{"work":{"id":"a1989e1b-d66d-4533-be3a-fb9c5fd62290","openalex_id":"https://openalex.org/W4297676498","doi":"10.48550/arxiv.2209.03003","arxiv_id":"2209.03003","raw_key":null,"title":"Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow","authors":null,"authors_text":"Xingchao Liu, Chengyue Gong, Qiang Liu","year":2022,"venue":"cs.LG","abstract":"We present rectified flow, a surprisingly simple approach to learning (neural) ordinary differential equation (ODE) models to transport between two empirically observed distributions \\pi_0 and \\pi_1, hence providing a unified solution to generative modeling and domain transfer, among various other tasks involving distribution transport. The idea of rectified flow is to learn the ODE to follow the straight paths connecting the points drawn from \\pi_0 and \\pi_1 as much as possible. This is achieved by solving a straightforward nonlinear least squares optimization problem, which can be easily scaled to large models without introducing extra parameters beyond standard supervised learning. The straight paths are special and preferred because they are the shortest paths between two points, and can be simulated exactly without time discretization and hence yield computationally efficient models. We show that the procedure of learning a rectified flow from data, called rectification, turns an arbitrary coupling of \\pi_0 and \\pi_1 to a new deterministic coupling with provably non-increasing convex transport costs. In addition, recursively applying rectification allows us to obtain a sequence of flows with increasingly straight paths, which can be simulated accurately with coarse time discretization in the inference phase. In empirical studies, we show that rectified flow performs superbly on image generation, image-to-image translation, and domain adaptation. In particular, on image generation and translation, our method yields nearly straight flows that give high quality results even with a single Euler discretization step.","external_url":"https://arxiv.org/abs/2209.03003","cited_by_count":78,"metadata_source":"pith","metadata_fetched_at":"2026-08-05T02:28:24.338817+00:00","pith_arxiv_id":"2209.03003","created_at":"2026-05-09T05:55:32.160780+00:00","updated_at":"2026-08-05T02:28:24.338817+00:00","title_quality_ok":true,"display_title":"Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow","render_title":"Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow"},"hub":{"state":{"work_id":"a1989e1b-d66d-4533-be3a-fb9c5fd62290","tier":"super_hub","tier_reason":"100+ Pith inbound or 10,000+ external citations","pith_inbound_count":420,"external_cited_by_count":78,"distinct_field_count":26,"first_pith_cited_at":"2022-09-30T16:30:31+00:00","last_pith_cited_at":"2026-07-09T10:21:06+00:00","author_build_status":"needed","summary_status":"needed","contexts_status":"needed","graph_status":"needed","ask_index_status":"needed","reader_status":"not_needed","recognition_status":"not_needed","updated_at":"2026-08-21T22:19:31.943846+00:00","tier_text":"super_hub"},"tier":"super_hub","role_counts":[{"context_role":"background","n":41},{"context_role":"method","n":32},{"context_role":"baseline","n":1},{"context_role":"dataset","n":1},{"context_role":"other","n":1}],"polarity_counts":[{"context_polarity":"background","n":40},{"context_polarity":"use_method","n":32},{"context_polarity":"unclear","n":2},{"context_polarity":"baseline","n":1},{"context_polarity":"use_dataset","n":1}],"runs":{"ask_index":{"job_type":"ask_index","status":"succeeded","result":{"title":"Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow","claims":[{"claim_text":"We present rectified flow, a surprisingly simple approach to learning (neural) ordinary differential equation (ODE) models to transport between two empirically observed distributions \\pi_0 and \\pi_1, hence providing a unified solution to generative modeling and domain transfer, among various other tasks involving distribution transport. The idea of rectified flow is to learn the ODE to follow the straight paths connecting the points drawn from \\pi_0 and \\pi_1 as much as possible. This is achieved by solving a straightforward nonlinear least squares optimization problem, which can be easily sca","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-13T23:23:54.269285+00:00"},"author_expand":{"job_type":"author_expand","status":"succeeded","result":{"authors_linked":[{"id":"9e2902f0-906a-432f-9a48-055a6c70b8a7","orcid":null,"display_name":"Xingchao Liu"},{"id":"d5f5c705-12b8-459e-9f5e-d7a217ab7387","orcid":null,"display_name":"Chengyue Gong"},{"id":"e47a7999-1017-4c7d-bbfc-697e5db881fb","orcid":null,"display_name":"Qiang Liu"}]},"error":null,"updated_at":"2026-05-13T23:13:54.479237+00:00"},"context_extract":{"job_type":"context_extract","status":"succeeded","result":{"enqueued_papers":25},"error":null,"updated_at":"2026-05-13T23:14:00.945487+00:00"},"graph_features":{"job_type":"graph_features","status":"succeeded","result":{"co_cited":[{"title":"Flow Matching for Generative Modeling","work_id":"6edb71c4-5d64-40af-a394-9757ea051a36","shared_citers":95},{"title":"Score-Based Generative Modeling through Stochastic Differential Equations","work_id":"d9110e53-a5d4-4794-a4c5-a575e91c31ad","shared_citers":45},{"title":"Denoising Diffusion Implicit Models","work_id":"8fa2128b-d18c-405c-ac92-0e669cf89ac0","shared_citers":31},{"title":"Wan: Open and Advanced Large-Scale Video Generative Models","work_id":"ad3ebc3b-4224-46c9-b61d-bcf135da0a7c","shared_citers":28},{"title":"Building Normalizing Flows with Stochastic Interpolants","work_id":"83654cbd-832b-4269-bec9-26faa87ececd","shared_citers":26},{"title":"Classifier-Free Diffusion Guidance","work_id":"acf2c588-c088-4a6c-938e-150ad7c666d7","shared_citers":22},{"title":"Mean Flows for One-step Generative Modeling","work_id":"07a52ad5-0f82-4095-9a66-559b09fea1ae","shared_citers":22},{"title":"Auto-Encoding Variational Bayes","work_id":"97d95295-30e1-42b4-bbf6-85f0fa4edb44","shared_citers":18},{"title":"Decoupled Weight Decay Regularization","work_id":"07ef7360-d385-4033-83f7-8384a6325204","shared_citers":18},{"title":"SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis","work_id":"8034c587-fba6-4941-87ba-c98f2ac962cb","shared_citers":18},{"title":"Stochastic Interpolants: A Unifying Framework for Flows and Diffusions","work_id":"c2c7dd8f-fbfb-4591-89ec-9a3a0e6744bd","shared_citers":18},{"title":"CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer","work_id":"f38fc088-12aa-4bf4-9ecd-08d3e797ccb7","shared_citers":16},{"title":"Proximal Policy Optimization Algorithms","work_id":"240c67fe-d14d-4520-91c1-38a4e272ca19","shared_citers":16},{"title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models","work_id":"c5006563-f3ec-438a-9e35-b7b484f34828","shared_citers":15},{"title":"Training Diffusion Models with Reinforcement Learning","work_id":"67684dda-3930-452a-b91a-36cbb8e2e219","shared_citers":15},{"title":"Flow-GRPO: Training Flow Matching Models via Online RL","work_id":"bf1e8e81-ff31-401a-a5dc-d9c49df168ab","shared_citers":14},{"title":"FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space","work_id":"5dfe19d5-3541-4803-8fe9-3c8b9e29b281","shared_citers":14},{"title":"Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesis","work_id":"40702548-f094-4c67-a5db-a62f426f852e","shared_citers":14},{"title":"HunyuanVideo: A Systematic Framework For Large Video Generative Models","work_id":"881efa7e-7e73-4c66-9cc3-2803e551061c","shared_citers":14},{"title":"Improving and generalizing flow-based generative models with minibatch optimal transport","work_id":"7b689852-603d-46be-aaa5-c38325ba2182","shared_citers":14},{"title":"Hierarchical Text-Conditional Image Generation with CLIP Latents","work_id":"0c6a768b-70b8-4242-bb0e-459f1008c9fc","shared_citers":13},{"title":"Qwen-Image Technical Report","work_id":"d06d7ecc-7579-4f89-a60b-4278a0f3c562","shared_citers":13},{"title":"DanceGRPO: Unleashing GRPO on Visual Generation","work_id":"7404dd36-8f9c-478f-b089-ef9f8189c711","shared_citers":12},{"title":"DINOv2: Learning Robust Visual Features without Supervision","work_id":"26b304e5-b54a-4f26-be7e-83299eca52e4","shared_citers":11}],"time_series":[{"n":1,"year":2022},{"n":2,"year":2023},{"n":1,"year":2024},{"n":11,"year":2025},{"n":122,"year":2026}]},"error":null,"updated_at":"2026-05-13T23:13:57.677854+00:00"},"identity_refresh":{"job_type":"identity_refresh","status":"succeeded","result":{"fixed":1,"items":[{"title":"Qwen3 Technical Report","work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e","resolver":"local_arxiv","confidence":0.98,"old_work_id":"25a4e30c-1232-48e7-9925-02fa12ba7c9e"}],"errors":[],"attempted":1},"error":null,"updated_at":"2026-05-13T23:13:53.558682+00:00"},"role_polarity":{"job_type":"role_polarity","status":"succeeded","result":{"title":"Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow","claims":[{"claim_text":"We present rectified flow, a surprisingly simple approach to learning (neural) ordinary differential equation (ODE) models to transport between two empirically observed distributions \\pi_0 and \\pi_1, hence providing a unified solution to generative modeling and domain transfer, among various other tasks involving distribution transport. The idea of rectified flow is to learn the ODE to follow the straight paths connecting the points drawn from \\pi_0 and \\pi_1 as much as possible. This is achieved by solving a straightforward nonlinear least squares optimization problem, which can be easily sca","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-13T23:13:49.705557+00:00"},"summary_claims":{"job_type":"summary_claims","status":"succeeded","result":{"title":"Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow","claims":[{"claim_text":"We present rectified flow, a surprisingly simple approach to learning (neural) ordinary differential equation (ODE) models to transport between two empirically observed distributions \\pi_0 and \\pi_1, hence providing a unified solution to generative modeling and domain transfer, among various other tasks involving distribution transport. The idea of rectified flow is to learn the ODE to follow the straight paths connecting the points drawn from \\pi_0 and \\pi_1 as much as possible. This is achieved by solving a straightforward nonlinear least squares optimization problem, which can be easily sca","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow because it crossed a citation-hub threshold.","role_counts":[]},"error":null,"updated_at":"2026-05-13T23:13:57.683235+00:00"}},"summary":{"title":"Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow","claims":[{"claim_text":"We present rectified flow, a surprisingly simple approach to learning (neural) ordinary differential equation (ODE) models to transport between two empirically observed distributions \\pi_0 and \\pi_1, hence providing a unified solution to generative modeling and domain transfer, among various other tasks involving distribution transport. The idea of rectified flow is to learn the ODE to follow the straight paths connecting the points drawn from \\pi_0 and \\pi_1 as much as possible. This is achieved by solving a straightforward nonlinear least squares optimization problem, which can be easily sca","claim_type":"abstract","evidence_strength":"source_metadata"}],"why_cited":"Pith tracks Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow because it crossed a citation-hub threshold.","role_counts":[]},"graph":{"co_cited":[{"title":"Flow Matching for Generative Modeling","work_id":"6edb71c4-5d64-40af-a394-9757ea051a36","shared_citers":95},{"title":"Score-Based Generative Modeling through Stochastic Differential Equations","work_id":"d9110e53-a5d4-4794-a4c5-a575e91c31ad","shared_citers":45},{"title":"Denoising Diffusion Implicit Models","work_id":"8fa2128b-d18c-405c-ac92-0e669cf89ac0","shared_citers":31},{"title":"Wan: Open and Advanced Large-Scale Video Generative Models","work_id":"ad3ebc3b-4224-46c9-b61d-bcf135da0a7c","shared_citers":28},{"title":"Building Normalizing Flows with Stochastic Interpolants","work_id":"83654cbd-832b-4269-bec9-26faa87ececd","shared_citers":26},{"title":"Classifier-Free Diffusion Guidance","work_id":"acf2c588-c088-4a6c-938e-150ad7c666d7","shared_citers":22},{"title":"Mean Flows for One-step Generative Modeling","work_id":"07a52ad5-0f82-4095-9a66-559b09fea1ae","shared_citers":22},{"title":"Auto-Encoding Variational Bayes","work_id":"97d95295-30e1-42b4-bbf6-85f0fa4edb44","shared_citers":18},{"title":"Decoupled Weight Decay Regularization","work_id":"07ef7360-d385-4033-83f7-8384a6325204","shared_citers":18},{"title":"SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis","work_id":"8034c587-fba6-4941-87ba-c98f2ac962cb","shared_citers":18},{"title":"Stochastic Interpolants: A Unifying Framework for Flows and Diffusions","work_id":"c2c7dd8f-fbfb-4591-89ec-9a3a0e6744bd","shared_citers":18},{"title":"CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer","work_id":"f38fc088-12aa-4bf4-9ecd-08d3e797ccb7","shared_citers":16},{"title":"Proximal Policy Optimization Algorithms","work_id":"240c67fe-d14d-4520-91c1-38a4e272ca19","shared_citers":16},{"title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models","work_id":"c5006563-f3ec-438a-9e35-b7b484f34828","shared_citers":15},{"title":"Training Diffusion Models with Reinforcement Learning","work_id":"67684dda-3930-452a-b91a-36cbb8e2e219","shared_citers":15},{"title":"Flow-GRPO: Training Flow Matching Models via Online RL","work_id":"bf1e8e81-ff31-401a-a5dc-d9c49df168ab","shared_citers":14},{"title":"FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Space","work_id":"5dfe19d5-3541-4803-8fe9-3c8b9e29b281","shared_citers":14},{"title":"Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesis","work_id":"40702548-f094-4c67-a5db-a62f426f852e","shared_citers":14},{"title":"HunyuanVideo: A Systematic Framework For Large Video Generative Models","work_id":"881efa7e-7e73-4c66-9cc3-2803e551061c","shared_citers":14},{"title":"Improving and generalizing flow-based generative models with minibatch optimal transport","work_id":"7b689852-603d-46be-aaa5-c38325ba2182","shared_citers":14},{"title":"Hierarchical Text-Conditional Image Generation with CLIP Latents","work_id":"0c6a768b-70b8-4242-bb0e-459f1008c9fc","shared_citers":13},{"title":"Qwen-Image Technical Report","work_id":"d06d7ecc-7579-4f89-a60b-4278a0f3c562","shared_citers":13},{"title":"DanceGRPO: Unleashing GRPO on Visual Generation","work_id":"7404dd36-8f9c-478f-b089-ef9f8189c711","shared_citers":12},{"title":"DINOv2: Learning Robust Visual Features without Supervision","work_id":"26b304e5-b54a-4f26-be7e-83299eca52e4","shared_citers":11}],"time_series":[{"n":1,"year":2022},{"n":2,"year":2023},{"n":1,"year":2024},{"n":11,"year":2025},{"n":122,"year":2026}]},"authors":[{"id":"d5f5c705-12b8-459e-9f5e-d7a217ab7387","orcid":null,"display_name":"Chengyue Gong","source":"manual","import_confidence":0.72},{"id":"e47a7999-1017-4c7d-bbfc-697e5db881fb","orcid":null,"display_name":"Qiang Liu","source":"manual","import_confidence":0.72},{"id":"9e2902f0-906a-432f-9a48-055a6c70b8a7","orcid":null,"display_name":"Xingchao Liu","source":"manual","import_confidence":0.72}]}}