Search Results for author: Jun Huang

Found 106 papers, 39 papers with code

Meta Distant Transfer Learning for Pre-trained Language Models

no code implementations EMNLP 2021 Chengyu Wang, Haojie Pan, Minghui Qiu, Jun Huang, Fei Yang, Yin Zhang

For tasks related to distant domains with different class label sets, PLMs may memorize non-transferable knowledge for the target domain and suffer from negative transfer.

Implicit Relations Meta-Learning +2

MTADiffusion: Mask Text Alignment Diffusion Model for Object Inpainting

no code implementations CVPR 2025 Jun Huang, Ting Liu, Yihang Wu, Xiaochao Qu, Luoqi Liu, Xiaolin Hu

Advancements in generative models have enabled image inpainting models to generate content within specific regions of an image based on provided prompts and masks.

Image Inpainting

Toward Safety-First Human-Like Decision Making for Autonomous Vehicles in Time-Varying Traffic Flow

no code implementations17 Jun 2025 Xiao Wang, Junru Yu, Jun Huang, Qiong Wu, Ljubo Vacic, Changyin Sun

Despite the recent advancements in artificial intelligence technologies have shown great potential in improving transport efficiency and safety, autonomous vehicles(AVs) still face great challenge of driving in time-varying traffic flow, especially in dense and interactive situations.

Autonomous Driving Decision Making

Infinity Parser: Layout Aware Reinforcement Learning for Scanned Document Parsing

1 code implementation1 Jun 2025 Baode Wang, Biao Wu, Weizhen Li, Meng Fang, Yanjie Liang, Zuming Huang, Haozhe Wang, Jun Huang, Ling Chen, Wei Chu, Yuan Qi

Automated parsing of scanned documents into richly structured, machine-readable formats remains a critical bottleneck in Document AI, as traditional multi-stage pipelines suffer from error propagation and limited adaptability to diverse layouts.

document understanding Language Modeling +5

EasyDistill: A Comprehensive Toolkit for Effective Knowledge Distillation of Large Language Models

no code implementations27 May 2025 Chengyu Wang, Junbing Yan, Wenrui Cai, Yuanhao Yue, Jun Huang

In this paper, we present EasyDistill, a comprehensive toolkit designed for effective black-box and white-box knowledge distillation (KD) of large language models (LLMs).

Knowledge Distillation

Reasoning with OmniThought: A Large CoT Dataset with Verbosity and Cognitive Difficulty Annotations

no code implementations16 May 2025 Wenrui Cai, Chengyu Wang, Junbing Yan, Jun Huang, Xiangzhong Fang

Based on the proposed OmniThought dataset, we further train and release a series of high-performing LRMs, specifically equipped with stronger reasoning abilities and optimal CoT output length and difficulty level.

Code Generation Mathematical Problem-Solving

Deep Learning in Renewable Energy Forecasting: A Cross-Dataset Evaluation of Temporal and Spatial Models

no code implementations6 May 2025 Lutfu Sua, Haibo Wang, Jun Huang

Unpredictability of renewable energy sources coupled with the complexity of those methods used for various purposes in this area calls for the development of robust methods such as DL models within the renewable energy domain.

Hyperparameter Optimization L2 Regularization

CoordField: Coordination Field for Agentic UAV Task Allocation In Low-altitude Urban Scenarios

no code implementations30 Apr 2025 Tengchao Zhang, Yonglin Tian, Fei Lin, Jun Huang, Patrik P. Süli, Rui Qin, Fei-Yue Wang

With the increasing demand for heterogeneous Unmanned Aerial Vehicle (UAV) swarms to perform complex tasks in urban environments, system design now faces major challenges, including efficient semantic understanding, flexible task planning, and the ability to dynamically adjust coordination strategies in response to evolving environmental conditions and continuously changing task requirements.

Task Planning

DistilQwen2.5: Industrial Practices of Training Distilled Open Lightweight Language Models

no code implementations21 Apr 2025 Chengyu Wang, Junbing Yan, Yuanhao Yue, Jun Huang

These distilled models exhibit enhanced instruction-following capabilities compared to the original models based on a series of distillation techniques that incorporate knowledge from much larger LLMs.

Computational Efficiency Instruction Following

Understanding Attention Mechanism in Video Diffusion Models

no code implementations16 Apr 2025 Bingyan Liu, Chengyu Wang, Tongtong Su, Huan Ten, Jun Huang, Kailing Guo, Kui Jia

Notably, high-entropy attention maps are often key elements linked to superior video quality, whereas low-entropy attention maps are associated with the video's intra-frame structure.

Video Editing

Training Small Reasoning LLMs with Cognitive Preference Alignment

no code implementations14 Apr 2025 Wenrui Cai, Chengyu Wang, Junbing Yan, Jun Huang, Xiangzhong Fang

We further propose the cognitive preference optimization (CogPO) algorithm to enhance the reasoning abilities of smaller models by aligning thoughts of these models with their cognitive capacities.

AirVista-II: An Agentic System for Embodied UAVs Toward Dynamic Scene Semantic Understanding

no code implementations13 Apr 2025 Fei Lin, Yonglin Tian, Tengchao Zhang, Jun Huang, Sangtian Guan, Fei-Yue Wang

Unmanned Aerial Vehicles (UAVs) are increasingly important in dynamic environments such as logistics transportation and disaster response.

Disaster Response Scheduling

Spatiotemporal Impact of Trade Policy Variables on Asian Manufacturing Hubs: Bayesian Global Vector Autoregression Model

no code implementations22 Mar 2025 Lutfu S. Sua, Haibo Wang, Jun Huang

A novel spatiotemporal framework using diverse econometric approaches is proposed in this research to analyze relationships among eight economy-wide variables in varying market conditions.

Time Series

Beyond Existance: Fulfill 3D Reconstructed Scenes with Pseudo Details

no code implementations6 Mar 2025 YiFei Gao, Jun Huang, Lei Wang, Ruiting Dai, Jun Cheng

The emergence of 3D Gaussian Splatting (3D-GS) has significantly advanced 3D reconstruction by providing high fidelity and fast training speeds across various scenarios.

3D Reconstruction

Renewable Energy Prediction: A Comparative Study of Deep Learning Models for Complex Dataset Analysis

no code implementations27 Jan 2025 Haibo Wang, Jun Huang, Lutfu Sua, Bahram Alidaee

The results demonstrate that the combination of early stopping, dropout, and L1 regularization provides the best performance to reduce overfitting in the CNN and TD-MLP models with larger training set, while the combination of early stopping, dropout, and L2 regularization is the most effective to reduce the overfitting in CNN-LSTM and AE models with smaller training set.

Hyperparameter Optimization L2 Regularization

UAVs Meet LLMs: Overviews and Perspectives Toward Agentic Low-Altitude Mobility

1 code implementation4 Jan 2025 Yonglin Tian, Fei Lin, Yiduo Li, Tengchao Zhang, Qiyao Zhang, Xuan Fu, Jun Huang, Xingyuan Dai, Yutong Wang, Chunwei Tian, Bai Li, Yisheng Lv, Levente Kovács, Fei-Yue Wang

Low-altitude mobility, exemplified by unmanned aerial vehicles (UAVs), has introduced transformative advancements across various domains, like transportation, logistics, and agriculture.

Fingerprinting Denoising Diffusion Probabilistic Models

no code implementations CVPR 2025 Huan Teng, Yuhui Quan, Chengyu Wang, Jun Huang, Hui Ji

Diffusion models, especially denoising diffusion probabilistic models (DDPMs), are prevalent tools in generative AI, making their intellectual property (IP) protection increasingly important.

Denoising

Encapsulated Composition of Text-to-Image and Text-to-Video Models for High-Quality Video Synthesis

no code implementations CVPR 2025 Tongtong Su, Chengyu Wang, Bingyan Liu, Jun Huang, Dongming Lu

By encapsulating the T2V temporal-only prior into the T2I generation process, EVS successfully leverages the strengths of both types of models, resulting in videos of improved imaging and motion quality.

Denoising

DocLayLLM: An Efficient Multi-modal Extension of Large Language Models for Text-rich Document Understanding

1 code implementation CVPR 2025 Wenhui Liao, Jiapeng Wang, Hongliang Li, Chengyu Wang, Jun Huang, Lianwen Jin

By lightly integrating visual patch tokens and 2D positional tokens into LLMs' input and encoding the document content using the LLMs themselves, we fully take advantage of the document comprehension capability of LLMs and enhance their perception of OCR information.

document understanding Optical Character Recognition (OCR)

Data-Juicer 2.0: Cloud-Scale Adaptive Data Processing for and with Foundation Models

2 code implementations23 Dec 2024 Daoyuan Chen, Yilun Huang, Xuchen Pan, Nana Jiang, Haibin Wang, Yilei Zhang, Ce Ge, Yushuo Chen, WenHao Zhang, Zhijian Ma, Jun Huang, Wei Lin, Yaliang Li, Bolin Ding, Jingren Zhou

The burgeoning field of foundation models necessitates advanced data processing mechanisms capable of harnessing vast and valuable data with various types used by these models.

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

no code implementations6 Dec 2024 Yuanhao Yue, Chengyu Wang, Jun Huang, Peng Wang

Specializing LLMs in various domain-specific tasks has emerged as a critical step towards achieving high performance.

Data Augmentation

An Open API Architecture to Discover the Trustworthy Explanation of Cloud AI Services

1 code implementation5 Nov 2024 Zerui Wang, Yan Liu, Jun Huang

This article presents the design of an open-API-based explainable AI (XAI) service to provide feature contribution explanations for cloud AI services.

Radical-mediated Electrical Enzyme Assay For At-home Clinical Test

no code implementations3 Nov 2024 Hyun-June Jang, Hyou-Arm Joung, Xiaoao Shi, Rui Ding, Justine Wagner, Erting Tang, Wen Zhuang, Byunghoon Ryu, Guanmin Chen, Kiang-Teck Jerry Yeo, Jun Huang, Junhong Chen

To meet the growing demand for accurate, rapid, and cost-effective at-home clinical testing, we developed a radical-mediated enzyme assay (REEA) integrated with a paper fluidic system and electrically read by a handheld field-effect transistor (FET) device.

Evaluating Semantic Variation in Text-to-Image Synthesis: A Causal Perspective

1 code implementation14 Oct 2024 Xiangru Zhu, Penglei Sun, Yaoxian Song, Yanghua Xiao, Zhixu Li, Chengyu Wang, Jun Huang, Bei Yang, Xiaoxiao Xu

To address these deficiencies, we propose a novel metric called SemVarEffect and a benchmark named SemVarBench, designed to evaluate the causality between semantic variations in inputs and outputs in T2I synthesis.

cross-modal alignment Image Generation

Spatio-Temporal Encoding and Decoding-Based Method for Future Human Activity Skeleton Synthesis

no code implementations8 Jul 2024 Tingyu Liu, Jun Huang, Chenyi Weng

Inferring future activity information based on observed activity data is a crucial step to improve the accuracy of early activity prediction.

Activity Prediction Decoder +1

On the Role of Long-tail Knowledge in Retrieval Augmented Large Language Models

no code implementations24 Jun 2024 Dongyang Li, Junbing Yan, Taolin Zhang, Chengyu Wang, Xiaofeng He, Longtao Huang, Hui Xue, Jun Huang

Retrieval augmented generation (RAG) exhibits outstanding performance in promoting the knowledge capabilities of large language models (LLMs) with retrieved documents related to user queries.

RAG Retrieval +2

Will Southeast Asia be the next global manufacturing hub? A multiway cointegration, causality, and dynamic connectedness analyses on factors influencing offshore decisions

no code implementations11 Jun 2024 Haibo Wang, Lutfu S. Sua, Jun Huang, Jaime Ortiz, Bahram Alidaee

The results of the cointegration, causality, and dynamic connectedness analyses show that a subset of Southeast Asian countries have spillover effects on each other.

DAFNet: Dynamic Auxiliary Fusion for Sequential Model Editing in Large Language Models

1 code implementation31 May 2024 Taolin Zhang, Qizhou Chen, Dongyang Li, Chengyu Wang, Xiaofeng He, Longtao Huang, Hui Xue, Jun Huang

(2) Considering that auxiliary parameters are required to store the knowledge for sequential editing, we construct a new dataset named \textbf{DAFSet}, fulfilling recent, popular, long-tail and robust properties to enhance the generality of sequential editing.

Hallucination Model Editing

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

no code implementations22 May 2024 Yuanhao Yue, Chengyu Wang, Jun Huang, Peng Wang

In addition, by incorporating curriculum planning, our approach systematically escalates the difficulty levels of tasks, progressively enhancing the student LLM's capabilities.

Code Generation Instruction Following +1

Exploring Graph-based Knowledge: Multi-Level Feature Distillation via Channels Relational Graph

no code implementations14 May 2024 Zhiwei Wang, Jun Huang, Longhua Ma, Chengyu Wu, Hongyu Ma

In visual tasks, large teacher models capture essential features and deep information, enhancing performance.

R4: Reinforced Retriever-Reorder-Responder for Retrieval-Augmented Large Language Models

no code implementations4 May 2024 Taolin Zhang, Dongyang Li, Qizhou Chen, Chengyu Wang, Longtao Huang, Hui Xue, Xiaofeng He, Jun Huang

The reordering learning process is divided into two steps according to the quality of the generated responses: document order adjustment and document representation enhancement.

Graph Attention Hallucination +5

Guided AbsoluteGrad: Magnitude of Gradients Matters to Explanation's Localization and Saliency

1 code implementation23 Apr 2024 Jun Huang, Yan Liu

This paper proposes a new gradient-based XAI method called Guided AbsoluteGrad for saliency map explanations.

AlphaFin: Benchmarking Financial Analysis with Retrieval-Augmented Stock-Chain Framework

1 code implementation19 Mar 2024 Xiang Li, Zhenyu Li, Chen Shi, Yong Xu, Qing Du, Mingkui Tan, Jun Huang, Wei Lin

The task of financial analysis primarily encompasses two key areas: stock trend prediction and the corresponding financial question answering.

Benchmarking Financial Analysis +5

DiffChat: Learning to Chat with Text-to-Image Synthesis Models for Interactive Image Creation

no code implementations8 Mar 2024 Jiapeng Wang, Chengyu Wang, Tingfeng Cao, Jun Huang, Lianwen Jin

We present DiffChat, a novel method to align Large Language Models (LLMs) to "chat" with prompt-as-input Text-to-Image Synthesis (TIS) models (e. g., Stable Diffusion) for interactive image creation.

Image Generation Instruction Following +1

Towards Understanding Cross and Self-Attention in Stable Diffusion for Text-Guided Image Editing

no code implementations CVPR 2024 Bingyan Liu, Chengyu Wang, Tingfeng Cao, Kui Jia, Jun Huang

Deep Text-to-Image Synthesis (TIS) models such as Stable Diffusion have recently gained significant popularity for creative Text-to-image generation.

Denoising text-guided-image-editing +1

Do Large Language Models Understand Logic or Just Mimick Context?

no code implementations19 Feb 2024 Junbing Yan, Chengyu Wang, Jun Huang, Wei zhang

Over the past few years, the abilities of large language models (LLMs) have received extensive attention, which have performed exceptionally well in complicated scenarios such as logical reasoning and symbolic inference.

counterfactual In-Context Learning +1

From Complex to Simple: Unraveling the Cognitive Tree for Reasoning with Small Language Models

no code implementations12 Nov 2023 Junbing Yan, Chengyu Wang, Taolin Zhang, Xiaofeng He, Jun Huang, Wei zhang

Reasoning is a distinctive human capacity, enabling us to address complex problems by breaking them down into a series of manageable cognitive steps.

Language Modelling Logical Reasoning

Sharing, Teaching and Aligning: Knowledgeable Transfer Learning for Cross-Lingual Machine Reading Comprehension

no code implementations12 Nov 2023 Tingfeng Cao, Chengyu Wang, Chuanqi Tan, Jun Huang, Jinhui Zhu

In cross-lingual language understanding, machine translation is often utilized to enhance the transferability of models across languages, either by translating the training data from the source language to the target, or from the target to the source to aid inference.

Cross-Lingual Transfer Machine Reading Comprehension +2

BeautifulPrompt: Towards Automatic Prompt Engineering for Text-to-Image Synthesis

no code implementations12 Nov 2023 Tingfeng Cao, Chengyu Wang, Bingyan Liu, Ziheng Wu, Jinhui Zhu, Jun Huang

Then, to ensure that our generated prompts can generate more beautiful images, we further propose a Reinforcement Learning with Visual AI Feedback technique to fine-tune our model to maximize the reward values of the generated prompts, where the reward values are calculated based on the PickScore and the Aesthetic Scores.

Prompt Engineering Text to Image Generation +1

Uncertainty-aware Parameter-Efficient Self-training for Semi-supervised Language Understanding

1 code implementation19 Oct 2023 Jianing Wang, Qiushi Sun, Nuo Chen, Chengyu Wang, Jun Huang, Ming Gao, Xiang Li

The recent success of large pre-trained language models (PLMs) heavily hinges on massive labeled data, which typically produces inferior performance in low-resource scenarios.

Hierarchical Side-Tuning for Vision Transformers

no code implementations9 Oct 2023 Weifeng Lin, Ziheng Wu, Wentao Yang, Mingxin Huang, Jun Huang, Lianwen Jin

In this paper, we introduce Hierarchical Side-Tuning (HST), an innovative PETL method facilitating the transfer of ViT models to diverse downstream tasks.

image-classification Image Classification +6

Twin Graph-based Anomaly Detection via Attentive Multi-Modal Learning for Microservice System

1 code implementation7 Oct 2023 Jun Huang, Yang Yang, Hang Yu, Jianguo Li, Xiao Zheng

The MST graph provides a virtual representation of the status and scheduling relationships among service instances of a real-world microservice system.

Anomaly Detection Scheduling

EasyPhoto: Your Smart AI Photo Generator

2 code implementations7 Oct 2023 Ziheng Wu, Jiaqi Xu, Xinyi Zou, Kunzhe Huang, Xing Shi, Jun Huang

By training a digital doppelganger of a specific user ID using 5 to 20 relevant images, the finetuned model (according to the trained LoRA model) allows for the generation of AI photos using arbitrary templates.

Knowledgeable In-Context Tuning: Exploring and Exploiting Factual Knowledge for In-Context Learning

no code implementations26 Sep 2023 Jianing Wang, Chengyu Wang, Chuanqi Tan, Jun Huang, Ming Gao

Large language models (LLMs) enable in-context learning (ICL) by conditioning on a few labeled training examples as a text-based prompt, eliminating the need for parameter updates and achieving competitive performance.

Few-Shot Learning In-Context Learning +3

Making Small Language Models Better Multi-task Learners with Mixture-of-Task-Adapters

no code implementations20 Sep 2023 Yukang Xie, Chengyu Wang, Junbing Yan, Jiyong Zhou, Feiqi Deng, Jun Huang

Recently, Large Language Models (LLMs) have achieved amazing zero-shot learning performance over a variety of Natural Language Processing (NLP) tasks, especially for text generative tasks.

Zero-Shot Learning

PAI-Diffusion: Constructing and Serving a Family of Open Chinese Diffusion Models for Text-to-image Synthesis on the Cloud

no code implementations11 Sep 2023 Chengyu Wang, Zhongjie Duan, Bingyan Liu, Xinyi Zou, Cen Chen, Kui Jia, Jun Huang

Text-to-image synthesis for the Chinese language poses unique challenges due to its large vocabulary size, and intricate character relationships.

Image Generation Style Transfer

TransPrompt v2: A Transferable Prompting Framework for Cross-task Text Classification

no code implementations29 Aug 2023 Jianing Wang, Chengyu Wang, Cen Chen, Ming Gao, Jun Huang, Aoying Zhou

We propose TransPrompt v2, a novel transferable prompting framework for few-shot learning across similar or distant text classification tasks.

Few-Shot Learning Few-Shot Text Classification +1

On the Trustworthiness Landscape of State-of-the-art Generative Models: A Survey and Outlook

no code implementations31 Jul 2023 Mingyuan Fan, Chengyu Wang, Cen Chen, Yang Liu, Jun Huang

Diffusion models and large language models have emerged as leading-edge generative models, revolutionizing various aspects of human life.

Fairness

Scale-Aware Modulation Meet Transformer

2 code implementations ICCV 2023 Weifeng Lin, Ziheng Wu, Jiayu Chen, Jun Huang, Lianwen Jin

Specifically, SMT with 11. 5M / 2. 4GFLOPs and 32M / 7. 7GFLOPs can achieve 82. 2% and 84. 3% top-1 accuracy on ImageNet-1K, respectively.

object-detection Object Detection +1

On the Robustness of Split Learning against Adversarial Attacks

no code implementations16 Jul 2023 Mingyuan Fan, Cen Chen, Chengyu Wang, Wenmeng Zhou, Jun Huang

Split learning enables collaborative deep learning model training while preserving data privacy and model security by avoiding direct sharing of raw data and model details (i. e., sever and clients only hold partial sub-networks and exchange intermediate computations).

Adversarial Attack

SLAMB: Accelerated Large Batch Training with Sparse Communication

1 code implementation The International Conference on Machine Learning (ICML) 2023 Hang Xu, Wenxuan Zhang, Jiawei Fei, Yuzhe Wu, Tingwen Xie, Jun Huang, Yuchen Xie, Mohamed Elhoseiny, Panos Kalnis

Distributed training of large deep neural networks requires frequent exchange of massive data between machines, thus communication efficiency is a major concern.

ConaCLIP: Exploring Distillation of Fully-Connected Knowledge Interaction Graph for Lightweight Text-Image Retrieval

no code implementations28 May 2023 Jiapeng Wang, Chengyu Wang, Xiaodan Wang, Jun Huang, Lianwen Jin

Large-scale pre-trained text-image models with dual-encoder architectures (such as CLIP) are typically adopted for various vision-language applications, including text-image retrieval.

Image Retrieval Knowledge Distillation +2

Optimal Linear Subspace Search: Learning to Construct Fast and High-Quality Schedulers for Diffusion Models

1 code implementation24 May 2023 Zhongjie Duan, Chengyu Wang, Cen Chen, Jun Huang, Weining Qian

In this paper, we first provide a detailed theoretical and empirical analysis of the generation process of the diffusion models based on schedulers.

Image Generation

Detection Transformer with Stable Matching

2 code implementations ICCV 2023 Shilong Liu, Tianhe Ren, Jiayu Chen, Zhaoyang Zeng, Hao Zhang, Feng Li, Hongyang Li, Jun Huang, Hang Su, Jun Zhu, Lei Zhang

We point out that the unstable matching in DETR is caused by a multi-optimization path problem, which is highlighted by the one-to-one matching design in DETR.

Decoder Position

Uncertainty-aware Self-training for Low-resource Neural Sequence Labeling

no code implementations17 Feb 2023 Jianing Wang, Chengyu Wang, Jun Huang, Ming Gao, Aoying Zhou

Neural sequence labeling (NSL) aims at assigning labels for input language tokens, which covers a broad range of applications, such as named entity recognition (NER) and slot filling, etc.

named-entity-recognition Named Entity Recognition +3

Exploiting Style Transfer-based Task Augmentation for Cross-Domain Few-Shot Learning

no code implementations19 Jan 2023 Shuzhen Rao, Jun Huang, Zengming Tang

Motivated by the observation that the domain shift between training tasks and target tasks usually can reflect in their style variation, we propose Task Augmented Meta-Learning (TAML) to conduct style transfer-based task augmentation to improve the domain generalization ability.

cross-domain few-shot learning Diversity +2

Revisiting and Advancing Chinese Natural Language Understanding with Accelerated Heterogeneous Knowledge Pre-training

1 code implementation11 Oct 2022 Taolin Zhang, Junwei DOng, Jianing Wang, Chengyu Wang, Ang Wang, Yinghui Liu, Jun Huang, Yong Li, Xiaofeng He

Recently, knowledge-enhanced pre-trained language models (KEPLMs) improve context-aware representations via learning from structured relations in knowledge graphs, and/or linguistic knowledge from syntactic or dependency analysis.

Knowledge Graphs Language Modeling +3

YOLOX-PAI: An Improved YOLOX, Stronger and Faster than YOLOv6

3 code implementations27 Aug 2022 Ziheng Wu, Xinyi Zou, Wenmeng Zhou, Jun Huang

We develop an all-in-one computer vision toolbox named EasyCV to facilitate the use of various SOTA computer vision methods.

object-detection Object Detection

P2ANet: A Dataset and Benchmark for Dense Action Detection from Table Tennis Match Broadcasting Videos

no code implementations26 Jul 2022 Jiang Bian, Xuhong LI, Tao Wang, Qingzhong Wang, Jun Huang, Chen Liu, Jun Zhao, Feixiang Lu, Dejing Dou, Haoyi Xiong

While deep learning has been widely used for video analytics, such as video classification and action detection, dense action detection with fast-moving subjects from sports videos is still challenging.

Action Detection Action Localization +2

KECP: Knowledge Enhanced Contrastive Prompting for Few-shot Extractive Question Answering

1 code implementation6 May 2022 Jianing Wang, Chengyu Wang, Minghui Qiu, Qiuhui Shi, Hongbin Wang, Jun Huang, Ming Gao

Extractive Question Answering (EQA) is one of the most important tasks in Machine Reading Comprehension (MRC), which can be solved by fine-tuning the span selecting heads of Pre-trained Language Models (PLMs).

Contrastive Learning Extractive Question-Answering +6

Making Pre-trained Language Models End-to-end Few-shot Learners with Contrastive Prompt Tuning

1 code implementation1 Apr 2022 Ziyun Xu, Chengyu Wang, Minghui Qiu, Fuli Luo, Runxin Xu, Songfang Huang, Jun Huang

Pre-trained Language Models (PLMs) have achieved remarkable performance for various language understanding tasks in IR systems, which require the fine-tuning process based on labeled training data.

Contrastive Learning

From Dense to Sparse: Contrastive Pruning for Better Pre-trained Language Model Compression

2 code implementations14 Dec 2021 Runxin Xu, Fuli Luo, Chengyu Wang, Baobao Chang, Jun Huang, Songfang Huang, Fei Huang

Unified in contrastive learning, CAP enables the pruned model to learn from the pre-trained model for task-agnostic knowledge, and fine-tuned model for task-specific knowledge.

Contrastive Learning Language Modeling +3

DKPLM: Decomposable Knowledge-enhanced Pre-trained Language Model for Natural Language Understanding

1 code implementation2 Dec 2021 Taolin Zhang, Chengyu Wang, Nan Hu, Minghui Qiu, Chengguang Tang, Xiaofeng He, Jun Huang

Knowledge-Enhanced Pre-trained Language Models (KEPLMs) are pre-trained models with relation triples injecting from knowledge graphs to improve language understanding abilities.

Knowledge Graphs Knowledge Probing +4

S-DCCRN: Super Wide Band DCCRN with learnable complex feature for speech enhancement

no code implementations16 Nov 2021 Shubo Lv, Yihui Fu, Mengtao Xing, Jiayao Sun, Lei Xie, Jun Huang, Yannan Wang, Tao Yu

In speech enhancement, complex neural network has shown promising performance due to their effectiveness in processing complex-valued spectrum.

16k Denoising +2

Millimeter-Wave NR-U and WiGig Coexistence: Joint User Grouping, Beam Coordination and Power Control

no code implementations11 Aug 2021 Xiaoxia Xu, Qimei Chen, Hao Jiang, Jun Huang

Our aim for the proposed coexistence network is to maximize the spectral efficiency while ensuring the strict NR-U delay requirement and the WiGig transmission performance in real time environments.

Multi-layered Semantic Representation Network for Multi-label Image Classification

1 code implementation22 Jun 2021 Xiwen Qu, Hao Che, Jun Huang, Linchuan Xu, Xiao Zheng

To this end, this paper designs a Multi-layered Semantic Representation Network (MSRN) which discovers both local and global semantics of labels through modeling label correlations and utilizes the label semantics to guide the semantic representations learning at multiple layers through an attention mechanism.

Classification image-classification +2

A heuristic resolution of the Abraham-Minkowski controversy

no code implementations4 Jan 2021 Guoxu Feng, Jun Huang

This paper reviews the history and origin of the Abraham-Minkowski controversy and points out that it is a continuation of the controversy over the speed of light in medium.

Optics

Learning to Expand: Reinforced Pseudo-relevance Feedback Selection for Information-seeking Conversations

no code implementations25 Nov 2020 Haojie Pan, Cen Chen, Chengyu Wang, Minghui Qiu, Liu Yang, Feng Ji, Jun Huang

More specifically, we propose a reinforced selector to extract useful PRF terms to enhance response candidates and a BERT-based response ranker to rank the PRF-enhanced responses.

EasyTransfer -- A Simple and Scalable Deep Transfer Learning Platform for NLP Applications

2 code implementations18 Nov 2020 Minghui Qiu, Peng Li, Chengyu Wang, Hanjie Pan, Ang Wang, Cen Chen, Xianyan Jia, Yaliang Li, Jun Huang, Deng Cai, Wei Lin

The literature has witnessed the success of leveraging Pre-trained Language Models (PLMs) and Transfer Learning (TL) algorithms to a wide range of Natural Language Processing (NLP) applications, yet it is not easy to build an easy-to-use and scalable TL toolkit for this purpose.

Compiler Optimization Conversational Question Answering +1

EasyASR: A Distributed Machine Learning Platform for End-to-end Automatic Speech Recognition

no code implementations14 Sep 2020 Chengyu Wang, Mengli Cheng, Xu Hu, Jun Huang

We present EasyASR, a distributed machine learning platform for training and serving large-scale Automatic Speech Recognition (ASR) models, as well as collecting and processing audio data at scale.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +2

One-shot Text Field Labeling using Attention and Belief Propagation for Structure Information Extraction

1 code implementation9 Sep 2020 Mengli Cheng, Minghui Qiu, Xing Shi, Jun Huang, Wei. Lin

Existing learning based methods for text labeling task usually require a large amount of labeled examples to train a specific model for each type of document.

One-Shot Learning Text Detection

Knowledge-Empowered Representation Learning for Chinese Medical Reading Comprehension: Task, Model and Resources

1 code implementation Findings (ACL) 2021 Taolin Zhang, Chengyu Wang, Minghui Qiu, Bite Yang, Xiaofeng He, Jun Huang

In this paper, we introduce a multi-target MRC task for the medical domain, whose goal is to predict answers to medical questions and the corresponding support sentences from medical information sources simultaneously, in order to ensure the high reliability of medical knowledge serving.

Machine Reading Comprehension Multi-Task Learning +1

Weakly Supervised Construction of ASR Systems with Massive Video Data

no code implementations4 Aug 2020 Mengli Cheng, Chengyu Wang, Xu Hu, Jun Huang, Xiaobo Wang

Building Automatic Speech Recognition (ASR) systems from scratch is significantly challenging, mostly due to the time-consuming and financially-expensive process of annotating a large amount of audio data with transcripts.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +4

Meta Fine-Tuning Neural Language Models for Multi-Domain Text Mining

2 code implementations EMNLP 2020 Chengyu Wang, Minghui Qiu, Jun Huang, Xiaofeng He

In this paper, we propose an effective learning procedure named Meta Fine-Tuning (MFT), served as a meta-learner to solve a group of similar NLP tasks for neural language models.

Few-Shot Learning Language Modeling +1

SwapText: Image Based Texts Transfer in Scenes

no code implementations CVPR 2020 Qiangpeng Yang, Hongsheng Jin, Jun Huang, Wei. Lin

First, a novel text swapping network is proposed to replace text labels only in the foreground image.

Image Generation Translation

KEML: A Knowledge-Enriched Meta-Learning Framework for Lexical Relation Classification

no code implementations25 Feb 2020 Chengyu Wang, Minghui Qiu, Jun Huang, Xiaofeng He

We further combine a meta-learning process over the auxiliary task distribution and supervised learning to train the neural lexical relation classifier.

General Classification Meta-Learning +2

AdaBERT: Task-Adaptive BERT Compression with Differentiable Neural Architecture Search

1 code implementation13 Jan 2020 Daoyuan Chen, Yaliang Li, Minghui Qiu, Zhen Wang, Bofang Li, Bolin Ding, Hongbo Deng, Jun Huang, Wei. Lin, Jingren Zhou

Motivated by the necessity and benefits of task-oriented BERT compression, we propose a novel compression method, AdaBERT, that leverages differentiable Neural Architecture Search to automatically compress BERT into task-adaptive small models for specific tasks.

Knowledge Distillation Neural Architecture Search

Learning to Selectively Transfer: Reinforced Transfer Learning for Deep Text Matching

no code implementations30 Dec 2018 Chen Qu, Feng Ji, Minghui Qiu, Liu Yang, Zhiyu Min, Haiqing Chen, Jun Huang, W. Bruce Croft

Specifically, the data selector "acts" on the source domain data to find a subset for optimization of the TL model, and the performance of the TL model can provide "rewards" in turn to update the selector.

Information Retrieval Natural Language Inference +5

Review Helpfulness Prediction with Embedding-Gated CNN

no code implementations29 Aug 2018 Cen Chen, Minghui Qiu, Yinfei Yang, Jun Zhou, Jun Huang, Xiaolong Li, Forrest Bao

Product reviews, in the form of texts dominantly, significantly help consumers finalize their purchasing decisions.

Prediction Sentence

Response Ranking with Deep Matching Networks and External Knowledge in Information-seeking Conversation Systems

1 code implementation1 May 2018 Liu Yang, Minghui Qiu, Chen Qu, Jiafeng Guo, Yongfeng Zhang, W. Bruce Croft, Jun Huang, Haiqing Chen

Our models and research findings provide new insights on how to utilize external knowledge with deep neural models for response selection and have implications for the design of the next generation of information-seeking conversation systems.

Knowledge Distillation Retrieval +1

Modelling Domain Relationships for Transfer Learning on Retrieval-based Question Answering Systems in E-commerce

no code implementations23 Nov 2017 Jianfei Yu, Minghui Qiu, Jing Jiang, Jun Huang, Shuangyong Song, Wei Chu, Haiqing Chen

In this paper, we study transfer learning for the PI and NLI problems, aiming to propose a general framework, which can effectively and efficiently adapt the shared knowledge learned from a resource-rich source domain to a resource- poor target domain.

Chatbot Natural Language Inference +5

AliMe Chat: A Sequence to Sequence and Rerank based Chatbot Engine

no code implementations ACL 2017 Minghui Qiu, Feng-Lin Li, Siyu Wang, Xing Gao, Yan Chen, Weipeng Zhao, Haiqing Chen, Jun Huang, Wei Chu

We propose AliMe Chat, an open-domain chatbot engine that integrates the joint results of Information Retrieval (IR) and Sequence to Sequence (Seq2Seq) based generation models.

Chatbot Information Retrieval +1

Cannot find the paper you are looking for? You can Submit a new open access paper.