AI Paper Atlas

Research map / Vision-language and multimodal / CLIP and zero-shot

Hystar: Hypernetwork-driven Style-adaptive Retrieval via Dynamic SVD Modulation

Yujia Cai, Boxuan Li, Chenghao Xu, Jiexi Yan

ICLR 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Open-vocabulary alignment from VLMs · 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 hypernetwork + static modulation together over either alone (ablation)Hystar on CLIP · DSR Sketch queryTop-1 accuracy: 77.0 (static) / 70.6 (hyper) -> 85.3Improved
add StyleNCE loss instead of InfoNCE (ablation)Hystar on CLIP · DSR Sketch queryTop-1 accuracy: 85.3 -> 90.2Improved
add StyleNCE loss instead of InfoNCE (ablation)Hystar on CLIP · DSR Low-Res queryTop-1 accuracy: 94.2 -> 98.0Improved
dynamic modulation on attention with static MLP rather than the reverse (ablation)Hystar on CLIP · DSR Sketch queryTop-1 accuracy: 71.2 -> 90.2Improved
dynamic modulation on attention only rather than on attention and MLP (ablation)Hystar on CLIP · DSR Sketch queryTop-1 accuracy: 69.9 -> 90.2Improved
inject hypernetwork at middle layers {4,7,10,13} instead of all layers 4-12 (ablation)Hystar on CLIP · DSR Sketch queryTop-1 accuracy: 76.5 -> 90.2Improved
inject at middle layers instead of only layer 1 (ablation)Hystar on CLIP · DSR Sketch queryTop-1 accuracy: 72.4 -> 90.2Improved
increase hypernetwork hidden width from 2048 to 4096 (ablation)Hystar on CLIP · DSR Sketch queryTop-1 accuracy: 90.2 -> 90.8No clear effect
add a third hypernetwork layer (depth 3 instead of 2) (ablation)Hystar on CLIP · DSR Sketch queryTop-1 accuracy: 90.2 -> 90.4No clear effect
use DINOv2 vs CLIP(self) as style extractor (ablation)Hystar on CLIP · DSR Top-1 averageTop-1 accuracy: 83.0 -> 83.6No clear effect
add style conditioning via hypernetwork over static-only modulation (ablation)Hystar on CLIP · DSR Top-1 averageTop-1 accuracy: 73.9 -> 83.6Improved
Hystar vs FreestyleRet on inference costCLIP · Inference latencyspeed (ms): 96 -> 108 ms; extra params 11.5% -> 3.5%Hurt

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