检索结果-内蒙古大学图书馆

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Li, Jiatong Hu, Renjun Huang, Kunzhe Zhuang, Yan Liu, Qi Zhu, Mengxiao Shi, Xing Lin, Wei University of Science and Technology of China China Alibaba Cloud Computing China

Expert-designed close-ended benchmarks are indispensable in assessing the knowledge capacity of large language models (LLMs). Despite their widespread use, concerns have mounted regarding their reliability due to limited test scenarios and an unavoidable risk of data contamination. To rectify this, we present PertEval, a toolkit devised for in-depth probing of LLMs' knowledge capacity through knowledge-invariant perturbations. These perturbations employ human-like restatement techniques to generate on-the-fly test samples from static benchmarks, meticulously retaining knowledge-critical content while altering irrelevant details. Our toolkit further includes a suite of response consistency analyses that compare performance on raw vs. perturbed test sets to precisely assess LLMs' genuine knowledge capacity. Six representative LLMs are re-evaluated using PertEval. Results reveal significantly inflated performance of the LLMs on raw benchmarks, including an absolute 25.8% overestimation for GPT-4. Additionally, through a nuanced response pattern analysis, we discover that PertEval retains LLMs' uncertainty to specious knowledge, and reveals their potential rote memorization to correct options which leads to overestimated performance. We also find that the detailed response consistency analyses by PertEval could illuminate various weaknesses in existing LLMs' knowledge mastery and guide the development of refinement. Our findings provide insights for advancing more robust and genuinely knowledgeable LLMs. Our code is available at https://***/aigc-apps/PertEval. © 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

VideoCLIP-XL: Advancing Long Description Understanding for Video CLIP Models

VideoCLIP-XL: Advancing Long Description Understanding for V...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Wang, Jiapeng Wang, Chengyu Huang, Kunzhe Huang, Jun Jin, Lianwen South China University of Technology Guangzhou China Alibaba Cloud Computing Hangzhou China

ISBN: (纸本)9798891761643

Contrastive Language-Image Pre-training (CLIP) has been widely studied and applied in numerous applications. However, the emphasis on brief summary texts during pre-training prevents CLIP from understanding long descriptions. This issue is particularly acute regarding videos given that videos often contain abundant detailed contents. In this paper, we propose the VideoCLIP-XL (eXtra Length) model, which aims to unleash the long-description understanding capability of video CLIP models. Firstly, we establish an automatic data collection system and gather a large-scale VILD pre-training dataset with VIdeo and Long-Description pairs. Then, we propose Text-similarity-guided Primary Component Matching (TPCM) to better learn the distribution of feature space while expanding the long description capability. We also introduce two new tasks namely Detail-aware Description Ranking (DDR) and Hallucination-aware Description Ranking (HDR) for further understanding improvement. Finally, we construct a Long Video Description Ranking (LVDR) benchmark for evaluating the long-description capability more comprehensively. Extensive experimental results on widely-used text-video retrieval benchmarks with both short and long descriptions and our LVDR benchmark can fully demonstrate the effectiveness of our method. © 2024 Association for Computational Linguistics.

关键词： Video analysis

来源：评论

学校读者我要写书评

暂无评论

Building a Family of Data Augmentation Models for Low-cost LLM Fine-tuning on the cloud 31

Building a Family of Data Augmentation Models for Low-cost L...

引用

31st International Conference on Computational Linguistics, COLING 2025

作者： Yue, Yuanhao Wang, Chengyu Huang, Jun Wang, Peng School of Computer Science Fudan University Shanghai China Alibaba Cloud Computing Hangzhou China

ISBN: (纸本)9798891761971

Specializing LLMs in various domain-specific tasks has emerged as a critical step towards achieving high performance. However, the construction and annotation of datasets in specific domains are always very costly. Apart from using superior and expensive closed-source LLM APIs to construct datasets, some open-source models have become strong enough to handle dataset construction in many scenarios. Thus, we present a family of data augmentation models designed to significantly improve the efficiency for model fine-tuning. These models, trained based on sufficiently small LLMs, support key functionalities with low inference costs: instruction expansion, instruction refinement, and instruction-response pair expansion. To fulfill this goal, we first construct an automatic data collection system with seed datasets generated from both public repositories and our in-house datasets. This system leverages powerful LLMs to expand, refine and re-write the instructions and responses, incorporating quality assessment techniques. Following this, we introduce the training process of our models, which effectively distills task-solving and text synthesis abilities from teacher LLMs. Finally, we demonstrate how we integrate these functionalities into a machine learning platform to support low-cost LLM fine-tuning from both dataset preparation and training perspectives for users. Experiments and an application study prove the effectiveness of our approach. ©2025 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Distilling Instruction-following Abilities of Large Language Models with Task-aware Curriculum Planning

Distilling Instruction-following Abilities of Large Language...

引用

2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024

作者： Yue, Yuanhao Wang, Chengyu Huang, Jun Wang, Peng School of Computer Science Fudan University Shanghai China Alibaba Cloud Computing Hangzhou China

ISBN: (纸本)9798891761681

Instruction tuning aims to align large language models (LLMs) with open-domain instructions and human-preferred responses. While several studies have explored autonomous approaches to distilling and annotating instructions from powerful proprietary LLMs, such as ChatGPT, they often neglect the impact of the distributions and characteristics of tasks, together with the varying difficulty of instructions in training sets. This oversight can lead to imbalanced knowledge capabilities and poor generalization powers of student LLMs. To address these challenges, we introduce Task-Aware Curriculum Planning for Instruction Refinement (TAPIR), a multi-round distillation framework that utilizes an oracle LLM to select instructions that are difficult for a student LLM to follow. To balance the student's capabilities, task distributions in training sets are adjusted with responses automatically refined according to their corresponding tasks. In addition, by incorporating curriculum planning, our approach systematically escalates the difficulty levels of tasks, progressively enhancing the student LLM's capabilities. We rigorously evaluate TAPIR using several widely recognized benchmarks (such as AlpacaEval 2.0, MT-Bench, etc.) and multiple student LLMs. Empirical results demonstrate that student LLMs, trained with our method and less training data, outperform larger instruction-tuned models and strong distillation baselines. © 2024 Association for Computational Linguistics.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

BOOSTING VANILLA LIGHTWEIGHT VISION TRANSFORMERS VIA RE-PARAMETERIZATION 12

BOOSTING VANILLA LIGHTWEIGHT VISION TRANSFORMERS VIA RE-PARA...

引用

12th International Conference on Learning Representations, ICLR 2024

作者： Tan, Zhentao Li, Xiaodan Wu, Yue Chu, Qi Lu, Le Yu, Nenghai Ye, Jieping Alibaba Cloud China Alibaba Group China University of Science and Technology of China China East China Normal University China

Large-scale Vision Transformers have achieved promising performance on downstream tasks through feature pre-training. However, the performance of vanilla lightweight Vision Transformers (ViTs) is still far from satisfactory compared to that of recent lightweight CNNs or hybrid networks. In this paper, we aim to unlock the potential of vanilla lightweight ViTs by exploring the adaptation of the widely-used re-parameterization technology to ViTs for improving learning ability during training without increasing the inference cost. The main challenge comes from the fact that CNNs perfectly complement with re-parameterization over convolution and batch normalization, while vanilla Transformer architectures are mainly comprised of linear and layer normalization layers. We propose to incorporate the linear ensemble into linear layers by expanding the depth of the linear layers with batch normalization and fusing multiple linear features with hierarchical representation ability through a pyramid structure. We also discover and solve a new transformer-specific distribution rectification problem caused by multi-branch re-parameterization. Finally, we propose our Two-Dimensional Re-parameterized Linear module (TDRL) for ViTs. Under the popular self-supervised pre-training and supervised fine-tuning strategy, our TDRL can be used in these two stages to enhance both generic and task-specific representation. Experiments demonstrate that our proposed method not only boosts the performance of vanilla Vit-Tiny on various vision tasks to new state-of-the-art (SOTA) but also shows promising generality ability on other networks. Code will be available. © 2024 12th International Conference on Learning Representations, ICLR 2024. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Inductive Relation Prediction by Disentangled Subgraph Structure

引用

Tsinghua science and technology 2024年第5期29卷 1566-1579页

作者： Guiduo Duan Rui Guo Wenlong Luo Guangchun Luo Tianxi Huang Laboratory of Intelligent Collaborative Computing University of Electronic Science and Technology of ChinaChengdu 611731Chinaand also with Trusted Cloud Computing and Big Data Key Laboratory of Sichuan ProvinceChengdu 611731China School of Computer Science and Engineering University of Electronic Science and Technology of ChinaChengdu 611731China Hangzhou NetEase Cloud Music Technology Co. Ltd.Hangzhou 310052China Department of Fundamental Courses Chengdu Textile CollegeChengdu 611700China

Currently,most existing inductive relation prediction approaches are based on subgraph structures,with subgraph features extracted using graph neural networks to predict ***,subgraphs may contain disconnected regions,which usually represent different semantic *** not all semantic information about the regions is helpful in relation prediction,we propose a relation prediction model based on a disentangled subgraph structure and implement a feature updating approach based on relevant semantic *** indirectly achieve the disentangled subgraph structure from a semantic perspective,the mapping of entity features into different semantic spaces and the aggregation of related semantics on each semantic space are *** disentangled model can focus on features having higher semantic relevance in the prediction,thus addressing a problem with existing approaches,which ignore the semantic differences in different subgraph ***,using a gated recurrent neural network,this model enhances the features of entities by sorting them by distance and extracting the path information in the ***,it is shown that when there are numerous disconnected regions in the subgraph,our model outperforms existing mainstream models in terms of both Area Under the Curve-Precision-Recall(AUC-PR)and Hits@*** prove that semantic differences in the knowledge graph can be effectively distinguished and verify the effectiveness of this method.

关键词： disentangled subgraph structure knowledge graph completion Gated Recurrent Unit(GRU) feature updating

来源：评论

学校读者我要写书评

暂无评论

Efficient Quality Controllable Neural Image Compression based on QD-Model

Efficient Quality Controllable Neural Image Compression base...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Wang, Shaokang Xiang, Guoqing Xu, Jinchang Zhang, Shanghang Xie, Xiaodong School of Computer Science Peking University Beijing China Alibaba Cloud Computing Beijing China

ISBN: (纸本)9798350368741

Neural image compression has achieved significant advancements, consistently outperforming traditional codecs in terms of performance. However, research on quality control algorithms for neural image compression is still lacking. In this paper, we propose a framework designed to control the quality of compressed images through a one-pass pre-analysis. First, we construct a foundational relationship between the quantization factor and compression distortion, utilizing variable rate neural image compression as the basis for quality control. Second, we introduce the image Content-Compression features-based Distortion Estimation Network (C2DEN) to efficiently fit the sample-adaptive Quantization-Distortion (QD) model. Leveraging the QD model, we convert the target quality into a quantization factor to control the compression model, enabling quality-controllable compression of samples. Experimental results show that the average quality errors on four different datasets are only 0.89%, 1.79%, 1.73%, and 1.61%. Compared with existing control methods, our method reduces the algorithm time complexity by 98.58%, 98.52%, 98.85%, and 98.50% while ensuring accuracy, which further demonstrates the superiority of our method. © 2025 IEEE.

关键词： linear model Neural image compression quality control quantization factor

来源：评论

学校读者我要写书评

暂无评论

Plan-on-Graph: Self-Correcting Adaptive Planning of Large Language Model on Knowledge Graphs 38

Plan-on-Graph: Self-Correcting Adaptive Planning of Large La...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Chen, Liyi Tong, Panrong Jin, Zhongming Sun, Ying Ye, Jieping Xiong, Hui University of Science and Technology of China China Alibaba Cloud Computing China Thrust of Artificial Intelligence The Hong Kong University of Science and Technology Guangzhou China Department of Computer Science and Engineering The Hong Kong University of Science and Technology Hong Kong

Large Language Models (LLMs) have shown remarkable reasoning capabilities on complex tasks, but they still suffer from out-of-date knowledge, hallucinations, and opaque decision-making. In contrast, Knowledge Graphs (KGs) can provide explicit and editable knowledge for LLMs to alleviate these issues. Existing paradigm of KG-augmented LLM manually predefines the breadth of exploration space and requires flawless navigation in KGs. However, this paradigm cannot adaptively explore reasoning paths in KGs based on the question semantics and self-correct erroneous reasoning paths, resulting in a bottleneck in efficiency and effect. To address these limitations, we propose a novel self-correcting adaptive planning paradigm for KG-augmented LLM named Plan-on-Graph (PoG), which first decomposes the question into several sub-objectives and then repeats the process of adaptively exploring reasoning paths, updating memory, and reflecting on the need to self-correct erroneous reasoning paths until arriving at the answer. Specifically, three important mechanisms of Guidance, Memory, and Reflection are designed to work together, to guarantee the adaptive breadth of self-correcting planning for graph reasoning. Finally, extensive experiments on three real-world datasets demonstrate the effectiveness and efficiency of PoG. © 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

DARC: High-dimensional Diffusing Anomaly Detection and Root Cause Location in cloud computing Systems

DARC: High-dimensional Diffusing Anomaly Detection and Root ...

引用

2023 IEEE International Conference on Big Data, BigData 2023

作者： Sun, Wanning Ma, Xuhua Peng, Ruimin Xu, Yifan Wu, Dongrui Alibaba Cloud Alibaba Group Beijing China Alibaba Cloud Alibaba Group Hangzhou China Huazhong University of Science and Technology School of Artificial Intelligence and Automation Wuhan China

ISBN: (纸本)9798350324457

The modern cloud computing system has evolved into a highly dynamic and complex ecosystem with thousands of modules. Modifications and updates to these modules occur every day to accommodate customer needs. Unfortunately, these frequent changes may introduce anomalies to the system, whose diffusion can undermine the system performance and even cause system outage. Though very important, it is challenging to detect anomalies at the early stages and locate their root causes, due to the complexity of the cloud ecosystem and the huge number of attribute combinations. This paper proposes DARC for high-dimensional diffusing anomaly detection and root cause location in cloud computing systems. DARC uses first two-stage percentile analysis and Mann-Kendall score thresholding to detect rare anomalies, and then a bottom-up search strategy with three computational complexity reduction techniques to efficiently locate the root causes. Extensive experiments showed that DARC is able to accurately and efficiently locate root causes of diffusing anomalies. It has been successfully used in the daily practice of alibaba cloud, one of the world's largest cloud computing service providers. © 2023 IEEE.

关键词： cloud computing high-dimensional diffusing anomaly detection root cause location

来源：评论

学校读者我要写书评

暂无评论

Modelling and Temperature Control of Liquid Cooling Process for Lithium-Ion Battery

引用

Journal of Thermal science 2024年第5期33卷 1794-1808页

作者： LI Guanru ZHONG Yangfan BAO Chujin CHEN Yongping ZHANG Chengbin School of Energy and Environment Southeast UniversityNanjing 210096China Key Laboratory of Efficient Low-Carbon Energy Conversion and Utilization of Jiangsu Provincial Higher Education Institutions Suzhou University of Science and TechnologySuzhou 215009China Alibaba Cloud Computing Co.Ltd. Hangzhou 311121China Zhejiang Immertech Co.Ltd. Hangzhou 310000China

Efficient thermal management of lithium-ion battery,working under extremely rapid charging-discharging,is of widespread interest to avoid the battery degradation due to temperature rise,resulting in the enhanced ***,thermal management of lithium-ion battery has been performed via a liquid cooling theoretical model integrated with thermoelectric model of battery packs and single-phase heat *** to alleviate the battery temperature fluctuation by automatically manipulating the flow rate of working fluid,a nominal model-free controller,i.e.,fuzzy logic controller is *** optimized on-off controller based on pump speed optimization is introduced to serve as the comparative *** control simulations are conducted under regular operating and extreme operating conditions,and two controllers are applied to control battery temperature with proper intervals which is conducive to enhance the battery charge-discharge *** results indicate that,for any operating condition,the fuzzy logic controller shows excellence in terms of the tracking accuracy of set-point of battery *** to the establishment of fuzzy set and fuzzy behavioral rules,the battery temperature has been throughout maintained near the set point,and the temperature fluctuation amplitude is highly reduced,with better temperature control accuracy of~0.2℃(regular condition)and~0.5℃(extreme condition)compared with~1.1℃(regular condition)and~1.6℃(extreme condition)of optimized on-off *** in the case of extreme operating condition,the proposed optimized on-off controller manifests the hysteresis in temperature fluctuation,which is ascribed to the set of dead-band for the feedback *** simulation results cast new light on the utilization and development of model-free temperature controller for the thermal management of lithium-ion battery.

关键词： temperature control battery thermal management model-free controller fuzzy logic control

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：