AI Paper Atlas

Research map / Architectures and efficiency / Quantization

Shift-and-Sum Quantization for Visual Autoregressive Models

Jaehyeon Moon, Bumsub Ham

ICLR 2026 · Architectures and efficiency › Quantization · approach: Post-training quantization of ViTs · 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 shift-and-sum quantization (ablation)BRECQ on VAR-d16 · ImageNet W4/A6FID: 11.16 -> 10.15Improved
add calibration data resampling (ablation)BRECQ on VAR-d16 · ImageNet W4/A6FID: 11.16 -> 10.19Improved
add both shift-and-sum and resampling (ablation)BRECQ on VAR-d16 · ImageNet W4/A6FID: 11.16 -> 9.20Improved
add shift-and-sum quantization (ablation)BRECQ on VAR-d20 · ImageNet W4/A6FID: 6.74 -> 6.10Improved
add calibration data resampling (ablation)BRECQ on VAR-d20 · ImageNet W4/A6FID: 6.74 -> 5.86Improved
add both components (ablation)BRECQ on VAR-d20 · ImageNet W4/A6FID: 6.74 -> 5.36Improved
raise attention threshold theta from 0.05 to 0.2 (fewer shifted tokens) (ablation)Shift-and-sum VAR-d16 · ImageNet W4/A6FID: 10.15 -> 10.33Mixed
combine Ours with LiteVAROurs on VAR-d16 · ImageNet W4/A4FID: 24.57 -> 18.92Improved
combine Ours with LiteVAROurs on Infinity-2B · Text-to-image W4/A4ImageReward: 0.575 -> 0.748Improved
use Ours instead of BRECQVAR-d30 · ImageNet W4/A4FID: 15.71 -> 9.26Improved
use Ours instead of BRECQ at 8-bitVAR-d16 ONNX CPU · throughputimage/sec: 0.2195 -> 0.2147No clear effect
use Ours instead of BRECQ at 6/6 bitsVAR-d30 · inpainting/outpainting/editingLPIPS: 0.2852 -> 0.2843No 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