Jindong Gu

Do 3D Large Language Models Really Understand 3D Spatial Relationships?

A Guardrail for Safety Preservation: When Safety-Sensitive Subspace Meets Harmful-Resistant Null-Space

TRACEDET: HALLUCINATION DETECTION FROM THE DECODING TRACE OF DIFFUSION LARGE LANGUAGE MODELS

Safe Autoregressive Image Generation with Iterative Self-Improving Codebooks

DD-Ranking: Rethinking the Evaluation of Dataset Distillation

PostAlign: Multimodal Grounding as a Corrective Lens for MLLMs

GLARE: Towards Graph-less Retrieval for Retrieval Augmented Generation on Million-scale Knowledge Graphs

Bag of Tricks for Subverting Reasoning-based Safety Guardrails

MULTI-FACETED MULTIMODAL MONOSEMANTICITY

Can Knowledge-Graph-based Retrieval Augmented Generation Really Retrieve What You Need?

True Multimodal In-Context Learning Needs Attention to the Visual Context

Image Token Matters: Mitigating Hallucination in Discrete Tokenizer-based Large Vision-Language Models via Latent Editing

Improved Techniques for Optimization-Based Jailbreaking on Large Language Models

Primitive Vision: Improving Diagram Understanding in MLLMs

OT-Attack: Enhancing Adversarial Transferability of Vision-Language Models via Optimal Transport Optimization

Open Eyes, Then Reason: Fine-grained Visual Mathematical Understanding in MLLMs

VIDEOGUARD: PROTECTING VIDEO CONTENT FROM UNAUTHORIZED EDITING

Can Editing LLMs Inject Harm?

LLM Jailbreak Detection for (Almost) Free!

FedBiP: Heterogeneous One-Shot Federated Learning with Personalized Latent Diffusion Models

Typography Leads Semantic Diversifying: Amplifying Adversarial Transferability across Multimodal Large Language Models

Improving Adversarial Transferability in MLLMs via Dynamic Vision-Language Alignment Attack

Learning Visual Prompts for Guiding the Attention of Vision Transformers

Cracking the Collective Mind: Adversarial Manipulation in Multi-Agent Systems