AI Paper Atlas

Research map / Architectures and efficiency / Quantization

MemeBQ:Memory Efficient Binary Quantization of LLMs

Yuanhui Wang, Kunlong Liu, Minnan Pei, Zhangming Li, Peisong Wang, Qinghao Hu

AAAI 2026 · Architectures and efficiency › Quantization · approach: Post-training quantization of LLMs

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add group bitmap search (k-means within row groups) to ARB-LLM (ablation)ARB-LLM_RC · Llama-2-7B WikiText2perplexity: 16.44 -> 8.77Improved
add SimS row grouping (shared bitmap) without GBS (ablation)ARB-LLM_RC · Llama-2-7B WikiText2perplexity: 16.44 -> 21.67Hurt
add GBS on top of SimS (ablation)MemeBQ · Llama-2-7B WikiText2perplexity: 21.67 -> 16.73Improved
use greedy cosine-similarity grouping instead of merge-to-nearest (ablation)SimS on BiLLM · Llama-2-7B WikiText2perplexity: 70,99 -> 31.91Improved
use greedy search instead of k-means for row grouping (ablation)SimS on BiLLM · Llama-2-7B WikiText2perplexity: 37.68 -> 31.91 (cosine)Improved
use cosine similarity instead of euclidean distance for row similarity (ablation)SimS greedy search · Llama-2-7B WikiText2perplexity: 35.19 -> 31.91Improved
use cosine instead of euclidean for greedy grouping (ablation)SimS greedy search · Llama-2-7B C4perplexity: 36.15 -> 37.24Mixed
increase group size from 1 to 2 (halve extra bits) (ablation)MemeBQ · Llama-2-7B WikiText2perplexity: 8.77 -> 16.73Hurt
increase group size from 2 to 4 (0.25 extra bits) (ablation)MemeBQ · Llama-2-7B WikiText2perplexity: 16.73 -> 33.81Hurt
disable bitmap (group size 0) (ablation)MemeBQ · Llama-2-7B WikiText2perplexity: 8.77 -> 377.97Hurt
use MemeBQ instead of ARB-LLM_RC at the same extra bitsbinary PTQ · Llama-3-8B 7 zero-shot QA averageaverage accuracy: 52.27 -> 59.54Improved
MemeBQ with half the extra bits versus ARB-LLM_RCbinary PTQ · Llama-2-7B WikiText2perplexity: 16.44 -> 16.73No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy