Topo-R1: Detecting Topological Anomalies via Vision-Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xu, Meilong, Hu, Qingqiao, Hu, Xiaoling, Abousamra, Shahira, Yu, Xin, Lyu, Weimin, Qi, Kehan, Samaras, Dimitris, Chen, Chao
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913119553454080
author Xu, Meilong
Hu, Qingqiao
Hu, Xiaoling
Abousamra, Shahira
Yu, Xin
Lyu, Weimin
Qi, Kehan
Samaras, Dimitris
Chen, Chao
author_facet Xu, Meilong
Hu, Qingqiao
Hu, Xiaoling
Abousamra, Shahira
Yu, Xin
Lyu, Weimin
Qi, Kehan
Samaras, Dimitris
Chen, Chao
contents Topology is critical in tubular structures such as blood vessels, nerve fibers, and road networks, where connectivity and loop structure govern downstream functional analysis. Vision-Language Models (VLMs) are promising candidates for understanding such structures, given their reasoning and grounding capabilities. To probe their topological perception, we systematically evaluate leading closed- and open-source VLMs on localizing and classifying four canonical topological anomalies (broken/spurious connections, missing/extra branches) in tubular-network segmentation masks. They perform nearly at random, indicating that topology-aware perception is largely absent from current general-purpose VLMs. As no existing resource pairs segmentation masks with localized anomaly annotations, we build an automated, multi-domain data-curation pipeline that synthesizes diverse topological perturbations with verifiable Betti-number annotations across graduated difficulty levels, yielding the first systematic benchmark with a large-scale training set and held-out in-distribution (ID) and out-of-distribution (OOD) test suites. Building on this benchmark, we introduce Topo-R1, centered on a topology-aware composite reward that jointly scores localization, classification, and skeleton-level structural fidelity. Supervised fine-tuning cold-starts schema-compliant outputs, and Group Relative Policy Optimization (GRPO) then optimizes the policy against this reward, steering predictions toward topologically meaningful structure rather than superficial pixel overlap. Extensive experiments show that Topo-R1 substantially outperforms general-purpose VLMs and matches or exceeds supervised baselines across ID, OOD, and real-segmentation-output protocols, establishing a strong foundation for VLM-based topological understanding of structured visual data.
format Preprint
id arxiv_https___arxiv_org_abs_2603_13054
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Topo-R1: Detecting Topological Anomalies via Vision-Language Models
Xu, Meilong
Hu, Qingqiao
Hu, Xiaoling
Abousamra, Shahira
Yu, Xin
Lyu, Weimin
Qi, Kehan
Samaras, Dimitris
Chen, Chao
Computer Vision and Pattern Recognition
Topology is critical in tubular structures such as blood vessels, nerve fibers, and road networks, where connectivity and loop structure govern downstream functional analysis. Vision-Language Models (VLMs) are promising candidates for understanding such structures, given their reasoning and grounding capabilities. To probe their topological perception, we systematically evaluate leading closed- and open-source VLMs on localizing and classifying four canonical topological anomalies (broken/spurious connections, missing/extra branches) in tubular-network segmentation masks. They perform nearly at random, indicating that topology-aware perception is largely absent from current general-purpose VLMs. As no existing resource pairs segmentation masks with localized anomaly annotations, we build an automated, multi-domain data-curation pipeline that synthesizes diverse topological perturbations with verifiable Betti-number annotations across graduated difficulty levels, yielding the first systematic benchmark with a large-scale training set and held-out in-distribution (ID) and out-of-distribution (OOD) test suites. Building on this benchmark, we introduce Topo-R1, centered on a topology-aware composite reward that jointly scores localization, classification, and skeleton-level structural fidelity. Supervised fine-tuning cold-starts schema-compliant outputs, and Group Relative Policy Optimization (GRPO) then optimizes the policy against this reward, steering predictions toward topologically meaningful structure rather than superficial pixel overlap. Extensive experiments show that Topo-R1 substantially outperforms general-purpose VLMs and matches or exceeds supervised baselines across ID, OOD, and real-segmentation-output protocols, establishing a strong foundation for VLM-based topological understanding of structured visual data.
title Topo-R1: Detecting Topological Anomalies via Vision-Language Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.13054