检索结果-内蒙古大学图书馆

38th Conference on Neural Information Processing systems, NeurIPS 2024

作者： Guan, Jian Wu, Wei Wen, Zujie Xu, Peng Wang, Hongning Huang, Minlie The CoAI group DCST Institute for Artificial Intelligence China State Key Lab of Intelligent Technology and Systems China Beijing National Research Center for Information Science and Technology China Tsinghua University Beijing100084 China Ant Group China

The notable success of large language models (LLMs) has sparked an upsurge in building language agents to complete various complex tasks. We present AMOR, an agent framework based on open-source LLMs, which reasons with external knowledge bases and adapts to specific domains through human supervision to the reasoning process. AMOR builds reasoning logic over a finite state machine (FSM) that solves problems through autonomous executions and transitions over disentangled modules. This allows humans to provide direct feedback to the individual modules, and thus naturally forms process supervision. Based on this reasoning and feedback framework, we develop AMOR through two-stage fine-tuning: warm-up and adaptation. The former fine-tunes the LLM with examples automatically constructed from various public datasets, enabling AMOR to generalize across different knowledge environments, while the latter tailors AMOR to specific domains using process feedback. Extensive experiments across multiple domains demonstrate the advantage of AMOR to strong baselines, thanks to its FSM-based reasoning and process feedback mechanism. The code and data are publicly available at https://***/JianGuanTHU/AMOR. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

OneActor: Consistent Subject Generation via Cluster-Conditioned Guidance 38

OneActor: Consistent Subject Generation via Cluster-Conditio...

引用

38th Conference on Neural Information Processing systems, NeurIPS 2024

作者： Wang, Jiahao Yan, Caixia Lin, Haonan Zhang, Weizhan Wang, Mengmeng Gong, Tieliang Dai, Guang Sun, Hao School of Computer Science and Technology MOEKLINNS Xi'an Jiaotong University China State Key Laboratory of Communication Content Cognition China College of Computer Science and Technology Zhejiang University of Technology China SGIT AI Lab State Grid Corporation of China Switzerland China Telecom Artificial Intelligence Technology Co.Ltd. China

Text-to-image diffusion models benefit artists with high-quality image generation. Yet their stochastic nature hinders artists from creating consistent images of the same subject. Existing methods try to tackle this challenge and generate consistent content in various ways. However, they either depend on external restricted data or require expensive tuning of the diffusion model. For this issue, we propose a novel one-shot tuning paradigm, termed OneActor. It efficiently performs consistent subject generation solely driven by prompts via a learned semantic guidance to bypass the laborious backbone tuning. We lead the way to formalize the objective of consistent subject generation from a clustering perspective, and thus design a cluster-conditioned model. To mitigate the overfitting challenge shared by one-shot tuning pipelines, we augment the tuning with auxiliary samples and devise two inference strategies: semantic interpolation and cluster guidance. These techniques are later verified to significantly improve the generation quality. Comprehensive experiments show that our method outperforms a variety of baselines with satisfactory subject consistency, superior prompt conformity as well as high image quality. Our method is capable of multi-subject generation and compatible with popular diffusion extensions. Besides, we achieve a 4× faster tuning speed than tuning-based baselines and, if desired, avoid increasing the inference time. Furthermore, our method can be naturally utilized to pre-train a consistent subject generation network from scratch, which will implement this research task into more practical applications. Project page: https://***/OneActor-webpage/. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

An Individually Controlled Multitined Expandable Electrode Using Active Cannula-Based Shape Morphing for On-Demand Conformal Radiofrequency Ablation Lesions

ADVANCED INTELLIGENT SYSTEMS

引用

ADVANCED INTELLIGENT systemS 2022年第7期4卷

作者： Chai, Zhiping Lyu, Liangxiong Pu, Menghao Chen, Xianwen Zhu, Jiaqi Liang, Huageng Ding, Han Wu, Zhigang Soft Intelligence Lab State Key Laboratory of Digital Manufacturing Equipment and Technology Huazhong University of Science and Technology Wuhan China Department of Urology Union Hospital Tongji Medical College Huazhong University of Science and Technology Wuhan China

Being minimally invasive and highly effective, radiofrequency ablation (RFA) is widely used for small-sized malignant tumor treatment. However, in clinical practice, a large number of tumors are found in irregular shape, while the current RFA devices are hard to control the morphologic appearance of RFA lesions on demand, which usually ends up with unnecessarily excessive tissue ablation and subsequently often brings irreversible damage to the organs' functions. Herein, active cannulas for each of the individually controlled subelectrodes to achieve an on-demand shape morphing and thus conformal RFA lesion are introduced. The target shape as well as the length of inserted subelectrodes can be precisely controlled by tuning the active stylets and cannulas. What's more, owing to independent movement and energy control of each subelectrodes, the electrode is shown to be not only efficient enough to accomplish accurate trajectory control to target tissue in a single insertion, but also adaptive enough to ablate target tissues with diverse morphologic appearances and locations. On-demand conformal ablation of target tissue is demonstrated as well under the guidance of ultrasound imaging with the device. Potentially, the RFA electrode is a promising minimally invasive treatment of malignant tumors in future clinical practice. An interactive preprint version of the article can be found at: https://***/doi/full/10.22541/au.164019293.38729522.

关键词： active cannulas individually controlled electrodes multitined electrodes on-demand conformal lesions radiofrequency ablation shape morphing

来源：评论

学校读者我要写书评

暂无评论

Safe Reinforcement Learning using Finite-Horizon Gradient-based Estimation 41

Safe Reinforcement Learning using Finite-Horizon Gradient-ba...

引用

41st International Conference on Machine Learning, ICML 2024

作者： Dai, Juntao Yang, Yaodong Zheng, Qian Pan, Gang College of Computer Science and Technology Zhejiang University Hangzhou China The State Key Lab of Brain-Machine Intelligence Zhejiang University Hangzhou China Center for AI Safety and Governance Peking University Beijing China

A key aspect of Safe Reinforcement Learning (Safe RL) involves estimating the constraint condition for the next policy, which is crucial for guiding the optimization of safe policy updates. However, the existing Advantage-based Estimation (ABE) method relies on the infinite-horizon discounted advantage function. This dependence leads to catastrophic errors in finite-horizon scenarios with non-discounted constraints, resulting in safety-violation updates. In response, we propose the first estimation method for finite-horizon non-discounted constraints in deep Safe RL, termed Gradient-based Estimation (GBE), which relies on the analytic gradient derived along trajectories. Our theoretical and empirical analyses demonstrate that GBE can effectively estimate constraint changes over a finite horizon. Constructing a surrogate optimization problem with GBE, we developed a novel Safe RL algorithm called Constrained Gradient-based Policy Optimization (CGPO). CGPO identifies feasible optimal policies by iteratively resolving sub-problems within trust regions. Our empirical results reveal that CGPO, unlike baseline algorithms, successfully estimates the constraint functions of subsequent policies, thereby ensuring the efficiency and feasibility of each update. Copyright 2024 by the author(s)

关键词： Deep reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

DRIP: Unleashing Diffusion Priors for Joint Foreground and Alpha Prediction in Image Matting 38

DRIP: Unleashing Diffusion Priors for Joint Foreground and A...

引用

38th Conference on Neural Information Processing systems, NeurIPS 2024

作者： Li, Xiaodi Yang, Zongxin Quan, Ruijie Yang, Yi The State Key Lab of Brain-Machine Intelligence Zhejiang University Hangzhou China CCAI College of Computer Science and Technology Zhejiang University Hangzhou China DBMI HMS Harvard University Boston United States

Recovering the foreground color and opacity/alpha matte from a single image (i.e., image matting) is a challenging and ill-posed problem where data priors play a critical role in achieving precise results. Due to the limited matting datasets, traditional methods usually struggle to produce high-quality *** address this, we explore the potential of leveraging vision priors embedded in pre-trained latent diffusion models (LDM) for estimating foreground RGBA values in challenging scenarios and rare objects. We introduce Drip, a novel approach for image matting that harnesses the rich prior knowledge of LDM models. Our method incorporates a switcher and a cross-domain attention mechanism to extend the original LDM for joint prediction of the foreground color and opacity. This setup facilitates mutual information exchange and ensures high consistency across both modalities. To mitigate the inherent reconstruction errors of the LDM's VAE decoder, we propose a latent transparency decoder to align the RGBA prediction with the input image, thereby reducing discrepancies. Comprehensive experimental results demonstrate that our approach achieves state-of-the-art performance in foreground and alpha predictions and shows remarkable generalizability across various benchmarks. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-perspective Improvement of Knowledge Graph Completion with Large Language Models 30

Multi-perspective Improvement of Knowledge Graph Completion ...

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Xu, Derong Zhang, Ziheng Lin, Zhenxi Wu, Xian Zhu, Zhihong Xu, Tong Zhao, Xiangyu Zheng, Yefeng Chen, Enhong University of Science and Technology of China State Key Laboratory of Cognitive Intelligence China Jarvis Research Center Tencent YouTu Lab China City University of Hong Kong Hong Kong Peking University China

ISBN: (纸本)9782493814104

Knowledge graph completion (KGC) is a widely used method to tackle incompleteness in knowledge graphs (KGs) by making predictions for missing links. Description-based KGC leverages pre-trained language models to learn entity and relation representations with their names or descriptions, which shows promising results. However, the performance of description-based KGC is still limited by the quality of text and the incomplete structure, as it lacks sufficient entity descriptions and relies solely on relation names, leading to sub-optimal results. To address this issue, we propose MPIKGC, a general framework to compensate for the deficiency of contextualized knowledge and improve KGC by querying large language models (LLMs) from various perspectives, which involves leveraging the reasoning, explanation, and summarization capabilities of LLMs to expand entity descriptions, understand relations, and extract structures, respectively. We conducted extensive evaluation of the effectiveness and improvement of our framework based on four description-based KGC models and four datasets, for both link prediction and triplet classification tasks. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Knowledge graph

来源：评论

学校读者我要写书评

暂无评论

High-efficiency crystalline carbon nitride photocatalysts:Status and perspectives

引用

Nano Research 2024年第9期17卷 7840-7863页

作者： Wenji Pu Yunqiao Zhou Lingfeng Yang Haifeng Gong Yuhan Li Qingyu Yang Dieqing Zhang Engineering Research Center for Waste Oil Recovery Technology and Equipment Ministry of EducationChongqing Key Laboratory of Catalysis and New Environmental MaterialsChongqing Technology and Business UniversityChongqing 400067China The Education Ministry Key Lab of Resource Chemistry Joint International Research Laboratory of Resource Chemistry of Ministry of EducationShanghai Key Laboratory of Rare Earth Functional Materialsand Shanghai Frontiers Science Center of Biomimetic CatalysisShanghai Normal UniversityShanghai 200234China State Key Laboratory of Tibetan Plateau Earth System Resources and EnvironmentInstitute of Tibetan Plateau ResearchChinese Academy of SciencesBeijing 100101China

Crystallinity and crystal structure greatly influence the photocatalytic behavior of *** g-C3N4 produced by traditional thermal-induced polycondensation reaction bears low crystallinity and thus poor photoactivity,which originates from the incomplete polymerization of the precursor containing amine groups,abundant hydrogen bonds,and unreacted amino,as well as cyanide functional groups in the *** photocatalytic process,these residual functional groups often work as electron trap sites,which may hinder the transfer of electrons on the plane,resulting in low ***,crystalline carbon nitride(CCN)was reported as a promising photocatalyst because its increased crystallinity not only reduces the number of carriers recombination centers,but also increases charge conductivity and improves light utilization due to extendedπ-conjugated systems and delocalizedπ-*** such,we summarize the recent studies on CCN-based photocatalysts for the photoactivity ***,the unique structure and properties of CCN materials are ***,the preparation methods and modification strategies are well *** also sum up the applications of CCN-based materials in the environmental purification and energy ***,this review concerning CNN materials ends with prospects and challenges in the obtainment of high crystallinity by effective techniques,and the deep understanding of photocatalytic mechanism.

关键词： crystalline carbon nitride preparation method modification strategy photocatalytic application

来源：评论

学校读者我要写书评

暂无评论

Separation and bias of deep equilibrium models on expressivity and learning dynamics 24

Separation and bias of deep equilibrium models on expressivi...

引用

Proceedings of the 38th International Conference on Neural Information Processing systems

作者： Zhoutong Wu Yimu Zhang Cong Fang Zhouchen Lin Academy for Advanced Interdisciplinary Studies Peking University State Key Lab of General AI School of Intelligence Science and Technology Peking University State Key Lab of General AI School of Intelligence Science and Technology Peking University and Institute for Artificial Intelligence Peking University State Key Lab of General AI School of Intelligence Science and Technology Peking University and Institute for Artificial Intelligence Peking University and Pazhou Laboratory (Huangpu) Guangzhou Guangdong China

ISBN: (纸本)9798331314385

The deep equilibrium model (DEQ) generalizes the conventional feedforward neural network by fixing the same weights for each layer block and extending the number of layers to infinity. This novel model directly finds the fixed points of such a forward process as features for prediction. Despite empirical evidence showcasing its efficacy compared to feedforward neural networks, a theoretical understanding for its separation and bias is still limited. In this paper, we take a step by proposing some separations and studying the bias of DEQ in its expressive power and learning dynamics. The results include: (1) A general separation is proposed, showing the existence of a width-m DEQ that any fully connected neural networks (FNNs) with depth O(mα) for α ∈ (0,1) cannot approximate unless its width is sub-exponential in m; (2) DEQ with polynomially bounded size and magnitude can efficiently approximate certain steep functions (which has very large derivatives) in L∞ norm, whereas FNN with bounded depth and exponentially bounded width cannot unless its weights magnitudes are exponentially large; (3) The implicit regularization caused by gradient flow from a diagonal linear DEQ is characterized, with specific examples showing the benefits brought by such regularization. From the overall study, a high-level conjecture from our analysis and empirical validations is that DEQ has potential advantages in learning certain high-frequency components.

关键词：

来源：评论

学校读者我要写书评

暂无评论

A Theoretical Understanding of Self-Correction through In-context Alignment 38

A Theoretical Understanding of Self-Correction through In-co...

引用

38th Conference on Neural Information Processing systems, NeurIPS 2024

作者： Wang, Yifei Wu, Yuyang Wei, Zeming Jegelka, Stefanie Wang, Yisen MIT CSAIL United States School of EECS Peking University China School of Mathematical Sciences Peking University China State Key Lab of General Artificial Intelligence School of Intelligence Science and Technology Peking University China CIT MCML MDSI TU Munich Germany Institute for Artificial Intelligence Peking University China

Going beyond mimicking limited human experiences, recent studies show initial evidence that, like humans, large language models (LLMs) are capable of improving their abilities purely by self-correction, i.e., correcting previous responses through self-examination, as seen in models like OpenAI o1. Nevertheless, little is known about how such capabilities arise. In this work, based on a simplified setup akin to an alignment task, we theoretically analyze self-correction from an in-context learning perspective, showing that when LLMs give relatively accurate self-examinations as rewards, they are capable of refining responses in an in-context way. Notably, going beyond previous theories on over-simplified linear transformers, our theoretical construction underpins the roles of several key designs of realistic transformers for self-correction: softmax attention, multi-head attention, and the MLP block. We validate these findings extensively on synthetic datasets. Inspired by these findings, we propose a simple self-correction strategy, Checking as Context (CaC), which finds novel applications in alleviating social bias and defending against LLM jailbreaks. We believe that these findings will inspire further research on understanding, exploiting, and enhancing self-correction for building better foundation models. Code is at https://***/yifeiwang77/Self-Correction. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Minimizing Data Collection Latency for Coexisting Time-Critical Wireless Networks with Tree Topologies

引用

IEEE Transactions on Network and Service Management 2025年第3期22卷 2354-2369页

作者： Zhang, Jialin Liang, Wei Yang, Bo Shi, Huaguang Liang, Ying-Chang Chinese Academy of Sciences State Key Laboratory of Robotics Shenyang Institute of Automation Shenyang110016 China Chinese Academy of Sciences Key Laboratory of Networked Control System Shenyang Institute of Automation Shenyang110016 China Northwest A and F University College of Information Engineering Yangling712100 China Henan University School of Artificial Intelligence Zhengzhou450046 China University of Electronic Science and Technology of China Center for Intelligent Networking and Communications Chengdu611731 China

Time-Critical Wireless Network (TCWN) is a promising communication technology that can satisfy the low latency, high reliability, and deterministic requirements of mission-critical applications. Multiple TCWNs required by various applications inevitably coexist with each other. Most existing works aim to achieve acceptable latency or consider the simplest topology (i.e., line topology). As latency requirements become more stringent, exploring the minimum data collection latency becomes an interesting problem. In this paper, the coexisting system consists of multiple tree-topology-based TCWNs. We first establish a conversion framework to convert an arbitrary tree topology into multiple analogous line topologies to reduce the analysis complexity. We then propose a Time-Critical wireless network Scheduling (TCS) algorithm to minimize the data collection latency of coexisting TCWNs. The TCS algorithm consists of two phases. In the internetwork scheduling phase, we strictly derive a general expression to characterize the practical network requirements. In the intranetwork scheduling phase, we design two levels of priority assignment algorithms to accurately characterize the critical states and resource requirements of different nodes. We conduct extensive simulations to verify the effectiveness of the TCS algorithm. The evaluation results show that the TCS algorithm can achieve minimum data collection latency in more than 99.956% cases, and the maximum difference compared to the optimal value is one time slot. © 2004-2012 IEEE.

关键词： Scheduling algorithms

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：