Objective: Emergency medical dispatch (EMD) is a high-stakes process challenged by caller distress, ambiguity, and cognitive load. Large Language Models (LLMs) and Multi-Agent Systems (MAS) offer opportunities to augment dispatchers. This study aimed to develop and evaluate a taxonomy-grounded, LLM-powered multi-agent system for simulating realistic EMD scenarios. Methods: We constructed a clinical taxonomy (32 chief complaints, 6 caller identities from MIMIC-III) and a six-phase call protocol. Using this framework, we developed an AutoGen-based MAS with Caller and Dispatcher Agents. The system grounds interactions in a fact commons to ensure clinical plausibility and mitigate misinformation. We used a hybrid evaluation framework: four physicians assessed 100 simulated cases for "Guidance Efficacy" and "Dispatch Effectiveness," supplemented by automated linguistic analysis (sentiment, readability, politeness). Results: Human evaluation, with substantial inter-rater agreement (Gwe's AC1 > 0.70), confirmed the system's high performance. It demonstrated excellent Dispatch Effectiveness (e.g., 94 % contacting the correct potential other agents) and Guidance Efficacy (advice provided in 91 % of cases), both rated highly by physicians. Algorithmic metrics corroborated these findings, indicating a predominantly neutral affective profile (73.7 % neutral sentiment; 90.4 % neutral emotion), high readability (Flesch 80.9), and a consistently polite style (60.0 % polite; 0 % impolite). Conclusion: Our taxonomy-grounded MAS simulates diverse, clinically plausible dispatch scenarios with high fidelity. Findings support its use for dispatcher training, protocol evaluation, and as a foundation for real-time decision support. This work outlines a pathway for safely integrating advanced AI agents into emergency response workflows.

Access to humanities research databases is often hindered by the limitations of traditional interaction formats, particularly in the methods of searching and response generation. This study introduces an LLM-based smart assistant designed to facilitate natural language communication with digital humanities data. The assistant, developed in a chatbot format, leverages the RAG approach and integrates state-of-the-art technologies such as hybrid search, automatic query generation, text-to-SQL filtering, semantic database search, and hyperlink insertion. To evaluate the effectiveness of the system, experiments were conducted to assess the response quality of various language models. The testing was based on the Prozhito digital archive, which contains diary entries from predominantly Russian-speaking individuals who lived in the 20th century. The chatbot is tailored to support anthropology and history researchers, as well as non-specialist users with an interest in the field, without requiring prior technical training. By enabling researchers to query complex databases with natural language, this tool aims to enhance accessibility and efficiency in humanities research. The study highlights the potential of Large Language Models to transform the way researchers and the public interact with digital archives, making them more intuitive and inclusive. Additional materials are presented in GitHub repository: https://github.com/alekosus/talking-to-data-intersys2025.

In Romania, cardiovascular problems are the leading cause of death, accounting for nearly one-third of annual fatalities. The severity of this situation calls for innovative diagnosis method for cardiovascular diseases. This article aims to explore efficient, light-weight and rapid methods for arrhythmia diagnosis, in resource-constrained healthcare settings. Due to the lack of Romanian public medical data, we trained our systems using international public datasets, having in mind that the ECG signals are the same regardless the patients' nationality. Within this purpose, we combined multiple datasets, usually used in the field of arrhythmias classification: PTB-XL electrocardiography dataset , PTB Diagnostic ECG Database, China 12-Lead ECG Challenge Database, Georgia 12-Lead ECG Challenge Database, and St. Petersburg INCART 12-lead Arrhythmia Database. For the input data, we employed ECG signal processing methods, specifically a variant of the Pan-Tompkins algorithm, useful in arrhythmia classification because it provides a robust and efficient method for detecting QRS complexes in ECG signals. Additionally, we used machine learning techniques, widely used for the task of classification, including convolutional neural networks (1D CNNs, 2D CNNs, ResNet) and Vision Transformers (ViTs). The systems were evaluated in terms of accuracy and F1 score. We annalysed our dataset from two perspectives. First, we fed the systems with the ECG signals and the GRU-based 1D CNN model achieved the highest accuracy of 93.4% among all the tested architectures. Secondly, we transformed ECG signals into images and the CNN2D model achieved an accuracy of 92.16%.

Recent work has explored the capability of large language models (LLMs) to identify and correct errors in LLM-generated responses. These refinement approaches frequently evaluate what sizes of models are able to do refinement for what problems, but less attention is paid to what effective feedback for refinement looks like. In this work, we propose looking at refinement with feedback as a composition of three distinct LLM competencies: (1) identification of bad generations; (2) fine-grained natural language feedback generation; (3) refining with fine-grained feedback. The first step can be implemented with a high-performing discriminative model and steps 2 and 3 can be implemented either via prompted or fine-tuned LLMs. A key property of this approach is that the step 2 critique model can give fine-grained feedback about errors, made possible by offloading the discrimination to a separate model in step 1. We show that models of different capabilities benefit from refining with this approach on the task of improving factual consistency of document grounded summaries. Overall, our proposed method consistently outperforms existing end-to-end refinement approaches and current trained models not fine-tuned for factuality critiquing.

Continual relation extraction (CRE) aims to solve the problem of catastrophic forgetting when learning a sequence of newly emerging relations. Recent CRE studies have found that catastrophic forgetting arises from the model's lack of robustness against future analogous relations. To address the issue, we introduce rationale, i.e., the explanations of relation classification results generated by large language models (LLM), into CRE task. Specifically, we design the multi-task rationale tuning strategy to help the model learn current relations robustly. We also conduct contrastive rationale replay to further distinguish analogous relations. Experimental results on two standard benchmarks demonstrate that our method outperforms the state-of-the-art CRE models.

The modern foundation of the free speech clause of the First Amendment is the concept of the marketplace of ideas. The notion comes from John Stuart Mill who first drew the analogy to a market where ideas compete freely with one another and people form their own judgments. The analogy was first noted in Justice Oliver Wendell Holmes' famous dissent in Abrams v. United States (1919) when he wrote, "The best test of truth is the power of the thought to get itself accepted in the competition of the market." This free and open market of ideas is considered vital to the function and preservation of democracy. As Holmes wrote in another famous dissent in United States v. Schwimmer (1929), "If there is any principle of the Constitution that more imperatively calls for attachment than any other, it is the principle of free thought--not free thought for those who agree with us freedom for the thought we hate." Until recently, the Supreme Court had not cared much where those thoughts might come from, or whether their source must be human.

Russian President Vladimir Putin has ordered officials to tighten control of the border with Ukraine after a spate of drone attacks that Russian authorities blamed on Kyiv delivered a new challenge to Moscow a year after its full-scale invasion of its neighbour. One drone crashed on Tuesday just 100km (60 miles) southeast of Moscow in an alarming development for Russian defences. While Putin didn't refer to any specific attacks in a speech in the Russian capital, he stepped up border controls hours after drone attacks targeted several areas in southern and western Russia and authorities closed the airspace over St Petersburg in response to what some reports said was a drone. Also on Tuesday, several Russian television stations aired a missile attack warning that officials blamed on hacking. The drone attacks caused no casualties but provoked a security stir after the war in Ukraine marked its first anniversary last week.

An eerily human-like robot cashier who can serve drinks and chit-chat with customers will soon be up and running in Dubai, meaning baristas could become a thing of the past. Donna Cyber-Cafe is set to open in Dubai next year, with a'supermodel' robot serving coffees and ice creams to customers, without the help of any humans. Donna, who has been created to be the spitting image of Eastern European model Diana Gabdullina, will offer speedy service and will even be able to start conversations with customers, take selfies or tell a fairy tell for those who ask. The impressive new droid has been created to appear like a real person, allowing Donna to read customer's emotions and move in an eerily realistic way. Donna Cyber-Cafe will be opening in Dubai next year.

This document provides extensive guidelines and examples for Rhetorical Structure Theory (RST) annotation in Mandarin Chinese. The guideline is divided into three sections. We first introduce preprocessing steps to prepare data for RST annotation. Secondly, we discuss syntactic criteria to segment texts into Elementary Discourse Units (EDUs). Lastly, we provide examples to define and distinguish discourse relations in different genres. We hope that this reference manual can facilitate RST annotations in Chinese and accelerate the development of the RST framework across languages.

A chatbot is software that simulates human-like conversations with users via text messages on chat. Its key task is to help users by providing answers to their questions. If we dive deep, chatbots are pieces of conversational software powered by artificial intelligence that have the capability to engage in one-to-one chat with customers on their preferred chat platform such as Facebook Messenger, Whatsapp, Instagram, Telegram, Slack and many more conversational platforms. Chatbots, run by pre-programmed algorithms, natural language processing and/or machine learning and conversed in ways that mimicked human communication. Unlike other automated customer service solutions such as IVRS systems that were universally disliked for their robotic nature, Chatbots are seen to get closer to passing the Turing Test convincingly simulating a human conversational partner so well that it was difficult to sense one was chatting with a machine. British Al pioneer Alan Turing in 1950 proposed a test to determine whether machines could think. According to the Turing test, a computer could demonstrate intelligence if a human interviewer, conversing with an unseen human and an unseen computer, could not tell which was which. Although much work has been done in many of the subgroups that fall under the Al umbrella, critics believe that no computer can truly pass the Turing test.