arXiv Machine Learning

Tracing Generated Samples to Training-Data Clusters in Flow-Matching Models

arXiv Computer Vision
Oct 9

Attributing HOW, Not Just WHICH: Counterfactual Response Trajectories for Diffusion Models

The paper introduces CADT, a Concept Attribution method through Dynamic Trajectories, which attributes diffusion model outputs by tracking factor-induced internal response trajectories rather than scalar responses. CADT constructs matched counterfactual pairs at identical noisy states to isolate factor-specific representation displacements and models their directional and magnitude evolution across denoising as dynamic attribution signatures. Experiments on multiple public datasets demonstrate consistent improvements over existing diffusion attribution baselines for hierarchical, compositional, and style attribution tasks.

By Haoqian Zhang, Ziyuan Yang, Zerui Shao, Yi Zhang
arXiv AI
Oct 1

Distilling Diffusion Score Discrepancy for Efficient Training Data Attribution

The paper introduces a new method for training data attribution in diffusion models called TID, which uses a local score discrepancy measure and can be estimated without retraining. It further distills this approach into TIDE, a forward‑only student that reproduces the teacher’s rankings using internal activations, achieving comparable accuracy at dramatically lower query cost. Experiments on CIFAR‑10, ArtBench‑10, and MS‑COCO show that TID outperforms existing methods and TIDE attributes samples in milliseconds, faster than generation itself.

By Shixuan Liu, Joan Serr\`a, Kin Wai Cheuk, Jinju Kim, Woosung Choi, Yukara Ikemiya, Wei-Hsiang Liao, Jiaqi W. Ma, Yuki Mitsufuji
arXiv Computer Vision
Nov 9

Flow-Guided Target-Space Alignment via Path Consistency

FlowAlign is a flow‑guided target‑space alignment method that uses conditional flow matching as a training‑time signal to map source modality representations into a fixed target space. It preserves both the global structure of the target distribution and instance‑level correspondence in paired source–target data, unlike existing contrastive, distribution‑matching, or pointwise regression objectives. The authors provide theoretical analysis and demonstrate improved instance‑level alignment while maintaining target‑space structure on synthetic data, single‑cell multiome benchmarks, and image–text alignment tasks.

By Ruizhi Yuan, Zeqiu Yu, Wei Gao, Wei Chen, Xiwei Tang, Lu Tang
arXiv Computer Vision
Nov 2

Differential Vector Erasure: Unified Training-Free Concept Erasure for Flow Matching Models

The paper introduces Differential Vector Erasure (DVE), a training‑free method for removing unwanted concepts from flow matching image generation models. DVE exploits the directional structure of the velocity field that encodes semantic concepts, constructing a differential vector field to isolate and suppress target concepts during inference. Experiments on the FLUX model show that DVE outperforms existing baselines in tasks such as NSFW suppression, style removal, and object erasure while maintaining image quality and diversity.

By Zhiqi Zhang, Xinhao Zhong, Yi Sun, Shuoyang Sun, Jiahui Wang, Bin Chen, Shu-Tao Xia, Xuan Wang
arXiv AI
Sep 2

Training-Free Refinement of Flow Matching with Divergence-based Sampling

The paper introduces Flow Divergence Sampler (FDS), a training‑free method that refines intermediate states in flow‑matching models by using the divergence of the marginal velocity field to detect and correct misguidance toward low‑density regions. FDS operates during inference, requires no additional training, and can be applied as a plug‑and‑play module with standard solvers and existing flow backbones. Experiments show that FDS consistently improves fidelity in tasks such as text‑to‑image synthesis and inverse problems.

By Yeonwoo Cha, Jaehoon Yoo, Semin Kim, Yunseo Park, Jinhyeon Kwon, Seunghoon Hong