InjectFlow is a training-free method that fixes semantic degradation and bias in Flow Matching models by injecting orthogonal semantics into the velocity field.
March 24, 2026
Original Paper
InjectFlow: Weak Guides Strong via Orthogonal Injection for Flow Matching
arXiv · 2603.20303
The Takeaway
It addresses 'trajectory lock-in' where models drift toward majority modes, causing failures in minority-class or OOD generation. On GenEval, it successfully fixed 75% of prompt failures without requiring retraining or new random seeds.
From the abstract
Flow Matching (FM) has recently emerged as a leading approach for high-fidelity visual generation, offering a robust continuous-time alternative to ordinary differential equation (ODE) based models. However, despite their success, FM models are highly sensitive to dataset biases, which cause severe semantic degradation when generating out-of-distribution or minority-class samples. In this paper, we provide a rigorous mathematical formalization of the ``Bias Manifold'' within the FM framework. We