检索结果-内蒙古大学图书馆

IEEE Workshop on Applications of Computer Vision (WACV)

作者： Wendi Yang Zihang Jiang Shang Zhao S. Kevin Zhou Division of Life Sciences and Medicine School of Biomedical Engineering University of Science and Technology of China (USTC) Hefei Anhui China Center for Medical Imaging Robotics Analytic Computing & Learning (MIRACLE) Suzhou Institute for Advance Research USTC Suzhou Jiangsu China Key Laboratory of Precision and Intelligent Chemistry USTC Hefei Anhui China Key Laboratory of Intelligent Information Processing of Chinese Academy of Sciences (CAS) Institute of Computing Technology CAS

ISBN: (数字)9798331510831

ISBN: (纸本)9798331510848

With the recent advancements in single-image-based 3D human pose and shape estimation (3DHPSE), there is a growing amount of works that can achieve good results on standard benchmarks but struggle to yield accurate hu-man mesh in extreme scenarios like occlusion. Previous works propose to leverage 2D poses to help 3D HPSE model improve performance under occlusion, but usually rely on manual design to integrate 2D poses and only aim for specific kinds of occlusion. In this paper, we present PostoMETRO (Pose token enhanced MEsh TRansfOrmer), which integrates 2D pose prior knowledge as tokens into transformers to improve model's performance under occlusion. Using a VQ- VAE-based pose tokenizer; we efficiently represent 2D poses as tokens andfeed them to transformers together with image tokens. Subsequently, these tokens are queried by vertex andJoint tokens to decode 3D coordinates of mesh vertices and human Joints. Our proposed 2D poses integration strategy is manual-design-free and suitable for various kinds of occlusion. Experiments on both standard and occlusion-specific benchmarks demonstrate the effectiveness of PostoMETRO. Code will be made available 1 1 https://***/PostoMETRO/PostoMETRO-Paper.

关键词： Solid modeling Computer vision Three-dimensional displays Shape Estimation Manuals Benchmark testing Transformers Robustness Standards

来源：评论

学校读者我要写书评

暂无评论

Lifting the Answer: Reranking Candidates on Data Augmented Text-to-SQL

Lifting the Answer: Reranking Candidates on Data Augmented T...

引用

International Conference on Machine Learning and Cybernetics (ICMLC)

作者： Guang-Yu Cheng Shuo Yu Wen-Bin Jiang Shuai Ma Xiang Ao Qing He Henan Institutes of Advanced Technology Zhengzhou University Zhengzhou P.R. China Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS) Institute of Computing Technology CAS Beijing China Baidu inc Beijing China

The text-to-SQL task is a realistic and challenging job, which aims to translate natural language questions into corresponding SQL queries. When attempting to generate SQL queries in the text-to-SQL task, prevailing semantic parsing models employ beam search to generate several candidates. Based on our pilot study, we observe that the gold-standard SQL answer may exist in the n-best candidate list produced by the decoder, rather than the first. Hence, we aim to lift the correct answer given several candidates generated by semantic parsing models for text-to-SQL. To this end, we propose a reranking module that reorders the n-best list of candidate SQL queries by pair-wise hinge loss. Meanwhile, a data augmentation module is leveraged to enrich the inadequate training instances for providing candidates with better qualities. These two modules can be easily grafted onto text-to-SQL backbone networks, and extensive experiments on the cross-domain text-to-SQL benchmark Spider demonstrate that our method achieves 73.8% in accuracy on the Spider dataset, surpassing the base model by up to 3.5%.

关键词：

来源：评论

学校读者我要写书评

暂无评论

LEARNING MONOTONIC ATTENTION IN TRANSDUCER FOR STREAMING GENERATION

arXiv

引用

arXiv 2024年

作者： Ma, Zhengrui Feng, Yang Zhang, Min Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China Key Laboratory of AI Safety Chinese Academy of Sciences China University of Chinese Academy of Sciences China School of Future Science and Engineering Soochow University China

Streaming generation models are increasingly utilized across various fields, with the Transducer architecture being particularly popular in industrial applications. However, its input-synchronous decoding mechanism presents challenges in tasks requiring non-monotonic alignments, such as simultaneous translation, leading to suboptimal performance in these contexts. In this research, we address this issue by tightly integrating Transducer’s decoding with the history of input stream via a learnable monotonic attention mechanism. Our approach leverages the forward-backward algorithm to infer the posterior probability of alignments between the predictor states and input timestamps, which is then used to estimate the context representations of monotonic attention in training. This allows Transducer models to adaptively adjust the scope of attention based on their predictions, avoiding the need to enumerate the exponentially large alignment space. Extensive experiments demonstrate that our MonoAttn-Transducer significantly enhances the handling of non-monotonic alignments in streaming generation, offering a robust solution for Transducer-based frameworks to tackle more complex streaming generation tasks. Codes: https://***/ictnlp/MonoAttn-Transducer. Copyright © 2024, The Authors. All rights reserved.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

A Non-autoregressive Generation Framework for End-to-End Simultaneous Speech-to-Speech Translation

arXiv

引用

arXiv 2024年

作者： Ma, Zhengrui Fang, Qingkai Zhang, Shaolei Guo, Shoutao Feng, Yang Zhang, Min Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China Key Laboratory of AI Safety Chinese Academy of Sciences China University of Chinese Academy of Sciences China School of Future Science and Engineering Soochow University China

Simultaneous translation models play a crucial role in facilitating communication. However, existing research primarily focuses on text-to-text or speech-to-text models, necessitating additional cascade components to achieve speech-to-speech translation. These pipeline methods suffer from error propagation and accumulate delays in each cascade component, resulting in reduced synchronization between the speaker and listener. To overcome these challenges, we propose a novel non-autoregressive generation framework for simultaneous speech translation (NAST-S2x1), which integrates speech-to-text and speech-to-speech tasks into a unified end-to-end framework. We develop a non-autoregressive decoder capable of concurrently generating multiple text or acoustic unit tokens upon receiving fixed-length speech chunks. The decoder can generate blank or repeated tokens and employ CTC decoding to dynamically adjust its latency. Experimental results show that NAST-S2x outperforms state-of-the-art models in both speech-to-text and speech-to-speech tasks. It achieves high-quality simultaneous interpretation within a delay of less than 3 seconds and provides a 28× decoding speedup in offline generation. Copyright © 2024, The Authors. All rights reserved.

关键词： Decoding

来源：评论

学校读者我要写书评

暂无评论

Non-autoregressive Streaming Transformer for Simultaneous Translation

arXiv

引用

arXiv 2023年

作者： Ma, Zhengrui Zhang, Shaolei Guo, Shoutao Shao, Chenze Zhang, Min Feng, Yang Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China School of Future Science and Engineering Soochow University China

Simultaneous machine translation (SiMT) models are trained to strike a balance between latency and translation quality. However, training these models to achieve high quality while maintaining low latency often leads to a tendency for aggressive anticipation. We argue that such issue stems from the autoregressive architecture upon which most existing SiMT models are built. To address those issues, we propose non-autoregressive streaming Transformer (NAST) which comprises a unidirectional encoder and a non-autoregressive decoder with intra-chunk parallelism. We enable NAST to generate the blank token or repetitive tokens to adjust its READ/WRITE strategy flexibly, and train it to maximize the non-monotonic latent alignment with an alignment-based latency loss. Experiments on various SiMT benchmarks demonstrate that NAST outperforms previous strong autoregressive SiMT baselines. Source code is publicly available at https://***/ictnlp/NAST. Copyright © 2023, The Authors. All rights reserved.

关键词： Machine translation

来源：评论

学校读者我要写书评

暂无评论

MoCE: Adaptive Mixture of Contextualization Experts for Byte-based Neural Machine Translation

arXiv

引用

arXiv 2024年

作者： Huang, Langlin Bu, Mengyu Feng, Yang Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China Key Laboratory of AI Safety Chinese Academy of Sciences China University of Chinese Academy of Sciences China Washington University St. Louis United States

Byte-based machine translation systems have shown significant potential in massively multilingual settings. Unicode encoding, which maps each character to specific byte(s), eliminates the emergence of unknown words, even in new languages. This avoids out-of-vocabulary risk in multilingual translation and enables broad language scalability. However, byte-level tokenization results in sequences that are hard to interpret due to limited semantic information per byte. Local contextualization has proven effective in assigning initial semantics to tokens, improving sentence comprehension. Nevertheless, variations in encoding rules across languages necessitate an adaptive approach for effective contextualization. To this end, we propose Mixture of Contextualization Experts (MoCE), adaptively selecting and mixing attention heads, which are treated as contextualization experts. This enhances the flexibility of contextualization scales and allows models to search for better contextualization combinations. Experiment results show that our method outperforms existing methods without extensive manual adjustment of hyper-parameters and surpasses subword-based models with fewer parameters in Ted-59 dataset. Our code is available at https://***/ictnlp/MoCE. © 2024, CC BY-NC-ND.

关键词： Encoding (symbols)

来源：评论

学校读者我要写书评

暂无评论

HiDiff: Hybrid Diffusion Framework for Medical Image Segmentation

arXiv

引用

arXiv 2024年

作者： Chen, Tao Wang, Chenhui Chen, Zhihao Lei, Yiming Shan, Hongming Institute of Science and Technology for Brain-inspired Intelligence Fudan University Shanghai200433 China Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai200433 China

Medical image segmentation has been significantly advanced with the rapid development of deep learning (DL) techniques. Existing DL-based segmentation models are typically discriminative;i.e., they aim to learn a mapping from the input image to segmentation masks. However, these discriminative methods neglect the underlying data distribution and intrinsic class characteristics, suffering from unstable feature space. In this work, we propose to complement discriminative segmentation methods with the knowledge of underlying data distribution from generative models. To that end, we propose a novel hybrid diffusion framework for medical image segmentation, termed HiDiff, which can synergize the strengths of existing discriminative segmentation models and new generative diffusion models. HiDiff comprises two key components: discriminative segmentor and diffusion refiner. First, we utilize any conventional trained segmentation models as discriminative segmentor, which can provide a segmentation mask prior for diffusion refiner. Second, we propose a novel binary Bernoulli diffusion model (BBDM) as the diffusion refiner, which can effectively, efficiently, and interactively refine the segmentation mask by modeling the underlying data distribution. Third, we train the segmentor and BBDM in an alternate-collaborative manner to mutually boost each other. Extensive experimental results on abdomen organ, brain tumor, polyps, and retinal vessels segmentation datasets, covering four widely-used modalities, demonstrate the superior performance of HiDiff over existing medical segmentation algorithms, including the state-of-the-art transformer- and diffusion-based ones. In addition, HiDiff excels at segmenting small objects and generalizing to new datasets. Source codes are made available at https://***/takimailto/HiDiff. Copyright © 2024, The Authors. All rights reserved.

关键词： Diffusion

来源：评论

学校读者我要写书评

暂无评论

Beyond MLE: Convex Learning for Text Generation

arXiv

引用

arXiv 2023年

作者： Shao, Chenze Ma, Zhengrui Zhang, Min Feng, Yang Key Laboratory of Intelligent Information Processing Institute of Computing Technology Chinese Academy of Sciences China University of Chinese Academy of Sciences China School of Future Science and Engineering Soochow University China

Maximum likelihood estimation (MLE) is a statistical method used to estimate the parameters of a probability distribution that best explain the observed data. In the context of text generation, MLE is often used to train generative language models, which can then be used to generate new text. However, we argue that MLE is not always necessary and optimal, especially for closed-ended text generation tasks like machine translation. In these tasks, the goal of model is to generate the most appropriate response, which does not necessarily require it to estimate the entire data distribution with MLE. To this end, we propose a novel class of training objectives based on convex functions, which enables text generation models to focus on highly probable outputs without having to estimate the entire data distribution. We investigate the theoretical properties of the optimal predicted distribution when applying convex functions to the loss, demonstrating that convex functions can sharpen the optimal distribution, thereby enabling the model to better capture outputs with high probabilities. Experiments on various text generation tasks and models show the effectiveness of our approach. It enables autoregressive models to bridge the gap between greedy and beam search, and facilitates the learning of non-autoregressive models with a maximum improvement of 9+ BLEU points. Moreover, our approach also exhibits significant impact on large language models (LLMs), substantially enhancing their generative capability on various tasks. Source code is available at https://***/ictnlp/Convex-Learning. Copyright © 2023, The Authors. All rights reserved.

关键词： Maximum likelihood estimation

来源：评论

学校读者我要写书评

暂无评论

VANER: Leveraging Large Language Model for Versatile and Adaptive Biomedical Named Entity Recognition 27

VANER: Leveraging Large Language Model for Versatile and Ada...

引用

27th European Conference on Artificial Intelligence, ECAI 2024

作者： Bian, Junyi Zhai, Weiqi Huang, Xiaodi Zheng, Jiaxuan Zhu, Shanfeng School of Computer Science Fudan University Shanghai200433 China Institute of Science and Technology for Brain-Inspired Intelligence Fudan University China Ministry of Education Shanghai200433 China MOE Frontiers Center for Brain Science Fudan University Shanghai200433 China Zhangjiang Fudan International Innovation Center Shanghai200433 China Shanghai Key Lab of Intelligent Information Processing Fudan University Shanghai200433 China School of Computing and Mathematics Charles Sturt University AlburyNSW2640 Australia

ISBN: (纸本)9781643685489

The prevalent solution for BioNER involves using representation learning techniques combined with sequence ***, such methods are inherently task-specific, demonstrate poor generalizability, and often require a dedicated model for each *** leverage the versatile capabilities of recent large language models (LLMs), several approaches have explored generative techniques for entity ***, these approaches often fall short compared to previous sequence labeling *** this paper, we utilize the open-sourced LLM LLaMA2 as the backbone model, and design specific instructions to distinguish between different types of entities and *** combining the LLM's understanding of instructions with sequence labeling techniques, we train a model using a mix of datasets capable of extracting various types of *** that the backbone LLMs lacks specialized medical knowledge, we also integrate external entity knowledge bases and employ instruction tuning to enable the model to densely recognize curated *** parameter-efficient training model, VANER, significantly outperforms previous LLMs-based *** the first time, as an LLM-based model, VANER surpasses the majority of conventional state-of-the-art BioNER systems, achieving the highest F1 scores across three datasets. © 2024 The Authors.

关键词： labeled data

来源：评论

学校读者我要写书评

暂无评论

EC-SLAM: Effectively Constrained Neural RGB-D SLAM with Sparse TSDF Encoding and Global Bundle Adjustment

arXiv

引用

arXiv 2024年

作者： Li, Guanghao Chen, Qi Yan, Yuxiang Pu, Jian Institute of Science and Technology for Brain-Inspired Intelligence Fudan University Shanghai200433 China Shanghai Key Lab of Intelligent Information Processing School of Computer Science Fudan University Shanghai200433 China

We introduce EC-SLAM, a real-time dense RGB-D simultaneous localization and mapping (SLAM) system leveraging Neural Radiance Fields (NeRF). While recent NeRF-based SLAM systems have shown promising results, they have yet to fully exploit NeRF’s potential to constrain pose optimization. EC-SLAM addresses this by using sparse parametric encodings and Truncated Signed Distance Fields (TSDF) to represent the map, enabling efficient fusion, reducing model parameters, and accelerating convergence. Our system also employs a globally constrained Bundle Adjustment (BA) strategy that capitalizes on NeRF’s implicit loop closure correction capability, improving tracking accuracy by reinforcing constraints on keyframes most relevant to the current optimized frame. Furthermore, by integrating a feature-based and uniform sampling strategy that minimizes ineffective constraint points for pose optimization, we reduce the impact of random sampling in NeRF. Extensive evaluations on the Replica, ScanNet, and TUM datasets demonstrate state-of-the-art performance, with precise tracking and reconstruction accuracy achieved alongside real-time operation at up to 21 Hz. The source code is available at https://***/Lightingooo/EC-SLAM. Copyright © 2024, The Authors. All rights reserved.

关键词： Mapping

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：