{"total":12,"items":[{"citing_arxiv_id":"2605.22795","ref_index":7,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Finite-Particle Convergence Rates for Conservative and Non-Conservative Drifting Models","primary_cat":"stat.ML","submitted_at":"2026-05-21T17:49:09+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"Derives continuous-time finite-particle convergence rates for a new conservative KDE-gradient drifting method and the non-conservative Laplace kernel method in one-step generative modeling.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.17808","ref_index":5,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"A Unified Framework for Data-Free One-Step Sampling via Wasserstein Gradient Flows","primary_cat":"cs.LG","submitted_at":"2026-05-18T03:32:22+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A unified framework decomposes Wasserstein gradient flow velocity fields across f-divergences into a shared beta direction and divergence-specific weighting, enabling data-free one-step sampling.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.17244","ref_index":11,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Drift Flow Matching","primary_cat":"cs.LG","submitted_at":"2026-05-17T03:58:01+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"Drift Flow Matching connects direct transport maps from Drift Models with flow-based iterative refinement to enable adaptive computation in generative modeling.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.11755","ref_index":25,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"One-Step Generative Modeling via Wasserstein Gradient Flows","primary_cat":"cs.LG","submitted_at":"2026-05-12T08:29:44+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"W-Flow compresses a Wasserstein gradient flow defined via Sinkhorn divergence into a single-step neural generator, reporting 1.29 FID on ImageNet 256x256 with improved mode coverage.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Ablation setting.To enable an informative comparison in a controlled setting, we follow the default ablation setup of [11]. Specifically, all ablation experiments are conducted in the SD-V AE [48] latent space, using a B/2 DiT generator together with the pretrained latent-MAE feature encoder from [11]. The generator is trained for 100 epochs, and FID [25] is evaluated on 50K generated images. We set ε= 0.05by default with additional detailed hyperparameters provided in Appendix C. Energy functionals in gradient flow.We implement and compare three different potentials: squared MMD, KL-divergence, and Sinkhorn divergence (Eq. (8)). Results in Table 2a confirm that Sinkhorn divergence performs the best, whereas both MMD and KL divergence yield noticeably weaker"},{"citing_arxiv_id":"2605.07727","ref_index":22,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Drifting Field Policy: A One-Step Generative Policy via Wasserstein Gradient Flow","primary_cat":"cs.LG","submitted_at":"2026-05-08T13:34:27+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"DFP is a one-step generative policy using Wasserstein gradient flow on a drifting model backbone, with a top-K behavior cloning surrogate, that reaches SOTA on Robomimic and OGBench manipulation tasks.","context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Haarnoja, A. Zhou, P. Abbeel, and S. Levine. Soft actor-critic: Off-policy maximum entropy deep reinforcement learning with a stochastic actor. InICML, 2018. 10 [21] P. Hansen-Estruch, I. Kostrikov, M. Janner, J. G. Kuba, and S. Levine. Idql: Implicit q-learning as an actor-critic method with diffusion policies.arXiv preprint arXiv:2304.10573, 2023. [22] P. He, O. Khangaonkar, H. Pirsiavash, Y . Bai, and S. Kolouri. Sinkhorn-drifting generative models.arXiv preprint arXiv:2603.12366, 2026. [23] J. Ho, A. Jain, and P. Abbeel. Denoising diffusion probabilistic models. InNeurIPS, 2020. [24] M. Janner, Y . Du, J. B. Tenenbaum, and S. Levine. Planning with diffusion for flexible behavior synthesis. InICML, 2022."},{"citing_arxiv_id":"2605.07327","ref_index":5,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Teacher-Feature Drifting: One-Step Diffusion Distillation with Pretrained Diffusion Representations","primary_cat":"cs.CV","submitted_at":"2026-05-08T06:33:33+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":5.0,"formal_verification":"none","one_line_summary":"A simplified one-step diffusion distillation uses pretrained teacher features directly for drifting loss plus a mode coverage term, achieving FID 1.58 on ImageNet-64 and 18.4 on SDXL.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.06140","ref_index":53,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"SymDrift: One-Shot Generative Modeling under Symmetries","primary_cat":"cs.LG","submitted_at":"2026-05-07T12:38:44+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"SymDrift makes drifting models produce symmetry-invariant samples in one step via symmetrized coordinate drifts or G-invariant embeddings, outperforming prior one-shot baselines on molecular benchmarks and cutting compute by up to 40x.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2605.05118","ref_index":16,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"On the Wasserstein Gradient Flow Interpretation of Drifting Models","primary_cat":"cs.LG","submitted_at":"2026-05-06T16:48:46+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"The paper interprets GMD algorithms as limiting points of Wasserstein gradient flows on KL divergence with Parzen smoothing and on Sinkhorn divergence, while extending the approach to MMD, sliced Wasserstein, and GAN critics.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.24196","ref_index":12,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Identifiability and Stability of Generative Drifting with Companion-Elliptic Kernel Families","primary_cat":"stat.ML","submitted_at":"2026-04-27T08:56:39+00:00","verdict":null,"verdict_confidence":null,"novelty_score":null,"formal_verification":null,"one_line_summary":null,"context_count":1,"top_context_role":"background","top_context_polarity":"background","context_text":"Cao, Wei, and Liu placed drifting in a broader Wasserstein gradient-flow framework for KDE-approximated divergences [2, 9], while explicitly noting that the Laplace kernel does not belong to their kernel class [9]. He et al. proved identifiability for a Sinkhorn-normalized two-sided variant, thereby highlighting the gap between one-sided and two- sided formulations [12]. Within the one-sided raw drifting formulation of Deng et al. [1], the present paper establishes identifiability on the full space of Borel probability measures and characterizes the stability obstruction caused by mass escape. Our results cover both the Laplace kernel and the companion-elliptic kernel family that contains it. 3 Notation Notation for function spaces."},{"citing_arxiv_id":"2604.23944","ref_index":26,"ref_count":2,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Sliced-Regularized Optimal Transport","primary_cat":"stat.ML","submitted_at":"2026-04-27T01:36:29+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"SROT regularizes the OT transport plan toward a sliced OT reference, yielding better approximations of exact OT than entropic OT and improving on the sliced OT plan itself.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2604.15114","ref_index":36,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Amortized Optimal Transport from Sliced Potentials","primary_cat":"stat.ML","submitted_at":"2026-04-16T15:05:59+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":6.0,"formal_verification":"none","one_line_summary":"RA-OT and OA-OT amortize optimal transport by regressing or optimizing sliced-OT Kantorovich potentials to approximate full OT plans efficiently across multiple measure pairs.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null},{"citing_arxiv_id":"2603.25182","ref_index":18,"ref_count":1,"confidence":0.9,"is_internal_anchor":false,"paper_title":"Learning Monge maps with constrained drifting models","primary_cat":"math.OC","submitted_at":"2026-03-26T08:51:51+00:00","verdict":"UNVERDICTED","verdict_confidence":"LOW","novelty_score":7.0,"formal_verification":"none","one_line_summary":"A new constrained gradient flow on the space of transport maps converges to the OT map and enables more stable and accurate training of convexity-constrained neural networks for learning Monge maps.","context_count":0,"top_context_role":null,"top_context_polarity":null,"context_text":null}],"limit":50,"offset":0}