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
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use fine-grained caption aggregate instead of bubbles for S_novice (ablation) | S_delta · CUB human study | mCP: 53.39 vs 48.40 | Improved |
| use caption aggregate instead of drawings for S_novice (ablation) | S_delta · CUB human study | mCP: 53.39 vs 49.07 | Improved |
| use caption aggregate instead of beginner classifier for S_novice (ablation) | S_delta · CUB human study | mCP: 53.39 vs 50.04 | Improved |
| use caption aggregate instead of bubbles for S_novice (ablation) | S_delta · CUB human study | mWCP: 54.24 vs 46.98 | Improved |
| use expert-exclusive S_delta instead of S_expert (ablation) | knowledge guide · CUB human study | mCP: 47.05 -> 53.39 | Improved |
| use expert-exclusive S_delta instead of S_novice (ablation) | knowledge guide · CUB human study | mCP: 40.02 -> 53.39 | Improved |
| use expert-exclusive S_delta instead of S_expert (ablation) | knowledge guide · CUB human study | mWCP: 45.51 -> 54.24 | Improved |
| show 7 regions instead of 3 (ablation) | S_delta · CUB human study | CP | Hurt |
| use practical S-hat_delta instead of S_delta without fine-grained descriptions | S_delta · CUB human study | CP | No clear effect |
| add S_delta localisation to B-CNN | B-CNN · CUB-Bird-200 | accuracy: 87.16 -> 87.78 | Improved |
| add S_delta localisation to ViT-B-16 | ViT-B-16 · CUB-Bird-200 | accuracy: 90.33 -> 90.42 | Improved |
| add S_delta localisation to PMA | PMA · CUB-Bird-200 | accuracy: 88.70 -> 88.92 | Improved |
Closest papers by meaning
- CleverBirds: A Multiple-Choice Benchmark for Fine-grained Human Knowledge Tracing (NeurIPS 2025 · similarity 0.88)
- Human and AI Perceptual Differences in Image Classification Errors (AAAI 2025 · similarity 0.80)
- Multi-scale Activation, Refinement, and Aggregation: Exploring Diverse Cues for Fine-Grained Bird Recognition (AAAI 2025 · similarity 0.80)
- Prompt-CAM: Making Vision Transformers Interpretable for Fine-Grained Analysis (CVPR 2025 · similarity 0.80)
- Democratizing Fine-grained Visual Recognition with Large Language Models (ICLR 2024 · similarity 0.80)
- RealBirdID: Benchmarking Bird Species Identification in the Era of MLLMs (CVPR 2026 · similarity 0.80)
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.
