Yee Whye Teh, Julius Berner, Morteza Mardani, So Takao, Ricardo Baptista, Abbas Mammadov, Bohan Chen
We lifted 7 functions out of this paper's own repositories and ran 6 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.
| Repository | Role | Ran |
|---|---|---|
| abbasmammadov/VFM | — | 6 of 7 |
| Function | Status | Where it lives |
|---|---|---|
| adaptive | Ran | abbasmammadov/VFM/loss.py code served (permissive licence) · get_code("5d2f29a549efeacb") |
| compute_guidance | Ran | abbasmammadov/VFM/loss.py code served (permissive licence) · get_code("f373f40c9eee6bd2") |
| interpolant | Ran | abbasmammadov/VFM/loss.py code served (permissive licence) · get_code("ce1652c9ff63a153") |
| ln_sampler | Ran | abbasmammadov/VFM/loss.py code served (permissive licence) · get_code("76ad6b50bfae7ab3") |
| mean_flat | Ran | abbasmammadov/VFM/loss.py code served (permissive licence) · get_code("113e3a84f65fdd5c") |
| mse_loss | Ran | abbasmammadov/VFM/loss.py code served (permissive licence) · get_code("b766939802ffeabc") |
| AdapterLoss | Not yet run | abbasmammadov/VFM/loss.py code served (permissive licence) · get_code("18b1a579c2d4c244") |
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Flow maps enable high-quality image generation in a single forward pass. However, unlike iterative diffusion models, their lack of an explicit sampling trajectory impedes incorporating external constraints for conditional generation and solving inverse problems. We put forth Variational Flow Maps, a framework for conditional sampling that shifts the perspective of conditioning from "guiding a sampling path", to that of "learning the proper initial noise". Specifically, given an observation, we seek to learn a noise adapter model that outputs a noise distribution, so that after mapping to the data space via flow map, the samples respect the observation and data prior. To this end, we develop a principled variational objective that jointly trains the noise adapter and the flow map, improving noise-data alignment, such that sampling from complex data posterior is achieved with a simple adapter. Experiments on various inverse problems show that VFMs produce wellcalibrated conditional samples in a single (or few) steps. For ImageNet, VFM attains competitive fidelity while accelerating the sampling by orders of magnitude compared to alternative iterative diffusion/flow models. Code is available at https://github.com/abbasmammadov/VFM .
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2603.07276")
get_code_for_paper("2603.07276")
have("2603.07276")
Connect an agent — have() is free.