Research map / Vision-language and multimodal / Human and embodied modeling
Guide Me: Interacting With Deep Networks
Read the paperCheck 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 |
|---|---|---|---|
| Add iterative back-propagation guidance using pixel-class hints | FCN-8s · PascalVOC 2012 val | mIoU: 62.6 -> 81.0 | Helped |
| Add one text hint and guide at res5c | DeepLab with ResNet-101 · COCO-Stuff test | mIoU: 30.5% to 36.5% | Helped |
| Use the spatially extended guiding block instead of FiLM without a residual block (ablation) | DeepLab with ResNet-101 · COCO-Stuff test, res4a with find queries | mIoU: 33.31 -> 33.56 | Helped |
| Use the spatially extended guiding block instead of FiLM with a residual block (ablation) | DeepLab with ResNet-101 · COCO-Stuff test, res4a with find queries | mIoU: 33.08 -> 33.11 | No clear effect |
| Place the guide later in the network, at res5c rather than res3a (ablation) | DeepLab with ResNet-101 · COCO-Stuff test with find queries | mIoU: 32.21 -> 36.50 | Helped |
| Use find queries instead of remove queries (ablation) | DeepLab with ResNet-101 · COCO-Stuff test, res4a | mIoU: 31.53 -> 33.56 | Helped |
| Use find queries instead of remove queries (ablation) | DeepLab with ResNet-101 · COCO-Stuff test, res5a | mIoU: 32.56 -> 35.97 | Helped |
| Apply a second successive find or rmv hint rather than stopping after one (ablation) | DeepLab with ResNet-101 · COCO-Stuff test | mIoU: 34.04 -> 35.01 | Helped |
| Apply a third or fourth successive find or rmv hint rather than stopping after two (ablation) | DeepLab with ResNet-101 · COCO-Stuff test | mIoU: 35.01 -> 34.24; 31.44 | Hurt |
Closest papers by meaning
- Language Models as Black-Box Optimizers for Vision-Language Models (CVPR 2024 · similarity 0.87)
- IPO: Interpretable Prompt Optimization for Vision-Language Models (NeurIPS 2024 · similarity 0.86)
- VISion On Request: Enhanced VLLM efficiency with sparse, dynamically selected, vision-language interactions (CVPR 2026 · similarity 0.85)
- Memory-Space Visual Prompting for Efficient Vision-Language Fine-Tuning (ICML 2024 · similarity 0.85)
- VLsI: Verbalized Layers-to-Interactions from Large to Small Vision Language Models (CVPR 2025 · similarity 0.85)
- Language-Instructed Vision Embeddings for Controllable and Generalizable Perception (ICLR 2026 · 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.