While Large Language Models show remarkable performance in natural language understanding, their resource-intensive nature makes them less accessible. In contrast, smaller language models such as MiniCPM offer more sustainable scalability, but often underperform without specialized optimization. In this paper, we explore the enhancement of smaller language models through the improvement of their text embeddings. We select three language models, MiniCPM, Phi-2, and Gemma, to conduct contrastive fine-tuning on the NLI dataset. Our results demonstrate that this fine-tuning method enhances the quality of text embeddings for all three models across various benchmarks, with MiniCPM showing the most significant improvements of an average 56.33% performance gain. The contrastive fine-tuning code is publicly available at https://github.com/trapoom555/Language-Model-STS-CFT.

In an era defined by the explosive growth of data and rapid technological advancements, Multimodal Large Language Models (MLLMs) stand at the forefront of artificial intelligence (AI) systems. Designed to seamlessly integrate diverse data types-including text, images, videos, audio, and physiological sequences-MLLMs address the complexities of real-world applications far beyond the capabilities of single-modality systems. In this paper, we systematically sort out the applications of MLLM in multimodal tasks such as natural language, vision, and audio. We also provide a comparative analysis of the focus of different MLLMs in the tasks, and provide insights into the shortcomings of current MLLMs, and suggest potential directions for future research. Through these discussions, this paper hopes to provide valuable insights for the further development and application of MLLM.

Entity Alignment (EA) aims to match equivalent entities in different Knowledge Graphs (KGs), which is essential for knowledge fusion and integration. Recently, embedding-based EA has attracted significant attention and many approaches have been proposed. Early approaches primarily focus on learning entity embeddings from the structural features of KGs, defined by relation triples. Later methods incorporated entities' names and attributes as auxiliary information to enhance embeddings for EA. However, these approaches often used different techniques to encode structural and attribute information, limiting their interaction and mutual enhancement. In this work, we propose a dense entity retrieval framework for EA, leveraging language models to uniformly encode various features of entities and facilitate nearest entity search across KGs. Alignment candidates are first generated through entity retrieval, which are subsequently reranked to determine the final alignments. We conduct comprehensive experiments on both cross-lingual and monolingual EA datasets, demonstrating that our approach achieves state-of-the-art performance compared to existing EA methods.

Vision--Language Models (VLMs) have demonstrated success across diverse applications, yet their potential to assist in relevance judgments remains uncertain. This paper assesses the relevance estimation capabilities of VLMs, including CLIP, LLaVA, and GPT-4V, within a large-scale \textit{ad hoc} retrieval task tailored for multimedia content creation in a zero-shot fashion. Preliminary experiments reveal the following: (1) Both LLaVA and GPT-4V, encompassing open-source and closed-source visual-instruction-tuned Large Language Models (LLMs), achieve notable Kendall's $\tau \sim 0.4$ when compared to human relevance judgments, surpassing the CLIPScore metric. (2) While CLIPScore is strongly preferred, LLMs are less biased towards CLIP-based retrieval systems. (3) GPT-4V's score distribution aligns more closely with human judgments than other models, achieving a Cohen's $\kappa$ value of around 0.08, which outperforms CLIPScore at approximately -0.096. These findings underscore the potential of LLM-powered VLMs in enhancing relevance judgments.

This paper introduces a new LLM based agent framework for simulating electric vehicle (EV) charging behavior, integrating user preferences, psychological characteristics, and environmental factors to optimize the charging process. The framework comprises several modules, enabling sophisticated, adaptive simulations. Dynamic decision making is supported by continuous reflection and memory updates, ensuring alignment with user expectations and enhanced efficiency. The framework's ability to generate personalized user profiles and real-time decisions offers significant advancements for urban EV charging management. Future work could focus on incorporating more intricate scenarios and expanding data sources to enhance predictive accuracy and practical utility.

LLM-driven dialog systems are used in a diverse set of applications, ranging from healthcare to customer service. However, given their generalization capability, it is difficult to ensure that these chatbots stay within the boundaries of the specialized domains, potentially resulting in inaccurate information and irrelevant responses. This paper introduces an unsupervised approach for automatically inducing domain-specific dialog flows that can be used to constrain LLM-based chatbots. We introduce two variants of dialog flow based on the availability of in-domain conversation instances. Through human and automatic evaluation over various dialog domains, we demonstrate that our high-quality data-guided dialog flows achieve better domain coverage, thereby overcoming the need for extensive manual crafting of such flows.

Large language models (LLMs) are now being considered and even deployed for applications that support high-stakes decision-making, such as recruitment and clinical decisions. While several methods have been proposed for measuring bias, there remains a gap between predictions, which are what the proposed methods consider, and how they are used to make decisions. In this work, we introduce Rank-Allocational-Based Bias Index (RABBI), a model-agnostic bias measure that assesses potential allocational harms arising from biases in LLM predictions. We compare RABBI and current bias metrics on two allocation decision tasks. We evaluate their predictive validity across ten LLMs and utility for model selection. Our results reveal that commonly-used bias metrics based on average performance gap and distribution distance fail to reliably capture group disparities in allocation outcomes, whereas RABBI exhibits a strong correlation with allocation disparities. Our work highlights the need to account for how models are used in contexts with limited resource constraints.

The rapid advancement of Generative Artificial Intelligence (GenAI) presents both opportunities and challenges for English for Academic Purposes (EAP) instruction. This paper proposes an adaptation of the AI Assessment Scale (AIAS) specifically tailored for EAP contexts, termed the EAP-AIAS. This framework aims to provide a structured approach for integrating GenAI tools into EAP assessment practices while maintaining academic integrity and supporting language development. The EAP-AIAS consists of five levels, ranging from "No AI" to "Full AI", each delineating appropriate GenAI usage in EAP tasks. We discuss the rationale behind this adaptation, considering the unique needs of language learners and the dual focus of EAP on language proficiency and academic acculturation. This paper explores potential applications of the EAP-AIAS across various EAP assessment types, including writing tasks, presentations, and research projects. By offering a flexible framework, the EAP-AIAS seeks to empower EAP practitioners seeking to deal with the complexities of GenAI integration in education and prepare students for an AI-enhanced academic and professional future. This adaptation represents a step towards addressing the pressing need for ethical and pedagogically sound AI integration in language education.

This study introduces 'Psycho Analyst', a custom GPT model based on OpenAI's GPT-4, optimized for pre-screening mental health disorders. Enhanced with DSM-5, PHQ-8, detailed data descriptions, and extensive training data, the model adeptly decodes nuanced linguistic indicators of mental health disorders. It utilizes a dual-task framework that includes binary classification and a three-stage PHQ-8 score computation involving initial assessment, detailed breakdown, and independent assessment, showcasing refined analytic capabilities. Validation with the DAIC-WOZ dataset reveals F1 and Macro-F1 scores of 0.929 and 0.949, respectively, along with the lowest MAE and RMSE of 2.89 and 3.69 in PHQ-8 scoring. These results highlight the model's precision and transformative potential in enhancing public mental health support, improving accessibility, cost-effectiveness, and serving as a second opinion for professionals.

Instruction fine-tuning stands as a crucial advancement in leveraging large language models (LLMs) for enhanced task performance. However, the annotation of instruction datasets has traditionally been expensive and laborious, often relying on manual annotations or costly API calls of proprietary LLMs. To address these challenges, we introduce FANNO, a fully autonomous, open-sourced framework that revolutionizes the annotation process without the need for pre-existing annotated data. Utilizing a Mistral-7b-instruct model, FANNO efficiently produces diverse and high-quality datasets through a structured process involving document pre-screening, instruction generation, and response generation. Experiments on Open LLM Leaderboard and AlpacaEval benchmark show that the FANNO can generate high-quality data with diversity and complexity for free, comparable to human-annotated or cleaned datasets like Alpaca-GPT4-Cleaned.