Speech event detection is crucial for multimedia retrieval, involving the tagging of both semantic and acoustic events. Traditional ASR systems often overlook the interplay between these events, focusing solely on content, even though the interpretation of dialogue can vary with environmental context. This paper tackles two primary challenges in speech event detection: the continual integration of new events without forgetting previous ones, and the disentanglement of semantic from acoustic events. We introduce a new task, continual event detection from speech, for which we also provide two benchmark datasets. To address the challenges of catastrophic forgetting and effective disentanglement, Figure 1: In continual learning, learners incrementally acquire we propose a novel method, 'Double Mixture.' This method merges new event types and must evaluate all previously speech expertise with robust memory mechanisms to enhance learned types during testing. This process is particularly adaptability and prevent forgetting. Our comprehensive experiments challenging in speech-based scenarios due to the complex interplay show that this task presents significant challenges that are of semantic content (semantic event) and background not effectively addressed by current state-of-the-art methods in either sounds (acoustic event).

This work explores a novel data augmentation method based on Large Language Models (LLMs) for predicting item difficulty and response time of retired USMLE Multiple-Choice Questions (MCQs) in the BEA 2024 Shared Task. Our approach is based on augmenting the dataset with answers from zero-shot LLMs (Falcon, Meditron, Mistral) and employing transformer-based models based on six alternative feature combinations. The results suggest that predicting the difficulty of questions is more challenging. Notably, our top performing methods consistently include the question text, and benefit from the variability of LLM answers, highlighting the potential of LLMs for improving automated assessment in medical licensing exams. We make our code available https://github.com/ana-rogoz/BEA-2024.

Negative flips are errors introduced in a classification system when a legacy model is updated. Existing methods to reduce the negative flip rate (NFR) either do so at the expense of overall accuracy by forcing a new model to imitate the old models, or use ensembles, which multiply inference cost prohibitively. We analyze the role of ensembles in reducing NFR and observe that they remove negative flips that are typically not close to the decision boundary, but often exhibit large deviations in the distance among their logits. Based on the observation, we present a method, called Ensemble Logit Difference Inhibition (ELODI), to train a classification system that achieves paragon performance in both error rate and NFR, at the inference cost of a single model. The method distills a homogeneous ensemble to a single student model which is used to update the classification system. ELODI also introduces a generalized distillation objective, Logit Difference Inhibition (LDI), which only penalizes the logit difference of a subset of classes with the highest logit values. On multiple image classification benchmarks, model updates with ELODI demonstrate superior accuracy retention and NFR reduction.

This paper examines the impact of Generative Artificial Intelligence (GenAI) tools like ChatGPT on the creation and consumption of terminological definitions. From the terminologist's point of view, the strategic use of GenAI tools can streamline the process of crafting definitions, reducing both time and effort, while potentially enhancing quality. GenAI tools enable AI-assisted terminography, notably post-editing terminography, where the machine produces a definition that the terminologist then corrects or refines. However, the potential of GenAI tools to fulfill all the terminological needs of a user, including term definitions, challenges the very existence of terminological definitions and resources as we know them. Unlike terminological definitions, GenAI tools can describe the knowledge activated by a term in a specific context. However, a main drawback of these tools is that their output can contain errors. For this reason, users requiring reliability will likely still resort to terminological resources for definitions. Nevertheless, with the inevitable integration of AI into terminology work, the distinction between human-created and AI-created content will become increasingly blurred.

The paper presents a novel cloud-based digital twin learning platform for teaching and training concepts of cognitive robotics. Instead of forcing interested learners or students to install a new operating system and bulky, fragile software onto their personal laptops just to solve tutorials or coding assignments of a single lecture on robotics, it would be beneficial to avoid technical setups and directly dive into the content of cognitive robotics. To achieve this, the authors utilize containerization technologies and Kubernetes to deploy and operate containerized applications, including robotics simulation environments and software collections based on the Robot operating System (ROS). The web-based Integrated Development Environment JupyterLab is integrated with RvizWeb and XPRA to provide real-time visualization of sensor data and robot behavior in a user-friendly environment for interacting with robotics software. The paper also discusses the application of the platform in teaching Knowledge Representation, Reasoning, Acquisition and Retrieval, and Task-Executives. The authors conclude that the proposed platform is a valuable tool for education and research in cognitive robotics, and that it has the potential to democratize access to these fields. The platform has already been successfully employed in various academic courses, demonstrating its effectiveness in fostering knowledge and skill development.

Exploration efficiency poses a significant challenge in goal-conditioned reinforcement learning (GCRL) tasks, particularly those with long horizons and sparse rewards. A primary limitation to exploration efficiency is the agent's inability to leverage environmental structural patterns. In this study, we introduce a novel framework, GEASD, designed to capture these patterns through an adaptive skill distribution during the learning process. This distribution optimizes the local entropy of achieved goals within a contextual horizon, enhancing goal-spreading behaviors and facilitating deep exploration in states containing familiar structural patterns. Our experiments reveal marked improvements in exploration efficiency using the adaptive skill distribution compared to a uniform skill distribution. Additionally, the learned skill distribution demonstrates robust generalization capabilities, achieving substantial exploration progress in unseen tasks containing similar local structures.

Recent research on proactive conversational agents (PCAs) mainly focuses on improving the system's capabilities in anticipating and planning action sequences to accomplish tasks and achieve goals before users articulate their requests. This perspectives paper highlights the importance of moving towards building human-centered PCAs that emphasize human needs and expectations, and that considers ethical and social implications of these agents, rather than solely focusing on technological capabilities. The distinction between a proactive and a reactive system lies in the proactive system's initiative-taking nature. Without thoughtful design, proactive systems risk being perceived as intrusive by human users. We address the issue by establishing a new taxonomy concerning three key dimensions of human-centered PCAs, namely Intelligence, Adaptivity, and Civility. We discuss potential research opportunities and challenges based on this new taxonomy upon the five stages of PCA system construction. This perspectives paper lays a foundation for the emerging area of conversational information retrieval research and paves the way towards advancing human-centered proactive conversational systems.

We propose Iterative Facuality Refining on Informative Scientific Question-Answering (ISQA) feedback\footnote{Code is available at \url{https://github.com/lizekai-richard/isqa}}, a method following human learning theories that employs model-generated feedback consisting of both positive and negative information. Through iterative refining of summaries, it probes for the underlying rationale of statements to enhance the factuality of scientific summarization. ISQA does this in a fine-grained manner by asking a summarization agent to reinforce validated statements in positive feedback and fix incorrect ones in negative feedback. Our findings demonstrate that the ISQA feedback mechanism significantly improves the factuality of various open-source LLMs on the summarization task, as evaluated across multiple scientific datasets.

Higher education plays a critical role in driving an innovative economy by equipping students with knowledge and skills demanded by the workforce. While researchers and practitioners have developed data systems to track detailed occupational skills, such as those established by the U.S. Department of Labor (DOL), much less effort has been made to document skill development in higher education at a similar granularity. Here, we fill this gap by presenting a longitudinal dataset of skills inferred from over three million course syllabi taught at nearly three thousand U.S. higher education institutions. To construct this dataset, we apply natural language processing to extract from course descriptions detailed workplace activities (DWAs) used by the DOL to describe occupations. We then aggregate these DWAs to create skill profiles for institutions and academic majors. Our dataset offers a large-scale representation of college-educated workers and their role in the economy. To showcase the utility of this dataset, we use it to 1) compare the similarity of skills taught and skills in the workforce according to the US Bureau of Labor Statistics, 2) estimate gender differences in acquired skills based on enrollment data, 3) depict temporal trends in the skills taught in social science curricula, and 4) connect college majors' skill distinctiveness to salary differences of graduates. Overall, this dataset can enable new research on the source of skills in the context of workforce development and provide actionable insights for shaping the future of higher education to meet evolving labor demands especially in the face of new technologies.

We introduce CHARM, the first benchmark for comprehensively and in-depth evaluating the commonsense reasoning ability of large language models (LLMs) in Chinese, which covers both globally known and Chinese-specific commonsense. We evaluated 7 English and 12 Chinese-oriented LLMs on CHARM, employing 5 representative prompt strategies for improving LLMs' reasoning ability, such as Chain-of-Thought. Our findings indicate that the LLM's language orientation and the task's domain influence the effectiveness of the prompt strategy, which enriches previous research findings. We built closely-interconnected reasoning and memorization tasks, and found that some LLMs struggle with memorizing Chinese commonsense, affecting their reasoning ability, while others show differences in reasoning despite similar memorization performance. We also evaluated the LLMs' memorization-independent reasoning abilities and analyzed the typical errors. Our study precisely identified the LLMs' strengths and weaknesses, providing the clear direction for optimization. It can also serve as a reference for studies in other fields. We will release CHARM at https://github.com/opendatalab/CHARM .