# GuardTrace-VL: Detecting Unsafe Multimodel Reasoning via Iterative Safety Supervision

arXiv: https://arxiv.org/abs/2511.20994 · HTML: https://mlsanity.com/paper/2511.20994/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 34 of its 43 references that are in the map.
- First posted on arXiv 26 Nov 2025.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)

## Builds on (most cited first, 30 of 34)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks](https://mlsanity.com/paper/2312.14238/) (2023)
- [LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://mlsanity.com/paper/2403.13372/) (2024)
- [Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations](https://mlsanity.com/paper/2312.06674/) (2023)
- [BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Dataset](https://mlsanity.com/paper/2307.04657/) (2023)
- [WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs](https://mlsanity.com/paper/2406.18495/) (2024)
- [GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning](https://mlsanity.com/paper/2507.01006/) (2025)
- [FigStep: Jailbreaking Large Vision-Language Models via Typographic Visual Prompts](https://mlsanity.com/paper/2311.05608/) (2023)
- [A Holistic Approach to Undesired Content Detection in the Real World](https://mlsanity.com/paper/2208.03274/) (2022)
- [Visual Adversarial Examples Jailbreak Aligned Large Language Models](https://mlsanity.com/paper/2306.13213/) (2023)
- [MM-SafetyBench: A Benchmark for Safety Evaluation of Multimodal Large Language Models](https://mlsanity.com/paper/2311.17600/) (2023)
- [Jailbreak in pieces: Compositional Adversarial Attacks on Multi-Modal Language Models](https://mlsanity.com/paper/2307.14539/) (2023)
- [A New Generation of Perspective API: Efficient Multilingual Character-level Transformers](https://mlsanity.com/paper/2202.11176/) (2022)
- [Qwen3Guard Technical Report](https://mlsanity.com/paper/2510.14276/) (2025)
- [Images are Achilles' Heel of Alignment: Exploiting Visual Vulnerabilities for Jailbreaking Multimodal Large Language Models](https://mlsanity.com/paper/2403.09792/) (2024)
- [Llama Guard 3 Vision: Safeguarding Human-AI Image Understanding Conversations](https://mlsanity.com/paper/2411.10414/) (2024)
- [Safety Tax: Safety Alignment Makes Your Large Reasoning Models Less Reasonable](https://mlsanity.com/paper/2503.00555/) (2025)
- [H-CoT: Hijacking the Chain-of-Thought Safety Reasoning Mechanism to Jailbreak Large Reasoning Models, Including OpenAI o1/o3, DeepSeek-R1, and Gemini 2.0 Flash Thinking](https://mlsanity.com/paper/2502.12893/) (2025)
- [AIR-Bench 2024: A Safety Benchmark Based on Risk Categories from Regulations and Policies](https://mlsanity.com/paper/2407.17436/) (2024)
- [SafeBench: A Safety Evaluation Framework for Multimodal Large Language Models](https://mlsanity.com/paper/2410.18927/) (2024)
- [GuardReasoner-VL: Safeguarding VLMs via Reinforced Reasoning](https://mlsanity.com/paper/2505.11049/) (2025)
- [Distraction is All You Need for Multimodal Large Language Model Jailbreaking](https://mlsanity.com/paper/2502.10794/) (2025)
- [SafeMLRM: Demystifying Safety in Multi-modal Large Reasoning Models](https://mlsanity.com/paper/2504.08813/) (2025)
- [Safety Alignment for Vision Language Models](https://mlsanity.com/paper/2405.13581/) (2024)

## Built on by (most cited first, 4 of 4)

- [BraveGuard: From Open-World Threats to Safer Computer-Use Agents](https://mlsanity.com/paper/2606.01166/) (2026)
- [Multi2AV-Safety: Benchmarking Safety in Multimodal-to-Audio-Video Generation](https://mlsanity.com/paper/2608.26535/) (2026)
- [Taming CoT Obfuscation in VLMs: From Mechanistic Evidence to Activation Enforcement](https://mlsanity.com/paper/2609.24243/) (2026)
- [PlanGuard: A Guardrail for Multi-Step Plan Safety in Embodied Agents](https://mlsanity.com/paper/2609.32801/) (2026)

## Abstract (arXiv)

Multimodal large reasoning models (MLRMs) are increasingly deployed for vision-language tasks that produce explicit intermediate rationales. However, reasoning traces can contain unsafe content even when the final answer is non-harmful, creating deployment risks. Existing multimodal safety guards primarily evaluate only the input question and the final answer, neglecting the intermediate reasoning process. This oversight allows undetected harm, such as biased inferences or policy-violating use of visual context, to emerge during reasoning. We introduce GuardTrace-VL, a vision-aware safety auditor that monitors the full Question-Thinking-Answer (QTA) pipeline via joint image-text analysis, enabling detection of unsafe content as it emerges in the reasoning stage. To support training and evaluation, we construct the GuardTrace dataset, which is generated through diverse prompting strategies and refined via a MLRM- and human-based voting and verification pipeline. Furthermore, we propose a three-stage progressive training scheme combined with the data refinement process, enabling the model to learn nuanced and context-dependent safety preferences according to different risk levels. On our proposed test set covering both in-domain and out-of-domain scenarios, GuardTrace-VL model achieves an F1 score of 93.1% on unsafe reasoning detection tasks, representing a 13.5% improvement in F1 score compared to the previous strongest multimodal safety defense methods. The codes will be made publicly available.