Research map / Detection and segmentation / Open-set and few-shot detection
CoT-PL: Chain-of-Thought Pseudo-Labeling for Open-Vocabulary Object Detection
Hojun Choi, Youngsun Lim, Jaeyo Shin, Hyunjung Shim
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 SAM and MLLM 1-step pseudo-labeling (ablation) | BARON baseline · OV-COCO | AP50 novel: 34.0 -> 37.6 | Improved |
| use 3-step reasoning instead of 1-step (ablation) | MSPL · OV-COCO | AP50 novel: 37.6 -> 41.6 | Improved |
| add RTA (fine-grained linguistic attributes) (ablation) | MSPL · OV-COCO | AP50 novel: 41.6 -> 42.5 | Improved |
| add CBL (contrastive background learning) (ablation) | MSPL · OV-COCO | AP50 novel: 42.5 -> 43.4 | Improved |
| use SAM instead of Mask R-CNN proposals (ablation) | MSPL · OV-COCO | AP50 novel: 40.9 -> 43.4 | Improved |
| use SAM instead of MAVL proposals (ablation) | MSPL · OV-COCO | AP50 novel: 42.2 -> 43.4 | Improved |
| use MIN threshold for semantic anchors instead of RANDOM (ablation) | MSPL · OV-COCO | AP50 novel: 41.7 -> 43.4 | Improved |
| use MIN threshold instead of AVERAGE (ablation) | MSPL · OV-COCO | AP50 novel: 42.6 -> 43.4 | Improved |
| use Qwen2 7B instead of BLIP2 2.7B as MLLM (ablation) | MSPL · OV-COCO | AP50 novel: 39.6 -> 43.4 | Improved |
| use Qwen2 7B instead of InstructBLIP 7B (ablation) | MSPL · OV-COCO | AP50 novel: 42.6 -> 43.4 | No clear effect |
| use blur and gray context instead of black mask (ablation) | MSPL · OV-COCO | AP50 novel: 38.7 -> 43.4 | Improved |
| use blur and gray context instead of bounding box (ablation) | MSPL · OV-COCO | AP50 novel: 33.2 -> 43.4 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 4 of 4 on most shared results.
- CAKE — CAKE: Category Aware Knowledge Extraction for Open-Vocabulary Object Detection (AAAI 2025) · better on 2 of 2 shared results
- CORA — CORA: Adapting CLIP for Open-Vocabulary Detection With Region Prompting and Anchor Pre-Matching (CVPR 2023) · better on 2 of 2 shared results
- DeCo-DETR — DeCo-DETR: Decoupled Cognition DETR for efficient Open-Vocabulary Object Detection (ICLR 2026) · better on 1 of 1 shared result
- OV-DQUO — OV-DQUO: Open-Vocabulary DETR with Denoising Text Query Training and Open-World Unknown Objects Supervision (AAAI 2025) · better on 1 of 1 shared result
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.
Closest papers by meaning
- Taming Self-Training for Open-Vocabulary Object Detection (CVPR 2024 · similarity 0.89)
- DetCLIPv3: Towards Versatile Generative Open-vocabulary Object Detection (CVPR 2024 · similarity 0.89)
- Cyclic Contrastive Knowledge Transfer for Open-Vocabulary Object Detection (ICLR 2025 · similarity 0.89)
- How to Evaluate the Generalization of Detection? A Benchmark for Comprehensive Open-Vocabulary Detection (AAAI 2024 · similarity 0.88)
- Generative Region-Language Pretraining for Open-Ended Object Detection (CVPR 2024 · similarity 0.88)
- Open-Det: An Efficient Learning Framework for Open-Ended Detection (ICML 2025 · 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.
