AI Paper Atlas

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

VLOD-TTA: Test-Time Adaptation of Vision-Language Object Detectors

Atif Belal, Heitor Medeiros, Marco Pedersoli, Eric Granger

ECCV 2026 · Vision-language and multimodal › CLIP and zero-shot · approach: Test-time adaptation of VLMs · 5 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
replace standard entropy with IoU-weighted entropy (IWE) (ablation)Adapter (YOLO-World) · Watercolor, ClipArt, ComicmAP / AP50Improved
add image-conditioned prompt selection (IPS) on top of IWE (ablation)VLOD-TTA (YOLO-World) · Watercolor, ClipArt, ComicmAP / AP50Improved
use prompt averaging instead of ZS prompts (ablation)YOLO-World · Watercolor and ComicAP50: -0.35 on Watercolor, -0.30 on ComicHurt
use prompt selection instead of averaging (ablation)YOLO-World · Watercolor, ClipArt, ComicAP50: average gain of +1.0 over ZSImproved
use CLIP-style prompts instead of GPT prompts (ablation)VLOD-TTA (YOLO-World) · ClipArt, ComicAP50: -0.8 on ClipArt, -1.3 on ComicHurt
use dataset-specific prompts instead of dataset-agnostic GPT prompts (ablation)VLOD-TTA (YOLO-World) · Watercolor, ClipArt, ComicAP50: +0.9 -> +1.0 Watercolor, +1.0 -> +1.6 ClipArt, +1.2 -> +1.7 ComicImproved
adapt vision backbone with adapters (ablation)YOLO-World · style-shift datasets avgAP50: +4.4 AP50Improved
adapt text encoder with adapters (ablation)Grounding DINO · style-shift datasets avgAP50: +3.3 AP50Improved
adapt vision encoder or neck with adapters (ablation)Grounding DINO · style-shift datasets avgAP50Hurt
add single scale augmentation (ablation)VLOD-TTA (YOLO-World) · Watercolor, ClipArt, ComicAP50: +6.0, +6.6, +5.1Improved
add more than one augmentation (ablation)VLOD-TTA (YOLO-World) · Watercolor, ClipArt, ComicAP50No clear effect
increase input resolution from 640x640 to 1024x1024 (ablation)VLOD-TTA (YOLO-World) · CityscapesmAP gain over ZS: +0.6 mAP -> +2.7 mAPImproved

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