# From Sparse Decisions to Dense Reasoning: A Multi-attribute Trajectory Paradigm for Multimodal Moderation

arXiv: https://arxiv.org/abs/2602.02536 · HTML: https://mlsanity.com/paper/2602.02536/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Vision-language and text-to-image safety, one of 12 directions in LLM safety and security.
- Builds on 56 of its 66 references that are in the map.
- First posted on arXiv 28 Jan 2026.

## Placed in

- LLM safety and security / [Vision-language and text-to-image safety](https://mlsanity.com/safety/vision-language-and-text-to-image-safety/)

## Builds on (most cited first, 30 of 56)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [Improved Baselines with Visual Instruction Tuning](https://mlsanity.com/paper/2310.03744/) (2023)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [LLaVA-OneVision: Easy Visual Task Transfer](https://mlsanity.com/paper/2408.03326/) (2024)
- [InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks](https://mlsanity.com/paper/2312.14238/) (2023)
- [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://mlsanity.com/paper/2402.04249/) (2024)
- [Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations](https://mlsanity.com/paper/2312.06674/) (2023)
- [ShareGPT4V: Improving Large Multi-Modal Models with Better Captions](https://mlsanity.com/paper/2311.12793/) (2023)
- [A Survey on Multimodal Large Language Models](https://mlsanity.com/paper/2306.13549/) (2023)
- [Safe RLHF: Safe Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2310.12773/) (2023)
- [MiniGPT-v2: large language model as a unified interface for vision-language multi-task learning](https://mlsanity.com/paper/2310.09478/) (2023)
- [XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Models](https://mlsanity.com/paper/2308.01263/) (2023)
- [WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs](https://mlsanity.com/paper/2406.18495/) (2024)
- [FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts](https://mlsanity.com/paper/2311.05608/) (2023)
- [Deliberative Alignment: Reasoning Enables Safer Language Models](https://mlsanity.com/paper/2412.16339/) (2024)
- [On Evaluating Adversarial Robustness of Large Vision-Language Models](https://mlsanity.com/paper/2305.16934/) (2023)
- [MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Models](https://mlsanity.com/paper/2311.17600/) (2023)
- [ShieldGemma: Generative AI Content Moderation Based on Gemma](https://mlsanity.com/paper/2407.21772/) (2024)
- [ToxicChat: Unveiling Hidden Challenges of Toxicity Detection in Real-World User-AI Conversation](https://mlsanity.com/paper/2310.17389/) (2023)
- [Skywork-Reward-V2: Scaling Preference Data Curation via Human-AI Synergy](https://mlsanity.com/paper/2507.01352/) (2025)
- [Making LLaMA SEE and Draw with SEED Tokenizer](https://mlsanity.com/paper/2310.01218/) (2023)
- [Image Hijacks: Adversarial Images can Control Generative Models at Runtime](https://mlsanity.com/paper/2309.00236/) (2023)
- [Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Models](https://mlsanity.com/paper/2402.02207/) (2024)
- [Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learning](https://mlsanity.com/paper/2504.13914/) (2025)
- [RewardBench 2: Advancing Reward Model Evaluation](https://mlsanity.com/paper/2506.01937/) (2025)

## Abstract (arXiv)

Safety moderation is pivotal for identifying harmful content. Despite the success of textual safety moderation, its multimodal counterparts remain hindered by a dual sparsity of data and supervision. Conventional reliance on binary labels lead to shortcut learning, which obscures the intrinsic classification boundaries necessary for effective multimodal discrimination. Hence, we propose a novel learning paradigm (UniMod) that transitions from sparse decision-making to dense reasoning traces. By constructing structured trajectories encompassing evidence grounding, modality assessment, risk mapping, policy decision, and response generation, we reformulate monolithic decision tasks into a multi-dimensional boundary learning process. This approach forces the model to ground its decision in explicit safety semantics, preventing the model from converging on superficial shortcuts. To facilitate this paradigm, we develop a multi-head scalar reward model (UniRM). UniRM provides multi-dimensional supervision by assigning attribute-level scores to the response generation stage. Furthermore, we introduce specialized optimization strategies to decouple task-specific parameters and rebalance training dynamics, effectively resolving interference between diverse objectives in multi-task learning. Empirical results show UniMod achieves competitive textual moderation performance and sets a new multimodal benchmark using less than 40% of the training data used by leading baselines. Ablations further validate our multi-attribute trajectory reasoning, offering an effective and efficient framework for multimodal moderation. Supplementary materials are available at hrefhttps://trustworthylab.github.io/UniMod/project website.