Reasoning models have achieved remarkable performance on tasks like math and logical reasoning thanks to their ability to search during reasoning. However, they still suffer from overthinking, often performing unnecessary reasoning steps even after reaching the correct answer. This raises the question: can models evaluate the correctness of their intermediate answers during reasoning? In this work, we study whether reasoning models encode information about answer correctness through probing the model's hidden states. The resulting probe can verify intermediate answers with high accuracy and produces highly calibrated scores. Additionally, we find models' hidden states encode correctness of future answers, enabling early prediction of the correctness before the intermediate answer is fully formulated. We then use the probe as a verifier to decide whether to exit reasoning at intermediate answers during inference, reducing the number of inference tokens by 24\% without compromising performance. These findings confirm that reasoning models do encode a notion of correctness yet fail to exploit it, revealing substantial untapped potential to enhance their efficiency.

Shoaib Jameel University of Southampton Southampton, United Kingdom M.S.Jameel@southampton.ac.uk Abstract --Personality detection automatically identifies an individual's personality from various data sources, such as social media texts. However, as the parameter scale of language models continues to grow, the computational cost becomes increasingly difficult to manage. Fine-tuning also grows more complex, making it harder to justify the effort and reliably predict outcomes. We introduce a novel parameter-efficient fine-tuning framework, PersLLM, to address these challenges. By storing the features in the memory layer, we eliminate the need for repeated complex computations by the LLM. Meanwhile, the lightweight output network serves as a proxy for evaluating the overall effectiveness of the framework, improving the predictability of results. Experimental results on key benchmark datasets like Kaggle and Pandora show that PersLLM significantly reduces computational cost while maintaining competitive performance and strong adaptability. I NTRODUCTION Personality refers to the stable traits in an individual's emotions, thoughts, and behaviours that shape how they perceive, interpret, and interact with the world [1], [2]. It is a complex and multifaceted construct that combines various characteristics to form a person's unique identity.

Large Language Models (LLMs) have significantly advanced smart education in the Artificial General Intelligence (AGI) era. A promising application lies in the automatic generalization of instructional design for curriculum and learning activities, focusing on two key aspects: (1) Customized Generation: generating niche-targeted teaching content based on students' varying learning abilities and states, and (2) Intelligent Optimization: iteratively optimizing content based on feedback from learning effectiveness or test scores. Currently, a single large LLM cannot effectively manage the entire process, posing a challenge for designing intelligent teaching plans. To address these issues, we developed EduPlanner, an LLM-based multi-agent system comprising an evaluator agent, an optimizer agent, and a question analyst, working in adversarial collaboration to generate customized and intelligent instructional design for curriculum and learning activities. Taking mathematics lessons as our example, EduPlanner employs a novel Skill-Tree structure to accurately model the background mathematics knowledge of student groups, personalizing instructional design for curriculum and learning activities according to students' knowledge levels and learning abilities. Additionally, we introduce the CIDDP, an LLM-based five-dimensional evaluation module encompassing clarity, Integrity, Depth, Practicality, and Pertinence, to comprehensively assess mathematics lesson plan quality and bootstrap intelligent optimization. Experiments conducted on the GSM8K and Algebra datasets demonstrate that EduPlanner excels in evaluating and optimizing instructional design for curriculum and learning activities. Ablation studies further validate the significance and effectiveness of each component within the framework. Our code is publicly available at https://github.com/Zc0812/Edu_Planner

The use of AI in legal analysis and prediction (LegalAI) has gained widespread attention, with past research focusing on retrieval-based methods and fine-tuning large models. However, these approaches often require large datasets and underutilize the capabilities of modern large language models (LLMs). In this paper, inspired by the debate phase of real courtroom trials, we propose a novel legal judgment prediction model based on the Debate-Feedback architecture, which integrates LLM multi-agent debate and reliability evaluation models. Unlike traditional methods, our model achieves significant improvements in efficiency by minimizing the need for large historical datasets, thus offering a lightweight yet robust solution. Comparative experiments show that it outperforms several general-purpose and domain-specific legal models, offering a dynamic reasoning process and a promising direction for future LegalAI research.

This paper presents Thanos, a novel weight-pruning algorithm designed to reduce the memory footprint and enhance the computational efficiency of large language models (LLMs) by removing redundant weights while maintaining accuracy. Thanos introduces a block-wise pruning strategy with adaptive masks that dynamically adjust to weight importance, enabling flexible sparsity patterns and structured formats, such as $n:m$ sparsity, optimized for hardware acceleration. Experimental evaluations demonstrate that Thanos achieves state-of-the-art performance in structured pruning and outperforms existing methods in unstructured pruning. By providing an efficient and adaptable approach to model compression, Thanos offers a practical solution for deploying large models in resource-constrained environments.

Error detection (ED) in tabular data is crucial yet challenging due to diverse error types and the need for contextual understanding. Traditional ED methods often rely heavily on manual criteria and labels, making them labor-intensive. Large language models (LLM) can minimize human effort but struggle with errors requiring a comprehensive understanding of data context. In this paper, we propose ZeroED, a novel hybrid zero-shot error detection framework, which combines LLM reasoning ability with the manual label-based ED pipeline. ZeroED operates in four steps, i.e., feature representation, error labeling, training data construction, and detector training. Initially, to enhance error distinction, ZeroED generates rich data representations using error reason-aware binary features, pre-trained embeddings, and statistical features. Then, ZeroED employs LLM to label errors holistically through in-context learning, guided by a two-step reasoning process for detailed error detection guidelines. To reduce token costs, LLMs are applied only to representative data selected via clustering-based sampling. High-quality training data is constructed through in-cluster label propagation and LLM augmentation with verification. Finally, a classifier is trained to detect all errors. Extensive experiments on seven public datasets demonstrate that, ZeroED substantially outperforms state-of-the-art methods by a maximum 30% improvement in F1 score and up to 90% token cost reduction.

Recent advancements in Large Language Models (LLMs) have made them a popular information-seeking tool among end users. However, the statistical training methods for LLMs have raised concerns about their representation of under-represented topics, potentially leading to biases that could influence real-world decisions and opportunities. These biases could have significant economic, social, and cultural impacts as LLMs become more prevalent, whether through direct interactions--such as when users engage with chatbots or automated assistants--or through their integration into third-party applications (as agents), where the models influence decision-making processes and functionalities behind the scenes. Our study examines the biases present in LLMs recommendations of U.S. cities and towns across three domains: relocation, tourism, and starting a business. We explore two key research questions: (i) How similar LLMs responses are, and (ii) How this similarity might favor areas with certain characteristics over others, introducing biases. We focus on the consistency of LLMs responses and their tendency to over-represent or under-represent specific locations. Our findings point to consistent demographic biases in these recommendations, which could perpetuate a ``rich-get-richer'' effect that widens existing economic disparities.

Large Language Models (LLMs) excel in language comprehension and generation but are prone to hallucinations, producing factually incorrect or unsupported outputs. Retrieval Augmented Generation (RAG) systems address this issue by grounding LLM responses with external knowledge. This study evaluates the relationship between retriever effectiveness and hallucination reduction in LLMs using three retrieval approaches: sparse retrieval based on BM25 keyword search, dense retrieval using semantic search with Sentence Transformers, and a proposed hybrid retrieval module. The hybrid module incorporates query expansion and combines the results of sparse and dense retrievers through a dynamically weighted Reciprocal Rank Fusion score. Using the HaluBench dataset, a benchmark for hallucinations in question answering tasks, we assess retrieval performance with metrics such as mean average precision and normalised discounted cumulative gain, focusing on the relevance of the top three retrieved documents. Results show that the hybrid retriever achieves better relevance scores, outperforming both sparse and dense retrievers. Further evaluation of LLM-generated answers against ground truth using metrics such as accuracy, hallucination rate, and rejection rate reveals that the hybrid retriever achieves the highest accuracy on fails, the lowest hallucination rate, and the lowest rejection rate. These findings highlight the hybrid retriever's ability to enhance retrieval relevance, reduce hallucination rates, and improve LLM reliability, emphasising the importance of advanced retrieval techniques in mitigating hallucinations and improving response accuracy.

--Composed Image Retrieval (CIR) aims to search an image of interest using a combination of a reference image and modification text as the query. Despite recent advancements, this task remains challenging due to limited training data and laborious triplet annotation processes. T o address this issue, this paper proposes to synthesize the training triplets to augment the training resource for the CIR problem. During pretraining, we leverage the trained generator to directly create Modification Text-oriented Synthetic Triplets (MTST) conditioned on pairs of images. For fine-tuning, we first synthesize reverse modification text to connect the target image back to the reference image. Subsequently, we devise a two-hop alignment strategy to incre-mentally close the semantic gap between the multimodal pair and the target image. We initially learn an implicit prototype utilizing both the original triplet and its reversed version in a cycle manner, followed by combining the implicit prototype feature with the modification text to facilitate accurate alignment with the target image. Extensive experiments validate the efficacy of the generated triplets and confirm that our proposed methodology attains competitive recall on both the CIRR and FashionIQ benchmarks. Wang is with the School of Electronics and Information Engineering, Hefei University of Technology, Hefei 230009, China (e-mail: wangyx15@stu.xjtu.edu.cn). H. Lin is with the School of Artificial Intelligence, University of the Chinese Academy of Sciences, Beijing 101408, China (e-mail: haokun.lin@cripac.ia.ac.cn). Zhou are with the Department of Computer Science, City University of Hong Kong, Hong Kong 999077, China (e-mail: chenma@cityu.edu.hk). Z. Zheng is with Faculty of Science and Technology, and Institute of Collaborative Innovation, University of Macau, Macau 999078, China (e-mail: zhedongzheng@um.edu.mo).

Sequential recommendation is an advanced recommendation technique that utilizes the sequence of user behaviors to generate personalized suggestions by modeling the temporal dependencies and patterns in user preferences. However, it requires a server to centrally collect users' data, which poses a threat to the data privacy of different users. In recent years, federated learning has emerged as a distributed architecture that allows participants to train a global model while keeping their private data locally. This survey pioneers Federated Sequential Recommendation (FedSR), where each user joins as a participant in federated training to achieve a recommendation service that balances data privacy and model performance. We begin with an introduction to the background and unique challenges of FedSR. Then, we review existing solutions from two levels, each of which includes two specific techniques. Additionally, we discuss the critical challenges and future research directions in FedSR.