Research map / Vision-language and multimodal / Multimodal LLMs
WeatherReasonSeg: A Benchmark for Weather-Aware Reasoning Segmentation in Visual Language Models
Wanjun Du, Zifeng Yuan, Tingting Chen, Fucai Ke, Beibei Lin, Shunli Zhang
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 |
|---|---|---|---|
| increase synthetic weather severity from clean to heavy (ablation) | Seg-Zero* · WeatherReasonSeg synthetic | gIoU: 65.3 -> 47.7 | Hurt |
| heavy fog on synthetic ReasonSeg (ablation) | Seg-Zero-7B · WeatherReasonSeg synthetic fog test | gIoU: 57.5 -> 49.9 | Hurt |
| heavy snow on synthetic ReasonSeg (ablation) | Seg-Zero-7B · WeatherReasonSeg synthetic snow test | cIoU: 52 -> 39.72 | Hurt |
| heavy weather on grounding-based pipeline (ablation) | Grounded SAM · WeatherReasonSeg synthetic fog test | gIoU: 21.3 -> 12.8 | Hurt |
| increase severity light to heavy (ablation) | LISA-7B · WeatherReasonSeg synthetic fog test | gIoU: 41.8 -> 39.4 | Hurt |
| snow vs rain vs fog weather type (heavy) (ablation) | Seg-Zero-7B · WeatherReasonSeg synthetic test | gIoU: fog 49.9, rain 47.8, snow 45.7 | Mixed |
| real-world adverse weather instead of perception-only SAM2 | Seg-Zero-7B · WeatherReasonSeg real-world night | gIoU: 74.8 -> 28.9 | Hurt |
| real-world fog: reasoning segmentation vs SAM2 | Seg-Zero-7B · WeatherReasonSeg real-world fog | gIoU: 82.2 -> 40.6 | Hurt |
| Requirement vs Function queries under rain | Seg-Zero · WeatherReasonSeg real-world rain | cIoU: 41.2 -> 31.8 | Hurt |
| context-dependent query dimensions (Application, Requirement) vs observable (Structure, Function) | Seg-Zero · WeatherReasonSeg real-world night | cIoU: 31.2 -> 22.4 | Hurt |
Closest papers by meaning
- Semantic Segmentation in Multiple Adverse Weather Conditions with Domain Knowledge Retention (AAAI 2024 · similarity 0.89)
- Parsing All Adverse Scenes: Severity-Aware Semantic Segmentation with Mask-Enhanced Cross-Domain Consistency (AAAI 2024 · similarity 0.88)
- End-to-End Video Semantic Segmentation in Adverse Weather using Fusion Blocks and Temporal-Spatial Teacher-Student Learning (NeurIPS 2024 · similarity 0.87)
- VLRS-Bench: A Vision-Language Reasoning Benchmark for Remote Sensing (NeurIPS 2026 · similarity 0.86)
- LISAt: Language-Instructed Segmentation Assistant for Satellite Imagery (NeurIPS 2025 · similarity 0.86)
- ERF: A Benchmark Dataset for Robust Semantic Segmentation Under Extreme Rainfall Conditions (AAAI 2025 · similarity 0.86)
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.
