ThinkingGuard: Decoding Implicit Hazards via Step-by-Step Risk Attribution in Multimodal Large Language Models
cs.CV, cs.AI, cs.LG
Submitted: 2026-09-29
Updated: 2026-09-29
Code: https://github.com/FroggyChen/ThinkingGuard
Terminology
Sources
- Qwen3-VL Technical Report
- Llama Guard 3 Vision: Safeguarding Human-AI Image Understanding Conversations
- Agent Smith: A Single Image Can Jailbreak One Million Multimodal LLM Agents Exponentially Fast
- LlavaGuard: An Open VLM-based Framework for Safeguarding Vision Datasets and Models
- LoRA: Low-Rank Adaptation of Large Language Models
- Open-FinLLMs: Open Multimodal Large Language Models for Financial Applications
- Safety of Multimodal Large Language Models on Images and Texts
- GuardReasoner-VL: Safeguarding VLMs via Reinforced Reasoning
- JailDAM: Jailbreak Detection with Adaptive Memory for Vision-Language Model
- VLSU: Mapping the Limits of Joint Multimodal Understanding for AI Safety
- Direct Preference Optimization: Your Language Model is Secretly a Reward Model
- OpenAI GPT-5 System Card
- SaFeR-VLM: Toward Safety-aware Fine-grained Reasoning in Multimodal Models
- CrossGuard: Safeguarding MLLMs against Joint-Modal Implicit Malicious Attacks
- USB: A Comprehensive and Unified Safety Evaluation Benchmark for Multimodal Large Language Models
- Multimodal Situational Safety
- Safety Fine-Tuning at (Almost) No Cost: A Baseline for Vision Large Language Models
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models