Skip to main content
archive
Search Submit Donate Log in
Press Enter to search · Advanced search

Computer Science > Machine Learning

arXiv:2603.19880 (cs)
[Submitted on 20 Mar 2026 (v1), last revised 19 Apr 2026 (this version, v2)]

Title:What If Consensus Lies? Selective-Complementary Reinforcement Learning at Test Time

Authors:Dong Yan, Jian Liang, Yanbo Wang, Shuo Lu, Ran He, Tieniu Tan
View a PDF of the paper titled What If Consensus Lies? Selective-Complementary Reinforcement Learning at Test Time, by Dong Yan and 5 other authors
View PDF HTML (experimental)
Abstract:Test-Time Reinforcement Learning (TTRL) enables Large Language Models (LLMs) to enhance reasoning capabilities on unlabeled test streams by deriving pseudo-rewards from majority voting consensus. However, existing TTRL methods rely exclusively on positive pseudo-labeling strategies. Such reliance becomes vulnerable under challenging scenarios where answer distributions are highly dispersed, resulting in weak consensus that inadvertently reinforces incorrect trajectories as supervision signals. In this paper, we propose SCRL (Selective-Complementary Reinforcement Learning), a robust test-time reinforcement learning framework that effectively mitigates label noise amplification. SCRL develops Selective Positive Pseudo-Labeling, which enforces strict consensus criteria to filter unreliable majorities. Complementarily, SCRL introduces Entropy-Gated Negative Pseudo-Labeling, the first negative supervision mechanism in TTRL, to reliably prune incorrect trajectories based on generation uncertainty. Extensive experiments on multiple reasoning benchmarks demonstrate that SCRL achieves substantial improvements over baselines, while maintaining robust generalization and training stability under constrained rollout budgets. Our code is available at this https URL.
Comments: Accepted at ACL 2026 Main Conference
Subjects: Machine Learning (cs.LG); Artificial Intelligence (cs.AI)
Cite as: arXiv:2603.19880 [cs.LG]
  (or arXiv:2603.19880v2 [cs.LG] for this version)
  https://doi.org/10.48550/arXiv.2603.19880
arXiv-issued DOI via DataCite

Submission history

From: Dong Yan [view email]
[v1] Fri, 20 Mar 2026 11:47:12 UTC (337 KB)
[v2] Sun, 19 Apr 2026 02:15:51 UTC (339 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled What If Consensus Lies? Selective-Complementary Reinforcement Learning at Test Time, by Dong Yan and 5 other authors
  • View PDF
  • HTML (experimental)
  • TeX Source
view license

Current browse context:

cs.LG
< prev   |   next >
new | recent | 2026-03
Change to browse by:
cs
cs.AI

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar
Loading...

BibTeX formatted citation

Data provided by:

Bookmark

BibSonomy Reddit

Bibliographic and Citation Tools

Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)

Code, Data and Media Associated with this Article

alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
ScienceCast (What is ScienceCast?)

Demos

Replicate (What is Replicate?)
Hugging Face Spaces (What is Spaces?)
TXYZ.AI (What is TXYZ.AI?)

Recommenders and Search Tools

Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
IArxiv Recommender (What is IArxiv?)
  • Author
  • Venue
  • Institution
  • Topic

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?)
We gratefully acknowledge support from our major funders, member institutions, , and all contributors.
About · Help · Contact · Subscribe · Copyright · Privacy · Accessibility · Operational Status (opens in new tab)
Major funding support from
Simons Foundation Simons Foundation International Schmidt Sciences