Bin Zhu

LLMBind: A Unified Modality-Task Integration Framework

2024 · arXiv (Cornell University)

Despite recent progress in Multi-Modal Large Language Models (MLLMs), it remains challenging to integrate diverse tasks ranging from pixel-level perception to high-fidelity generation. Existing approaches often suffer from either restricted task extensibility or severe performance …

A Switch-Constrained Multiple Model Smoothing Algorithm for Maneuvering Target Tracking

2024

In most existing multiple model (MM) smoothing algorithms for maneuvering target tracking, the mode evolution process is usually described by a Markov chain, which implicitly assumes that the target mode may switch at all time …

Benchmarking Gaslighting Negation Attacks Against Multimodal Large Language Models

2025 · arXiv (Cornell University)

Multimodal Large Language Models (MLLMs) have exhibited remarkable advancements in integrating different modalities, excelling in complex understanding and generation tasks. Despite their success, MLLMs remain vulnerable to conversational adversarial inputs. In this paper, we systematically …

Benchmarking Gaslighting Negation Attacks Against Multimodal Large Language Models

2026

Multimodal Large Language Models (MLLMs) have exhibited remarkable advancements in integrating different modalities, excelling in complex understanding and generation tasks. Despite their success, MLLMs remain vulnerable to conversational adversarial inputs. In this paper, we systematically …