Search Results for author: Junjie Hu

Found 72 papers, 40 papers with code

XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalisation

2 code implementations ICML 2020 Junjie Hu, Sebastian Ruder, Aditya Siddhant, Graham Neubig, Orhan Firat, Melvin Johnson

However, these broad-coverage benchmarks have been mostly limited to English, and despite an increasing interest in multilingual models, a benchmark that enables the comprehensive evaluation of such methods on a diverse range of languages and tasks is still missing.

Retrieval Sentence +2

DeTox: Toxic Subspace Projection for Model Editing

1 code implementation22 May 2024 Rheeya Uppaal, Apratim De, Yiting He, Yiquao Zhong, Junjie Hu

Furthermore, these tuning-based methods require large-scale preference data for training and are susceptible to noisy preference data.

Language Modelling Model Editing

How does Multi-Task Training Affect Transformer In-Context Capabilities? Investigations with Function Classes

1 code implementation4 Apr 2024 Harmon Bhasin, Timothy Ossowski, Yiqiao Zhong, Junjie Hu

Large language models (LLM) have recently shown the extraordinary ability to perform unseen tasks based on few-shot examples provided as text, also known as in-context learning (ICL).

In-Context Learning Multi-Task Learning

Lookahead Exploration with Neural Radiance Representation for Continuous Vision-Language Navigation

1 code implementation2 Apr 2024 Zihan Wang, Xiangyang Li, Jiahao Yang, Yeqi Liu, Junjie Hu, Ming Jiang, Shuqiang Jiang

Vision-and-language navigation (VLN) enables the agent to navigate to a remote location following the natural language instruction in 3D environments.

Navigate Vision and Language Navigation +1

Data Augmentation using LLMs: Data Perspectives, Learning Paradigms and Challenges

no code implementations5 Mar 2024 Bosheng Ding, Chengwei Qin, Ruochen Zhao, Tianze Luo, Xinze Li, Guizhen Chen, Wenhan Xia, Junjie Hu, Anh Tuan Luu, Shafiq Joty

In the rapidly evolving field of machine learning (ML), data augmentation (DA) has emerged as a pivotal technique for enhancing model performance by diversifying training examples without the need for additional data collection.

Data Augmentation

Mitigating Fine-tuning Jailbreak Attack with Backdoor Enhanced Alignment

no code implementations22 Feb 2024 Jiongxiao Wang, Jiazhao Li, Yiquan Li, Xiangyu Qi, Junjie Hu, Yixuan Li, Patrick McDaniel, Muhao Chen, Bo Li, Chaowei Xiao

Despite the general capabilities of Large Language Models (LLMs) like GPT-4 and Llama-2, these models still request fine-tuning or adaptation with customized data when it comes to meeting the specific business demands and intricacies of tailored use cases.

Chatbot Meets Pipeline: Augment Large Language Model with Definite Finite Automaton

no code implementations6 Feb 2024 Yiyou Sun, Junjie Hu, Wei Cheng, Haifeng Chen

This paper introduces the Definite Finite Automaton augmented large language model (DFA-LLM), a novel framework designed to enhance the capabilities of conversational agents using large language models (LLMs).

Chatbot Language Modelling +2

Learning Label Hierarchy with Supervised Contrastive Learning

1 code implementation31 Jan 2024 Ruixue Lian, William A. Sethares, Junjie Hu

This paper introduces a family of Label-Aware SCL methods (LASCL) that incorporates hierarchical information to SCL by leveraging similarities between classes, resulting in creating a more well-structured and discriminative feature space.

Contrastive Learning text-classification +1

Prompting Large Vision-Language Models for Compositional Reasoning

1 code implementation20 Jan 2024 Timothy Ossowski, Ming Jiang, Junjie Hu

Vision-language models such as CLIP have shown impressive capabilities in encoding texts and images into aligned embeddings, enabling the retrieval of multimodal data in a shared embedding space.

Retrieval Visual Reasoning

The Wisdom of Partisan Crowds: Comparing Collective Intelligence in Humans and LLM-based Agents

no code implementations16 Nov 2023 Yun-Shiuan Chuang, Siddharth Suresh, Nikunj Harlalka, Agam Goyal, Robert Hawkins, Sijia Yang, Dhavan Shah, Junjie Hu, Timothy T. Rogers

Human groups are able to converge on more accurate beliefs through deliberation, even in the presence of polarization and partisan bias -- a phenomenon known as the "wisdom of partisan crowds."

Simulating Opinion Dynamics with Networks of LLM-based Agents

1 code implementation16 Nov 2023 Yun-Shiuan Chuang, Agam Goyal, Nikunj Harlalka, Siddharth Suresh, Robert Hawkins, Sijia Yang, Dhavan Shah, Junjie Hu, Timothy T. Rogers

Accurately simulating human opinion dynamics is crucial for understanding a variety of societal phenomena, including polarization and the spread of misinformation.

Misinformation Prompt Engineering

Evolving Domain Adaptation of Pretrained Language Models for Text Classification

no code implementations16 Nov 2023 Yun-Shiuan Chuang, Yi Wu, Dhruv Gupta, Rheeya Uppaal, Ananya Kumar, Luhang Sun, Makesh Narsimhan Sreedhar, Sijia Yang, Timothy T. Rogers, Junjie Hu

Adapting pre-trained language models (PLMs) for time-series text classification amidst evolving domain shifts (EDS) is critical for maintaining accuracy in applications like stance detection.

Domain Adaptation Stance Detection +3

Automatic Personalized Impression Generation for PET Reports Using Large Language Models

2 code implementations18 Sep 2023 Xin Tie, Muheon Shin, Ali Pirasteh, Nevein Ibrahim, Zachary Huemann, Sharon M. Castellino, Kara M. Kelly, John Garrett, Junjie Hu, Steve Y. Cho, Tyler J. Bradshaw

Twelve language models were trained on a corpus of PET reports using the teacher-forcing algorithm, with the report findings as input and the clinical impressions as reference.

Single Sequence Prediction over Reasoning Graphs for Multi-hop QA

no code implementations1 Jul 2023 Gowtham Ramesh, Makesh Sreedhar, Junjie Hu

Recent generative approaches for multi-hop question answering (QA) utilize the fusion-in-decoder method~\cite{izacard-grave-2021-leveraging} to generate a single sequence output which includes both a final answer and a reasoning path taken to arrive at that answer, such as passage titles and key facts from those passages.

Decoder Multi-hop Question Answering +1

Multimodal Prompt Retrieval for Generative Visual Question Answering

1 code implementation30 Jun 2023 Timothy Ossowski, Junjie Hu

Recent years have witnessed impressive results of pre-trained vision-language models on knowledge-intensive tasks such as visual question answering (VQA).

Domain Adaptation Generative Visual Question Answering +3

Benchmarking LLM-based Machine Translation on Cultural Awareness

no code implementations23 May 2023 Binwei Yao, Ming Jiang, Diyi Yang, Junjie Hu

Furthermore, we devise a novel evaluation metric to assess the understandability of translations in a reference-free manner by GPT-4.

Benchmarking In-Context Learning +3

Is Fine-tuning Needed? Pre-trained Language Models Are Near Perfect for Out-of-Domain Detection

1 code implementation22 May 2023 Rheeya Uppaal, Junjie Hu, Yixuan Li

Fine-tuning with pre-trained language models has been a de facto procedure to derive OOD detectors with respect to in-distribution (ID) data.

Out of Distribution (OOD) Detection

Explicit Attention-Enhanced Fusion for RGB-Thermal Perception Tasks

1 code implementation28 Mar 2023 Mingjian Liang, Junjie Hu, Chenyu Bao, Hua Feng, Fuqin Deng, Tin Lun Lam

Specifically, we consider the following cases: i) both RGB data and thermal data, ii) only one of the types of data, and iii) none of them generate discriminative features.

Crowd Counting object-detection +4

Lifelong-MonoDepth: Lifelong Learning for Multi-Domain Monocular Metric Depth Estimation

1 code implementation9 Mar 2023 Junjie Hu, Chenyou Fan, Liguang Zhou, Qing Gao, Honghai Liu, Tin Lun Lam

With the rapid advancements in autonomous driving and robot navigation, there is a growing demand for lifelong learning models capable of estimating metric (absolute) depth.

Autonomous Driving Depth Prediction +2

ConTEXTual Net: A Multimodal Vision-Language Model for Segmentation of Pneumothorax

1 code implementation2 Mar 2023 Zachary Huemann, Xin Tie, Junjie Hu, Tyler J. Bradshaw

ConTEXTual Net was trained on the CANDID-PTX dataset consisting of 3, 196 positive cases of pneumothorax with segmentation annotations from 6 different physicians as well as clinical radiology reports.

Descriptive Image Captioning +5

Attentional Graph Convolutional Network for Structure-aware Audio-Visual Scene Classification

no code implementations31 Dec 2022 Liguang Zhou, Yuhongze Zhou, Xiaonan Qi, Junjie Hu, Tin Lun Lam, Yangsheng Xu

Then, to build multi-scale hierarchical information of input features, we utilize an attention fusion mechanism to aggregate features from multiple layers of the backbone network.

Scene Classification Scene Recognition +1

Peer Learning for Unbiased Scene Graph Generation

no code implementations31 Dec 2022 Liguang Zhou, Junjie Hu, Yuhongze Zhou, Tin Lun Lam, Yangsheng Xu

Unbiased scene graph generation (USGG) is a challenging task that requires predicting diverse and heavily imbalanced predicates between objects in an image.

Graph Generation Unbiased Scene Graph Generation

Beyond Counting Datasets: A Survey of Multilingual Dataset Construction and Necessary Resources

no code implementations28 Nov 2022 Xinyan Velocity Yu, Akari Asai, Trina Chatterjee, Junjie Hu, Eunsol Choi

While the NLP community is generally aware of resource disparities among languages, we lack research that quantifies the extent and types of such disparity.

Progressive Self-Distillation for Ground-to-Aerial Perception Knowledge Transfer

1 code implementation29 Aug 2022 Junjie Hu, Chenyou Fan, Mete Ozay, Hua Feng, Yuan Gao, Tin Lun Lam

In this paper, we introduce the ground-to-aerial perception knowledge transfer and propose a progressive semi-supervised learning framework that enables drone perception using only labeled data of ground viewpoint and unlabeled data of flying viewpoints.

Autonomous Driving Knowledge Distillation +1

Dense Depth Distillation with Out-of-Distribution Simulated Images

no code implementations26 Aug 2022 Junjie Hu, Chenyou Fan, Mete Ozay, Hualie Jiang, Tin Lun Lam

We study data-free knowledge distillation (KD) for monocular depth estimation (MDE), which learns a lightweight model for real-world depth perception tasks by compressing it from a trained teacher model while lacking training data in the target domain.

Data-free Knowledge Distillation Image Classification +1

MIA 2022 Shared Task: Evaluating Cross-lingual Open-Retrieval Question Answering for 16 Diverse Languages

no code implementations NAACL (MIA) 2022 Akari Asai, Shayne Longpre, Jungo Kasai, Chia-Hsuan Lee, Rui Zhang, Junjie Hu, Ikuya Yamada, Jonathan H. Clark, Eunsol Choi

We present the results of the Workshop on Multilingual Information Access (MIA) 2022 Shared Task, evaluating cross-lingual open-retrieval question answering (QA) systems in 16 typologically diverse languages.

Question Answering Retrieval

Deep Depth Completion from Extremely Sparse Data: A Survey

no code implementations11 May 2022 Junjie Hu, Chenyu Bao, Mete Ozay, Chenyou Fan, Qing Gao, Honghai Liu, Tin Lun Lam

Depth completion aims at predicting dense pixel-wise depth from an extremely sparse map captured from a depth sensor, e. g., LiDARs.

3D Reconstruction Autonomous Driving +2

DEEP: DEnoising Entity Pre-training for Neural Machine Translation

no code implementations ACL 2022 Junjie Hu, Hiroaki Hayashi, Kyunghyun Cho, Graham Neubig

It has been shown that machine translation models usually generate poor translations for named entities that are infrequent in the training corpus.

Denoising Multi-Task Learning +3

Advanced Statistical Learning on Short Term Load Process Forecasting

no code implementations19 Oct 2021 Junjie Hu, Brenda López Cabrera, Awdesch Melzer

The predictive information is fundamental for the risk and production management of electricity consumers.

Decision Making Management +1

GlobalWoZ: Globalizing MultiWoZ to Develop Multilingual Task-Oriented Dialogue Systems

1 code implementation ACL 2022 Bosheng Ding, Junjie Hu, Lidong Bing, Sharifah Mahani Aljunied, Shafiq Joty, Luo Si, Chunyan Miao

Much recent progress in task-oriented dialogue (ToD) systems has been driven by available annotation data across multiple domains for training.

Task-Oriented Dialogue Systems

PLNet: Plane and Line Priors for Unsupervised Indoor Depth Estimation

1 code implementation12 Oct 2021 Hualie Jiang, Laiyan Ding, Junjie Hu, Rui Huang

Unsupervised learning of depth from indoor monocular videos is challenging as the artificial environment contains many textureless regions.

Depth Estimation

Learn2Agree: Fitting with Multiple Annotators without Objective Ground Truth

no code implementations8 Sep 2021 Chongyang Wang, Yuan Gao, Chenyou Fan, Junjie Hu, Tin Lun Lam, Nicholas D. Lane, Nadia Bianchi-Berthouze

For such issues, we propose a novel Learning to Agreement (Learn2Agree) framework to tackle the challenge of learning from multiple annotators without objective ground truth.

Networks of News and Cross-Sectional Returns

no code implementations12 Aug 2021 Junjie Hu, Wolfgang Karl Härdle

We uncover networks from news articles to study cross-sectional stock returns.

Boosting Light-Weight Depth Estimation Via Knowledge Distillation

2 code implementations13 May 2021 Junjie Hu, Chenyou Fan, Hualie Jiang, Xiyue Guo, Yuan Gao, Xiangyong Lu, Tin Lun Lam

However, this KD process can be challenging and insufficient due to the large model capacity gap between the teacher and the student.

Computational Efficiency Knowledge Distillation +1

A Two-stage Unsupervised Approach for Low light Image Enhancement

no code implementations19 Oct 2020 Junjie Hu, Xiyue Guo, Junfeng Chen, Guanqi Liang, Fuqin Deng, Tin Lun Lam

However, most of them suffer from the following problems: 1) the need of pairs of low light and normal light images for training, 2) the poor performance for dark images, 3) the amplification of noise.

Low-Light Image Enhancement Simultaneous Localization and Mapping +1

Semantic Histogram Based Graph Matching for Real-Time Multi-Robot Global Localization in Large Scale Environment

3 code implementations19 Oct 2020 Xiyue Guo, Junjie Hu, Junfeng Chen, Fuqin Deng, Tin Lun Lam

The core problem of visual multi-robot simultaneous localization and mapping (MR-SLAM) is how to efficiently and accurately perform multi-robot global localization (MR-GL).

Graph Matching Simultaneous Localization and Mapping

Explicit Alignment Objectives for Multilingual Bidirectional Encoders

no code implementations NAACL 2021 Junjie Hu, Melvin Johnson, Orhan Firat, Aditya Siddhant, Graham Neubig

Pre-trained cross-lingual encoders such as mBERT (Devlin et al., 2019) and XLMR (Conneau et al., 2020) have proven to be impressively effective at enabling transfer-learning of NLP systems from high-resource languages to low-resource languages.

Retrieval Sentence +4

On Learning Language-Invariant Representations for Universal Machine Translation

no code implementations ICML 2020 Han Zhao, Junjie Hu, Andrej Risteski

The goal of universal machine translation is to learn to translate between any pair of languages, given a corpus of paired translated documents for \emph{a small subset} of all pairs of languages.

Machine Translation Sentence +1

XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalization

4 code implementations24 Mar 2020 Junjie Hu, Sebastian Ruder, Aditya Siddhant, Graham Neubig, Orhan Firat, Melvin Johnson

However, these broad-coverage benchmarks have been mostly limited to English, and despite an increasing interest in multilingual models, a benchmark that enables the comprehensive evaluation of such methods on a diverse range of languages and tasks is still missing.

Cross-Lingual Transfer Retrieval +2

Risk of Bitcoin Market: Volatility, Jumps, and Forecasts

no code implementations11 Dec 2019 Junjie Hu, Wolfgang Karl Härdle, Weiyu Kuo

Cryptocurrency, the most controversial and simultaneously the most interesting asset, has attracted many investors and speculators in recent years.

Analysis of Deep Networks for Monocular Depth Estimation Through Adversarial Attacks with Proposal of a Defense Method

no code implementations20 Nov 2019 Junjie Hu, Takayuki Okatani

However, the prediction of saliency maps is itself vulnerable to the attacks, even though it is not the direct target of the attacks.

Monocular Depth Estimation

What Makes A Good Story? Designing Composite Rewards for Visual Storytelling

1 code implementation11 Sep 2019 Junjie Hu, Yu Cheng, Zhe Gan, Jingjing Liu, Jianfeng Gao, Graham Neubig

Previous storytelling approaches mostly focused on optimizing traditional metrics such as BLEU, ROUGE and CIDEr.

Visual Storytelling

REO-Relevance, Extraness, Omission: A Fine-grained Evaluation for Image Captioning

1 code implementation IJCNLP 2019 Ming Jiang, Junjie Hu, Qiuyuan Huang, Lei Zhang, Jana Diesner, Jianfeng Gao

In this study, we present a fine-grained evaluation method REO for automatically measuring the performance of image captioning systems.

Image Captioning

Handling Syntactic Divergence in Low-resource Machine Translation

1 code implementation IJCNLP 2019 Chunting Zhou, Xuezhe Ma, Junjie Hu, Graham Neubig

Despite impressive empirical successes of neural machine translation (NMT) on standard benchmarks, limited parallel data impedes the application of NMT models to many language pairs.

Data Augmentation Machine Translation +2

A Hybrid Retrieval-Generation Neural Conversation Model

1 code implementation19 Apr 2019 Liu Yang, Junjie Hu, Minghui Qiu, Chen Qu, Jianfeng Gao, W. Bruce Croft, Xiaodong Liu, Yelong Shen, Jingjing Liu

In this paper, we propose a hybrid neural conversation model that combines the merits of both response retrieval and generation methods.

Retrieval Text Generation +1

Visualization of Convolutional Neural Networks for Monocular Depth Estimation

1 code implementation ICCV 2019 Junjie Hu, Yan Zhang, Takayuki Okatani

We formulate it as an optimization problem of identifying the smallest number of image pixels from which the CNN can estimate a depth map with the minimum difference from the estimate from the entire image.

Interpretable Machine Learning

compare-mt: A Tool for Holistic Comparison of Language Generation Systems

2 code implementations NAACL 2019 Graham Neubig, Zi-Yi Dou, Junjie Hu, Paul Michel, Danish Pruthi, Xinyi Wang, John Wieting

In this paper, we describe compare-mt, a tool for holistic analysis and comparison of the results of systems for language generation tasks such as machine translation.

Machine Translation Sentence +2

The ARIEL-CMU Systems for LoReHLT18

no code implementations24 Feb 2019 Aditi Chaudhary, Siddharth Dalmia, Junjie Hu, Xinjian Li, Austin Matthews, Aldrian Obaja Muis, Naoki Otani, Shruti Rijhwani, Zaid Sheikh, Nidhi Vyas, Xinyi Wang, Jiateng Xie, Ruochen Xu, Chunting Zhou, Peter J. Jansen, Yiming Yang, Lori Levin, Florian Metze, Teruko Mitamura, David R. Mortensen, Graham Neubig, Eduard Hovy, Alan W. black, Jaime Carbonell, Graham V. Horwood, Shabnam Tafreshi, Mona Diab, Efsun S. Kayi, Noura Farra, Kathleen McKeown

This paper describes the ARIEL-CMU submissions to the Low Resource Human Language Technologies (LoReHLT) 2018 evaluations for the tasks Machine Translation (MT), Entity Discovery and Linking (EDL), and detection of Situation Frames in Text and Speech (SF Text and Speech).

Machine Translation Translation

Contextual Encoding for Translation Quality Estimation

1 code implementation WS 2018 Junjie Hu, Wei-Cheng Chang, Yuexin Wu, Graham Neubig

In this paper, propose a method to effectively encode the local and global contextual information for each target word using a three-part neural network approach.

Sentence Translation

Rapid Adaptation of Neural Machine Translation to New Languages

1 code implementation EMNLP 2018 Graham Neubig, Junjie Hu

This paper examines the problem of adapting neural machine translation systems to new, low-resourced languages (LRLs) as effectively and rapidly as possible.

Machine Translation Translation

Automatic Estimation of Simultaneous Interpreter Performance

1 code implementation ACL 2018 Craig Stewart, Nikolai Vogler, Junjie Hu, Jordan Boyd-Graber, Graham Neubig

Simultaneous interpretation, translation of the spoken word in real-time, is both highly challenging and physically demanding.

Machine Translation Translation

Revisiting Single Image Depth Estimation: Toward Higher Resolution Maps with Accurate Object Boundaries

4 code implementations23 Mar 2018 Junjie Hu, Mete Ozay, Yan Zhang, Takayuki Okatani

Experimental results show that these two improvements enable to attain higher accuracy than the current state-of-the-arts, which is given by finer resolution reconstruction, for example, with small objects and object boundaries.

Decoder Monocular Depth Estimation

Principled Hybrids of Generative and Discriminative Domain Adaptation

no code implementations ICLR 2018 Han Zhao, Zhenyao Zhu, Junjie Hu, Adam Coates, Geoff Gordon

This provides us a very general way to interpolate between generative and discriminative extremes through different choices of priors.

Domain Adaptation

Structural Embedding of Syntactic Trees for Machine Comprehension

no code implementations EMNLP 2017 Rui Liu, Junjie Hu, Wei Wei, Zi Yang, Eric Nyberg

Deep neural networks for machine comprehension typically utilizes only word or character embeddings without explicitly taking advantage of structured linguistic information such as constituency trees and dependency trees.

Question Answering Reading Comprehension

Semi-Supervised QA with Generative Domain-Adaptive Nets

no code implementations ACL 2017 Zhilin Yang, Junjie Hu, Ruslan Salakhutdinov, William W. Cohen

In this framework, we train a generative model to generate questions based on the unlabeled text, and combine model-generated questions with human-generated questions for training question answering models.

Domain Adaptation Question Answering +2

Words or Characters? Fine-grained Gating for Reading Comprehension

1 code implementation6 Nov 2016 Zhilin Yang, Bhuwan Dhingra, Ye Yuan, Junjie Hu, William W. Cohen, Ruslan Salakhutdinov

Previous work combines word-level and character-level representations using concatenation or scalar weighting, which is suboptimal for high-level tasks like reading comprehension.

Question Answering Reading Comprehension +1

Learning Lexical Entries for Robotic Commands using Crowdsourcing

no code implementations8 Sep 2016 Junjie Hu, Jean Oh, Anatole Gershman

Robotic commands in natural language usually contain various spatial descriptions that are semantically similar but syntactically different.

Machine Translation Translation

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