AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

Robust onion: Peeling Open Vocab Object Detectors Under Noise

Priyank Pathak, Mukilan Karuppasamy, Aaditya Baranwal, Shruti Vyas, Yogesh Rawat

ECCV 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Biologically inspired robust vision · 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 trainable spatial tokens (TK0) to frozen GLIP-T backboneGLIP-T · BDD-100K trained, AllmAP/AP50 (%): 31.4 -> 45.1Improved
add TK0 tokens to frozen modelGLIP-T · COCO zero-shotmAP (%): 46.6 -> 43.2Hurt
end-to-end finetuning instead of frozen modelGLIP-T · COCO zero-shotmAP (%): 46.6 -> 3.0Hurt
train only fusion transformer (Fuse) instead of full E2EGLIP-T · BDD-100K trained, AllmAP/AP50 (%): 66.4 -> 65.1No clear effect
combine NN cross-exchange block with TK0 (ablation)GLIP-T · BDD-100K trained, AllmAP/AP50 (%): 45.1 -> 55.2Improved
combine NN and TK0 versus TK0 alone (ablation)GLIP-T · COCO zero-shotmAP (%): 43.2 -> 33.4Hurt
NN & TK0 instead of zero-shotGLIP-T · Foggy Cityscapes zero-shotmAP (%): 21.4 -> 28.2Improved
NN & TK0 instead of zero-shotGLIP-T · Virtual KITTI 2 zero-shot (All)mAP (%): 20.9 -> 29.9Improved
NN & TK0 instead of zero-shotGLIP-T · Flickr30k -> WiderFace APAP: 11.98 -> 13.23Improved
NN & TK0 instead of zero-shotGLIP-T · UAVDT -> VisDRONE APAP: 20.44 -> 19.34Hurt
NN & TK0 (2.41M params) instead of E2E (231.76M)GLIP-T · trainable parametersparams (M): 231.76 -> 2.41Improved
add LLM-modified noise-aware captions at evaluationGLIP variants · Flickr30k pixelationRecall@1No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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