Skip to content
AI News HubLIVE
Original source2 min read

PrefO2O: Zero-Shot Preference-Based Reinforcement Learning via Offline-to-Online Latent Preference Transfer

Summary

PrefO2O introduces an offline-to-online latent preference transfer framework that enables zero-shot preference-based reinforcement learning on unseen target tasks without any human feedback. It learns a task-invariant latent preference space from limited offline preference data and uses confidence-weighted optimal transport alignment to generate pseudo-preference supervision, outperforming preference-transfer baselines and matching or exceeding PbRL trained with real human feedback on robotic manipulation and locomotion benchmarks.

SourcearXiv RoboticsAuthor: Ruiqi Wang, Dezhong Zhao, Xuesu Xiao, Erdem Biyik, Byung-Cheol Min, Guohua Chen
PrefO2O: Zero-Shot Preference-Based Reinforcement Learning via Offline-to-Online Latent Preference Transfer
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[Submitted on 22 Oct 2026]

Title:PrefO2O: Zero-Shot Preference-Based Reinforcement Learning via Offline-to-Online Latent Preference Transfer

View a PDF of the paper titled PrefO2O: Zero-Shot Preference-Based Reinforcement Learning via Offline-to-Online Latent Preference Transfer, by Ruiqi Wang and 5 other authors

View PDF HTML (experimental)

Abstract:Preference-based reinforcement learning (PbRL) has emerged as a promising paradigm for learning robot behaviors aligned with human intent. However, its scalability is fundamentally limited by the need to collect expensive human preference labels for every new task. In this work, we present PrefO2O, an offline-to-online latent preference transfer framework that enables zero-shot PbRL on unseen target tasks without requiring any human feedback. The core idea is to learn a task-invariant latent preference space from a small amount of offline preference data on source tasks, where transferable preference structures are disentangled from task-specific variations. This learned manifold allows PrefO2O to transfer preference structures and reward priors to new tasks under distribution shift, while confidence-weighted optimal transport alignment generates reliable pseudo-preference supervision for online reward adaptation. Extensive experiments on robotic manipulation and locomotion benchmarks demonstrate that PrefO2O substantially outperforms existing preference-transfer baselines and matches or exceeds PbRL methods trained with real human feedback, while generalizing across multi-to-one and one-to-one transfer settings as well as across different robot embodiments.

Subjects:

Robotics (cs.RO); Artificial Intelligence (cs.AI)

Cite as: arXiv:2610.31812 [cs.RO]

(or arXiv:2610.31812v1 [cs.RO] for this version)

https://doi.org/10.48550/arXiv.2610.31812

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Ruiqi Wang [view email] [v1] Thu, 22 Oct 2026 18:49:10 UTC (9,983 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled PrefO2O: Zero-Shot Preference-Based Reinforcement Learning via Offline-to-Online Latent Preference Transfer, by Ruiqi Wang and 5 other authors

View PDF

HTML (experimental)

TeX Source

view license

Additional Features

Audio Summary

Current browse context:

cs.RO

new | recent | 2026-10

Change to browse by:

cs cs.AI

References & Citations

SciX

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

ResearchersAdvanced

Key points

  • PrefO2O transfers preference structures and reward priors from source tasks to target tasks for zero-shot PbRL.
  • It disentangles transferable preference structure from task-specific variation using limited offline preference data.
  • Confidence-weighted optimal transport alignment produces reliable pseudo-preference supervision for online reward adaptation.
  • Experiments show gains over preference-transfer baselines and parity or better versus PbRL with real human feedback, across multi-to-one, one-to-one, and cross-embodiment settings.

Highlights and analysis are generated automatically and may contain errors. Check the original source.