Large Language Models (LLMs) are widely used in Automated Essay Scoring (AES) due to their ability to capture semantic meaning. Traditional fine-tuning approaches required technical expertise, limiting accessibility for educators with limited technical backgrounds. However, prompt-based tools like ChatGPT have made AES more accessible, enabling educators to obtain machine-generated scores using natural-language prompts (i.e., the prompt-based paradigm). Despite advancements, prior studies have shown bias in fine-tuned LLMs, particularly against disadvantaged groups. It remains unclear whether such biases persist or are amplified in the prompt-based paradigm with cutting-edge tools. Since such biases are believed to stem from the demographic information embedded in pre-trained models (i.e., the ability of LLMs' text embeddings to predict demographic attributes), this study explores the relationship between the model's predictive power of students' demographic attributes based on their written works and its predictive bias in the scoring task in the prompt-based paradigm. Using a publicly available dataset of over 25,000 students' argumentative essays, we designed prompts to elicit demographic inferences (i.e., gender, first-language background) from GPT-4o and assessed fairness in automated scoring. Then we conducted multivariate regression analysis to explore the impact of the model's ability to predict demographics on its scoring outcomes. Our findings revealed that (i) prompt-based LLMs can somewhat infer students' demographics, particularly their first-language backgrounds, from their essays; (ii) scoring biases are more pronounced when the LLM correctly predicts students' first-language background than when it does not; and (iii) scoring error for non-native English speakers increases when the LLM correctly identifies them as non-native.

Large language models (LLMs) exhibit probabilistic output characteristics, yet conventional evaluation frameworks rely on deterministic scalar metrics. This study introduces a Bayesian approach for LLM capability assessment that integrates prior knowledge through probabilistic inference, addressing limitations under limited-sample regimes. By treating model capabilities as latent variables and leveraging a curated query set to induce discriminative responses, we formalize model ranking as a Bayesian hypothesis testing problem over mutually exclusive capability intervals. Experimental evaluations with GPT-series models demonstrate that the proposed method achieves superior discrimination compared to conventional evaluation methods. Results indicate that even with reduced sample sizes, the approach maintains statistical robustness while providing actionable insights, such as probabilistic statements about a model's likelihood of surpassing specific baselines. This work advances LLM evaluation methodologies by bridging Bayesian inference with practical constraints in real-world deployment scenarios.

We consider the problem of causal discovery (a.k.a., causal structure learning) in a multi-domain setting. We assume that the causal functions are invariant across the domains, while the distribution of the exogenous noise may vary. Under causal sufficiency (i.e., no confounders exist), we show that the causal diagram can be discovered under less restrictive functional assumptions compared to previous work. What enables causal discovery in this setting is bijective generation mechanisms (BGM), which ensures that the functional relation between the exogenous noise $E$ and the endogenous variable $Y$ is bijective and differentiable in both directions at every level of the cause variable $X = x$. BGM generalizes a variety of models including additive noise model, LiNGAM, post-nonlinear model, and location-scale noise model. Further, we derive a statistical test to find the parents set of the target variable. Experiments on various synthetic and real-world datasets validate our theoretical findings.

Large Language Models (LLMs) currently struggle to sequentially add new memories and integrate new knowledge. These limitations contrast with the human ability to continuously learn from new experiences and acquire knowledge throughout life. Most existing approaches add memories either through large context windows or external memory buffers (e.g., Retrieval-Augmented Generation), and studies on knowledge injection rarely test scenarios resembling everyday life events. In this work, we introduce a continual learning framework, Memory Embedded in Gated LLMs (MEGa), which injects event memories directly into the weights of LLMs. Each memory is stored in a dedicated set of gated low-rank weights. During inference, a gating mechanism activates relevant memory weights by matching query embeddings to stored memory embeddings. This enables the model to both recall entire memories and answer related questions. On two datasets - fictional characters and Wikipedia events - MEGa outperforms baseline approaches in mitigating catastrophic forgetting. Our model draws inspiration from the complementary memory system of the human brain.

Chain-of-Thought (CoT) significantly enhances formal reasoning capabilities in Large Language Models (LLMs) by training them to explicitly generate intermediate reasoning steps. While LLMs readily benefit from such techniques, improving reasoning in Small Language Models (SLMs) remains challenging due to their limited model capacity. Recent work by Deepseek-R1 demonstrates that distillation from LLM-generated synthetic data can substantially improve the reasoning ability of SLM. However, the detailed modeling recipe is not disclosed. In this work, we present a systematic training recipe for SLMs that consists of four steps: (1) large-scale mid-training on diverse distilled long-CoT data, (2) supervised fine-tuning on high-quality long-CoT data, (3) Rollout DPO leveraging a carefully curated preference dataset, and (4) Reinforcement Learning (RL) with Verifiable Reward. We apply our method on Phi-4-Mini, a compact 3.8B-parameter model. The resulting Phi-4-Mini-Reasoning model exceeds, on math reasoning tasks, much larger reasoning models, e.g., outperforming DeepSeek-R1-Distill-Qwen-7B by 3.2 points and DeepSeek-R1-Distill-Llama-8B by 7.7 points on Math-500. Our results validate that a carefully designed training recipe, with large-scale high-quality CoT data, is effective to unlock strong reasoning capabilities even in resource-constrained small models.

Despite the recent advances in Large Language Models, benchmarks for evaluating legal writing remain scarce due to the inherent complexity of assessing open-ended responses in this domain. One of the key challenges in evaluating language models on domain-specific tasks is finding test datasets that are public, frequently updated, and contain comprehensive evaluation guidelines. The Brazilian Bar Examination meets these requirements. We introduce oab-bench, a benchmark comprising 105 questions across seven areas of law from recent editions of the exam. The benchmark includes comprehensive evaluation guidelines and reference materials used by human examiners to ensure consistent grading. We evaluate the performance of four LLMs on oab-bench, finding that Claude-3.5 Sonnet achieves the best results with an average score of 7.93 out of 10, passing all 21 exams. We also investigated whether LLMs can serve as reliable automated judges for evaluating legal writing. Our experiments show that frontier models like OpenAI's o1 achieve a strong correlation with human scores when evaluating approved exams, suggesting their potential as reliable automated evaluators despite the inherently subjective nature of legal writing assessment. The source code and the benchmark -- containing questions, evaluation guidelines, model-generated responses, and their respective automated evaluations -- are publicly available.

--Federated domain generalization (FedDG) aims to learn a globally generalizable model from decentralized clients with heterogeneous data while preserving privacy. Recent studies have introduced prompt learning to adapt vision-language models (VLMs) in FedDG by learning a single global prompt. However, such a one-prompt-fits-all learning paradigm typically leads to performance degradation on personalized samples. Although the mixture of experts (MoE) offers a promising solution for specialization, existing MoE-based methods suffer from coarse image-level expert assignment and high communication costs from parameterized routers. T o address these limitations, we propose TRIP, a T oken-level pRompt mIxture with Parameter-free routing framework for FedDG, which treats multiple prompts as distinct experts. Unlike existing image-level routing designs, TRIP assigns different tokens within an image to specific experts. T o ensure communication efficiency, TRIP incorporates a parameter-free routing mechanism based on token clustering and optimal transport. The instance-specific prompt is then synthesized by aggregating experts, weighted by the number of tokens assigned to each. Additionally, TRIP develops an unbiased learning strategy for prompt experts, leveraging the VLM's zero-shot generalization capability. Extensive experiments across four benchmarks demonstrate that TRIP achieves optimal generalization results, with communicating only 1K parameters per round. HE exponential growth of data from diverse and decentralized sources has significantly accelerated advancements in machine learning. However, traditional machine learning paradigms typically operate in a centralized manner, which requires all decentralized data to be processed on a central server, raising serious privacy concerns. This work was supported by the Shandong Provincial Natural Science Foundation under Grant ZR2020KF015, and by the Taishan Scholar Program of Shandong Province under Grant tsqn202211199 and Grant tstp20221137. S. Gong, C. Cui, and X. Dong are with the School of Computing and Artificial Intelligence, Shandong University of Finance and Economics, Jinan 250014, China (e-mail: gsh8210@163.com; X. Nie is with the School of Computer Science and Technology, Shandong Jianzhu University, Jinan 250101, China (e-mail: niexsh@hotmail.com). L. Zhu is with the College of Electronics and Information Engineering, Tongji University, Shanghai 201804, China (e-mail: leizhu0608@gmail.com). X. Chang is with the School of Information Science and Technology, University of Science and Technology of China, Anhui 230026, China.

Multi-Agent Reinforcement Learning (MARL) has become a powerful framework for numerous real-world applications, modeling distributed decision-making and learning from interactions with complex environments. Resource Allocation Optimization (RAO) benefits significantly from MARL's ability to tackle dynamic and decentralized contexts. MARL-based approaches are increasingly applied to RAO challenges across sectors playing pivotal roles to Industry 4.0 developments. This survey provides a comprehensive review of recent MARL algorithms for RAO, encompassing core concepts, classifications, and a structured taxonomy. By outlining the current research landscape and identifying primary challenges and future directions, this survey aims to support researchers and practitioners in leveraging MARL's potential to advance resource allocation solutions.

Artificial Intelligence (AI)-generated feedback in educational settings has garnered considerable attention due to its potential to enhance learning outcomes. However, a comprehensive understanding of the linguistic characteristics of AI-generated feedback, including readability, lexical richness, and adaptability across varying challenge levels, remains limited. This study delves into the linguistic and structural attributes of feedback generated by Google's Gemini 1.5-flash text model for computer science multiple-choice questions (MCQs). A dataset of over 1,200 MCQs was analyzed, considering three difficulty levels (easy, medium, hard) and three feedback tones (supportive, neutral, challenging). Key linguistic metrics, such as length, readability scores (Flesch-Kincaid Grade Level), vocabulary richness, and lexical density, were computed and examined. A fine-tuned RoBERTa-based multi-task learning (MTL) model was trained to predict these linguistic properties, achieving a Mean Absolute Error (MAE) of 2.0 for readability and 0.03 for vocabulary richness. The findings reveal significant interaction effects between feedback tone and question difficulty, demonstrating the dynamic adaptation of AI-generated feedback within diverse educational contexts. These insights contribute to the development of more personalized and effective AI-driven feedback mechanisms, highlighting the potential for improved learning outcomes while underscoring the importance of ethical considerations in their design and deployment.

FULL PAPER Motion Generation for Food Topping Challenge 2024: Serving Salmon Roe Bowl and Picking Fried Chicken Koki Inami a, Masashi Konosu a, Koki Yamane a, Nozomu Masuya a, Yunhan Li a, Yu-Han Shu a, Hiroshi Sato a, Shinnosuke Homma a, and Sho Sakaino b a Intelligent and Mechanical Interaction Systems, Degree Programs in Systems and Information and Engineering, Graduate School of Science and Technology, University of Tsukuba, Japan; b Department of Intelligent Interaction Technologies, Institute of Systems and Information Engineering, University of Tsukuba, Japan; ARTICLE HISTORY Compiled May 1, 2025 ABSTRACT Although robots have been introduced in many industries, food production robots are yet to be widely employed because the food industry requires not only delicate movements to handle food but also complex movements that adapt to the environment. Force control is important for handling delicate objects such as food. In addition, achieving complex movements is possible by making robot motions based on human teachings. Four-channel bilateral control is proposed, which enables the simultaneous teaching of position and force information. Moreover, methods have been developed to reproduce motions obtained through human teachings and generate adaptive motions using learning. We demonstrated the effectiveness of these methods for food handling tasks in the Food Topping Challenge at the 2024 IEEE International Conference on Robotics and Automation (ICRA 2024). For the task of serving salmon roe on rice, we achieved the best performance because of the high reproducibility and quick motion of the proposed method. Further, for the task of picking fried chicken, we successfully picked the most pieces of fried chicken among all participating teams. This paper describes the implementation and performance of these methods. KEYWORDS Bilateral control; motion-copying system; imitation learning; 1. Introduction Robot automation has progressed in recent years, especially in the manufacturing industry because robots excel at precise repetitive movements in a well-equipped environment. However, their use in the food industry has been limited because of the complexity of the work and the need to generate movements that adapt to the work environment.