The advanced function-calling capabilities of foundation models open up new possibilities for deploying agents to perform complex API tasks. However, managing large amounts of data and interacting with numerous APIs makes function calling hardware-intensive and costly, especially on edge devices. Current Large Language Models (LLMs) struggle with function calling at the edge because they cannot handle complex inputs or manage multiple tools effectively. This results in low task-completion accuracy, increased delays, and higher power consumption. In this work, we introduce Less-is-More, a novel fine-tuning-free function-calling scheme for dynamic tool selection. Our approach is based on the key insight that selectively reducing the number of tools available to LLMs significantly improves their function-calling performance, execution time, and power efficiency on edge devices. Experimental results with state-of-the-art LLMs on edge hardware show agentic success rate improvements, with execution time reduced by up to 70% and power consumption by up to 40%.

The fast advancement of Large Vision-Language Models (LVLMs) has shown immense potential. These models are increasingly capable of tackling abstract visual tasks. Geometric structures, particularly graphs with their inherent flexibility and complexity, serve as an excellent benchmark for evaluating these models' predictive capabilities. While human observers can readily identify subtle visual details and perform accurate analyses, our investigation reveals that state-of-the-art LVLMs exhibit consistent limitations in specific visual graph scenarios, especially when confronted with stylistic variations. In response to these challenges, we introduce VisGraphVar (Visual Graph Variability), a customizable benchmark generator able to produce graph images for seven distinct task categories (detection, classification, segmentation, pattern recognition, link prediction, reasoning, matching), designed to systematically evaluate the strengths and limitations of individual LVLMs. We use VisGraphVar to produce 990 graph images and evaluate six LVLMs, employing two distinct prompting strategies, namely zero-shot and chain-of-thought. The findings demonstrate that variations in visual attributes of images (e.g., node labeling and layout) and the deliberate inclusion of visual imperfections, such as overlapping nodes, significantly affect model performance. This research emphasizes the importance of a comprehensive evaluation across graph-related tasks, extending beyond reasoning alone. VisGraphVar offers valuable insights to guide the development of more reliable and robust systems capable of performing advanced visual graph analysis.

Machine Learning and data mining techniques (i.e. supervised and unsupervised techniques) are used across domains to detect user safety violations. Examples include classifiers used to detect whether an email is spam or a web-page is requesting bank login information. However, existing ML/DM classifiers are limited in their ability to understand natural languages w.r.t the context and nuances. The aforementioned challenges are overcome with the arrival of Gen-AI techniques, along with their inherent ability w.r.t translation between languages, fine-tuning between various tasks and domains. In this manuscript, we provide a comprehensive overview of the various work done while using Gen-AI techniques w.r.t user safety. In particular, we first provide the various domains (e.g. phishing, malware, content moderation, counterfeit, physical safety) across which Gen-AI techniques have been applied. Next, we provide how Gen-AI techniques can be used in conjunction with various data modalities i.e. text, images, videos, audio, executable binaries to detect violations of user-safety. Further, also provide an overview of how Gen-AI techniques can be used in an adversarial setting. We believe that this work represents the first summarization of Gen-AI techniques for user-safety.

As machine intelligence evolves, the need to test and compare the problem-solving abilities of different AI models grows. However, current benchmarks are often simplistic, allowing models to perform uniformly well and making it difficult to distinguish their capabilities. Additionally, benchmarks typically rely on static question-answer pairs that the models might memorize or guess. To address these limitations, we introduce Dynamic Intelligence Assessment (DIA), a novel methodology for testing AI models using dynamic question templates and improved metrics across multiple disciplines such as mathematics, cryptography, cybersecurity, and computer science. The accompanying dataset, DIA-Bench, contains a diverse collection of challenge templates with mutable parameters presented in various formats, including text, PDFs, compiled binaries, visual puzzles, and CTF-style cybersecurity challenges. Our framework introduces four new metrics to assess a model's reliability and confidence across multiple attempts. These metrics revealed that even simple questions are frequently answered incorrectly when posed in varying forms, highlighting significant gaps in models' reliability. Notably, API models like GPT-4o often overestimated their mathematical capabilities, while ChatGPT-4o demonstrated better performance due to effective tool usage. In self-assessment, OpenAI's o1-mini proved to have the best judgement on what tasks it should attempt to solve. We evaluated 25 state-of-the-art LLMs using DIA-Bench, showing that current models struggle with complex tasks and often display unexpectedly low confidence, even with simpler questions. The DIA framework sets a new standard for assessing not only problem-solving but also a model's adaptive intelligence and ability to assess its limitations. The dataset is publicly available on the project's page: https://github.com/DIA-Bench.

Formal Property Verification (FPV), using SystemVerilog Assertions (SVA), is crucial for ensuring the completeness of design with respect to the specification. However, writing SVA is a laborious task and has a steep learning curve. In this work, we present a large language model (LLM) -based flow to automatically generate high-quality SVA from the design specification documents, named \ToolName. We introduce a novel sub-task-focused fine-tuning approach that effectively addresses functionally incorrect assertions produced by baseline LLMs, leading to a remarkable 7.3-fold increase in the number of functionally correct assertions. Recognizing the prevalence of syntax and semantic errors, we also developed an iterative refinement method that enhances the LLM's initial outputs by systematically re-prompting it to correct identified issues. This process is further strengthened by a custom compiler that generates meaningful error messages, guiding the LLM towards improved accuracy. The experiments demonstrate a 26\% increase in the number of assertions free from syntax errors using this approach, showcasing its potential to streamline the FPV process.

Large language models (LLMs) have demonstrated remarkable capabilities across a wide range of natural language processing tasks. However, their tendency to exhibit sycophantic behavior - excessively agreeing with or flattering users - poses significant risks to their reliability and ethical deployment. This paper provides a technical survey of sycophancy in LLMs, analyzing its causes, impacts, and potential mitigation strategies. We review recent work on measuring and quantifying sycophantic tendencies, examine the relationship between sycophancy and other challenges like hallucination and bias, and evaluate promising techniques for reducing sycophancy while maintaining model performance. Key approaches explored include improved training data, novel fine-tuning methods, post-deployment control mechanisms, and decoding strategies. We also discuss the broader implications of sycophancy for AI alignment and propose directions for future research. Our analysis suggests that mitigating sycophancy is crucial for developing more robust, reliable, and ethically-aligned language models.

Graphic designs are an effective medium for visual communication. They range from greeting cards to corporate flyers and beyond. Off-late, machine learning techniques are able to generate such designs, which accelerates the rate of content production. An automated way of evaluating their quality becomes critical. Towards this end, we introduce Design-o-meter, a data-driven methodology to quantify the goodness of graphic designs. Further, our approach can suggest modifications to these designs to improve its visual appeal. To the best of our knowledge, Design-o-meter is the first approach that scores and refines designs in a unified framework despite the inherent subjectivity and ambiguity of the setting. Our exhaustive quantitative and qualitative analysis of our approach against baselines adapted for the task (including recent Multimodal LLM-based approaches) brings out the efficacy of our methodology. We hope our work will usher more interest in this important and pragmatic problem setting.

Adversarial audio attacks pose a significant threat to the growing use of large language models (LLMs) in voice-based human-machine interactions. While existing research has primarily focused on model-specific adversarial methods, real-world applications demand a more generalizable and universal approach to audio adversarial attacks. In this paper, we introduce the Chat-Audio Attacks (CAA) benchmark including four distinct types of audio attacks, which aims to explore the the vulnerabilities of LLMs to these audio attacks in conversational scenarios. To evaluate the robustness of LLMs, we propose three evaluation strategies: Standard Evaluation, utilizing traditional metrics to quantify model performance under attacks; GPT-4o-Based Evaluation, which simulates real-world conversational complexities; and Human Evaluation, offering insights into user perception and trust. We evaluate six state-of-the-art LLMs with voice interaction capabilities, including Gemini-1.5-Pro, GPT-4o, and others, using three distinct evaluation methods on the CAA benchmark. Our comprehensive analysis reveals the impact of four types of audio attacks on the performance of these models, demonstrating that GPT-4o exhibits the highest level of resilience.

Denoising with a Joint-Embedding Predictive Architecture (D-JEPA), an autoregressive model, has demonstrated outstanding performance in class-conditional image generation. However, the application of next-token prediction in high-resolution text-to-image generation remains underexplored. In this paper, we introduce D-JEPA$\cdot$T2I, an extension of D-JEPA incorporating flow matching loss, designed to enable data-efficient continuous resolution learning. D-JEPA$\cdot$T2I leverages a multimodal visual transformer to effectively integrate textual and visual features and adopts Visual Rotary Positional Embedding (VoPE) to facilitate continuous resolution learning. Furthermore, we devise a data feedback mechanism that significantly enhances data utilization efficiency. For the first time, we achieve state-of-the-art \textbf{high-resolution} image synthesis via next-token prediction. The experimental code and pretrained models will be open-sourced at \url{https://d-jepa.github.io/t2i}.

Large language models (LLMs) have been widely adopted in applications such as automated content generation and even critical decision-making systems. However, the risk of prompt injection allows for potential manipulation of LLM outputs. While numerous attack methods have been documented, achieving full control over these outputs remains challenging, often requiring experienced attackers to make multiple attempts and depending heavily on the prompt context. Recent advancements in gradient-based white-box attack techniques have shown promise in tasks like jailbreaks and system prompt leaks. Our research generalizes gradient-based attacks to find a trigger that is (1) Universal: effective irrespective of the target output; (2) Context-Independent: robust across diverse prompt contexts; and (3) Precise Output: capable of manipulating LLM inputs to yield any specified output with high accuracy. We propose a novel method to efficiently discover such triggers and assess the effectiveness of the proposed attack. Furthermore, we discuss the substantial threats posed by such attacks to LLM-based applications, highlighting the potential for adversaries to taking over the decisions and actions made by AI agents.