VIPER Strike: Defeating Visual Reasoning CAPTCHAs via Structured Vision-Language Inference

2026-01-10

arXiv.org

AI Summary

This work addresses the challenge of effectively breaking visual reasoning CAPTCHAs (VRCs), which exhibit high layout diversity and demand fine-grained perceptual understanding that general-purpose methods struggle to handle. To this end, the authors propose ViPer, a unified attack framework that integrates structured multi-object visual parsing with adaptive large language model (LLM) reasoning through a modular pipeline for visual layout parsing, attribute-semantic alignment, and target coordinate inference. ViPer is compatible with multiple mainstream LLMs—including GPT, Grok, DeepSeek, and Kimi—and achieves an average success rate of 93.2% across six major VRC platforms such as VTT and Geetest, substantially outperforming existing approaches while maintaining over 90% success across different LLM backends. Additionally, the study introduces Template Space Randomization (TSR) as a defense strategy to enhance human–machine distinguishability.

Technology Category

• Computer Vision: Large Vision Models
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Visual Reasoning CAPTCHAs (VRCs) combine visual scenes with natural-language queries that demand compositional inference over objects, attributes, and spatial relations. They are increasingly deployed as a primary defense against automated bots. Existing solvers fall into two paradigms: vision-centric, which rely on template-specific detectors but fail on novel layouts, and reasoning-centric, which leverage LLMs but struggle with fine-grained visual perception. Both lack the generality needed to handle heterogeneous VRC deployments. We present ViPer, a unified attack framework that integrates structured multi-object visual perception with adaptive LLM-based reasoning. ViPer parses visual layouts, grounds attributes to question semantics, and infers target coordinates within a modular pipeline. Evaluated on six major VRC providers (VTT, Geetest, NetEase, Dingxiang, Shumei, Xiaodun), ViPer achieves up to 93.2% success, approaching human-level performance across multiple benchmarks. Compared to prior solvers, GraphNet (83.2%), Oedipus (65.8%), and the Holistic approach (89.5%), ViPer consistently outperforms all baselines. The framework further maintains robustness across alternative LLM backbones (GPT, Grok, DeepSeek, Kimi), sustaining accuracy above 90%. To anticipate defense, we further introduce Template-Space Randomization (TSR), a lightweight strategy that perturbs linguistic templates without altering task semantics. TSR measurably reduces solver (i.e., attacker) performance. Our proposed design suggests directions for human-solvable but machine-resistant CAPTCHAs.

Problem

• Visual Reasoning CAPTCHAs
• compositional inference
• visual perception
• generalization
• bot defense

Innovation

• structured vision-language inference
• visual reasoning CAPTCHAs
• modular perception-reasoning pipeline
• template-space randomization
• LLM-based visual grounding