AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Positional Bias in Multimodal Embedding Models: Do They Favor the Beginning, the Middle, or the End?

Kebin Wu, Fatima Albreiki

AAAI 2026 · Vision-language and multimodal › Human and embodied modeling · approach: Multimodal retrieval-augmented modeling · 2 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
train on shuffled sub-captions instead of original caption order (ablation)Long-CLIP (ViT-B/16) · Urban1K text encoder (T2I)Recall@1 drop when first segment moves start to end: drop 0.303 (0.581 -> 0.278) vs 0.199 (0.581 -> 0.382)Improved
use relative positional encoding (RoPE, TULIP) instead of absolute encoding (Long-CLIP) (ablation)CLIP-style text encoder · Urban1K text encoder (T2I)Recall@1 when first segment moves from position 0 to 2: TULIP 0.66 -> 0.065 vs Long-CLIP 0.582 -> 0.485Hurt
use sigmoid loss (SigLIP) instead of softmax contrastive loss (ablation)CLIP-style models · Image-text retrieval positional biaspositional bias patternNo clear effect
scale model from ViT-B/16 to ViT-L/14 (ablation)CLIP · Urban1K/COCO positional biascoefficient of variation across positionsHurt
increase image resolution from 224 to 336 (ViT-L/14) (ablation)CLIP image encoder · Urban1K image positional biascoefficient of variation across positions: lower in 5 of 7 casesImproved
use smaller patch size (B/16 instead of B/32) (ablation)CLIP image encoder · Urban1K image positional biascoefficient of variation across positionsImproved
use ResNet-50 CNN image encoder instead of ViT (ablation)CLIP · Urban1K/ImageNet positional biaspositional bias presenceNo clear effect
train vision encoder with image-text pairs (CLIP) instead of image-only (ablation)ResNet-50 / ViT-B/16 image encoder · ImageNet classification with shifted segmenttop-1 accuracy variation across positionsHurt
short captions (COCO) instead of long captions (ablation)CLIP-ViT-B/16 text encoder · COCORecall@10 positional biasNo clear effect

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