Research map / Vision-language and multimodal / CLIP and zero-shot
Thinking Beyond Labels: Vocabulary-Free Fine-Grained Recognition using Reasoning-Augmented LMMs
Dmitry Demidov, Muhammad Zaigham Zaheer, Zongyan Han, Omkar Thawakar, Rao Anwer
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 |
|---|---|---|---|
| add meta-category information to the prompt (ablation) | FiNDR (Qwen2.5-VL-72B) · Oxford Pets | sACC: 74.02 -> 82.96 | Improved |
| add meta-category information to the prompt (ablation) | FiNDR (Qwen2.5-VL-72B) · CUB-200 | cACC: 57.84 -> 58.90 | Improved |
| add expert preamble to the prompt (ablation) | FiNDR (Qwen2.5-VL-72B) · Oxford Pets | sACC: 82.96 -> 84.46 | Improved |
| add expert preamble to the prompt (ablation) | FiNDR (Qwen2.5-VL-72B) · Stanford Cars | cACC: 55.95 -> 55.19 | Mixed |
| use FiNDR prompt instead of base prompt (ablation) | Qwen2.5-VL · CUB-200 | sACC: 74.59 -> 76.65 | Improved |
| add internal reasoning (thinking) to explicit prompting (ablation) | Gemini2.5-Flash · CUB-200 | sACC: 76.86 -> 78.04 | Improved |
| use FiNDR prompt instead of base prompt (ablation) | Gemini2.5-Flash · CUB-200 | sACC: 73.46 -> 76.86 | Improved |
| use FiNDR over E-FineR (reasoning LMM discovery) | FiNDR · 5 fine-grained datasets (avg) | cACC: 58.4 -> 67.9 | Improved |
| use FiNDR over E-FineR | FiNDR · Oxford Flowers | sACC: 54.0 -> 56.5 | Improved |
| use FiNDR over E-FineR | FiNDR · Stanford Cars | sACC: 64.0 -> 64.4 | No clear effect |
| vocabulary-free FiNDR vs zero-shot CLIP with ground-truth names | FiNDR · Average of 5 datasets | cACC: 65.8 -> 67.9 | Improved |
Closest papers by meaning
- Democratizing Fine-grained Visual Recognition with Large Language Models (ICLR 2024 · similarity 0.91)
- Fine-R1: Make Multi-modal LLMs Excel in Fine-Grained Visual Recognition by Chain-of-Thought Reasoning (ICLR 2026 · similarity 0.90)
- Analyzing and Boosting the Power of Fine-Grained Visual Recognition for Multi-modal Large Language Models (ICLR 2025 · similarity 0.89)
- Benchmarking Large Vision-Language Models on Fine-Grained Image Tasks: A Comprehensive Evaluation (ICLR 2026 · similarity 0.89)
- OVMR: Open-Vocabulary Recognition with Multi-Modal References (CVPR 2024 · similarity 0.89)
- From Coarse to Fine-Grained Open-Set Recognition (CVPR 2024 · similarity 0.88)
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.
