Work

IREASONER: Trajectory-Aware Intrinsic Reasoning Supervision for Self-Evolving Large Multimodal Models

Year: 2026

Type: conference-paper

Abstract: Recent work shows that large multimodal models (LMMs) can self-improve from unlabeled data via self-play and intrinsic feedback.Yet existing self-evolving frameworks mainly reward final outcomes, leaving intermediate reasoning weakly constrained despite its importance for visually grounded decision making.We propose IREASONER, a self-evolving framework that improves an LMM's implicit reasoning by explicitly eliciting chain-of-thought (CoT) and rewarding its internal agreement.In a Proposer-Solver loop over unlabeled images, IREASONER augments outcome-level intrinsic rewards with a trajectory-aware signal defined over intermediate reasoning steps, providing learning signals that distinguish reasoning paths leading to the same answer without ground-truth labels or external judges.Starting from Qwen2.5-VL-7B,IREASONER yields up to +2.1 points across diverse multimodal reasoning benchmarks under fully unsupervised post-training.We hope this work serves as a starting point for reasoning-aware selfimprovement in LMMs in purely unsupervised settings.Our code is available here.

Source: Findings of the Association for Computational Linguistics: ACL (Findings of ACL)

DOI: 10.18653/v1/2026.findings-acl.1468

Authors: Meghana Sunil, Manikandarajan Venmathimaran, Muthu Subash Kavitha

Institutions: Nagasaki University, Vellore Institute of Technology University, Loughborough University

Language: English

Topic: Multimodal Machine Learning Applications

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Funder: Japan Society for the Promotion of Science

Award: A data-saving and self-supervised deep learning system for continuous ischemic stroke assessment