MuPHI: Learning Implicit Multimodal Harm Reasoning via Semantically Grounded Reward Optimization

• URL: http://arxiv.org/abs/2605.29951v1
• Date: Thu, 28 May 2026 13:58:15 GMT
• Title: MuPHI: Learning Implicit Multimodal Harm Reasoning via Semantically Grounded Reward Optimization
• Abstract summary: We introduce Multimodal Pragmatic Harm Interpretation (MuPHI), a dataset containing image-text pairs where harm is encoded in subtle multimodal cues.<n>MuPHIRM is a reasoning-augmented training framework which learns joint semantics by optimizing multi-perspective rewards.<n>Our findings suggest that reasoning-oriented reward optimization offers a promising direction towards building multimodal systems that generalize beyond benchmark-specific shortcuts.
• Score: 14.713115117792503
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Understanding how harm emerges from interaction between otherwise benign image-text pairs requires intent-aware cross-modal reasoning beyond surface-level features. Existing vision-language models (VLMs) excel at literal reasoning over perceptual cues but often fail to derive harmful semantics that rely on implicit, context-dependent reasoning. To evaluate VLMs on compositional harm detection and reasoning, we introduce Multimodal Pragmatic Harm Interpretation (MuPHI), a dataset containing image-text pairs where harm is encoded in subtle multimodal cues. MuPHI spans diverse harm categories and includes annotated harm rationales for assessing VLM reasoning chains. To improve both detection and reasoning in VLMs, we propose MuPHIRM, a reasoning-augmented training framework which learns joint semantics by optimizing multi-perspective rewards. MuPHIRM improves both harm detection and reasoning quality of VLMs while demonstrating superior out-of-distribution robustness compared to both trained and inference-time baselines. Our findings suggest that reasoning-oriented reward optimization offers a promising direction towards building multimodal systems that generalize beyond benchmark-specific shortcuts.

Related papers

• Enhancing Multimodal In-Context Learning via Inductive-Deductive Reasoning [65.15766304205657]
  In-context learning (ICL) allows large models to adapt to tasks using a few examples, yet its extension to vision-language models (VLMs) remains fragile.<n>Our analysis reveals that the fundamental limitation lies in an inductive gap, models often produce correct answers from flawed reasoning.<n>We introduce a framework that restructures multimodal ICL as a principled inductive-deductive process.
  arXiv  Detail & Related papers  (2026-05-04T09:18:19Z)
• Walk the Talk: Bridging the Reasoning-Action Gap for Thinking with Images via Multimodal Agentic Policy Optimization [89.68681087743876]
  Multimodal Large Language Models (MLLMs) have incentivized models to think with images'' by actively invoking visual tools during multi-turn reasoning.<n>The common Reinforcement Learning practice of relying on outcome-based rewards ignores the fact that textual plausibility often masks executive failure.<n>We introduce Multimodal Agentic Policy Optimization (MAPO), bridging the gap between textual reasoning and visual actions generated by models within their Multimodal Chain-of-Thought.
  arXiv  Detail & Related papers  (2026-04-08T07:48:07Z)
• Do MLLMs Really See It: Reinforcing Visual Attention in Multimodal LLMs [55.61018839017648]
  Chain-of-thought (CoT) reasoning has substantially improved multimodal large language models (MLLMs) on complex reasoning tasks.<n>Existing approaches largely rely on long textual reasoning trajectories and provide limited mechanisms for learning stable visual attention policies.<n>We propose SAYO, a visual reasoning model trained with a reinforcement learning framework that introduces a region-level visual attention-based reward.
  arXiv  Detail & Related papers  (2026-02-09T03:33:23Z)
• Analyzing Reasoning Consistency in Large Multimodal Models under Cross-Modal Conflicts [74.47786985522762]
  We identify a critical failure mode termed textual inertia, where models tend to blindly adhere to the erroneous text while neglecting conflicting visual evidence.<n>We propose the LogicGraph Perturbation Protocol that structurally injects perturbations into the reasoning chains of diverse LMMs.<n>Results reveal that models successfully self-correct in less than 10% of cases and predominantly succumb to blind textual error propagation.
  arXiv  Detail & Related papers  (2026-01-07T16:39:34Z)
• Rationale-Enhanced Decoding for Multi-modal Chain-of-Thought [11.538345159297839]
  Chain-of-thought (CoT) prompting has been adapted for large vision-language models (LLMs) to enhance multi-modal reasoning.<n>Existing LVLMs often ignore the contents of generated rationales in CoT reasoning.<n>We propose rationale-enhanced decoding (RED), a novel plug-and-play inference-time decoding strategy.
  arXiv  Detail & Related papers  (2025-07-10T12:07:13Z)
• Multimodal Mathematical Reasoning with Diverse Solving Perspective [65.07953438724105]
  We introduce MathV-DP, a novel dataset that captures multiple diverse solution trajectories for each image-question pair.<n>We propose Qwen-VL-DP, a model built upon Qwen-VL, fine-tuned with supervised learning and enhanced via group relative policy optimization.<n>Our method emphasizes learning from varied reasoning perspectives and distinguishing between correct yet distinct solutions.
  arXiv  Detail & Related papers  (2025-07-03T17:07:20Z)
• CF-VLM:CounterFactual Vision-Language Fine-tuning [10.299136720220416]
  CounterFactual Vision-Language Fine-tuning (CF-VLM) is a novel framework that enhances causal reasoning capabilities of vision-language models.<n>CF-VLM introduces three complementary training objectives: maintaining foundational cross-modal alignment, reinforcing the uniqueness and stability of factual scene representations, and sharpening the model's sensitivity to minimal but critical causal edits.
  arXiv  Detail & Related papers  (2025-06-10T20:20:05Z)
• Perception-R1: Advancing Multimodal Reasoning Capabilities of MLLMs via Visual Perception Reward [77.34936657745578]
  We propose Perception-R1, which introduces a novel visual perception reward that explicitly encourages MLLMs to perceive the visual content accurately.<n>We show that Perception-R1 achieves state-of-the-art performance on most benchmarks using only 1,442 training data.
  arXiv  Detail & Related papers  (2025-06-08T16:48:42Z)
• Reasoning-Aligned Perception Decoupling for Scalable Multi-modal Reasoning [95.44766931218896]
  Multi-modal large language models (MLLMs) still lag behind text-based reasoning.<n>We introduce Perception-Reasoning Decoupling, which modularizes the MLLM's reasoning component and makes it easily replaceable.<n>We propose a novel reinforcement learning algorithm called Visual Perception Optimization (VPO) to align the MLLM's perceptual output with the final reasoning task.
  arXiv  Detail & Related papers  (2025-06-05T02:28:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.