TINA: Text-Free Inversion Attack for Unlearned Text-to-Image Diffusion Models
TINA: Text-Free Inversion Attack for Unlearned Text-to-Image Diffusion Models explores TINA is a novel attack method that uncovers erased concepts in text-to-image diffusion models by leveraging visual-only probes.. Commercial viability score: 5/10 in Adversarial Attacks.
BUILDER'S SANDBOX
Build This Paper
Use an AI coding agent to implement this research.
OpenAI CodexAI Agent
Lightweight coding agent in your terminal.
Claude CodeAI Agent
Agentic coding tool for terminal workflows.
AntiGravity IDEScaffolding
AI agent mindset installer and workflow scaffolder.
CursorIDE
AI-first code editor built on VS Code.
VS CodeIDE
Free, open-source editor by Microsoft.
Recommended Stack
Startup Essentials
Estimated $9K - $13K over 6-10 weeks.
See exactly what it costs to build this -- with 3 comparable funded startups.
7-day free trial. Cancel anytime.
Discover the researchers behind this paper and find similar experts.
7-day free trial. Cancel anytime.
Related Papers
- Builds On ThisWhy Instruction-Based Unlearning Fails in Diffusion Models?Viability: 3.0
- Prior WorkLocalized Concept Erasure in Text-to-Image Diffusion Models via High-Level Representation MisdirectionViability: 5.0
- Higher ViabilityA Concept is More Than a Word: Diversified Unlearning in Text-to-Image Diffusion ModelsViability: 7.0
- Higher ViabilityZ-Erase: Enabling Concept Erasure in Single-Stream Diffusion TransformersViability: 7.0
- Higher ViabilityPrototype-Guided Concept Erasure in Diffusion Models