The large language model (LLM) community focuses almost exclusively on decoder-only language models, since they are easier to use for text generation. However, a large subset of the community still uses encoder-only models for tasks such as classification or retrieval. Previous work has attempted to compare these architectures, but is forced to make comparisons with models that have different numbers of parameters, training techniques, and datasets. We introduce the SOTA open-data Ettin suite of models: paired encoder-only and decoder-only models ranging from 17 million parameters to 1 billion, trained on up to 2 trillion tokens. Using the same recipe for both encoder-only and decoder-only models produces SOTA recipes in both categories for their respective sizes, beating ModernBERT as an encoder and Llama 3.2 and SmolLM2 as decoders. Like previous work, we find that encoder-only models excel at classification and retrieval tasks while decoders excel at generative tasks. However, we show that adapting a decoder model to encoder tasks (and vice versa) through continued training is subpar compared to using only the reverse objective (i.e. a 400M encoder outperforms a 1B decoder on MNLI, and vice versa for generative tasks). We open-source all artifacts of this study including training data, training order segmented by checkpoint, and 200+ checkpoints to allow future work to analyze or extend all aspects of training.

Learning new information without forgetting prior knowledge is central to human intelligence. In contrast, neural network models suffer from catastrophic forgetting: a significant degradation in performance on previously learned tasks when acquiring new information. The Complementary Learning Systems (CLS) theory offers an explanation for this human ability, proposing that the brain has distinct systems for pattern separation (encoding distinct memories) and pattern completion (retrieving complete memories from partial cues). To capture these complementary functions, we leverage the representational generalization capabilities of variational autoencoders (VAEs) and the robust memory storage properties of Modern Hopfield networks (MHNs), combining them into a neurally plausible continual learning model. We evaluate this model on the Split-MNIST task, a popular continual learning benchmark, and achieve close to state-of-the-art accuracy (~90%), substantially reducing forgetting. Representational analyses empirically confirm the functional dissociation: the VAE underwrites pattern completion, while the MHN drives pattern separation. By capturing pattern separation and completion in scalable architectures, our work provides a functional template for modeling memory consolidation, generalization, and continual learning in both biological and artificial systems.

Previous literature has largely shown that Large Language Models (LLMs) perpetuate social biases learnt from their pre-training data. Given the notable lack of resources for social bias evaluation in languages other than English, and for social contexts outside of the United States, this paper introduces the Spanish and the Catalan Bias Benchmarks for Question Answering (EsBBQ and CaBBQ). Based on the original BBQ, these two parallel datasets are designed to assess social bias across 10 categories using a multiple-choice QA setting, now adapted to the Spanish and Catalan languages and to the social context of Spain. We report evaluation results on different LLMs, factoring in model family, size and variant. Our results show that models tend to fail to choose the correct answer in ambiguous scenarios, and that high QA accuracy often correlates with greater reliance on social biases.

Large Language Models (LLMs) enable new possibilities for qualitative research at scale, including coding and data annotation. While multi-agent systems (MAS) can emulate human coding workflows, their benefits over single-agent coding remain poorly understood. We conducted an experimental study of how agent persona and temperature shape consensus-building and coding accuracy of dialog segments based on a codebook with 8 codes. Our open-source MAS mirrors deductive human coding through structured agent discussion and consensus arbitration. Using six open-source LLMs (with 3 to 32 billion parameters) and 18 experimental configurations, we analyze over 77,000 coding decisions against a gold-standard dataset of human-annotated transcripts from online math tutoring sessions. Temperature significantly impacted whether and when consensus was reached across all six LLMs. MAS with multiple personas (including neutral, assertive, or empathetic), significantly delayed consensus in four out of six LLMs compared to uniform personas. In three of those LLMs, higher temperatures significantly diminished the effects of multiple personas on consensus. However, neither temperature nor persona pairing lead to robust improvements in coding accuracy. Single agents matched or outperformed MAS consensus in most conditions. Only one model (OpenHermesV2:7B) and code category showed above-chance gains from MAS deliberation when temperature was 0.5 or lower and especially when the agents included at least one assertive persona. Qualitative analysis of MAS collaboration for these configurations suggests that MAS may nonetheless aid in narrowing ambiguous code applications that could improve codebooks and human-AI coding. We contribute new insight into the limits of LLM-based qualitative methods, challenging the notion that diverse MAS personas lead to better outcomes. We open-source our MAS and experimentation code.

We present a robust ensemble-based system for multilingual multimodal reasoning, designed for the ImageCLEF 2025 EXAMS V challenge. Our approach integrates Gemini 2.5 Flash for visual description, Gemini 1.5 Pro for caption refinement and consistency checks, and Gemini 2.5 Pro as a reasoner which handles final answer selection, all coordinated through carefully engineered few-shot and zero-shot prompts. We conducted an extensive ablation study, training several large language models (Gemini 2.5 Flash, Phi 4, Gemma 3, Mistral) on an English dataset and its multilingual augmented version. Additionally, we evaluated Gemini 2.5 Flash in a zero-shot setting for comparison and found it to substantially outperform the trained models. Prompt design also proved critical: enforcing concise, language-normalized formats and prohibiting explanatory text boosted model accuracy on the English validation set from 55.9% to 61.7%. On the official leaderboard, our system (Team MSA) achieved first place overall in the multilingual track with 81.4% accuracy, and led 11 out of 13 individual language tracks, with top results such as 95.07% for Croatian and 92.12% for Italian. These findings highlight that lightweight OCR-VLM ensembles, when paired with precise prompt strategies and cross-lingual augmentation, can outperform heavier end-to-end models in high-stakes, multilingual educational settings.

We introduce ExRec, a general framework for personalized exercise recommendation with semantically-grounded knowledge tracing. Our method builds on the observation that existing exercise recommendation approaches simulate student performance via knowledge tracing (KT) but they often overlook two key aspects: (a) the semantic content of questions and (b) the sequential, structured progression of student learning. To address this, our ExRec presents an end-to-end pipeline, from annotating the KCs of questions and learning their semantic representations to training KT models and optimizing several reinforcement learning (RL) methods. Moreover, we improve standard Q-learning-based continuous RL methods via a tailored model-based value estimation (MVE) approach that directly leverages the components of KT model in estimating cumulative knowledge improvement. We validate the effectiveness of our ExRec using various RL methods across four real-world tasks with different educational goals in online math learning. We further show that ExRec generalizes robustly to new, unseen questions and that it produces interpretable student learning trajectories. Together, our findings highlight the promise of KT-guided RL for effective personalization in education.

ABSTRACT Large language models, with their strong reasoning ability and rich knowledge, have brought revolution to many tasks of AI, but their impact on sign language generation remains limited due to its complexity and unique rules. In this paper, we propose TEAch Me Sign (TEAM-Sign), treating sign language as another natural language. By fine-tuning an LLM, we enable it to learn the correspondence between text and sign language, and facilitate generation. Considering the differences between sign and spoken language, we employ a stepwise prompting strategy to extract the inherent sign language knowledge within the LLM, thereby supporting the learning and generation process. Experimental results on How2Sign and Phoenix14T datasets demonstrate that our approach effectively leverages both the sign language knowledge and reasoning capabilities of LLM to align the different distribution and grammatical rules between sign and spoken language.

--Conversational agents, such as chatbots and virtual assistants, have become essential in software development, boosting productivity, collaboration, and automating various tasks. This paper examines the role of adaptive AI-powered conversational agents in software development, highlighting their ability to offer dynamic, context-aware assistance to developers. Unlike traditional rule-based systems, adaptive AI agents use machine learning and natural language processing to learn from interactions and improve over time, providing more personalized and responsive help. We look at how these tools have evolved from simple query-based systems to advanced AI-driven solutions like GitHub Copilot and Microsoft T eams bots. We also explore the challenges of integrating adaptive AI into software development processes. The study aims to assess the benefits and limitations of these systems, address concerns like data privacy and ethical issues, and offer insights into their future use in the field. Ultimately, adaptive AI chatbots have great potential to revolutionize software development by delivering real-time, customized support and enhancing the efficiency of development cycles. Conversational agents (CAs), including chatbots, dialogue systems, and virtual assistants, are software-based systems designed to process natural language and simulate intelligent dialogue with users [1].

This paper introduces MISS-QA, the first benchmark specifically designed to evaluate the ability of models to interpret schematic diagrams within scientific literature. MISS-QA comprises 1,500 expert-annotated examples over 465 scientific papers. In this benchmark, models are tasked with interpreting schematic diagrams that illustrate research overviews and answering corresponding information-seeking questions based on the broader context of the paper. We assess the performance of 18 frontier multimodal foundation models, including o4-mini, Gemini-2.5-Flash, and Qwen2.5-VL. We reveal a significant performance gap between these models and human experts on MISS-QA. Our analysis of model performance on unanswerable questions and our detailed error analysis further highlight the strengths and limitations of current models, offering key insights to enhance models in comprehending multimodal scientific literature.

Large Language Models (LLMs) have emerged as a promising paradigm for time series analytics, leveraging their massive parameters and the shared sequential nature of textual and time series data. However, a cross-modality gap exists between time series and textual data, as LLMs are pre-trained on textual corpora and are not inherently optimized for time series. In this tutorial, we provide an up-to-date overview of LLM-based cross-modal time series analytics. We introduce a taxonomy that classifies existing approaches into three groups based on cross-modal modeling strategies, e.g., conversion, alignment, and fusion, and then discuss their applications across a range of downstream tasks. In addition, we summarize several open challenges. This tutorial aims to expand the practical application of LLMs in solving real-world problems in cross-modal time series analytics while balancing effectiveness and efficiency. Participants will gain a thorough understanding of current advancements, methodologies, and future research directions in cross-modal time series analytics.