# Researchers introduce ConicFlow for constrained flow matching model sampling

Digest AI · Research · published 2026-10-14T04:00:00Z

Canonical: https://digestai.news/story/researchers-introduce-conicflow-for-constrained-flow-matching-model-sa

## Summary

It reframes constrained sampling from flow matching models as an **optimal control problem** at inference time. The approach balances deviations from the pretrained trajectory against a reward function while enforcing hard terminal constraints.

ConicFlow uses **receding-horizon control**, optimizing only short trajectory windows for efficiency. A **primal-dual algorithm** refines dual multipliers to guide generation toward feasible final samples. A **terminal-estimate variant** reduces local truncation errors by evaluating rewards and constraints at the predicted endpoint. Under standard assumptions, the method guarantees finite-horizon suboptimality, with sharper bounds for convex-affine structures. Experiments in low- and high-dimensional settings show it outperforms recent baselines in reward, constraint satisfaction, and adherence to pretrained flows.

## Key points

- ConicFlow reframes constrained sampling as an optimal control problem with hard terminal constraints
- Uses receding-horizon control and primal-dual algorithms for computational efficiency and feasibility
- Terminal-estimate variant reduces truncation errors and improves theoretical guarantees in experiments

## Why it matters

ConicFlow could improve constrained generation in flow matching models, useful for applications requiring precise control over outputs like drug discovery or robotics. Its theoretical guarantees and empirical performance may set new benchmarks for constrained sampling methods.

## Sources

1. [ConicFlow: constrained Inference-time optimal control for flow matching models](https://arxiv.org/abs/2610.14307) (arXiv cs.AI, 2026-10-14, primary source)

## Cite

Digest AI, "Researchers introduce ConicFlow for constrained flow matching model sampling", 14 October 2026, https://digestai.news/story/researchers-introduce-conicflow-for-constrained-flow-matching-model-sa

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/researchers-introduce-conicflow-for-constrained-flow-matching-model-sa.json