Continual generalized category discovery has been introduced and studied in the literature as a method that aims to continuously discover and learn novel categories in incoming data batches while avoiding catastrophic forgetting of previously learned categories. A key component in addressing this challenge is the model's ability to separate novel samples, where Extreme Value Theory (EVT) has been effectively employed. In this work, we propose a novel method that integrates EVT with proxy anchors to define boundaries around proxies using a probability of inclusion function, enabling the rejection of unknown samples. Additionally, we introduce a novel EVT-based loss function to enhance the learned representation, achieving superior performance compared to other deep-metric learning methods in similar settings. Using the derived probability functions, novel samples are effectively separated from previously known categories. However, category discovery within these novel samples can sometimes overestimate the number of new categories. To mitigate this issue, we propose a novel EVT-based approach to reduce the model size and discard redundant proxies. We also incorporate experience replay and knowledge distillation mechanisms during the continual learning stage to prevent catastrophic forgetting. Experimental results demonstrate that our proposed approach outperforms state-of-the-art methods in continual generalized category discovery scenarios.

Automated scoring (AS) systems are increasingly used for evaluating L2 writing, but require ongoing refinement for construct validity. While prior work suggested lexical bundles (LBs) - recurrent multi-word sequences satisfying certain frequency criteria - could inform assessment, their empirical integration into AS models needs further investigation. This study tested the impact of incorporating LB frequency features into an AS model for TOEFL independent writing tasks. Analyzing a sampled subcorpus (N=1,225 essays, 9 L1s) from the TOEFL11 corpus, scored by ETS-trained raters (Low, Medium, High), 3- to 9-word LBs were extracted, distinguishing prompt-specific from non-prompt types. A baseline Support Vector Machine (SVM) scoring model using established linguistic features (e.g., mechanics, cohesion, sophistication) was compared against an extended model including three aggregate LB frequency features (total prompt, total non-prompt, overall total). Results revealed significant, though generally small-effect, relationships between LB frequency (especially non-prompt bundles) and proficiency (p < .05). Mean frequencies suggested lower proficiency essays used more LBs overall. Critically, the LB-enhanced model improved agreement with human raters (Quadratic Cohen's Kappa +2.05%, overall Cohen's Kappa +5.63%), with notable gains for low (+10.1% exact agreement) and medium (+14.3% Cohen's Kappa) proficiency essays. These findings demonstrate that integrating aggregate LB frequency offers potential for developing more linguistically informed and accurate AS systems, particularly for differentiating developing L2 writers.

The field of spoken language processing is undergoing a shift from training custom-built, task-specific models toward using and optimizing spoken language models (SLMs) which act as universal speech processing systems. This trend is similar to the progression toward universal language models that has taken place in the field of (text) natural language processing. SLMs include both "pure" language models of speech -- models of the distribution of tokenized speech sequences -- and models that combine speech encoders with text language models, often including both spoken and written input or output. Work in this area is very diverse, with a range of terminology and evaluation settings. This paper aims to contribute an improved understanding of SLMs via a unifying literature survey of recent work in the context of the evolution of the field. Our survey categorizes the work in this area by model architecture, training, and evaluation choices, and describes some key challenges and directions for future work.

Traditionally, authorship attribution (AA) tasks relied on statistical data analysis and classification based on stylistic features extracted from texts. In recent years, pre-trained language models (PLMs) have attracted significant attention in text classification tasks. However, although they demonstrate excellent performance on large-scale short-text datasets, their effectiveness remains under-explored for small samples, particularly in AA tasks. Additionally, a key challenge is how to effectively leverage PLMs in conjunction with traditional feature-based methods to advance AA research. In this study, we aimed to significantly improve performance using an integrated integrative ensemble of traditional feature-based and modern PLM-based methods on an AA task in a small sample. For the experiment, we used two corpora of literary works to classify 10 authors each. The results indicate that BERT is effective, even for small-sample AA tasks. Both BERT-based and classifier ensembles outperformed their respective stand-alone models, and the integrated ensemble approach further improved the scores significantly. For the corpus that was not included in the pre-training data, the integrated ensemble improved the F1 score by approximately 14 points, compared to the best-performing single model. Our methodology provides a viable solution for the efficient use of the ever-expanding array of data processing tools in the foreseeable future.

Signal separation in the passive underwater acoustic domain has heavily relied on deep learning techniques to isolate ship radiated noise. However, the separation networks commonly used in this domain stem from speech separation applications and may not fully consider the unique aspects of underwater acoustics beforehand, such as the influence of different propagation media, signal frequencies and modulation characteristics. This oversight highlights the need for tailored approaches that account for the specific characteristics of underwater sound propagation. This study introduces a novel temporal network designed to separate ship radiated noise by employing a dual-path model and a feature decoupling approach. The mixed signals' features are transformed into a space where they exhibit greater independence, with each dimension's significance decoupled. Subsequently, a fusion of local and global attention mechanisms is employed in the separation layer. Extensive comparisons showcase the effectiveness of this method when compared to other prevalent network models, as evidenced by its performance in the ShipsEar and DeepShip datasets.

--The increasing availability of multimodal data across text, tables, and images presents new challenges for developing models capable of complex cross-modal reasoning. Existing methods for Multimodal Multi-hop Question Answering (MMQA) often suffer from limited reasoning capabilities, reliance on modality conversion (e.g., image-to-text), and inadequate alignment between visual and textual representations. T o address these limitations, this paper introduces Vision-Language Multimodal Transformer (VLMT), a unified architecture that integrates a transformer-based vision encoder with a sequence-to-sequence language model. VLMT employs a direct token-level injection mechanism to fuse visual and textual inputs within a shared embedding space, eliminating the need for intermediate projection layers. T o enhance cross-modal alignment and reasoning, a three-stage pretraining strategy is proposed to progressively align vision-language representations and improve the model's capacity for multimodal understanding. Based on the pretrained backbone, two task-specific modules are instantiated to form a two-stage MMQA framework: a multimodal reranker that predicts document relevance scores and utilizes a relative threshold with top-k strategy for context retrieval, and a mul-timodal question answering model that generates contextually grounded answers based on the retrieved evidence. Comprehensive experiments on two benchmark datasets demonstrate the effectiveness of the proposed approach. These results highlight VLMT's strong capabilities in multimodal reasoning and its potential to advance real-world information retrieval and question answering systems. The exponential growth of information in today's digital ecosystem has led to the proliferation of multimodal data--comprising text, tables, and images--across a wide range of platforms. Qi Zhi Lim is with the Faculty of Information Science and Technology, Multimedia University, Jalan A yer Keroh Lama, 75450 Melaka, Malaysia (email: 1181103589@student.mmu.edu.my). Chin Poo Lee is with the School of Computer Science, University of Nottingham Ningbo China, 199 Taikang East Road, Yinzhou District, Ningbo, Zhejiang Province, 315100, China (e-mail: leechinpoo@outlook.com). Kian Ming Lim is with the School of Computer Science, University of Nottingham Ningbo China, 199 Taikang East Road, Yinzhou District, Ningbo, Zhejiang Province, 315100, China (e-mail: Kian-Ming.Lim@nottingham.edu.cn). Multimodal Multi-hop Question Answering (MMQA) [1], [2] has emerged as a representative task in this domain, reflecting real-world information-seeking behavior where relevant evidence is scattered across multiple sources and modalities. MMQA requires models to perform two interdependent operations: retrieving relevant multimodal context and reasoning over the retrieved information to produce accurate and coherent answers. Early solutions to MMQA have largely followed modular paradigms.

Spectral Normalization for Lipschitz-Constrained Policies on Learning Humanoid Locomotion Jaeyong Shin 1, Woohyun Cha 1, Donghyeon Kim 1, 2, Junhyeok Cha 1, and Jaeheung Park 3 Abstract -- Reinforcement learning (RL) has shown great potential in training agile and adaptable controllers for legged robots, enabling them to learn complex locomotion behaviors directly from experience. However, policies trained in simulation often fail to transfer to real-world robots due to unrealistic assumptions such as infinite actuator bandwidth and the absence of torque limits. These conditions allow policies to rely on abrupt, high-frequency torque changes, which are infeasible for real actuators with finite bandwidth. Traditional methods address this issue by penalizing aggressive motions through regularization rewards, such as joint velocities, accelerations, and energy consumption, but they require extensive hyperparameter tuning. Alternatively, Lipschitz-Constrained Policies (LCP) enforce finite bandwidth action control by penalizing policy gradients, but their reliance on gradient calculations introduces significant GPU memory overhead. T o overcome this limitation, this work proposes Spectral Normalization (SN) as an efficient replacement for enforcing Lipschitz continuity. By constraining the spectral norm of network weights, SN effectively limits high-frequency policy fluctuations while significantly reducing GPU memory usage. Experimental evaluations in both simulation and real-world humanoid robot show that SN achieves performance comparable to gradient penalty methods while enabling more efficient parallel training. I. INTRODUCTION Reinforcement learning (RL) has emerged as a powerful framework for developing locomotion policies, leading to significant advancements in legged robots.

How Good Are Large Language Models for Course Recommendation in MOOCs? Shin'ichi Konomi Kyushu University, Japan konomi@artsci.kyushu-u.ac.jp ABSTRACT Large Language Models (LLMs) have made significant strides in natural language processing and are increasingly being integrated into recommendation systems. However, their potential in educational recommendation systems has yet to be fully explored. This paper investigates the use of LLMs as a general-purpose recommendation model, leveraging their vast knowledge derived from large-scale corpora for course recommendation tasks. We explore a variety of approaches, ranging from prompt-based methods to more advanced fine-tuning techniques, and compare their performance against traditional recommendation models. Extensive experiments were conducted on a real-world MOOC dataset, evaluating using LLMs as course recommendation systems across key dimensions such as accuracy, diversity, and novelty. Our results demonstrate that LLMs can achieve good performance comparable to traditional models, highlighting their potential to enhance educational recommendation systems.

Children can acquire language from less than 100 million words of input. Large language models are far less data-efficient: they typically require 3 or 4 orders of magnitude more data and still do not perform as well as humans on many evaluations. These intensive resource demands limit the ability of researchers to train new models and use existing models as developmentally plausible cognitive models. The BabyLM Challenge is a communal effort in which participants compete to optimize language model training on a fixed data budget. Submissions are compared on various evaluation tasks targeting grammatical ability, downstream task performance, and generalization. Participants can submit to up to three tracks with progressively looser data restrictions. From over 30 submissions, we extract concrete recommendations on how best to train data-efficient language models, and on where future efforts should (and perhaps should not) focus. The winning submissions using the LTG-BERT architecture (Samuel et al., 2023) outperformed models trained on trillions of words. Other submissions achieved strong results through training on shorter input sequences or training a student model on a pretrained teacher. Curriculum learning attempts, which accounted for a large number of submissions, were largely unsuccessful, though some showed modest improvements.

--Facial expressiveness plays a crucial role in a robot's ability to engage and interact with children. Prior research has shown that expressive robots can enhance child engagement during human-robot interactions. However, many robots used in therapy settings feature non-personalized, static faces designed with traditional facial feature considerations, which can limit the depth of interactions and emotional connections. Digital faces offer opportunities for personalization, yet the current landscape of robot face design lacks a dynamic, user-centered approach. Specifically, there is a significant research gap in designing robot faces based on child preferences. Instead, most robots in child-focused therapy spaces are developed from an adult-centric perspective. We present a novel study investigating the influence of child-drawn digital faces in child-robot interactions. This approach focuses on a design activity with children instructed to draw their own custom robot faces. We compare the perceptions of social intelligence (PSI) of two implementations: a generic digital face and a robot face, personalized using the user's drawn robot faces. The results of this study show the perceived social intelligence of a child-drawn robot was significantly higher compared to a generic face.