Faithful-First Reasoning, Planning, and Acting for Multimodal LLMs

arXiv:2511.08409v4 Announce Type: replace Abstract: Multimodal Large Language Models (MLLMs) frequently suffer from unfaithfulness, generating reasoning chains that drift from visual evidence or contradict final predictions. We propose Faithful-First Reasoning, Planning, and Acting (RPA) framework in which FaithEvi provides step-wise and chain-level supervision by evaluating the faithfulness of intermediate reasoning, and FaithAct uses these signals to plan and execute faithfulness-aware actions during inference. Experiments across multiple multimodal reasoning benchmarks show that faithful-first RPA improves perceptual faithfulness by up to 24% over prompt-based and tool-augmented reasoning frameworks, without degrading task accuracy. Our analysis shows that treating faithfulness as a guiding principle perceptually faithful reasoning trajectories and mitigates hallucination behavior. This work thereby establishes a unified framework for both evaluating and enforcing faithfulness in multimodal reasoning. Code is at https://github.com/lijunxian111/Faithful-First-RPA.

Related

• Stabilizing Unsupervised Self-Evolution of MLLMs via Continuous Softened Retracing reSampling
• Countering the Over-Reliance Trap: Mitigating Object Hallucination for LVLMs via a Self-Validation Framework
• Reasoning Fails Where Step Flow Breaks
• Hallucination as output-boundary misclassification: a composite abstention architecture for language models

Source: arXiv cs.AI | 2026-04-10