Mathematical error detection in educational settings presents a significant challenge for Multimodal Large Language Models (MLLMs), requiring a sophisticated understanding of both visual and textual mathematical content along with complex reasoning capabilities. Though effective in mathematical problem-solving, MLLMs often struggle with the nuanced task of identifying and categorizing student errors in multimodal mathematical contexts. Therefore, we introduce MathAgent, a novel Mixture-of-Math-Agent framework designed specifically to address these challenges. Our approach decomposes error detection into three phases, each handled by a specialized agent: an image-text consistency validator, a visual semantic interpreter, and an integrative error analyzer. This architecture enables more accurate processing of mathematical content by explicitly modeling relationships between multimodal problems and student solution steps. We evaluate MathAgent on real-world educational data, demonstrating approximately 5% higher accuracy in error step identification and 3% improvement in error categorization compared to baseline models. Besides, MathAgent has been successfully deployed in an educational platform that has served over one million K-12 students, achieving nearly 90% student satisfaction while generating significant cost savings by reducing manual error detection.

Personalization in large language models (LLMs) seeks to tailor models to individual user or user group preferences. Prompt-based methods augment queries with user preference information, whereas training-based methods directly encode preferences into model parameters for more effective personalization. Despite achieving some success in personalizing LLMs, prior methods often fail to directly optimize task-specific metrics and lack explicit privacy-preservation mechanisms. To address these limitations, we propose Privacy-Preserving Model Merging via Evolutionary Algorithms (PriME), a novel approach to personalization that employs gradient-free methods to directly optimize task-specific metrics while preserving user privacy. By incorporating privacy preservation into optimization, PriME produces a personalized module that effectively captures the target user's preferences while minimizing the privacy risks for the users sharing their private information. Experiments on the LaMP benchmark show that PriME outperforms both prompt-based and training-based methods, achieving up to a 45% performance improvement over the prior art. Further analysis shows that PriME achieves a significantly better privacy-utility trade-off, highlighting the potential of evolutionary approaches for privacy-preserving LLM personalization.

Transfer learning leverages the abundance of English data to address the scarcity of resources in modeling non-English languages, such as Korean. In this study, we explore the potential of Phrase Aligned Data (PAD) from standardized Statistical Machine Translation (SMT) to enhance the efficiency of transfer learning. Through extensive experiments, we demonstrate that PAD synergizes effectively with the syntactic characteristics of the Korean language, mitigating the weaknesses of SMT and significantly improving model performance. Moreover, we reveal that PAD complements traditional data construction methods and enhances their effectiveness when combined. This innovative approach not only boosts model performance but also suggests a cost-efficient solution for resource-scarce languages.

Large Language Models (LLMs) have demonstrated impressive capabilities on a broad array of NLP tasks, but their tendency to produce hallucinations$\unicode{x2013}$plausible-sounding but factually incorrect content$\unicode{x2013}$poses severe challenges in high-stakes domains. Existing hallucination detection methods either bear the computational cost of multiple inference passes or sacrifice accuracy for efficiency with single-pass approaches, neither of which is ideal in resource-constrained environments such as edge devices. We propose the Shannon Entropy Distribution Hallucination Detector (ShED-HD), a novel hallucination detection framework that bridges this gap by classifying sequence-level entropy patterns using a lightweight BiLSTM architecture with single-headed attention. In contrast to prior approaches, ShED-HD efficiently detects distinctive uncertainty patterns across entire output sequences, preserving contextual awareness. Through in-depth evaluation on three datasets (BioASQ, TriviaQA, and Jeopardy Questions), we show that ShED-HD significantly outperforms other computationally efficient approaches in the out-of-distribution setting, while achieving comparable performance in the in-distribution setting. ShED-HD facilitates hallucination detection that is low-cost, accurate, and generalizable, improving the credibility of content generated by LLMs in resource-constrained environments where trustworthy AI functionality is crucial.

Large Language Models (LLMs) have demonstrated exceptional performance on a range of downstream NLP tasks by generating text that closely resembles human writing. However, the ease of achieving this similarity raises concerns from potential malicious uses at scale by bad actors, as LLM-generated text becomes increasingly difficult to discern from human text. Although detection methods have been developed to address this issue, bad actors can further manipulate LLM-generated texts to make them less detectable. In this work, we study how further editing texts with Reinforcement Learning from Human Feedback (RLHF), which aligns model outputs with human preferences, affects (a) the quality of generated texts for two tasks, and (b) the performance of LLM-generated text detectors, looking at both training-based and zero-shot detection methods. Although RLHF improves the quality of LLM-generated texts, we find that it also tends to produce more detectable, lengthy, and repetitive outputs. Additionally, we observe that training-based detectors are vulnerable to short texts and to texts that incorporate code, whereas zero-shot detectors exhibit greater robustness.

Reinforcement learning from human feedback (RLHF) is an effective method for aligning large language models (LLMs) with human values. However, reward over-optimization remains an open challenge leading to discrepancies between the performance of LLMs under the reward model and the true human objectives. A primary contributor to reward over-optimization is the extrapolation error that arises when the reward model evaluates out-of-distribution (OOD) responses. However, current methods still fail to prevent the increasing frequency of OOD response generation during the reinforcement learning (RL) process and are not effective at handling extrapolation errors from OOD responses. In this work, we propose the Behavior-Supported Policy Optimization (BSPO) method to mitigate the reward over-optimization issue. Specifically, we define behavior policy as the next token distribution of the reward training dataset to model the in-distribution (ID) region of the reward model. Building on this, we introduce the behavior-supported Bellman operator to regularize the value function, penalizing all OOD values without impacting the ID ones. Consequently, BSPO reduces the generation of OOD responses during the RL process, thereby avoiding overestimation caused by the reward model's extrapolation errors. Theoretically, we prove that BSPO guarantees a monotonic improvement of the supported policy until convergence to the optimal behavior-supported policy. Empirical results from extensive experiments show that BSPO outperforms baselines in preventing reward over-optimization due to OOD evaluation and finding the optimal ID policy.

Vision-Language Models (VLMs) demand substantial computational resources during inference, largely due to the extensive visual input tokens for representing visual information. Previous studies have noted that visual tokens tend to receive less attention than text tokens, suggesting their lower importance during inference and potential for pruning. However, their methods encounter several challenges: reliance on greedy heuristic criteria for token importance and incompatibility with FlashAttention and KV cache. To address these issues, we introduce \textbf{TopV}, a compatible \textbf{TO}ken \textbf{P}runing with inference Time Optimization for fast and low-memory \textbf{V}LM, achieving efficient pruning without additional training or fine-tuning. Instead of relying on attention scores, we formulate token pruning as an optimization problem, accurately identifying important visual tokens while remaining compatible with FlashAttention. Additionally, since we only perform this pruning once during the prefilling stage, it effectively reduces KV cache size. Our optimization framework incorporates a visual-aware cost function considering factors such as Feature Similarity, Relative Spatial Distance, and Absolute Central Distance, to measure the importance of each source visual token, enabling effective pruning of low-importance tokens. Extensive experiments demonstrate that our method outperforms previous token pruning methods, validating the effectiveness and efficiency of our approach.

Emotion regulation is a crucial skill for managing emotions in everyday life, yet finding a constructive and accessible method to support these processes remains challenging due to their cognitive demands. In this study, we explore how regular interactions with a social robot, conducted in a structured yet familiar environment within university halls and departments, can provide effective support for emotion regulation through cognitive reappraisal. Twenty-one students participated in a five-session study at a university hall or department, where the robot facilitated structured conversations, encouraging the students to reinterpret emotionally charged situations that they shared with the robot. Quantitative and qualitative results indicate significant improvements in emotion self-regulation, with participants reporting better understanding and control of their emotions. The intervention led to significant changes in constructive emotion regulation tendencies and positive effects on mood and sentiment after each session. The findings also demonstrate that repeated interactions with the robot encouraged greater emotional expressiveness, including longer speech disclosures, increased use of affective language, and heightened facial arousal. Notably, expressiveness followed structured patterns aligned with the reappraisal process, with expression peaking during key reappraisal moments, particularly when participants were prompted to reinterpret negative experiences. The qualitative feedback further highlighted how the robot fostered introspection and provided a supportive space for discussing emotions, enabling participants to confront long-avoided emotional challenges. These findings demonstrate the potential of robots to effectively assist in emotion regulation in familiar environments, offering both emotional support and cognitive guidance.

Over the last few decades, Artificial Intelligence (AI) scientists have been conducting investigations to attain human-level performance by a machine in accomplishing a cognitive task. Within machine learning, the ultimate aspiration is to attain Artificial General Intelligence (AGI) through a machine. This pursuit has led to the exploration of two distinct AI paradigms. Symbolic AI, also known as classical or GOFAI (Good Old-Fashioned AI) and Connectionist (Sub-symbolic) AI, represented by Neural Systems, are two mutually exclusive paradigms. Symbolic AI excels in reasoning, explainability, and knowledge representation but faces challenges in processing complex real-world data with noise. Conversely, deep learning (Black-Box systems) research breakthroughs in neural networks are notable, yet they lack reasoning and interpretability. Neuro-symbolic AI (NeSy), an emerging area of AI research, attempts to bridge this gap by integrating logical reasoning into neural networks, enabling them to learn and reason with symbolic representations. While a long path, this strategy has made significant progress towards achieving common sense reasoning by systems. This article conducts an extensive review of over 977 studies from prominent scientific databases (DBLP, ACL, IEEExplore, Scopus, PubMed, ICML, ICLR), thoroughly examining the multifaceted capabilities of Neuro-Symbolic AI, with a particular focus on its healthcare applications, particularly in drug discovery, and Protein engineering research. The survey addresses vital themes, including reasoning, explainability, integration strategies, 41 healthcare-related use cases, benchmarking, datasets, current approach limitations from both healthcare and broader perspectives, and proposed novel approaches for future experiments.

Tibetan, a minority language in China, features a highly intricate grammatical structure, characterized by four verb tenses and a tense system with frequent irregularities, contributing to its extensive inflectional diversity. Recently, advances in Large Language Models (LLMs) have transformed the paradigm in many domains. Despite the success in other fields, current LLMs often fall short in catering to the needs of domain experts like Tibetans, and the potential of LLMs for Tibetan culture is under-explored. The intrinsic reasons are the immense and intricate nature of Tibetan culture as well as the necessity for higher granularity and richness in knowledge. Simultaneously, the complexity and uniqueness of its grammatical structure, coupled with its status as a minority ethnic language, contribute to data scarcity, which remains a fundamental challenge. To alleviate these issues, we introduce Llama-Sunshine (Sun-Shine), the first large language model for Tibetan culture, which is expert in various Tibetan language processing tasks. Sun-Shine incorporates state-of-the-art model architectures optimized for Tibetan's linguistic features. We also propose TIB-STC, a comprehensive dataset comprising diverse Tibetan texts such as literature, religious scripts, news, and conversational data, which is also the first large-scale dataset for Tibetan culture. Though comprehensive experiments, Sun-Shine not only demonstrates a higher level of knowledge expertise for Tibetan culture but also gains preliminary embodied intelligence capabilities in Tibetan language processing tasks, like language modeling, text classification, machine translation, and syntactic analysis. Moreover, it excels in low-resource scenarios, showcasing strong generalization capabilities.