AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

Making a Bird AI Expert Work for You and Me

Dongliang Chang, Kaiyue Pang, Ruoyi Du, Yujun Tong, Yi-Zhe Song, Zhanyu Ma, Jun Guo

TPAMI 2023 · Vision-language and multimodal › Multimodal LLMs · approach: Multimodal LLM benchmarks · 22 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
use fine-grained caption aggregate instead of bubbles for S_novice (ablation)S_delta · CUB human studymCP: 53.39 vs 48.40Improved
use caption aggregate instead of drawings for S_novice (ablation)S_delta · CUB human studymCP: 53.39 vs 49.07Improved
use caption aggregate instead of beginner classifier for S_novice (ablation)S_delta · CUB human studymCP: 53.39 vs 50.04Improved
use caption aggregate instead of bubbles for S_novice (ablation)S_delta · CUB human studymWCP: 54.24 vs 46.98Improved
use expert-exclusive S_delta instead of S_expert (ablation)knowledge guide · CUB human studymCP: 47.05 -> 53.39Improved
use expert-exclusive S_delta instead of S_novice (ablation)knowledge guide · CUB human studymCP: 40.02 -> 53.39Improved
use expert-exclusive S_delta instead of S_expert (ablation)knowledge guide · CUB human studymWCP: 45.51 -> 54.24Improved
show 7 regions instead of 3 (ablation)S_delta · CUB human studyCPHurt
use practical S-hat_delta instead of S_delta without fine-grained descriptionsS_delta · CUB human studyCPNo clear effect
add S_delta localisation to B-CNNB-CNN · CUB-Bird-200accuracy: 87.16 -> 87.78Improved
add S_delta localisation to ViT-B-16ViT-B-16 · CUB-Bird-200accuracy: 90.33 -> 90.42Improved
add S_delta localisation to PMAPMA · CUB-Bird-200accuracy: 88.70 -> 88.92Improved

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