Skip to content
AI Atlas
PaperActive

FlowCPO: A Unified Divergence View of Preference Alignment for Flow Models

arxiv.org/abs/2609.09905

Updated 51 min ago · first seen 11 Sept 2026

paper_01M294GN0HGPMV24XZD3XEZA51

Published
11 Sept 2026
T1 · 51 min ago
arXiv
2609.09905
T1 · 51 min ago
Category
stat.ML
T1 · 51 min ago

Abstract

Preference alignment for flow and diffusion models now spans online reinforcement learning and offline preference optimization, but the relation between these methods remains unclear. In particular, existing forward-process alignment methods require fresh samples from the current model, while offline methods based on fixed preference pairs rely primarily on positive-only fine-tuning or DPO-style likelihood-ratio surrogates. We organize these approaches through a divergence-based framework and introduce FlowCPO, an offline forward-KL objective that uses both preferred and dispreferred samples without online rollouts. For linear interpolation, we show under explicit regularity conditions that the forward-KL objective is bounded by a contrastive flow matching loss, yielding a tractable surrogate on fixed data. We further show that this loss is nonnegative, whereas the signed regression loss of simplified FlowDPO can be unbounded below. In the in-domain setting, FlowCPO achieves higher mean GenEval and OCR scores than the evaluated baselines, reaching 0.84 and 0.87 versus 0.81 and 0.74 for FlowDPO at CFG 3.0. In the out-of-domain setting, the results are mixed, with the best GenEval result but lower reward scores than RFT on several metrics.

Authors 7

Yansen Han, Shengyi Liao, Peng Sun, Deyuan Liu, Yuanxing Zhang, Pengfei Wan, Tao Lin

Specification

Official page

Source:arXiv (Atom API + RSS)T1observed 51 min agohigh

Arxiv announce type
cross

Source:arXiv (Atom API + RSS)T1observed 51 min agohigh

arXiv id
2609.09905

Source:arXiv (Atom API + RSS)T1observed 51 min agohigh

Categories
stat.ML, cs.AI, cs.CV, cs.LG

Source:arXiv (Atom API + RSS)T1observed 51 min agohigh

PDF

Source:arXiv (Atom API + RSS)T1observed 51 min agohigh

Primary category
stat.ML

Source:arXiv (Atom API + RSS)T1observed 51 min agohigh

Published
11 Sept 2026

Source:arXiv (Atom API + RSS)T1observed 51 min agohigh

Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →

Provenance

Attributed facts

9

Source tiers

T19

Freshest observation

51 min ago

Conflicts

None