← 研踪 PaperTrailarXiv:2605.09439 · v2
Inverse Design for Conditional Distribution Matching
首次提交 2026-05-10 17:27
版本更新 2026-11-18 03:08
雷达观测 2026-11-19 19:08
摘要
Generative models are powerful tools for sampling from a learned distribution , and inverse-design methods invert this map to find an input that produces a desired point output . However, many design goals are naturally distributional rather than pointwise, incorporating the inherent uncertainty of and targeting a specific form for it, a task not addressed by standard inverse design. To address this issue we introduce Conditional Distribution Matching (CDM), a new inverse-design problem class in generative modeling: given a joint distribution and a target distribution , find an input whose induced conditional distribution matches . We formally define two variants: Conditional Distribution Matching Sampling (CDMS) and Conditional Distribution Matching Optimization (CDMO). To solve these problems, we propose MLGD-F (Matching-Loss Guided Diffusion with a Fast inner sampler), a plug-and-play inference-time algorithm that combines a pretrained score-based diffusion model with a pretrained fast conditional sampler, requiring no additional training or fine-tuning. By leveraging single-step conditional sampling, MLGD-F enables tractable gradient computation, making the estimation of both memory-efficient and computationally lightweight. We validate MLGD-F on synthetic benchmarks, structured image transformations, and generative editing optimization, demonstrating reliable recovery of inputs whose conditional distributions match diverse user-specified targets, including discrete mixtures and continuous low-rank supports.
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v2 · 2026-11-18 03:08 · 最新观测版本
Inverse Design for Conditional Distribution Matching
Generative models are powerful tools for sampling from a learned distribution , and inverse-design methods invert this map to find an input that produces a desired point output . However, many design goals are naturally distributional rather than pointwise, incorporating the inherent uncertainty of and targeting a specific form for it, a task not addressed by standard inverse design. To address this issue we introduce Conditional Distribution Matching (CDM), a new inverse-design problem class in generative modeling: given a joint distribution and a target distribution , find an input whose induced conditional distribution matches . We formally define two variants: Conditional Distribution Matching Sampling (CDMS) and Conditional Distribution Matching Optimization (CDMO). To solve these problems, we propose MLGD-F (Matching-Loss Guided Diffusion with a Fast inner sampler), a plug-and-play inference-time algorithm that combines a pretrained score-based diffusion model with a pretrained fast conditional sampler, requiring no additional training or fine-tuning. By leveraging single-step conditional sampling, MLGD-F enables tractable gradient computation, making the estimation of both memory-efficient and computationally lightweight. We validate MLGD-F on synthetic benchmarks, structured image transformations, and generative editing optimization, demonstrating reliable recovery of inputs whose conditional distributions match diverse user-specified targets, including discrete mixtures and continuous low-rank supports.
查看此版本解读 →查看此版本原文 ↗