Research map / Label-efficient and robust learning / Recognition and long-tail
LoCo: Selective Local Competition for Discriminative Open-Vocabulary Multi-Label Recognition
Zhuoming Li, Bo Han, Xiaoyu Wang, Yuheng Jia
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes LoCo, a selective local competition method for open-vocabulary multi-label recognition that uses local prediction as a discriminative branch.
Problem
Open-vocabulary multi-label recognition (identifying all queried concepts in an image) has focused on global image-text matching and text-side representations, leaving local prediction under-explored.
Approach
Revisits local prediction and shows it can act as a strong discriminative branch once spatial consistency and related properties are handled; introduces selective local competition to achieve this.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace CLIP with ClearCLIP dense features for local prediction (ablation) | LoCo local branch · OV-MLR avg of 4 | F1 / mAP: 46.4 -> 52.5 F1; 55.0 -> 59.8 mAP | Improved |
| use SC-CLIP instead of ClearCLIP (ablation) | LoCo local branch · OV-MLR avg of 4 | mAP: 59.8 -> 62.4 | Improved |
| add Selective Softmax (ablation) | LoCo (SC-CLIP, 14 crops) · OV-MLR avg of 4 | F1-score: 56.6 -> 58.5 | Improved |
| add Selective Softmax (ablation) | LoCo (SC-CLIP, 14 crops) · OV-MLR avg of 4 | mAP: 65.5 -> 66.0 | Improved |
| add 14-crop multi-scale inference (ablation) | LoCo (SC-CLIP + Sel) · OV-MLR avg of 4 | mAP: 62.9 -> 66.0 | Improved |
| use raw logits (no competition) instead of Selective Softmax (ablation) | LoCo · OV-MLR avg of 4 | F1-score: 58.5 -> 53.1 | Hurt |
| use full softmax instead of Selective Softmax (ablation) | LoCo · OV-MLR avg of 4 | F1-score: 58.5 -> 56.6 | Hurt |
| use random compatibility prior instead of LLM prior (ablation) | LoCo · OV-MLR avg of 4 | F1-score: 58.5 -> 52.9 | Hurt |
| use WordNet prior instead of LLM prior (ablation) | LoCo · OV-MLR avg of 4 | F1-score: 58.5 -> 57.0 | Hurt |
| use LLM consensus prior instead of single GPT-5.5 prior (ablation) | LoCo · OV-MLR avg of 4 | F1-score: 58.5 -> 58.6 | No clear effect |
| add multi-scale inference to TaI-DPT with SC-CLIP (ablation) | TaI-DPT local prediction · OV-MLR avg of 4 | F1-score: 55.8 -> 55.7 | No clear effect |
| apply LoCo with CLIP backbone instead of SC-CLIP (ablation) | LoCo · OV-MLR avg of 4 | mAP: 66.0 -> 55.3 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- TagCLIP — TagCLIP: A Local-to-Global Framework to Enhance Open-Vocabulary Multi-Label Classification of CLIP without Training (AAAI 2024) · better on 6 of 7 shared results
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.
Surprising findings
- Local prediction, often overlooked, can serve as a strong discriminative branch for OV-MLR once spatial consistency and related properties are properly handled.
Closest papers by meaning
- Recover and Match: Open-Vocabulary Multi-Label Recognition through Knowledge-Constrained Optimal Transport (CVPR 2025 · similarity 0.89)
- OVMR: Open-Vocabulary Recognition with Multi-Modal References (CVPR 2024 · similarity 0.88)
- Image-to-Image Matching via Foundation Models: A New Perspective for Open-Vocabulary Semantic Segmentation (CVPR 2024 · similarity 0.86)
- Language-conditioned Detection Transformer (CVPR 2024 · similarity 0.86)
- TagCLIP: A Local-to-Global Framework to Enhance Open-Vocabulary Multi-Label Classification of CLIP without Training (AAAI 2024 · similarity 0.86)
- Training-free Open-Vocabulary Semantic Segmentation via Diverse Prototype Construction and Sub-region Matching (AAAI 2025 · similarity 0.85)
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.
