CAMLIS / PMLR · November 14, 2025

Text2VLM: Adapting Text-Only Datasets to Evaluate Visual Language Models

video Prompt Injection Model Evaluation Adversarial ML AI Red Teaming

Text2VLM: Adapting Text-Only Datasets to Evaluate Visual Language Models video thumbnail

Why it matters

Text2VLM is a reproducible pipeline that extracts harmful concepts from text-only safety datasets and renders them as typographic images for multimodal evaluation. Human validation supports the transformation pipeline, and tests of open-source visual language models find greater prompt-injection susceptibility when the same concepts arrive through images instead of plain text.

My takeaway: Transform the existing text attack suite into screenshots, documents, typographic images, and mixed text-image cases, then measure results separately by modality and layout. Preserve the source prompt and transformation metadata, validate generated cases with humans, and test OCR, perception, policy, and downstream tool behavior as distinct failure points.

Keep exploring

More curated notes connected through Prompt Injection and Model Evaluation.

OpenAI News · framework

OpenAI’s Frontier Governance Framework

OpenAI's 22-page Frontier Governance Framework maps its frontier-model processes to California's Transparency in Frontier AI Act and the EU AI Act's general-purpose AI code. It documents lifecycle risk assessment, cyber-offense and other risk tiers, mitigation and residual-risk decisions, critical-incident handling, security risk management, model reporting, external review, responsibility allocation, and change control.

OWASP GenAI Security Project · guide

OWASP Top 10 for Agentic Applications for 2026

OWASP's community guide organizes agentic-system risk into ten categories, including goal hijacking, tool misuse, identity and privilege abuse, memory poisoning, insecure inter-agent communication, cascading failures, and rogue-agent behavior. It provides a shared taxonomy and mitigation starting point rather than a certification checklist or evidence that a deployed system is secure.

OECD.AI Wonk · guide

A five-step roadmap to closing the AI evaluation gap

The roadmap addresses evaluation results that overstate real-world performance or fail to transfer across deployment contexts. Its five steps balance standardized and local tests, evaluate throughout the lifecycle, build qualified assurance and communication capacity, tailor tests to each value-chain actor and technology, and use a coordinated, trusted process for updating methods.