With over 781 billion website visits globally each month [51], their popularity highlights the growing need for developers to maintain high standards of quality and functionality. Traditional manual web testing approaches, however, can be time-consuming and challenging [8], leading to the increased adoption of automated testing methodologies to streamline the quality assurance process [5, 12, 13, 19, 24, 27, 30, 44, 48, 53, 56, 64]. Despite these advances, conventional testing tools may exhibit challenges and shortcomings regarding testing coverage, potentially overlooking critical bugs and usability issues [18, 19]. The discrepancy between tests generated by conventional methods and real user interactions further compounds these challenges [63], resulting in suboptimal testing outcomes. Web applications typically encompass a spectrum of actions, including form submissions, button clicks, and navigation through various pages. Automated testing tools for web applications encounter challenges stemming from the intricate and dynamic nature of modern web interfaces, which can feature diverse layouts, interactions, and non-deterministic states [3]. To address these challenges and mitigate the limitations of the traditional test generation methods, there has been a growing interest in leveraging deep learning (DL) [12, 13] and reinforcement learning (RL) [22, 23, 26, 27, 30, 31, 48, 64] techniques for automated testing in web applications. By assimilating insights from human testers' behavior, such automated testing approaches aim to emulate human-like interactions with web interfaces, thereby improving the comprehensiveness and effectiveness of testing. However, these DL and RL-based methodologies are not without their constraints.

As the demand for human-like interactions with LLMs continues to grow, so does the interest in manipulating their personality traits, which has emerged as a key area of research. Methods like prompt-based In-Context Knowledge Editing (IKE) and gradient-based Model Editor Networks (MEND) have been explored but show irregularity and variability. IKE depends on the prompt, leading to variability and sensitivity, while MEND yields inconsistent and gibberish outputs. To address this, we employed Opinion QA Based Parameter-Efficient Fine-Tuning (PEFT), specifically Quantized Low-Rank Adaptation (QLORA), to manipulate the Big Five personality traits: Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism. After PEFT, models such as Mistral-7B-Instruct and Llama-2-7B-chat began generating emojis, despite their absence in the PEFT data. For instance, Llama-2-7B-chat generated emojis in 99.5% of extraversion-related test instances, while Mistral-8B-Instruct did so in 92.5% of openness-related test instances. Explainability analysis indicated that the LLMs used emojis intentionally to express these traits. This paper provides a number of novel contributions.

Robotic assembly tasks are open challenges due to the long task horizon and complex part relations. Behavior trees (BTs) are increasingly used in robot task planning for their modularity and flexibility, but manually designing them can be effort-intensive. Large language models (LLMs) have recently been applied in robotic task planning for generating action sequences, but their ability to generate BTs has not been fully investigated. To this end, We propose LLM as BT-planner, a novel framework to leverage LLMs for BT generation in robotic assembly task planning and execution. Four in-context learning methods are introduced to utilize the natural language processing and inference capabilities of LLMs to produce task plans in BT format, reducing manual effort and ensuring robustness and comprehensibility. We also evaluate the performance of fine-tuned, fewer-parameter LLMs on the same tasks. Experiments in simulated and real-world settings show that our framework enhances LLMs' performance in BT generation, improving success rates in BT generation through in-context learning and supervised fine-tuning.

We propose a scaling law hypothesis for multimodal models processing text, audio, images, and video within a shared token and embedding space. Our framework predicts model performance based on modality-specific compression and tokenization efficiency, extending established scaling laws from text-based decoder models to mixed-modality systems. We explore whether leveraging more training data in multiple modalities can reduce the size of the multimodal model, enabling efficient deployment on resource-constrained devices.

Natural language interfaces to databases allow non-SQL experts to query relational databases more conveniently. Text-to-SQL, which automatically maps natural language questions to SQL queries [1, 2] has therefore emerged as an important problem, especially due to generative AI. Early Text-to-SQL systems were domain-specific with limited user interaction, often relying on rule-based approaches to parse input questions [3, 4, 5, 6]. Recent advancements have shifted towards greater domain independence by introducing supervised models trained on various cross-domain datasets [7, 8], and transformer-based models fine-tuned with built-in modules and constraints [9, 10, 11, 12]. Unlike retrieval-augmented generation (RAG) [13], which uses transformer-based language models fine-tuned on external knowledge, Text-to-SQL reduces potential hallucinations in domain-specific or knowledge-intensive tasks because the answer is from querying the database rather than being generated directly by a model. Recent developments in Text-to-SQL use large language models (LLMs) with zero-shot [14, 15] and few-shot prompting [16, 17], demonstrating that LLMs can serve as strong baselines with minimal demonstration of questions and schemas and no fine-tuning.

Recent advancements in AI safety have led to increased efforts in training and red-teaming large language models (LLMs) to mitigate unsafe content generation. However, these safety mechanisms may not be comprehensive, leaving potential vulnerabilities unexplored. This paper introduces MathPrompt, a novel jailbreaking technique that exploits LLMs' advanced capabilities in symbolic mathematics to bypass their safety mechanisms. By encoding harmful natural language prompts into mathematical problems, we demonstrate a critical vulnerability in current AI safety measures. Our experiments across 13 state-of-the-art LLMs reveal an average attack success rate of 73.6\%, highlighting the inability of existing safety training mechanisms to generalize to mathematically encoded inputs. Analysis of embedding vectors shows a substantial semantic shift between original and encoded prompts, helping explain the attack's success. This work emphasizes the importance of a holistic approach to AI safety, calling for expanded red-teaming efforts to develop robust safeguards across all potential input types and their associated risks.

Prompt optimization algorithms for Large Language Models (LLMs) excel in multi-step reasoning but still lack effective uncertainty estimation. This paper introduces a benchmark dataset to evaluate uncertainty metrics, focusing on Answer, Correctness, Aleatoric, and Epistemic Uncertainty.

Recent advancements in automatic code generation using large language models (LLMs) have brought us closer to fully automated secure software development. However, existing approaches often rely on a single agent for code generation, which struggles to produce secure, vulnerability-free code. Traditional program synthesis with LLMs has primarily focused on functional correctness, often neglecting critical dynamic security implications that happen during runtime. To address these challenges, we propose AutoSafeCoder, a multi-agent framework that leverages LLM-driven agents for code generation, vulnerability analysis, and security enhancement through continuous collaboration. The framework consists of three agents: a Coding Agent responsible for code generation, a Static Analyzer Agent identifying vulnerabilities, and a Fuzzing Agent performing dynamic testing using a mutation-based fuzzing approach to detect runtime errors. Our contribution focuses on ensuring the safety of multi-agent code generation by integrating dynamic and static testing in an iterative process during code generation by LLM that improves security. Experiments using the SecurityEval dataset demonstrate a 13% reduction in code vulnerabilities compared to baseline LLMs, with no compromise in functionality.

Large Language Model (LLM)-based recommendation systems provide more comprehensive recommendations than traditional systems by deeply analyzing content and user behavior. However, these systems often exhibit biases, favoring mainstream content while marginalizing non-traditional options due to skewed training data. This study investigates the intricate relationship between bias and LLM-based recommendation systems, with a focus on music, song, and book recommendations across diverse demographic and cultural groups. Through a comprehensive analysis conducted over different LLM-models, this paper evaluates the impact of bias on recommendation outcomes. Our findings reveal that bias is so deeply ingrained within these systems that even a simpler intervention like prompt engineering can significantly reduce bias, underscoring the pervasive nature of the issue. Moreover, factors like intersecting identities and contextual information, such as socioeconomic status, further amplify these biases, demonstrating the complexity and depth of the challenges faced in creating fair recommendations across different groups.

Implicit feedback, often used to build recommender systems, unavoidably confronts noise due to factors such as misclicks and position bias. Previous studies have attempted to alleviate this by identifying noisy samples based on their diverged patterns, such as higher loss values, and mitigating the noise through sample dropping or reweighting. Despite the progress, we observe existing approaches struggle to distinguish hard samples and noise samples, as they often exhibit similar patterns, thereby limiting their effectiveness in denoising recommendations. To address this challenge, we propose a Large Language Model Enhanced Hard Sample Denoising (LLMHD) framework. Specifically, we construct an LLM-based scorer to evaluate the semantic consistency of items with the user preference, which is quantified based on summarized historical user interactions. The resulting scores are used to assess the hardness of samples for the pointwise or pairwise training objectives. To ensure efficiency, we introduce a variance-based sample pruning strategy to filter potential hard samples before scoring. Besides, we propose an iterative preference update module designed to continuously refine summarized user preference, which may be biased due to false-positive user-item interactions. Extensive experiments on three real-world datasets and four backbone recommenders demonstrate the effectiveness of our approach.