric Flatness is introduced to quantify outlier distributions in quantization.
- ▪BDQ achieves less than 1% accuracy drop in W4A4 quantization on the LLaMA-3-8B model.
arXiv cs.AI files mainly under ai research. We currently carry 1,128 of its stories.
Story provenance
Source · retrieval · rights · ranking — open for full record
inspect →
Story provenance
Attribution is not the same as permission. This drawer separates discovery metadata, excerpts, WeSearch-generated summaries, reuse status, and whether the publisher receives the visit. Nothing here claims a legal grant the publisher has not made.
Record
| Original publisher | arXiv cs.AI |
| Canonical URL | https://arxiv.org/abs/2605.18800 |
| Publication time | Wed, 20 May 2026 00:00:00 -0400 |
| Retrieval time | 2026-05-20T04:04:59.484Z |
| Last seen | 2026-05-20T04:04:59.484Z |
| Headline source | Publisher (no WeSearch rewrite) |
| Excerpt source | publisher body |
| Excerpt method | First ~120 words (~800 chars) of extracted publisher body, fair-use limited. |
| Summary | WeSearch · cerebras-chat (WeSearch summarizer) |
| Summary source text | contentText |
| Citation coverage | Summary is a WeSearch-generated derivative; primary citation is the original publisher URL. |
| Cluster | BKPDoTEDDo-t |
| Cluster logic | Grouped by semantic title/content similarity across sources within a rolling window. Same-publisher template collisions are excluded from coverage comparison. |
| Ranking reason | Story pages are not engagement-ranked. Hub feeds use recency, with optional source-diversified chronological ordering (cap consecutive stories per source). No personalized ranking. |
| Publisher visit | Yes — open original |
| Substitutes article? | No — link-out required for full text |
Rights status (four layers)
WeSearch handling by dimension
| Indexing | May the item be indexed (stored, ranked, made findable)? | Allowed |
| Snippet | May a short excerpt of the publisher's text be shown? | Allowed |
| AI summary | May WeSearch generate its own short summary of the article? | Limited |
| Retrieval / RAG | May the content be exposed for third-party retrieval-augmented generation? | Not asserted |
| Model training | May the content be used to train AI models? | Not asserted |
| Commercial reuse | May the content be reused commercially? | Not permitted |
Basis: Derived from the published RSS/Atom feed. Contact: [email protected]. Reviewed: 2026-07-24.
Opening excerpt (first ~120 words) tap to expand
Computer Science > Machine Learning arXiv:2605.18800 (cs) [Submitted on 11 May 2026] Title:Theory-optimal Quantization Based on Flatness Authors:Xiusheng Huang, Zhe Li, Xuanwu Yin, Lu Wang, Yequan Wang, Dong Li, Emad Barsoum, Kang Liu View a PDF of the paper titled Theory-optimal Quantization Based on Flatness, by Xiusheng Huang and 6 other authors View PDF HTML (experimental) Abstract:Post-training quantization has emerged as a widely adopted technique for compressing and accelerating the inference of Large Language Models (LLMs). The primary challenges in LLMs quantization stem from activation outliers, which significantly degrade model performance especially at lower bit precision.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at arXiv cs.AI.
Discussion
0 commentsMore from arXiv cs.AI
