Research map / Label-efficient and robust learning / Recognition and long-tail
MPBR: Multimodal Progressive Bidirectional Reasoning for Open-Set Fine-Grained Recognition
Junfu Tan, Peiguang Jing, Yu Zhu, Yu Liu
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 MPBR instead of HAN-OSFGR for fine-grained open-set recognition | MPBR · iNat21-OSR-Aves 1-hop | AUROC: 74.69 -> 77.17 | Improved |
| use MPBR instead of HAN-OSFGR on Insecta 1-hop | MPBR · iNat21-OSR-Insecta 1-hop | AUROC: 79.65 -> 80.69 | Improved |
| add progressive bidirectional reasoning (PBR) without text (ablation) | Swin backbone · iNat21-OSR-Aves 1-hop | AUROC: 73.78 -> 74.98 | Improved |
| add textual embeddings to PBR (ablation) | Backbone + PBR · iNat21-OSR-Aves 1-hop | AUROC: 74.98 -> 75.53 | Improved |
| add Multi-Granularity Aggregation (MGA) with textual cues (ablation) | Backbone + PBR · iNat21-OSR-Aves 1-hop | AUROC: 75.53 -> 77.17 | Improved |
| add PBR beyond 1-hop (ablation) | Swin backbone · iNat21-OSR-Aves hops 2-7 | AUROC | Mixed |
| use MPBR in coarse-grained 5-7 hop splits versus HAN-OSFGR | MPBR · iNat21-OSR-Aves 7-hop | AUROC: 98.50 -> 98.46 | No clear effect |
| use MPBR versus HAN-OSFGR on Insecta 5-hop | MPBR · iNat21-OSR-Insecta 5-hop | OSCR: 88.17 -> 88.09 | Hurt |
| add textual embedding to MPBR (ablation) | MPBR · iNat2021-OSR-Aves | Semantic Drift Index: 0.539 -> 0.546 | Improved |
| add MPBR modules to backbone (ablation) | Swin backbone · iNat2021-OSR-Aves | Semantic Drift Index: 0.378 -> 0.546 | Improved |
Closest papers by meaning
- Specificity-aware reinforcement learning for fine-grained open-world classification (CVPR 2026 · similarity 0.87)
- Fine-R1: Make Multi-modal LLMs Excel in Fine-Grained Visual Recognition by Chain-of-Thought Reasoning (ICLR 2026 · similarity 0.86)
- Fill the GAP: A Granular Alignment Paradigm for Visual Reasoning in Multimodal Large Language Models (NeurIPS 2026 · similarity 0.86)
- Democratizing Fine-grained Visual Recognition with Large Language Models (ICLR 2024 · similarity 0.86)
- From Coarse to Fine-Grained Open-Set Recognition (CVPR 2024 · similarity 0.86)
- DiVE-k: DIFFERENTIAL VISUAL REASONING FOR FINE-GRAINED IMAGE RECOGNITION (ICLR 2026 · similarity 0.86)
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.
