Multimodal

The paper introduces MuPHI, a dataset and MuPHIRM, a reasoning-augmented training framework, to improve Vision-Language Models' ability to detect and reason about subtle, context-dependent multimodal…

The paper introduces Text-Conditioned Layer-wise Internal Alignment (TC-LIA), a model-agnostic method that significantly improves the detection of 'mirage'—when Vision-Language Models confidently answ…

The paper introduces MedPMC, a framework that transforms permissively licensed literature into high-fidelity infrastructure for medical multimodal models, resulting in improved performance on various…

This paper introduces MonoIR-RS, a large-scale infrared remote-sensing vision-language dataset and benchmark for understanding infrared imagery.

This paper introduces CFMME, a comprehensive Chinese financial multimodal benchmark, and evaluates current Large Vision-Language Models (LVLMs), finding that while state-of-the-art models perform mode…

The paper introduces 3DCodeBench, a systematic benchmark and platform for evaluating Vision-Language Model (VLM) agents' ability to generate procedural 3D models from text and images using code.

The paper proposes EAGLE, a novel evidence-aligned multi-agent framework, demonstrating that requiring shared visual evidence among agents is crucial for achieving reliable and trustworthy consensus i…

The paper demonstrates that adversarial examples can be used to manipulate Vision-Language Models (VLMs) into confidently providing authoritative but incorrect information, a process termed 'AI author…

The paper demonstrates that clinical vision-language models (VLMs) pose a significant privacy risk by allowing de-identified images to be re-linked to original reports, and proposes a targeted differe…

The paper proposes an end-to-end forensic pipeline using steganographic attribution and multimodal harm detection to reliably trace and attribute harmful misuse of AI-generated imagery on social platf…

This paper investigates how the structural properties of controlled vocabularies like the Art and Architecture Thesaurus impact the performance of vision-language models like CLIP for content-based im…

The paper introduces CrossMPI, a novel cross-modal prompt injection attack that uses image-only perturbations to steer the interpretation of both textual and visual inputs in Large Vision-Language Mod…

The paper proposes a Variational Adapter (VACSR) to improve cross-modal similarity representation by treating fine-grained image-text matching as a variational inference problem, thereby mitigating th…

This paper details the systematic construction and training of a high-performing Romanian Vision-Language Model (VLM), demonstrating that language-specific adaptation significantly boosts performance…

The paper evaluates the adversarial robustness of two open-source Vision-Language Models (LLaVA and Qwen2.5-VL) in a simulated e-commerce environment, finding that while LLaVA is vulnerable to gradien…

The paper introduces Pocket-Dentist, an efficiency-aware benchmark and model that demonstrates that compact, smaller Vision-Language Models (VLMs) can outperform larger models in accuracy while drasti…

The paper proposes a method to identify 'hub texts' that exploit vulnerabilities in cross-modal encoders, demonstrating that a single text can achieve unrealistically high similarity scores across div…

The paper proposes AsyMoE, a novel Mixture of Experts architecture for Large Vision-Language Models that explicitly models the inherent asymmetry between visual and linguistic modalities, achieving si…

The paper evaluates the performance of Vision-Language Models (VLMs) in a collaborative dialogue task requiring spatial reconstruction, finding that while detailed text representations improve results…

Researchers demonstrate how malicious providers can embed architectural backdoors into Vision--Language Model (VLM) supply chains through representation steering.