PathoSage is a three-stage framework addressing evidence adjudication in pathology, explicitly separating knowledge retrieval, evidence collection, and evidence adjudication. It employs an agentic system with Structured Evidence Deliberation to independently evaluate heterogeneous evidence and reduce anchoring bias.
WorldBench, a new multimodal benchmark from MIT researchers, evaluates 15 MLLMs on visually diverse images, revealing fundamental gaps in visual understanding with the top model scoring only 64.0% accuracy. The benchmark prioritizes visual diversity over various task types to expose these shortcomings.
This paper proposes the LAMO framework to address the challenge of deploying lightweight MLLM-powered autonomous GUI agents on resource-constrained devices. LAMO enhances lightweight MLLMs with GUI-specific knowledge and task scalability through multi-role orchestration.
AMBER introduces a new LLM-free, multi-dimensional benchmark designed to rigorously evaluate hallucination in Multimodal Large Language Models (MLLMs). This research aims to provide a comprehensive tool for assessing the reliability and accuracy of MLLM outputs.
This paper identifies and formalizes
SaFeR-Steer is a novel framework designed to improve the safety alignment of Multi-modal Large Language Models (MLLMs) in multi-turn dialogues, addressing challenges like escalating unsafe intent and long-context safety decay. It employs synthetic bootstrapping and feedback dynamics, while also releasing the STEER dataset for training and evaluation.
The paper proposes ICG, a novel framework for personalized cover image generation that integrates MLLM-based prompting with preference alignment. It utilizes semantic features and user embeddings to contextualize the diffusion model and adopts a multi-reward learning strategy to address the lack of labeled supervision.