Research map / Architectures and efficiency / Quantization
Layer-Wise Searching for 1-Bit Detectors
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 layer-wise differentiable binarization search instead of Sign, RSign, or random search (ablation) | 1-bit detector training · VOC test2007, Faster-RCNN with ResNet-18 | mAP: 65.1, 68.9, and 64.1 -> 73.2 | Helped |
| Use layer-wise differentiable binarization search instead of Sign, RSign, or random search (ablation) | 1-bit detector training · VOC test2007, SSD with VGG-16 | mAP: 65.9, 68.1, and 60.1 -> 71.4 | Helped |
| Apply LWS-Det rather than ReActNet | 1-bit Faster-RCNN with ResNet-18 · VOC test2007 | mAP: 69.6 -> 73.2 | Helped |
| Apply LWS-Det rather than ReActNet | 1-bit Faster-RCNN with ResNet-34 · VOC test2007 | mAP: 72.3 -> 75.8 | Helped |
| Apply LWS-Det rather than ReActNet | 1-bit Faster-RCNN with ResNet-50 · VOC test2007 | mAP: 73.1 -> 76.9 | Helped |
| Apply LWS-Det rather than ReActNet | 1-bit SSD with VGG-16 · VOC test2007 | mAP: 68.4 -> 71.4 | Helped |
| Apply LWS-Det rather than ReActNet | 1-bit Faster-RCNN with ResNet-18 · COCO minival | mAP@[.5, .95]: 21.1 -> 26.9 | Helped |
| Apply LWS-Det rather than ReActNet | 1-bit Faster-RCNN with ResNet-34 · COCO minival | mAP@[.5, .95]: 23.4 -> 29.9 | Helped |
| Apply LWS-Det rather than ReActNet | 1-bit Faster-RCNN with ResNet-50 · COCO minival | mAP@[.5, .95]: 26.1 -> 31.7 | Helped |
| Apply LWS-Det rather than ReActNet | 1-bit SSD with VGG-16 · COCO minival | mAP@[.5, .95]: 15.3 -> 17.1 | Helped |
| Use 1-bit LWS-Det rather than the real-valued model | Faster-RCNN with ResNet-50 · COCO minival | mAP@[.5, .95]: 37.7 -> 31.7 | Hurt |
Closest papers by meaning
- AQ-DETR: Low-Bit Quantized Detection Transformer with Auxiliary Queries (AAAI 2024 · similarity 0.82)
- Analog Foundation Models (NeurIPS 2025 · similarity 0.82)
- Learning 1-Bit Tiny Object Detector with Discriminative Feature Refinement (ICML 2024 · similarity 0.82)
- STBLLM: Breaking the 1-Bit Barrier with Structured Binary LLMs (ICLR 2025 · similarity 0.82)
- ScaleBITS: Scalable Bitwidth Search for Hardware-Aligned Mixed-Precision LLMs (NeurIPS 2026 · similarity 0.82)
- QBB: Quantization with Binary Bases for LLMs (NeurIPS 2024 · similarity 0.81)
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.