AI Paper Atlas

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

Joint Implicit and Explicit Language Learning for Pedestrian Attribute Recognition

Yukang Zhang, Lei Tan, Yang Lu, Yan Yan, Hanzi Wang

AAAI 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Fine-grained improvements to CLIP · 1 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 ILGL pseudo-word tokens (ablation)baseline · PETAmA: 86.84 -> 88.94Improved
add ILGL pseudo-word tokens (ablation)baseline · RAPv1mA: 81.25 -> 84.88Improved
add MIM on top of ILGL (ablation)JGEL · PETAmA: 88.94 -> 89.56Improved
add MIM on top of ILGL (ablation)JGEL · RAPv1Recall: 87.65 -> 88.38Improved
add EAEL alone to baseline (ablation)baseline · PETAmA: 86.84 -> 87.14Improved
add EAEL to ILGL + MIM (ablation)JGEL · PETAmA: 89.56 -> 89.90Improved
add EAEL to ILGL + MIM (ablation)JGEL · RAPv1Recall: 88.38 -> 89.96Improved
add TIM on top of ILGL, MIM, EAEL (ablation)JGEL · PETAmA: 89.90 -> 90.27Improved
add TIM on top of ILGL, MIM, EAEL (ablation)JGEL · RAPv1Recall: 89.96 -> 90.64Improved
add MIM to EAEL without ILGL (ablation)EAEL · PETAmA: 87.14 -> 88.31Improved
increase MIM self-attention layers from 2 to 4 (ablation)JGEL MIM · RAPv1Recall: 88.38 -> 87.64Hurt
increase MIM self-attention layers from 2 to 4 (ablation)JGEL MIM · PETAmA / Recall: mA 89.56 -> 89.68; Recall 92.36 -> 92.20Mixed

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