Recent advancements in large language model(LLM) performance on medical multiple choice question (MCQ) benchmarks have stimulated interest from healthcare providers and patients globally. Particularly in low-and middle-income countries (LMICs) facing acute physician shortages and lack of specialists, LLMs offer a potentially scalable pathway to enhance healthcare access and reduce costs. However, their effectiveness in the Global South, especially across the African continent, remains to be established. In this work, we introduce AfriMed-QA, the first large scale Pan-African English multi-specialty medical Question-Answering (QA) dataset, 15,000 questions (open and closed-ended) sourced from over 60 medical schools across 16 countries, covering 32 medical specialties. We further evaluate 30 LLMs across multiple axes including correctness and demographic bias. Our findings show significant performance variation across specialties and geographies, MCQ performance clearly lags USMLE (MedQA). We find that biomedical LLMs underperform general models and smaller edge-friendly LLMs struggle to achieve a passing score. Interestingly, human evaluations show a consistent consumer preference for LLM answers and explanations when compared with clinician answers.

The influence of contextual input on the behavior of large language models (LLMs) has prompted the development of context attribution methods that aim to quantify each context span's effect on an LLM's generations. The leave-one-out (LOO) error, which measures the change in the likelihood of the LLM's response when a given span of the context is removed, provides a principled way to perform context attribution, but can be prohibitively expensive to compute for large models. In this work, we introduce AttriBoT, a series of novel techniques for efficiently computing an approximation of the LOO error for context attribution. Specifically, AttriBoT uses cached activations to avoid redundant operations, performs hierarchical attribution to reduce computation, and emulates the behavior of large target models with smaller proxy models. Taken together, AttriBoT can provide a >300x speedup while remaining more faithful to a target model's LOO error than prior context attribution methods. This stark increase in performance makes computing context attributions for a given response 30x faster than generating the response itself, empowering real-world applications that require computing attributions at scale. We release a user-friendly and efficient implementation of AttriBoT to enable efficient LLM interpretability as well as encourage future development of efficient context attribution methods.

Semantic Knowledge Graphs (SKG) face challenges with scalability, flexibility, contextual understanding, and handling unstructured or ambiguous information. However, they offer formal and structured knowledge enabling highly interpretable and reliable results by means of reasoning and querying. Large Language Models (LLMs) overcome those limitations making them suitable in open-ended tasks and unstructured environments. Nevertheless, LLMs are neither interpretable nor reliable. To solve the dichotomy between LLMs and SKGs we envision Logic Augmented Generation (LAG) that combines the benefits of the two worlds. LAG uses LLMs as Reactive Continuous Knowledge Graphs that can generate potentially infinite relations and tacit knowledge on-demand. SKGs are key for injecting a discrete heuristic dimension with clear logical and factual boundaries. We exemplify LAG in two tasks of collective intelligence, i.e., medical diagnostics and climate projections. Understanding the properties and limitations of LAG, which are still mostly unknown, is of utmost importance for enabling a variety of tasks involving tacit knowledge in order to provide interpretable and effective results.

In this paper we present an approach to reduce hallucinations in Large Language Models (LLMs) by incorporating Knowledge Graphs (KGs) as an additional modality. Our method involves transforming input text into a set of KG embeddings and using an adapter to integrate these embeddings into the language model space, without relying on external retrieval processes. To facilitate this, we created WikiEntities, a dataset containing over 3 million Wikipedia texts annotated with entities from Wikidata and their corresponding embeddings from PyTorch-BigGraph. This dataset serves as a valuable resource for training Entity Linking models and adapting the described method to various LLMs using specialized adapters. Our method does not require fine-tuning of the language models themselves; instead, we only train the adapter. This ensures that the model's performance on other tasks is not affected. We trained an adapter for the Mistral 7B, LLaMA 2-7B (chat), and LLaMA 3-8B (instruct) models using this dataset and demonstrated that our approach improves performance on the HaluEval, True-False benchmarks and FEVER dataset. The results indicate that incorporating KGs as a new modality can effectively reduce hallucinations and improve the factual accuracy of language models, all without the need for external retrieval.

The evaluation of mathematical reasoning capabilities is essential for advancing Artificial General Intelligence (AGI). While Large Language Models (LLMs) have shown impressive performance in solving mathematical problems, existing benchmarks such as GSM8K and MATH present limitations, including narrow problem definitions with specific numbers and reliance on predetermined rules that hinder accurate assessments of reasoning and generality. This paper introduces the UTMath Benchmark, a robust evaluation framework designed to assess LLMs through extensive unit tests, with a focus on both the accuracy and generality of model responses. It comprises 1,053 cutting-edge problems spanning nine mathematical domains, with an average of 68 test cases per problem. UTMath is highly challenging, with the best-performing model, o1-mini, solving only 32.57\% of the problems, followed by o1-preview at 27.16\%, and GPT-4o at 26.93\%. Furthermore, we present the Reasoning-to-Coding of Thoughts (RCoT) approach, which encourages LLMs to engage in explicit reasoning prior to code generation, thereby facilitating the production of more sophisticated solutions and enhancing overall performance and efficiency. Additionally, we also release the UTMath-Train training dataset (more than 70k samples), to support the community in further exploring mathematical reasoning. Our benchmark can be accessed via the following link: https://github.com/UTMathGroup/UTMath

Large Language Models (LLMs) have demonstrated remarkable capabilities across various domains, particularly in task generalization for both text and vision data. While fine-tuning these models can significantly enhance their performance on specific downstream tasks, it often requires high-quality data that cannot be shared due to privacy concerns. Federated Learning (FL) offers a promising solution for collaborative training without direct data sharing. However, many parameter-efficient fine-tuning strategies for LLMs in FL, particularly those based on Low-Rank Adaptation (LoRA), face limitations. In this paper, we critically analyze the convergence and performance guarantees of popular FL frameworks utilizing LoRA, highlighting its suboptimal nature due to constrained subspace learning of low-rank matrices. This limitation hinders effective fine-tuning of LLMs in federated settings. Through rigorous analytical and empirical evaluations, we demonstrate that direct weight averaging outperforms LoRA-based strategies, leading to superior performance for fine-tuned models. Our comprehensive comparison unmasks inefficiencies in LoRA approaches and underscores the advantages of direct weight aggregation. We extend our analysis to low-rank gradient-based optimizers, such as GaLore, used during local training steps. Our findings show that GaLore along with direct-weight aggregation is a more effective approach, outperforming federated LoRA methods like FlexLoRA and FFA-LoRA across both text and image modalities. While privacy remains paramount in FL discourse, our focus is on assessing performance outcomes of federated fine-tuned models and evaluating various FL frameworks from both theoretical and empirical perspectives. Our findings advocate reassessing the reliance on LoRA within FL contexts, paving the way for more efficient training methodologies.

Remote sensing imagery is dense with objects and contextual visual information. There is a recent trend to combine paired satellite images and text captions for pretraining performant encoders for downstream tasks. However, while contrastive image-text methods like CLIP enable vision-language alignment and zero-shot classification ability, vision-only downstream performance tends to degrade compared to image-only pretraining, such as MAE. In this paper, we propose FLAVARS, a pretraining method that combines the best of both contrastive learning and masked modeling, along with geospatial alignment via contrastive location encoding. We find that FLAVARS significantly outperforms a baseline of SkyCLIP for vision-only tasks such as KNN classification and semantic segmentation, +6\% mIOU on SpaceNet1, while retaining the ability to perform zero-shot classification, unlike MAE pretrained methods.

Recent work has proposed automated red-teaming methods for testing the vulnerabilities of a given target large language model (LLM). These methods use red-teaming LLMs to uncover inputs that induce harmful behavior in a target LLM. In this paper, we study red-teaming strategies that enable a targeted security assessment. We propose an optimization framework for red-teaming with proximity constraints, where the discovered prompts must be similar to reference prompts from a given dataset. This dataset serves as a template for the discovered prompts, anchoring the search for test-cases to specific topics, writing styles, or types of harmful behavior. We show that established auto-regressive model architectures do not perform well in this setting. We therefore introduce a black-box red-teaming method inspired by text-diffusion models: Diffusion for Auditing and Red-Teaming (DART). DART modifies the reference prompt by perturbing it in the embedding space, directly controlling the amount of change introduced. We systematically evaluate our method by comparing its effectiveness with established methods based on model fine-tuning and zero- and few-shot prompting. Our results show that DART is significantly more effective at discovering harmful inputs in close proximity to the reference prompt.

Visual Commonsense Reasoning, which is regarded as one challenging task to pursue advanced visual scene comprehension, has been used to diagnose the reasoning ability of AI systems. However, reliable reasoning requires a good grasp of the scene's details. Existing work fails to effectively exploit the real-world object relationship information present within the scene, and instead overly relies on knowledge from training memory. Based on these observations, we propose a novel scene-graph-enhanced visual commonsense reasoning generation method named \textit{\textbf{G2}}, which first utilizes the image patches and LLMs to construct a location-free scene graph, and then answer and explain based on the scene graph's information. We also propose automatic scene graph filtering and selection strategies to absorb valuable scene graph information during training. Extensive experiments are conducted on the tasks and datasets of scene graph constructing and visual commonsense answering and explaining, respectively. Experimental results and ablation analysis demonstrate the effectiveness of our proposed framework.

The rapid advancement of Large Vision-Language Models (LVLMs) has enhanced capabilities offering potential applications from content creation to productivity enhancement. Despite their innovative potential, LVLMs exhibit vulnerabilities, especially in generating potentially toxic or unsafe responses. Malicious actors can exploit these vulnerabilities to propagate toxic content in an automated (or semi-) manner, leveraging the susceptibility of LVLMs to deception via strategically crafted prompts without fine-tuning or compute-intensive procedures. Despite the red-teaming efforts and inherent potential risks associated with the LVLMs, exploring vulnerabilities of LVLMs remains nascent and yet to be fully addressed in a systematic manner. This study systematically examines the vulnerabilities of open-source LVLMs, including LLaVA, InstructBLIP, Fuyu, and Qwen, using adversarial prompt strategies that simulate real-world social manipulation tactics informed by social theories. Our findings show that (i) toxicity and insulting are the most prevalent behaviors, with the mean rates of 16.13% and 9.75%, respectively; (ii) Qwen-VL-Chat, LLaVA-v1.6-Vicuna-7b, and InstructBLIP-Vicuna-7b are the most vulnerable models, exhibiting toxic response rates of 21.50%, 18.30% and 17.90%, and insulting responses of 13.40%, 11.70% and 10.10%, respectively; (iii) prompting strategies incorporating dark humor and multimodal toxic prompt completion significantly elevated these vulnerabilities. Despite being fine-tuned for safety, these models still generate content with varying degrees of toxicity when prompted with adversarial inputs, highlighting the urgent need for enhanced safety mechanisms and robust guardrails in LVLM development.