AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

DAMEX: Dataset-aware Mixture-of-Experts for visual understanding of mixture-of-datasets

Yash Jain, Harkirat Behl, Zsolt Kira, Vibhav Vineet

NeurIPS 2023 · Detection and segmentation › Open-set and few-shot detection · approach: Object-level representation learning · 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
route tokens to dataset-mapped expertsDINO trained on UODB · UODBMean AP: +2.0 APImproved
use dataset-aware routing rather than vanilla MoE routingDINO-MoE on UODB · UODBMean AP: +1.3 APImproved
route dataset-specific features to specialized expertsDINO trained on DOTA and DeepLesion · DOTA, DeepLesionAPImproved
use dataset-aware routingDINO trained on KITTI · UODBAPHurt
use DAMEX rather than non-MoE DINOKITTI and Kitchen mixture with 50-shot KITTI · KITTIAP: 14.6 -> 16.9Improved
use DAMEX rather than vanilla MoEKITTI and Kitchen mixture with 100-shot KITTI · KITTIAP: 20.9 -> 23.5Improved
use DAMEX rather than vanilla MoEKITTI and Kitchen mixture with 1000-shot KITTI · KITTIAP: 45.2 -> 46.1Improved
use DAMEX rather than vanilla MoEWatercolor and Comic mixture · Watercolor, ComicAP: +2.9 APImproved
use DAMEX rather than vanilla MoEKITTI and LISA mixture · KITTI, LISAAP: +1.7 AP on KITTI; +2.9 AP on LISAImproved
use human-prior dataset-to-expert mapping instead of random mapping (ablation)DAMEX on UODB · UODBMean AP: 43.7 -> 44.3Improved
use 4 experts rather than 2 experts (ablation)DAMEX on KITTI, DOTA, VOC, and Clipart · KITTI, DOTA, VOC, ClipartMean AP: 48.6 -> 49.7Improved
use 8 experts rather than 4 experts (ablation)DAMEX on KITTI, DOTA, VOC, and Clipart · KITTI, DOTA, VOC, ClipartMean AP: 49.7 -> 48.7Hurt

Later papers that compared with this method

1 later accepted paper reports DAMEX as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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