Research map / Image generation / Fast diffusion sampling
Guidance and alignment for flow matching
Improve guidance, stability, and reward alignment in flow matching models.
34 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2025–2026) · an approach in Fast diffusion sampling, keyword cluster “flow · matching · rectified”
State History Tried Checklist Preprints Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
0 papers in 2023–24 and 34 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic grew, from 0.0% to 2.6%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2025 · GoalFlow: Goal-Driven Flow Matching for Multimodal Trajectories Generation in End-to-End Autonomous Driving (CVPR 2025 · 30 citations)
- 2026 · Stable Velocity: A Variance Perspective on Flow Matching (ICML 2026 · 0 citations)
Earliest papers here
- GoalFlow: Goal-Driven Flow Matching for Multimodal Trajectories Generation in End-to-End Autonomous Driving (CVPR 2025 · 30 citations)
- FlowRAM: Grounding Flow Matching Policy with Region-Aware Mamba Framework for Robotic Manipulation (CVPR 2025 · 4 citations)
- Training-Free Guidance Beyond Differentiability: Scalable Path Steering with Tree Search in Diffusion and Flow Models (NeurIPS 2025 · 4 citations)
What has been tried
7 papers here report 79 measured changes (a component added, removed or swapped, compared with a baseline): helped 70%, no clear effect 6%, hurt 24%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| add a distance score map to guide rectified flow | Navsim Test · SPDM | 85.6 -> 88.5 | Helped | GoalFlow: Goal-Driven Flow Matching for Multimodal Trajectories Generation in End-to-End Autonomous Driving (CVPR 2025 · 30 citations) |
| add dynamic radius scheduling | RLBench multi-task · average success rate | 79.1 -> 82.3 | Helped | FlowRAM: Grounding Flow Matching Policy with Region-Aware Mamba Framework for Robotic Manipulation (CVPR 2025 · 4 citations) |
| Increase guidance strength from γ=20 to γ=200 | Class 2 · Prob | 0.313 ± 0.343 -> 0.951 ± 0.110 | Helped | Training-Free Guidance Beyond Differentiability: Scalable Path Steering with Tree Search in Diffusion and Flow Models (NeurIPS 2025 · 4 citations) |
| optimize translations only instead of translations and rotations | Peptide design test set · IMP (%) | 15.0 -> 14.4 | No clear effect | Training Free Guided Flow-Matching with Optimal Control (ICLR 2025 · 1 citation) |
| Use a higher reward temperature β | Aesthetic Score · reward convergence speed | Helped | Value Gradient Guidance for Flow Matching Alignment (NeurIPS 2025 · 0 citations) | |
| Add predictor to the method | MS-COCO 1K · FID | 73.6981 -> 72.9745 | Helped | Rectified CFG++ for Flow Based Models (NeurIPS 2025 · 0 citations) |
What reviewers will expect
Datasets and baselines named in the 14 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- CelebA-HQ · 2 of 14 papers
Recent preprints
Loading recent arXiv preprints…
Use it
Download BibTeX of 20 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Flow matching distillation (95 papers)
- Rectified flow straightening (70 papers)
Elsewhere in the topic
- Training-free diffusion acceleration (257 papers)
- Diffusion theory and design space (252 papers)
- Diffusion models as perception tools (198 papers)
- Masked diffusion language model decoding (147 papers)
- Guidance for conditional diffusion (123 papers)
- Diffusion theory and learned samplers (121 papers)
- Fast diffusion ODE solvers (120 papers)
- Distributional control of diffusion (109 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).