We calculate the agreement of human judgment and our automatic evaluation (i.e., ConvBenchEval()) and find it reaches 81.83% (seeing Table 3 - 6 for detailed agreement of each turn of overall). It demonstrates the effectiveness of ConvBenchEval(), which uses ChatGPT. The agreement between ChatGPT and GPT4 is very high at 87.38%. It demonstrates that using different LLMs as judges slightly influences the evaluation results. ConvBenchEval() armed with ChatGPT can is reliable and low-cost. From the above tables, we also observe that though GPT4V is expensive and can capture images, its judgment performs worse than GPT4's judgment.

We calculate the agreement of human judgment and our automatic evaluation (i.e., ConvBenchEval()) and find it reaches 81.83% (seeing Table 3 - 6 for detailed agreement of each turn of overall). It demonstrates the effectiveness of ConvBenchEval(), which uses ChatGPT. The agreement between ChatGPT and GPT4 is very high at 87.38%. It demonstrates that using different LLMs as judges slightly influences the evaluation results. ConvBenchEval() armed with ChatGPT can is reliable and low-cost. From the above tables, we also observe that though GPT4V is expensive and can capture images, its judgment performs worse than GPT4's judgment.

Multi-turn visual conversation is an important ability of real-world AI assistants. However, the related evaluation benchmark is missed.

Due to the rapid advancements in multimodal large language models, evaluating their multimodal mathematical capabilities continues to receive wide attention. Despite the datasets like MathVista proposed benchmarks for assessing mathematical capabilities in multimodal scenarios, there is still a lack of corresponding evaluation tools and datasets for fine-grained assessment in the context of K12 education in Chinese language. To systematically evaluate the capability of multimodal large models in solving Chinese multimodal mathematical problems, we propose a Chinese Multi-modal Math Skill Evaluation Benchmark, named CMMaTH, contraining 23k multimodal K12 math related questions, forming the largest Chinese multimodal mathematical problem benchmark to date. CMMaTH questions from elementary to high school levels, provide increased diversity in problem types, solution objectives, visual elements, detailed knowledge points, and standard solution annotations. We have constructed an open-source tool GradeGPT integrated with the CMMaTH dataset, facilitating stable, rapid, and cost-free model evaluation. Our data and code are available.

DeepFakes, which refer to AI-generated media content, have become an increasing concern due to their use as a means for disinformation. Detecting DeepFakes is currently solved with programmed machine learning algorithms. In this work, we investigate the capabilities of multimodal large language models (LLMs) in DeepFake detection. We conducted qualitative and quantitative experiments to demonstrate multimodal LLMs and show that they can expose AI-generated images through careful experimental design and prompt engineering. This is interesting, considering that LLMs are not inherently tailored for media forensic tasks, and the process does not require programming. We discuss the limitations of multimodal LLMs for these tasks and suggest possible improvements.

The advent of Vision Language Models (VLM) has allowed researchers to investigate the visual understanding of a neural network using natural language. Beyond object classification and detection, VLMs are capable of visual comprehension and common-sense reasoning. This naturally led to the question: How do VLMs respond when the image itself is inherently unreasonable? To this end, we present IllusionVQA: a diverse dataset of challenging optical illusions and hard-to-interpret scenes to test the capability of VLMs in two distinct multiple-choice VQA tasks - comprehension and soft localization. GPT4V, the best-performing VLM, achieves 62.99% accuracy (4-shot) on the comprehension task and 49.7% on the localization task (4-shot and Chain-of-Thought). Human evaluation reveals that humans achieve 91.03% and 100% accuracy in comprehension and localization. We discover that In-Context Learning (ICL) and Chain-of-Thought reasoning substantially degrade the performance of GeminiPro on the localization task. Tangentially, we discover a potential weakness in the ICL capabilities of VLMs: they fail to locate optical illusions even when the correct answer is in the context window as a few-shot example.

The field of computational pathology has witnessed remarkable progress in the development of both task-specific predictive models and task-agnostic self-supervised vision encoders. However, despite the explosive growth of generative artificial intelligence (AI), there has been limited study on building general purpose, multimodal AI assistants tailored to pathology. Here we present PathChat, a vision-language generalist AI assistant for human pathology using an in-house developed foundational vision encoder pretrained on 100 million histology images from over 100,000 patient cases and 1.18 million pathology image-caption pairs. The vision encoder is then combined with a pretrained large language model and the whole system is finetuned on over 250,000 diverse disease agnostic visual language instructions. We compare PathChat against several multimodal vision language AI assistants as well as GPT4V, which powers the commercially available multimodal general purpose AI assistant ChatGPT-4. When relevant clinical context is provided with the histology image, PathChat achieved a diagnostic accuracy of 87% on multiple-choice questions based on publicly available cases of diverse tissue origins and disease models. Additionally, using open-ended questions and human expert evaluation, we found that overall PathChat produced more accurate and pathologist-preferable responses to diverse queries related to pathology. As an interactive and general vision language AI assistant that can flexibly handle both visual and natural language inputs, PathChat can potentially find impactful applications in pathology education, research, and human-in-the-loop clinical decision making.

This work focuses on the potential of Vision LLMs (VLLMs) in visual reasoning. Different from prior studies, we shift our focus from evaluating standard performance to introducing a comprehensive safety evaluation suite, covering both out-of-distribution (OOD) generalization and adversarial robustness. For the OOD evaluation, we present two novel VQA datasets, each with one variant, designed to test model performance under challenging conditions. In exploring adversarial robustness, we propose a straightforward attack strategy for misleading VLLMs to produce visual-unrelated responses. Moreover, we assess the efficacy of two jailbreaking strategies, targeting either the vision or language component of VLLMs. Our evaluation of 21 diverse models, ranging from open-source VLLMs to GPT-4V, yields interesting observations: 1) Current VLLMs struggle with OOD texts but not images, unless the visual information is limited; and 2) These VLLMs can be easily misled by deceiving vision encoders only, and their vision-language training often compromise safety protocols. We release this safety evaluation suite at https://github.com/UCSC-VLAA/vllm-safety-benchmark.

In the intricate landscape of modern healthcare, medical image classification emerges as a pivotal task, driving crucial decisions in diagnosis, treatment planning, and patient management. This process involves the systematic categorization of various types of medical imagery--including X-rays, CT scans, MRIs, and ultrasound--into distinct classes that assist healthcare professionals in identifying anomalies, understanding physiological phenomena, and detecting diseases at early stages. The reliability and precision of image classification are paramount, given that these determinations form the bedrock upon which medical practitioners build their diagnostic and therapeutic strategies, directly impacting patient outcomes. With an increasing influx of complex imaging data and a growing need for rapid, accurate interpretation, the medical sector faces significant pressure to evolve beyond traditional analysis methods, necessitating innovative solutions that enhance the efficiency and accuracy of image classification. The advent of large foundation models in artificial intelligence has ushered in a transformative era of computational capabilities. These models, characterized by their extensive scale, diverse training datasets, and impressive adaptability, have demonstrated profound impacts across various domains.