Searcharxiv⌕ Search

arXiv subjects

Jie Zhou

Publications and source records attributed to Jie Zhou.

At least 523 records · Page 29Linked to original sources
arXiv2023

Human-M3: A Multi-view Multi-modal Dataset for 3D Human Pose Estimation in Outdoor Scenes

3D human pose estimation in outdoor environments has garnered increasing attention recently. However, prevalent 3D human pose datasets pertaining to outdoor scenes lack diversity, as they predominantly utilize only one type of modality (RGB image or pointcloud), and often feature only one individual within each scene. This limited scope of dataset infrastructure considerably hinders the variability of available data. In this article, we propose Human-M3, an outdoor multi-modal multi-view multi-person human pose database which includes not only multi-view RGB videos of outdoor scenes but also corresponding pointclouds. In order to obtain accurate human poses, we propose an algorithm based on multi-modal data input to generate ground truth annotation. This benefits from robust pointcloud detection and tracking, which solves the problem of inaccurate human localization and matching ambiguity that may exist in previous multi-view RGB videos in outdoor multi-person scenes, and generates reliable ground truth annotations. Evaluation of multiple different modalities algorithms has shown that this database is challenging and suitable for future research. Furthermore, we propose a 3D human pose estimation algorithm based on multi-modal data input, which demonstrates the advantages of multi-modal data input for 3D human pose estimation. Code and data will be released on https://github.com/soullessrobot/Human-M3-Dataset.

cs.CV↗
arXiv2023

EduChat: A Large-Scale Language Model-based Chatbot System for Intelligent Education

EduChat (https://www.educhat.top/) is a large-scale language model (LLM)-based chatbot system in the education domain. Its goal is to support personalized, fair, and compassionate intelligent education, serving teachers, students, and parents. Guided by theories from psychology and education, it further strengthens educational functions such as open question answering, essay assessment, Socratic teaching, and emotional support based on the existing basic LLMs. Particularly, we learn domain-specific knowledge by pre-training on the educational corpus and stimulate various skills with tool use by fine-tuning on designed system prompts and instructions. Currently, EduChat is available online as an open-source project, with its code, data, and model parameters available on platforms (e.g., GitHub https://github.com/icalk-nlp/EduChat, Hugging Face https://huggingface.co/ecnu-icalk ). We also prepare a demonstration of its capabilities online (https://vimeo.com/851004454). This initiative aims to promote research and applications of LLMs for intelligent education.

cs.CL↗
arXiv2023

A Confidence-based Partial Label Learning Model for Crowd-Annotated Named Entity Recognition

Existing models for named entity recognition (NER) are mainly based on large-scale labeled datasets, which always obtain using crowdsourcing. However, it is hard to obtain a unified and correct label via majority voting from multiple annotators for NER due to the large labeling space and complexity of this task. To address this problem, we aim to utilize the original multi-annotator labels directly. Particularly, we propose a Confidence-based Partial Label Learning (CPLL) method to integrate the prior confidence (given by annotators) and posterior confidences (learned by models) for crowd-annotated NER. This model learns a token- and content-dependent confidence via an Expectation-Maximization (EM) algorithm by minimizing empirical risk. The true posterior estimator and confidence estimator perform iteratively to update the true posterior and confidence respectively. We conduct extensive experimental results on both real-world and synthetic datasets, which show that our model can improve performance effectively compared with strong baselines.

cs.CL↗
arXiv2023

Partial Linear Cox Model with Deep ReLU Networks for Interval-Censored Failure Time Data

The partial linear Cox model for interval-censoring is well-studied under the additive assumption but is still under-investigated without this assumption. In this paper, we propose to use a deep ReLU neural network to estimate the nonparametric components of a partial linear Cox model for interval-censored data. This model not only retains the nice interpretability of the parametric component but also improves the predictive power compared to the partial linear additive Cox model. We derive the convergence rate of the proposed estimator and show that it can break the curse of dimensionality under some certain smoothness assumptions. Based on such rate, the asymptotic normality and the semiparametric efficiency are also established. Intensive simulation studies are carried out to demonstrate the finite sample performance on both estimation and prediction. The proposed estimation procedure is illustrated on a real dataset.

stat.ME↗
arXiv2023

Soft Language Clustering for Multilingual Model Pre-training

Multilingual pre-trained language models have demonstrated impressive (zero-shot) cross-lingual transfer abilities, however, their performance is hindered when the target language has distant typology from source languages or when pre-training data is limited in size. In this paper, we propose XLM-P, which contextually retrieves prompts as flexible guidance for encoding instances conditionally. Our XLM-P enables (1) lightweight modeling of language-invariant and language-specific knowledge across languages, and (2) easy integration with other multilingual pre-training methods. On the tasks of XTREME including text classification, sequence labeling, question answering, and sentence retrieval, both base- and large-size language models pre-trained with our proposed method exhibit consistent performance improvement. Furthermore, it provides substantial advantages for low-resource languages in unsupervised sentence retrieval and for target languages that differ greatly from the source language in cross-lingual transfer.

cs.CL↗
arXiv2023

Triple Sequence Learning for Cross-domain Recommendation

Cross-domain recommendation (CDR) aims to leverage the correlation of users' behaviors in both the source and target domains to improve the user preference modeling in the target domain. Conventional CDR methods typically explore the dual-relations between the source and target domains' behaviors. However, this may ignore the informative mixed behaviors that naturally reflect the user's global preference. To address this issue, we present a novel framework, termed triple sequence learning for cross-domain recommendation (Tri-CDR), which jointly models the source, target, and mixed behavior sequences to highlight the global and target preference and precisely model the triple correlation in CDR. Specifically, Tri-CDR independently models the hidden representations for the triple behavior sequences and proposes a triple cross-domain attention (TCA) method to emphasize the informative knowledge related to both user's global and target-domain preference. To comprehensively explore the cross-domain correlations, we design a triple contrastive learning (TCL) strategy that simultaneously considers the coarse-grained similarities and fine-grained distinctions among the triple sequences, ensuring the alignment while preserving information diversity in multi-domain. We conduct extensive experiments and analyses on six cross-domain settings. The significant improvements of Tri-CDR with different sequential encoders verify its effectiveness and universality. The code will be released upon acceptance.

cs.IR↗
arXiv2023

Dynamic Spatial Sparsification for Efficient Vision Transformers and Convolutional Neural Networks

In this paper, we present a new approach for model acceleration by exploiting spatial sparsity in visual data. We observe that the final prediction in vision Transformers is only based on a subset of the most informative tokens, which is sufficient for accurate image recognition. Based on this observation, we propose a dynamic token sparsification framework to prune redundant tokens progressively and dynamically based on the input to accelerate vision Transformers. Specifically, we devise a lightweight prediction module to estimate the importance score of each token given the current features. The module is added to different layers to prune redundant tokens hierarchically. While the framework is inspired by our observation of the sparse attention in vision Transformers, we find the idea of adaptive and asymmetric computation can be a general solution for accelerating various architectures. We extend our method to hierarchical models including CNNs and hierarchical vision Transformers as well as more complex dense prediction tasks that require structured feature maps by formulating a more generic dynamic spatial sparsification framework with progressive sparsification and asymmetric computation for different spatial locations. By applying lightweight fast paths to less informative features and using more expressive slow paths to more important locations, we can maintain the structure of feature maps while significantly reducing the overall computations. Extensive experiments demonstrate the effectiveness of our framework on various modern architectures and different visual recognition tasks. Our results clearly demonstrate that dynamic spatial sparsification offers a new and more effective dimension for model acceleration. Code is available at https://github.com/raoyongming/DynamicViT

cs.CV↗
arXiv2023

Monocrystalline Si/$β$-Ga$_2$O$_3$ p-n heterojunction diodes fabricated via grafting

The $β$-Ga$_2$O$_3$ has exceptional electronic properties with vast potential in power and RF electronics. Despite the excellent demonstrations of high-performance unipolar devices, the lack of p-type doping in $β$-Ga$_2$O$_3$ has hindered the development of Ga$_2$O$_3$-based bipolar devices. The approach of p-n diodes formed by polycrystalline p-type oxides with n-type $β$-Ga$_2$O$_3$ can face severe challenges in further advancing the $β$-Ga$_2$O$_3$ bipolar devices due to their unfavorable band alignment and the poor p-type oxide crystal quality. In this work, we applied the semiconductor grafting approach to fabricate monocrystalline Si/$β$-Ga$_2$O$_3$ p-n diodes for the first time. With enhanced concentration of oxygen atoms at the interface of Si/$β$-Ga$_2$O$_3$, double side surface passivation was achieved for both Si and $β$-Ga$_2$O$_3$ with an interface Dit value of 1-3 x 1012 /cm2 eV. A Si/$β$-Ga$_2$O$_3$ p-n diode array with high fabrication yield was demonstrated along with a diode rectification of 1.3 x 107 at +/- 2 V, a diode ideality factor of 1.13 and avalanche reverse breakdown characteristics. The diodes C-V shows frequency dispersion-free characteristics from 10 kHz to 2 MHz. Our work has set the foundation toward future development of $β$-Ga$_2$O$_3$-based transistors.

physics.app-ph↗
arXiv2023

Stochastic Bridges as Effective Regularizers for Parameter-Efficient Tuning

Parameter-efficient tuning methods (PETs) have achieved promising results in tuning large pre-trained language models (PLMs). By formalizing frozen PLMs and additional tunable parameters as systems and controls respectively, PETs can be theoretically grounded to optimal control and further viewed as optimizing the terminal cost and running cost in the optimal control literature. Despite the elegance of this theoretical grounding, in practice, existing PETs often ignore the running cost and only optimize the terminal cost, i.e., focus on optimizing the loss function of the output state, regardless of the running cost that depends on the intermediate states. Since it is non-trivial to directly model the intermediate states and design a running cost function, we propose to use latent stochastic bridges to regularize the intermediate states and use the regularization as the running cost of PETs. As the first work to propose regularized PETs that use stochastic bridges as the regularizers (running costs) for the intermediate states, we show the effectiveness and generality of this regularization across different tasks, PLMs and PETs. In view of the great potential and capacity, we believe more sophisticated regularizers can be designed for PETs and better performance can be achieved in the future. The code is released at \url{https://github.com/thunlp/stochastic-bridge-pet/tree/main}.

cs.CL↗
arXiv2023

VisionLLM: Large Language Model is also an Open-Ended Decoder for Vision-Centric Tasks

Large language models (LLMs) have notably accelerated progress towards artificial general intelligence (AGI), with their impressive zero-shot capacity for user-tailored tasks, endowing them with immense potential across a range of applications. However, in the field of computer vision, despite the availability of numerous powerful vision foundation models (VFMs), they are still restricted to tasks in a pre-defined form, struggling to match the open-ended task capabilities of LLMs. In this work, we present an LLM-based framework for vision-centric tasks, termed VisionLLM. This framework provides a unified perspective for vision and language tasks by treating images as a foreign language and aligning vision-centric tasks with language tasks that can be flexibly defined and managed using language instructions. An LLM-based decoder can then make appropriate predictions based on these instructions for open-ended tasks. Extensive experiments show that the proposed VisionLLM can achieve different levels of task customization through language instructions, from fine-grained object-level to coarse-grained task-level customization, all with good results. It's noteworthy that, with a generalist LLM-based framework, our model can achieve over 60\% mAP on COCO, on par with detection-specific models. We hope this model can set a new baseline for generalist vision and language models. The demo shall be released based on https://github.com/OpenGVLab/InternGPT. The code shall be released at https://github.com/OpenGVLab/VisionLLM.

cs.CV↗
arXiv2023

WebCPM: Interactive Web Search for Chinese Long-form Question Answering

Long-form question answering (LFQA) aims at answering complex, open-ended questions with detailed, paragraph-length responses. The de facto paradigm of LFQA necessitates two procedures: information retrieval, which searches for relevant supporting facts, and information synthesis, which integrates these facts into a coherent answer. In this paper, we introduce WebCPM, the first Chinese LFQA dataset. One unique feature of WebCPM is that its information retrieval is based on interactive web search, which engages with a search engine in real time. Following WebGPT, we develop a web search interface. We recruit annotators to search for relevant information using our interface and then answer questions. Meanwhile, the web search behaviors of our annotators would be recorded. In total, we collect 5,500 high-quality question-answer pairs, together with 14,315 supporting facts and 121,330 web search actions. We fine-tune pre-trained language models to imitate human behaviors for web search and to generate answers based on the collected facts. Our LFQA pipeline, built on these fine-tuned models, generates answers that are no worse than human-written ones in 32.5% and 47.5% of the cases on our dataset and DuReader, respectively.

cs.CL↗
arXiv2023

Farewell to Aimless Large-scale Pretraining: Influential Subset Selection for Language Model

Pretrained language models have achieved remarkable success in various natural language processing tasks. However, pretraining has recently shifted toward larger models and larger data, and this has resulted in significant computational and energy costs. In this paper, we propose Influence Subset Selection (ISS) for language model, which explicitly utilizes end-task knowledge to select a tiny subset of the pretraining corpus. Specifically, the ISS selects the samples that will provide the most positive influence on the performance of the end-task. Furthermore, we design a gradient matching based influence estimation method, which can drastically reduce the computation time of influence. With only 0.45% of the data and a three-orders-of-magnitude lower computational cost, ISS outperformed pretrained models (e.g., RoBERTa) on eight datasets covering four domains.

cs.CL↗
arXiv2023

Mitigating Catastrophic Forgetting in Task-Incremental Continual Learning with Adaptive Classification Criterion

Task-incremental continual learning refers to continually training a model in a sequence of tasks while overcoming the problem of catastrophic forgetting (CF). The issue arrives for the reason that the learned representations are forgotten for learning new tasks, and the decision boundary is destructed. Previous studies mostly consider how to recover the representations of learned tasks. It is seldom considered to adapt the decision boundary for new representations and in this paper we propose a Supervised Contrastive learning framework with adaptive classification criterion for Continual Learning (SCCL), In our method, a contrastive loss is used to directly learn representations for different tasks and a limited number of data samples are saved as the classification criterion. During inference, the saved data samples are fed into the current model to obtain updated representations, and a k Nearest Neighbour module is used for classification. In this way, the extensible model can solve the learned tasks with adaptive criteria of saved samples. To mitigate CF, we further use an instance-wise relation distillation regularization term and a memory replay module to maintain the information of previous tasks. Experiments show that SCCL achieves state-of-the-art performance and has a stronger ability to overcome CF compared with the classification baselines.

cs.LG↗
arXiv2023

GFDC: A Granule Fusion Density-Based Clustering with Evidential Reasoning

Currently, density-based clustering algorithms are widely applied because they can detect clusters with arbitrary shapes. However, they perform poorly in measuring global density, determining reasonable cluster centers or structures, assigning samples accurately and handling data with large density differences among clusters. To overcome their drawbacks, this paper proposes a granule fusion density-based clustering with evidential reasoning (GFDC). Both local and global densities of samples are measured by a sparse degree metric first. Then information granules are generated in high-density and low-density regions, assisting in processing clusters with significant density differences. Further, three novel granule fusion strategies are utilized to combine granules into stable cluster structures, helping to detect clusters with arbitrary shapes. Finally, by an assignment method developed from Dempster-Shafer theory, unstable samples are assigned. After using GFDC, a reasonable clustering result and some identified outliers can be obtained. The experimental results on extensive datasets demonstrate the effectiveness of GFDC.

cs.LG↗
arXiv2023

Generalized Iterative Formula for Bell Inequalities

Bell inequalities are a vital tool to detect the nonlocal correlations, but the construction of them for multipartite systems is still a complicated problem. In this work, inspired via a decomposition of $(n+1)$-partite Bell inequalities into $n$-partite ones, we present a generalized iterative formula to construct nontrivial $(n+1)$-partite ones from the $n$-partite ones. Our iterative formulas recover the well-known Mermin-Ardehali-Belinski{\uı}-Klyshko (MABK) and other families in the literature as special cases. Moreover, a family of ``dual-use'' Bell inequalities is proposed, in the sense that for the generalized Greenberger-Horne-Zeilinger states these inequalities lead to the same quantum violation as the MABK family and, at the same time, the inequalities are able to detect the non-locality in the entire entangled region. Furthermore, we present generalizations of the the I3322 inequality to any $n$-partite case which are still tight, and of the $46$ Śliwa's inequalities to the four-partite tight ones, by applying our iteration method to each inequality and its equivalence class.

quant-ph↗
arXiv2023

DC-MBR: Distributional Cooling for Minimum Bayesian Risk Decoding

Minimum Bayesian Risk Decoding (MBR) emerges as a promising decoding algorithm in Neural Machine Translation. However, MBR performs poorly with label smoothing, which is surprising as label smoothing provides decent improvement with beam search and improves generality in various tasks. In this work, we show that the issue arises from the un-consistency of label smoothing on the token-level and sequence-level distributions. We demonstrate that even though label smoothing only causes a slight change in the token-level, the sequence-level distribution is highly skewed. We coin the issue \emph{autoregressive over-smoothness}. To address this issue, we propose a simple and effective method, Distributional Cooling MBR (DC-MBR), which manipulates the entropy of output distributions by tuning down the Softmax temperature. We theoretically prove the equivalence between pre-tuning label smoothing factor and distributional cooling. Extensive experiments on NMT benchmarks validate that distributional cooling improves MBR in various settings.

cs.CL↗
arXiv2023

Unified Model Learning for Various Neural Machine Translation

Existing neural machine translation (NMT) studies mainly focus on developing dataset-specific models based on data from different tasks (e.g., document translation and chat translation). Although the dataset-specific models have achieved impressive performance, it is cumbersome as each dataset demands a model to be designed, trained, and stored. In this work, we aim to unify these translation tasks into a more general setting. Specifically, we propose a ``versatile'' model, i.e., the Unified Model Learning for NMT (UMLNMT) that works with data from different tasks, and can translate well in multiple settings simultaneously, and theoretically it can be as many as possible. Through unified learning, UMLNMT is able to jointly train across multiple tasks, implementing intelligent on-demand translation. On seven widely-used translation tasks, including sentence translation, document translation, and chat translation, our UMLNMT results in substantial improvements over dataset-specific models with significantly reduced model deployment costs. Furthermore, UMLNMT can achieve competitive or better performance than state-of-the-art dataset-specific methods. Human evaluation and in-depth analysis also demonstrate the superiority of our approach on generating diverse and high-quality translations. Additionally, we provide a new genre translation dataset about famous aphorisms with 186k Chinese->English sentence pairs.

cs.CL↗
arXiv2023

WeLM: A Well-Read Pre-trained Language Model for Chinese

Large Language Models pre-trained with self-supervised learning have demonstrated impressive zero-shot generalization capabilities on a wide spectrum of tasks. In this work, we present WeLM: a well-read pre-trained language model for Chinese that is able to seamlessly perform different types of tasks with zero or few-shot demonstrations. WeLM is trained with 10B parameters by "reading" a curated high-quality corpus covering a wide range of topics. We show that WeLM is equipped with broad knowledge on various domains and languages. On 18 monolingual (Chinese) tasks, WeLM can significantly outperform existing pre-trained models with similar sizes and match the performance of models up to 25 times larger. WeLM also exhibits strong capabilities in multi-lingual and code-switching understanding, outperforming existing multilingual language models pre-trained on 30 languages. Furthermore, We collected human-written prompts for a large set of supervised datasets in Chinese and fine-tuned WeLM with multi-prompted training. The resulting model can attain strong generalization on unseen types of tasks and outperform the unsupervised WeLM in zero-shot learning. Finally, we demonstrate that WeLM has basic skills at explaining and calibrating the decisions from itself, which can be promising directions for future research. Our models can be applied from https://welm.weixin.qq.com/docs/api/.

cs.CL↗