Skip to content
AI News HubLIVE
Original source2 min read

PRO-Step: Step-level Process Reward Optimization for Retrieval-Augmented Generation

Summary

PRO-Step addresses error propagation in multi-hop RAG by training a generative process reward model that judges each step on logical validity and evidential grounding, building preference pairs via PRM-guided value tree search, and optimizing the policy with step-level DPO. It achieves the best average EM and F1 across five single- and multi-hop QA benchmarks, with code, models, and training data released publicly.

SourcearXiv Computational LinguisticsAuthor: MinKeon Kim, Namjun Lee, Jaekwang Kim
PRO-Step: Step-level Process Reward Optimization for Retrieval-Augmented Generation
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 31 Aug 2026]

Title:PRO-Step: Step-level Process Reward Optimization for Retrieval-Augmented Generation

View a PDF of the paper titled PRO-Step: Step-level Process Reward Optimization for Retrieval-Augmented Generation, by MinKeon Kim and 2 other authors

View PDF HTML (experimental)

Abstract:Retrieval-Augmented Generation enhances Large Language Models by grounding responses in external knowledge, but multi-hop reasoning remains vulnerable to error propagation, where early retrieval failures confound subsequent steps. Standard outcome-based optimization only rewards the final answer, leaving intermediate retrieval and reasoning errors undetected. While existing process-based methods introduce step-level signals, they still score each step against the final answer, rewarding spurious successes where flawed retrieval coincidentally produces the correct answer. Step-level supervision in RAG requires evaluating both logical validity and evidential grounding at each step. We introduce PRO-STEP: we train a generative PRM that evaluates both dimensions, employ PRM-guided value tree search to construct preference pairs contrasting valid steps against flawed ones, and optimize the policy via step-level Direct Preference Optimization. Experiments on single and multi-hop QA datasets demonstrate that PRO-STEP achieves the best average EM and F1 across five benchmarks. Code, models, and training data are publicly available at this https URL.

Comments: 22 pages, 7 figures, 23 tables. Accepted to EMNLP 2026

Subjects:

Computation and Language (cs.CL); Artificial Intelligence (cs.AI)

Cite as: arXiv:2609.01658 [cs.CL]

(or arXiv:2609.01658v1 [cs.CL] for this version)

https://doi.org/10.48550/arXiv.2609.01658

arXiv-issued DOI via DataCite

Submission history

From: MinKeon Kim [view email] [v1] Mon, 31 Aug 2026 14:34:10 UTC (1,133 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled PRO-Step: Step-level Process Reward Optimization for Retrieval-Augmented Generation, by MinKeon Kim and 2 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CL

new | recent | 2026-09

Change to browse by:

cs cs.AI

References & Citations

NASA ADS

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

EngineersAdvanced

Key points

  • Existing process-based methods still score each step against the final answer, rewarding spurious successes from flawed retrieval
  • A generative PRM evaluates both logical validity and evidential grounding at every step
  • PRM-guided value tree search builds preference pairs contrasting valid and flawed steps for step-level DPO
  • Best average EM and F1 across five single- and multi-hop QA benchmarks, with code, models, and data public

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