检索结果-内蒙古大学图书馆

ToolEyes: Fine-Grained Evaluation for Tool Learning Capabilities of Large Language Models in Real-world Scenarios 31

学校读者我要写书评

暂无评论

ToolEyes: Fine-Grained Evaluation for Tool Learning Capabili...

31st International Conference on Computational Linguistics, COLING 2025

作者： Ye, Junjie Li, Guanyu Gao, Songyang Huang, Caishuang Wu, Yilong Li, Sixian Fan, Xiaoran Dou, Shihan Ji, Tao Zhang, Qi Gui, Tao Huang, Xuanjing School of Computer Science Fudan University China Institute of Modern Languages and Linguistics Fudan University China Research Institute of Intelligent Complex Systems Fudan University China Shanghai Key Laboratory of Intelligent Information Processing China Pengcheng Laboratory China

ISBN: (纸本)9798891761964

Existing evaluations of tool learning primarily focus on validating the alignment of selected tools (e.g., various APIs) for large language models (LLMs) with expected outcomes. However, these approaches rely on a limited set of scenarios where answers can be pre-determined. Furthermore, a sole emphasis on outcomes disregards the complex capabilities required for LLMs to effectively use tools. To tackle this issue, we propose ToolEyes, a fine-grained system tailored for the evaluation of the LLMs' tool learning capabilities in authentic scenarios. The system meticulously examines seven real-world scenarios, analyzing five dimensions crucial to LLMs in tool learning: format alignment, intent comprehension, behavior planning, tool selection, and answer organization. Additionally, ToolEyes incorporates a tool library boasting approximately 600 tools, serving as an intermediary between LLMs and the physical world. Evaluations involving ten LLMs across three categories reveal a preference for specific scenarios and limited cognitive abilities in tool learning. Intriguingly, expanding the model size even exacerbates the hindrance to tool learning. The code and data are available at https://***/Junjie-Ye/ToolEyes. © 2025 Association for Computational Linguistics.

关键词： Computational linguistics

Fuzzy-Model-Based Fault-Tolerant Control for Stochastic Re-Entrant Manufacturing systems

学校读者我要写书评

暂无评论

IEEE Transactions on systems, Man, and Cybernetics: systems 2025年第5期55卷 3542-3556页

作者： Zhang, Kexin Gao, Qing Ding, Steven X. Lu, Jinhu Qiu, Jianbin Guo, Yige Beihang University School of Cyber Science and Technology Beijing100191 China Beihang University School of Automation Science and Electrical Engineering Beijing100191 China Hangzhou Innovation Institute Beihang University Hangzhou310051 China Zhongguancun Laboratory Beijing100095 China Institute for Automatic Control and Complex Systems University of Duisburg-Essen Duisburg47057 Germany Research Institute of Intelligent Control and Systems Harbin Institute of Technology Harbin150001 China Zhongguancun Laboratory Beijing China

This study addresses the problem of guaranteed cost fault-tolerant fuzzy control for multiline re-entrant manufacturing systems (RMSs) against stochastic disturbances and workstation faults. Initially, a nonlinear hyperbolic impulsive partial differential equation model is employed to describe the complex and hybrid dynamics of RMSs suffering from unexpected faults within the working stations, and then the corresponding approximation T-S fuzzy model is constructed. In what follows, with the aid of the parallel distributed compensation fuzzy control scheme, the main results of stability analysis and controller synthesis for the closed-loop re-entrant manufacturing control system are derived using a timer-dependent Lyapunov functional with spatio-temporal auxiliary variables. It is found that by means of the proposed fault-tolerant control approach, the RMS can be effectively and robustly driven to a desired production mode with steady feeding and production rates while the upper bound of a quadratic cost function is minimized. Finally, the effectiveness of the proposed control approach is validated through numerical simulations. © 2013 IEEE.

关键词： Fuzzy control

Beyond Boundaries: Learning a Universal Entity Taxonomy across Datasets and Languages for Open Named Entity Recognition 31

学校读者我要写书评

暂无评论

Beyond Boundaries: Learning a Universal Entity Taxonomy acro...

31st International Conference on Computational Linguistics, COLING 2025

作者： Yang, Yuming Zhao, Wantong Huang, Caishuang Ye, Junjie Wang, Xiao Zheng, Huiyuan Nan, Yang Wang, Yuran Xu, Xueying Huang, Kaixin Zhang, Yunke Gui, Tao Zhang, Qi Huang, Xuanjing School of Computer Science Fudan University China Honor Device Co. Ltd. Institute of Modern Languages and Linguistics Fudan University China Pengcheng Laboratory China Research Institute of Intelligent Complex Systems Fudan University China Shanghai Key Laboratory of Intelligent Information Processing China

ISBN: (纸本)9798891761964

Open Named Entity Recognition (NER), which involves identifying arbitrary types of entities from arbitrary domains, remains challenging for Large Language Models (LLMs). Recent studies suggest that fine-tuning LLMs on extensive NER data can boost their performance. However, training directly on existing datasets neglects their inconsistent entity definitions and redundant data, limiting LLMs to dataset-specific learning and hindering out-of-domain adaptation. To address this, we present B2NERD, a compact dataset designed to guide LLMs' generalization in Open NER under a universal entity taxonomy. B2NERD is refined from 54 existing English and Chinese datasets using a two-step process. First, we detect inconsistent entity definitions across datasets and clarify them by distinguishable label names to construct a universal taxonomy of 400+ entity types. Second, we address redundancy using a data pruning strategy that selects fewer samples with greater category and semantic diversity. Comprehensive evaluation shows that B2NERD significantly enhances LLMs' Open NER capabilities. Our B2NER models, trained on B2NERD, outperform GPT-4 by 6.8-12.0 F1 points and surpass previous methods in 3 out-of-domain benchmarks across 15 datasets and 6 languages. The data, models, and code are publicly available at https://***/UmeanNever/B2NER. © 2025 Association for Computational Linguistics.

关键词： Taxonomies

Finite-Time Consensus of Second-Order Multiagent systems With Input Saturation via Hybrid Sliding-Mode Control

学校读者我要写书评

暂无评论

IEEE Transactions on Automation science and Engineering 2025年 22卷 14623-14632页

作者： Jiang, Xiaowei Jiao, Ranran Li, Bo Zhang, Xianhe Yan, Huaicheng China University of Geosciences School of Automation Wuhan430079 China Ministry of Education Hubei Key Laboratory of Advanced Control and Intelligent Automation of Complex Systems Engineering Research Center of Intelligent Geodetection Technology Wuhan430074 China Ministry of Education Key Laboratory of System Control and Information Processing Shanghai200240 China Anhui University of Finance and Economics School of Finance Bengbu233030 China Hubei Normal University School of Electrical Engineering and Automation Huangshi435002 China Ministry of Education East China University of Science and Technology Key Laboratory of Smart Manufacturing in Energy Chemical Process Shanghai200237 China

This paper addresses the finite-time consensus (FTC) issue for second-order multi-agent systems (MASs) with nonlinear disturbances. To tackle the challenges posed by increasingly complex communication environments, an innovative integral sliding-mode surface is designed using event-triggered control. Furthermore, the study explores communication constraints between the leader and followers, employing impulsive control to facilitate communication only at specific intervals. A novel hybrid integral sliding-mode control protocol is advanced for the second-order MASs, which effectively eliminates the "Zeno phenomenon" and notably reduces both communication frequency and energy consumption, thereby enhancing overall communication efficiency. Ultimately, the efficacy of the put forward protocol is verified through simulation examples including comparison experiment. © 2004-2012 IEEE.

关键词： Protocols Sliding mode control Event detection Multi-agent systems Costs Training Topology Switches Sufficient conditions Stability analysis

Efficient Algorithms for Minimizing the Kirchhoff Index via Adding Edges

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Zhou, Xiaotian Zehmakan, Ahad N. Zhang, Zhongzhi Shanghai Key Laboratory of Intelligent Information Processing School of Computer Science Fudan University Shanghai200433 China Research Institute of Intelligent Complex Systems Fudan University Shanghai200433 China School of Computing Australian National University Canberra Australia

关键词： Consensus algorithm

Fine-Tuning the BERT Model to Predict Depression and Anxiety Using Multi-Labeled Twitter Data

学校读者我要写书评

暂无评论

Fine-Tuning the BERT Model to Predict Depression and Anxiety...

Sentiment Analysis and Deep Learning (ICSADL), International Conference on

作者： Lamia Bendebane Zakaria Laboudi Asma Saighi Seif Eddine Bouziane Research Laboratory on Computer Science's Complex Systems (ReLa(CS)2) University of Oum El Bouaghi Algeria Laboratory of Artificial Intelligence and Autonomous Things (LIAOA) University of Oum El Bouaghi Algeria Department of Intelligent Systems Engineering National School of Artificial Intelligence Algiers Algeria

ISBN: (数字)9798331523923

ISBN: (纸本)9798331523930

Studying mental health through social media data has become an emerging area of research, notably for the detection of depression and anxiety. In this regard, many researches have been conducted, yielding very satisfactory results (e.g., [1]–[3]). However, most of these studies have addressed each of these two mental disorders separately. This is due to the overlap of symptoms associated with depression and anxiety that makes distinguishing between them significantly challenging. Based on this context, this work leverages pretrained large language models (LLMs) to develop efficient multi-class models for predicting both depression and anxiety. For this purpose, a small, multi-labeled Twitter dataset is first constructed. Then, a pre-trained BERT-based uncased model is fine-tuned for six epochs on the dataset, resulting in a model referred to as DAC-BERT. Finally, the DAC-BERT model is evaluated and compared against both hybrid deep learning models and other LLM-based models. The obtained results show that DAC-BERT model outperforms existing approaches, achieving accuracies of up to 97.20% on a multi-labeled dataset containing normal, depressive, and anxious tweets, and up to 96.50% on a dataset with only normal and depressive tweets. These findings highlight the promising potential of fine-tuning LLMs for predicting mental health disorders, particularly depression and anxiety.

关键词： Deep learning Sentiment analysis Social networking (online) Large language models Anxiety disorders Blogs Mental health Predictive models Depression Data models

Adapting Human Mesh Recovery with Vision-Language Feedback

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Xu, Chongyang Huang, Buzhen Zhang, Chengfang Feng, Ziliang Wang, Yangang College of Computer Science Sichuan University Chengdu610065 China Key Laboratory of Measurement and Control of Complex Systems of Engineering Ministry of Education The School of Automation Southeast University Nanjing210096 China Intelligent Policing Key Laboratory of Sichuan Province Sichuan Police College Luzhou646000 China

Human mesh recovery can be approached using either regression-based or optimization-based methods. Regression models achieve high pose accuracy but struggle with model-to-image alignment due to the lack of explicit 2D-3D correspondences. In contrast, optimization-based methods align 3D models to 2D observations but are prone to local minima and depth ambiguity. In this work, we leverage large vision-language models (VLMs) to generate interactive body part descriptions, which serve as implicit constraints to enhance 3D perception and limit the optimization space. Specifically, we formulate monocular human mesh recovery as a distribution adaptation task by integrating both 2D observations and language descriptions. To bridge the gap between text and 3D pose signals, we first train a text encoder and a pose VQ-VAE, aligning texts to body poses in a shared latent space using contrastive learning. Subsequently, we employ a diffusion-based framework to refine the initial parameters guided by gradients derived from both 2D observations and text descriptions. Finally, the model can produce poses with accurate 3D perception and image consistency. Experimental results on multiple benchmarks validate its effectiveness. The code will be made publicly available. © 2025, CC BY.

关键词： Mesh generation

Human-like conceptual representations emerge from language prediction

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Xu, Ningyu Zhang, Qi Du, Chao Luo, Qiang Qiu, Xipeng Huang, Xuanjing Zhang, Menghan School of Computer Science Fudan University Shanghai China Institute of Modern Languages and Linguistics Fudan University Shanghai China Shanghai Key Laboratory of Intelligent Information Processing Shanghai China Research Institute of Intelligent Complex Systems Fudan University Shanghai China Shanghai Collaborative Innovation Center of Intelligent Visual Computing Shanghai China Ministry of Education Key Laboratory of Contemporary Anthropology Fudan University Shanghai China

People acquire concepts through rich physical and social experiences and use them to understand the world. In contrast, large language models (LLMs), trained exclusively through next-token prediction over language data, exhibit remarkably human-like behaviors. Are these models developing concepts akin to humans, and if so, how are such concepts represented and organized? To address these questions, we reframed the classic reverse dictionary task to simulate human concept inference in context and investigated the emergence of human-like conceptual representations within LLMs. Our results demonstrate that LLMs can flexibly derive concepts from linguistic descriptions in relation to contextual cues about other concepts. The derived representations converged towards a shared, context-independent structure that effectively predicted human behavior across key psychological phenomena, including computation of similarities, categories and semantic scales. Moreover, these representations aligned well with neural activity patterns in the human brain, even in response to visual rather than linguistic stimuli, providing evidence for biological plausibility. These findings establish that structured, human-like conceptual representations can naturally emerge from language prediction without real-world grounding. More broadly, our work positions LLMs as promising computational tools for understanding complex human cognition and paves the way for better alignment between artificial and human intelligence. Copyright © 2025, The Authors. All rights reserved.

关键词： Semantics

Reinforcement Learning based Constrained Optimal Control: an Interpretable Reward Design

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Ni, Jingjie Li, Fangfei Jin, Xin Peng, Xianlun Tang, Yang School of Mathematics East China University of Science and Technology Shanghai200237 China Key Laboratory of Smart Manufacturing in Energy Chemical Process Ministry of Education East China University of Science and Technology Shanghai200237 China Research Institute of Intelligent Complex Systems Fudan University Shanghai200433 China

This paper presents an interpretable reward design framework for reinforcement learning based constrained optimal control problems with state and terminal constraints. The problem is formalized within a standard partially observable Markov decision process framework. The reward function is constructed from four weighted components: a terminal constraint reward, a guidance reward, a penalty for state constraint violations, and a cost reduction incentive reward. A theoretically justified reward design is then presented, which establishes bounds on the weights of the components. This approach ensures that constraints are satisfied and objectives are optimized while mitigating numerical instability. Acknowledging the importance of prior knowledge in reward design, we sequentially solve two subproblems, using each solution to inform the reward design for the subsequent problem. Subsequently, we integrate reinforcement learning with curriculum learning, utilizing policies derived from simpler subproblems to assist in tackling more complex challenges, thereby facilitating convergence. The framework is evaluated against original and randomly weighted reward designs in a multi-agent particle environment. Experimental results demonstrate that the proposed approach significantly enhances satisfaction of terminal and state constraints and optimization of control cost. Copyright © 2025, The Authors. All rights reserved.

关键词： Markov processes