Skip to content

Tilted Schr\"odinger Bridge Matching

Sep 2026 · 1 citation · 49 references
Computer Science

TL;DR

Tilted Schrodinger Bridge Matching (TSBM) is introduced, a post-training method for fine-tuning a learned bridge between source and target toward a reward-tilted target while preserving source, and theoretical justification and practical algorithm are derived.

Abstract

Schr\"odinger bridges provide an entropy-regularized framework and a principled solution for unpaired domain translation. In practice, a pretrained bridge may need to be adapted to human preferences or physical constraints through a reward a problem closely related to reward tilting in diffusion models but underexplored for Schr\"odinger bridges. We introduce Tilted Schr\"odinger Bridge Matching (TSBM), a post-training method for fine-tuning a learned bridge $P$ between source $p_0$ and target $p_1$ toward a reward-tilted target $p_1^r\propto p_1e^r$, while preserving source $p_0$. We formulate this adaptation as alternating optimization initialized from $P$, provide theoretical justification, and derive a practical algorithm based on Adjoint Matching. We evaluate TSBM on unpaired image-to-image translation targeting digit properties in MNIST and facial attributes in CelebA.

View source

Similar papers

#machine learning Preprint Oct 2026

BridgeTwist: Twisting Schr\"odinger Bridges for Training-Free Conditional Sampling

Generative models are often required to serve many downstream conditional tasks. In the training-free setting, a pretrained unconditional model is reused as a prior and combined with a task-specific likelihood to sample from the corresponding posterior. While such training-free samplers have been developed for diffusio...

Meng-Yu Li, Q. Qu, Jun S. Liu · 0 citations
#machine learning Preprint Oct 2026

Cost-augmented Schr\"odinger bridges on graphs are exactly solvable: a Feynman-Kac tilt replaces learned control

The generalized Schr\"odinger bridge on a graph moves mass between two distributions while charging a cost for the states visited. It has been approached by learning the rates of a controlled continuous-time Markov chain, with a temporal-difference penalty that restores the cost. A state cost folds into the reference p...

Akshay Balsubramani · 0 citations
Preprint Sep 2026

Quantum-Enhanced Sampling of Schr\"odinger Bridges

It is shown that exponential reweighting extends both the Schr\"odinger bridge formulation and the Gibbs spectral-gap estimate to models with additive running costs.

Tom Lollier, Eyal Neuman · 1 citation
Preprint Oct 2026

Gradient Discrete Schrodinger Bridge Matching for Peptide Generation

The optimal control of a discrete Schr\"odinger bridge is a discrete gradient, a difference of vertex potentials, but recent models such as MadSBM instead learn a free control on the edges by cross-entropy. We prove that this leaves the action uncontrolled: cross-entropy identifies the control only up to a state-depend...

A. Benabdallah, Mohsen Dlala · 0 citations

The Risk-Sensitive Schrödinger Bridge: Is Not a KL Projection

The Schr\"odinger bridge owes its computational power to a single structural fact: by Girsanov's theorem the controlled problem is a Kullback--Leibler (KL) projection onto a fixed reference measure, solvable by alternating projections. This letter shows that the fact does not survive risk sensitivity. When the expected...

Hamidreza Behjoo · 0 citations

Related blog posts

MIT News · Artificial Intelligence Nov 17, 2026

Learning the language of machine learning

By studying the computational foundations of language learning, Associate Professor Jacob Andreas builds intelligent systems that can communicate effectively with humans.

Microsoft Research Blog Nov 16, 2026

TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient

AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on individual tasks. Yet developers return to the same […] The post TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient appeared first on Microsoft Research.

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.