AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Fourier Compressor: Frequency-Domain Visual Token Compression for Vision-Language Models

Huanyu Wang, Jushi Kai, Haoli Bai, Lu Hou, Bo Jiang, Ziwei He, Zhouhan Lin

ECCV 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Attention-based visual token pruning · 13 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
apply Fourier Compressor to 256 tokens (55.6% compression)LLaVA-v1.5-7B · Image benchmarks avgaverage score: 64.6 -> 64.6No clear effect
apply Fourier Compressor to 144 tokens (75% compression)LLaVA-v1.5-7B · Image benchmarks avgaverage score: 64.6 -> 64.6No clear effect
compress to 36 tokens (93.75%)LLaVA-v1.5-7B · Image benchmarks avgaverage score: 64.6 -> 62.1Hurt
use Fourier Compressor instead of MQT-LLaVALLaVA-v1.5-7B · Image benchmarks avg (144 tokens)average score: 62.8 -> 64.6Improved
use Fourier Compressor instead of MQT-LLaVALLaVA-v1.5-7B · Image benchmarks avg (36 tokens)average score: 61.1 -> 62.1Improved
apply Fourier Compressor to 144 tokensLLaVA-v1.5-13B · Image benchmarks avgaverage score: 67.3 -> 65.9Hurt
use Fourier Compressor instead of PruMergeLLaVA-v1.5-13B · Image benchmarks (144 tokens)POPE: 84.4 -> 85.4Improved
apply Fourier Compressor (57.3% fewer tokens)Qwen2-VL-2B · Image benchmarks avgaverage score: 66.8 -> 65.7Hurt
apply Fourier Compressor (57.3% fewer tokens)Qwen2.5-VL-3B · Image benchmarks avgaverage score: 71.1 -> 72.6Improved
apply Fourier Compressor (57.3% fewer tokens)Qwen2.5-VL-3B · RealWorldQAaccuracy: 59.6 -> 64.3Improved
apply Fourier Compressor to 36 tokensLLaVA-v1.5-7B · FLOPsFLOPs (T): 8.54 -> 1.38Improved
apply Fourier Compressor (87.5% fewer video tokens)LLaVA-v1.5-7B · Video benchmark avg (zero-shot)average score: 45.6 -> 44.0Hurt

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