AI Paper Atlas

Research map / Architectures and efficiency / Quantization

Rethinking 1-bit Optimization Leveraging Pre-trained Large Language Models

Zhijun Tu, Jian Li, Yuanyuan Xi, Siqi Liu, Chuanjian Liu, Hanting Chen, Jie Hu, Yunhe Wang

ICML 2026 · Architectures and efficiency › Quantization · approach: Ultra-low-bit LLM quantization · 0 citations (OpenAlex)

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 consistent progressive training to vanilla 1-bit (ablation)1-bit SmolLM-135M · Wikitext2perplexity: 30.4 -> 27.4Improved
add consistent progressive training to vanilla 1-bit (ablation)1-bit SmolLM-135M · Zero-shot 7-task avgaccuracy: 41.0 -> 42.0Improved
add IR-Net style progressive training to vanilla 1-bit (ablation)1-bit SmolLM-135M · Wikitext2perplexity: 30.4 -> 29.8Improved
use dual scaling S_l x S_a instead of S_a only (ablation)1-bit Pythia-70M · Wikitext2perplexity: 73.1 -> 68.7Improved
use dual scaling S_l x S_a instead of S_a only (ablation)1-bit Pythia-70M · Average of Wiki2, C4, PTBperplexity: 75.5 -> 70.9Improved
replace fixed S_a with learnable S_l initialized from S_a (ablation)1-bit Pythia-70M · Average of Wiki2, C4, PTBperplexity: 75.5 -> 77.0Hurt
start from stronger full-precision model (SmolLM-135M instead of Pythia-160M) (ablation)BinaryLLM · Zero-shot 7-task avgaccuracy: Pythia +0.3 vs SmolLM -6.7 average change versus FP16Mixed
use quantization-aware BinaryLLM instead of 1-bit PTQ ARB-LLM-XLLaMA3-1B · Wikitext2perplexity: 115.4 -> 14.7Improved
use quantization-aware BinaryLLM instead of 1-bit PTQ ARB-LLM-XLLaMA3-3B · Wikitext2perplexity: 49.2 -> 12.4Improved
initialize 1-bit model from pre-trained SmolLM (20B tokens) instead of training FBI-LLM from scratch (1.26T tokens)130M-scale 1-bit LLM · Zero-shot 7-task avgaccuracy: 40.4 -> 42.8Improved

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