no code implementations • 31 May 2024 • Yuanjiang Luo, Hongxiang Li, Xuan Wu, Meng Cao, Xiaoshuang Huang, Zhihong Zhu, Peixi Liao, Hu Chen, Yi Zhang
Existing mainstream approaches follow the encoder-decoder paradigm for generating radiology reports.
no code implementations • 30 May 2024 • Xuan Wu, Hongxiang Li, Yuanjiang Luo, Xuxin Cheng, Xianwei Zhuang, Meng Cao, Keren Fu
Sign language video retrieval plays a key role in facilitating information access for the deaf community.
no code implementations • 29 May 2024 • Meng Cao, Haoran Tang, Jinfa Huang, Peng Jin, Can Zhang, Ruyang Liu, Long Chen, Xiaodan Liang, Li Yuan, Ge Li
Text-Video Retrieval (TVR) aims to align relevant video content with natural language queries.
1 code implementation • 3 Apr 2024 • Xiaoshuang Huang, Hongxiang Li, Meng Cao, Long Chen, Chenyu You, Dong An
Recent developments underscore the potential of textual information in enhancing learning models for a deeper understanding of medical visual semantics.
1 code implementation • 27 Mar 2024 • Lei Yu, Meng Cao, Jackie Chi Kit Cheung, Yue Dong
Our study investigates the mechanistic causes of hallucination, specifically non-factual ones where the LM incorrectly predicts object attributes in response to subject-relation queries.
1 code implementation • 14 Mar 2024 • Kaichao You, Runsheng Bai, Meng Cao, Jianmin Wang, Ion Stoica, Mingsheng Long
PyTorch \texttt{2. x} introduces a compiler designed to accelerate deep learning programs.
no code implementations • 4 Mar 2024 • Liang Zhang, Jionghao Lin, Conrad Borchers, John Sabatini, John Hollander, Meng Cao, Xiangen Hu
This research is motivated by the potential of LLMs to predict learning performance based on its inherent reasoning and computational capabilities.
no code implementations • 19 Feb 2024 • Aiwei Liu, Haoping Bai, Zhiyun Lu, Xiang Kong, Simon Wang, Jiulong Shan, Meng Cao, Lijie Wen
In this paper, we propose a method to evaluate the response preference by using the output probabilities of response pairs under contrastive prompt pairs, which could achieve better performance on LLaMA2-7B and LLaMA2-13B compared to RLAIF.
no code implementations • 5 Feb 2024 • Meng Cao, Hussain Hussain, Sandipan Sikdar, Denis Helic, Markus Strohmaier, Roman Kern
We further study how interventions on network properties influence fairness by examining counterfactual scenarios with alternative evolution outcomes and differing network properties.
no code implementations • 29 Jan 2024 • Liang Zhang, Jionghao Lin, Conrad Borchers, Meng Cao, Xiangen Hu
Learning performance data (e. g., quiz scores and attempts) is significant for understanding learner engagement and knowledge mastery level.
no code implementations • 14 Jan 2024 • Meng Cao, Lei Shu, Lei Yu, Yun Zhu, Nevan Wichers, Yinxiao Liu, Lei Meng
We investigate this approach under two different settings: one where the policy model is smaller and is paired with a more powerful critic model, and another where a single language model fulfills both roles.
no code implementations • 18 Nov 2023 • Yu Lu Liu, Meng Cao, Su Lin Blodgett, Jackie Chi Kit Cheung, Alexandra Olteanu, Adam Trischler
We focus on how, which, and when responsible AI issues are covered, which relevant stakeholders are considered, and mismatches between stated and realized research goals.
no code implementations • 7 Nov 2023 • Peilin Zhou, Meng Cao, You-Liang Huang, Qichen Ye, Peiyan Zhang, Junling Liu, Yueqi Xie, Yining Hua, Jaeboum Kim
Large Multimodal Models (LMMs) have demonstrated impressive performance across various vision and language tasks, yet their potential applications in recommendation tasks with visual assistance remain unexplored.
no code implementations • 3 Nov 2023 • Meng Cao, Mehdi Fatemi, Jackie Chi Kit Cheung, Samira Shabanian
While large language models (LLMs) have achieved impressive performance in generating fluent and realistic text, controlling the generated text so that it exhibits properties such as safety, factuality, and non-toxicity remains challenging.
no code implementations • 25 Oct 2023 • Ji Jiang, Meng Cao, Tengtao Song, Long Chen, Yi Wang, Yuexian Zou
Video Referring Expression Comprehension (REC) aims to localize a target object in videos based on the queried natural language.
1 code implementation • 13 Oct 2023 • Qichen Ye, Junling Liu, Dading Chong, Peilin Zhou, Yining Hua, Fenglin Liu, Meng Cao, ZiMing Wang, Xuxin Cheng, Zhu Lei, Zhenhua Guo
In the CPT and SFT phases, Qilin-Med achieved 38. 4% and 40. 0% accuracy on the CMExam test set, respectively.
1 code implementation • 11 Oct 2023 • Zhengfeng Lai, Haotian Zhang, BoWen Zhang, Wentao Wu, Haoping Bai, Aleksei Timofeev, Xianzhi Du, Zhe Gan, Jiulong Shan, Chen-Nee Chuah, Yinfei Yang, Meng Cao
For example, VeCLIP achieves up to +25. 2% gain in COCO and Flickr30k retrieval tasks under the 12M setting.
1 code implementation • ICCV 2023 • Hongxiang Li, Meng Cao, Xuxin Cheng, Yaowei Li, Zhihong Zhu, Yuexian Zou
Due to two annoying issues in video grounding: (1) the co-existence of some visual entities in both ground truth and other moments, \ie semantic overlapping; (2) only a few moments in the video are annotated, \ie sparse annotation dilemma, vanilla contrastive learning is unable to model the correlations between temporally distant moments and learned inconsistent video representations.
1 code implementation • 6 Jul 2023 • Xiaozhong Lyu, Stefan Grafberger, Samantha Biegel, Shaopeng Wei, Meng Cao, Sebastian Schelter, Ce Zhang
There are exponentially many terms in the multilinear extension, and one key contribution of this paper is a polynomial time algorithm that computes exactly, given a retrieval-augmented model with an additive utility function and a validation set, the data importance of data points in the retrieval corpus using the multilinear extension of the model's utility function.
no code implementations • 25 Jun 2023 • Yangjun Mao, Jun Xiao, Dong Zhang, Meng Cao, Jian Shao, Yueting Zhuang, Long Chen
A recent DIC method proposes to generate distinctive captions by comparing the target image with a set of semantic-similar reference images, i. e., reference-based DIC (Ref-DIC).
no code implementations • 13 Jun 2023 • Haoping Bai, Shancong Mou, Tatiana Likhomanenko, Ramazan Gokberk Cinbis, Oncel Tuzel, Ping Huang, Jiulong Shan, Jianjun Shi, Meng Cao
We introduce the VISION Datasets, a diverse collection of 14 industrial inspection datasets, uniquely poised to meet these challenges.
1 code implementation • 19 May 2023 • Kaichao You, Guo Qin, Anchang Bao, Meng Cao, Ping Huang, Jiulong Shan, Mingsheng Long
Subsequently, we propose a novel Tune mode to bridge the gap between Eval mode and Deploy mode.
1 code implementation • 27 Feb 2023 • Meng Cao, Mehdi Fatemi, Jackie Chi Kit Cheung, Samira Shabanian
Other methods rely on rule-based or prompt-based token elimination, which are limited as they dismiss future tokens and the overall meaning of the complete discourse.
no code implementations • 24 Feb 2023 • Shancong Mou, Xiaoyi Gu, Meng Cao, Haoping Bai, Ping Huang, Jiulong Shan, Jianjun Shi
In this paper, we propose a Robust GAN-inversion (RGI) method with a provable robustness guarantee to achieve image restoration under unknown \textit{gross} corruptions, where a small fraction of pixels are completely corrupted.
1 code implementation • 16 Feb 2023 • Meng Cao, Yue Dong, Jingyi He, Jackie Chi Kit Cheung
State-of-the-art abstractive summarization systems frequently hallucinate content that is not supported by the source document, mainly due to noise in the training dataset.
no code implementations • 15 Jan 2023 • Hongxiang Li, Meng Cao, Xuxin Cheng, Zhihong Zhu, Yaowei Li, Yuexian Zou
Video grounding aims to locate a moment of interest matching the given query sentence from an untrimmed video.
no code implementations • CVPR 2023 • Meng Cao, Fangyun Wei, Can Xu, Xiubo Geng, Long Chen, Can Zhang, Yuexian Zou, Tao Shen, Daxin Jiang
Weakly-Supervised Video Grounding (WSVG) aims to localize events of interest in untrimmed videos with only video-level annotations.
1 code implementation • 6 Oct 2022 • Ji Jiang, Meng Cao, Tengtao Song, Yuexian Zou
To this end, we introduce two new datasets (i. e., VID-Entity and VidSTG-Entity) by augmenting the VIDSentence and VidSTG datasets with the explicitly referred words in the whole sentence, respectively.
1 code implementation • 13 Sep 2022 • Hussain Hussain, Meng Cao, Sandipan Sikdar, Denis Helic, Elisabeth Lex, Markus Strohmaier, Roman Kern
We hope our findings raise awareness about this issue in our community and lay a foundation for the future development of GNN models that are more robust to such attacks.
no code implementations • 29 Aug 2022 • Pengfei Zhu, Xinjie Yao, Yu Wang, Meng Cao, Binyuan Hui, Shuai Zhao, QinGhua Hu
Multi-view learning has progressed rapidly in recent years.
1 code implementation • 21 Jul 2022 • Meng Cao, Tianyu Yang, Junwu Weng, Can Zhang, Jue Wang, Yuexian Zou
To further enhance the temporal reasoning ability of the learned feature, we propose a context projection head and a temporal aware contrastive loss to perceive the contextual relationships.
1 code implementation • 21 Jul 2022 • Meng Cao, Ji Jiang, Long Chen, Yuexian Zou
Extensive experiments demonstrate that our DCNet achieves state-of-the-art performance on both video and image REC benchmarks.
no code implementations • 20 Apr 2022 • Meng Cao
Automatic summarization is the process of shortening a set of textual data computationally, to create a subset (a summary) that represents the most important pieces of information in the original text.
no code implementations • 7 Apr 2022 • Weikai Li, Meng Cao, Songcan Chen
Unsupervised Source (data) Free domain adaptation (USFDA) aims to transfer knowledge from a well-trained source model to a related but unlabeled target domain.
no code implementations • 28 Mar 2022 • Shancong Mou, Meng Cao, Haoping Bai, Ping Huang, Jianjun Shi, Jiulong Shan
To combine the best of both worlds, we present an unsupervised patch autoencoder based deep image decomposition (PAEDID) method for defective region segmentation.
1 code implementation • CVPR 2022 • Can Zhang, Tianyu Yang, Junwu Weng, Meng Cao, Jue Wang, Yuexian Zou
These pre-trained models can be sub-optimal for temporal localization tasks due to the inherent discrepancy between video-level classification and clip-level localization.
no code implementations • 3 Mar 2022 • Shancong Mou, Meng Cao, Zhendong Hong, Ping Huang, Jiulong Shan, Jianjun Shi
Display front-of-screen (FOS) quality inspection is essential for the mass production of displays in the manufacturing process.
1 code implementation • ICLR 2022 • Wentao Zhang, Yexin Wang, Zhenbang You, Meng Cao, Ping Huang, Jiulong Shan, Zhi Yang, Bin Cui
Graph Neural Networks (GNNs) have achieved great success in various tasks, but their performance highly relies on a large number of labeled nodes, which typically requires considerable human effort.
no code implementations • 22 Nov 2021 • Haoping Bai, Meng Cao, Ping Huang, Jiulong Shan
On active learning task, our method achieves 97. 0% Top-1 Accuracy on CIFAR10 with 0. 1% annotated data, and 83. 9% Top-1 Accuracy on CIFAR100 with 10% annotated data.
1 code implementation • NeurIPS 2021 • Wentao Zhang, Yexin Wang, Zhenbang You, Meng Cao, Ping Huang, Jiulong Shan, Zhi Yang, Bin Cui
Message passing is the core of most graph models such as Graph Convolutional Network (GCN) and Label Propagation (LP), which usually require a large number of clean labeled data to smooth out the neighborhood over the graph.
no code implementations • EMNLP 2021 • Meng Cao, Long Chen, Mike Zheng Shou, Can Zhang, Yuexian Zou
Almost all existing video grounding methods fall into two frameworks: 1) Top-down model: It predefines a set of segment candidates and then conducts segment classification and regression.
1 code implementation • ACL 2022 • Meng Cao, Yue Dong, Jackie Chi Kit Cheung
State-of-the-art abstractive summarization systems often generate \emph{hallucinations}; i. e., content that is not directly inferable from the source text.
Abstractive Text Summarization Reinforcement Learning (RL) +1
no code implementations • 12 Aug 2021 • Meng Cao, Can Zhang, Long Chen, Mike Zheng Shou, Yuexian Zou
In this paper, we analyze that the motion cues behind the optical flow features are complementary informative.
Optical Flow Estimation Weakly-supervised Temporal Action Localization +1
no code implementations • 12 Aug 2021 • Meng Cao, HaoZhi Huang, Hao Wang, Xuan Wang, Li Shen, Sheng Wang, Linchao Bao, Zhifeng Li, Jiebo Luo
Compared with the state-of-the-art facial image editing methods, our framework generates video portraits that are more photo-realistic and temporally smooth.
no code implementations • 24 Jun 2021 • Meng Cao, Can Zhang, Dongming Yang, Yuexian Zou
Compared to the traditional single-stage segmentation network, our NASK conducts the detection in a coarse-to-fine manner with the first stage segmentation spotting the rectangle text proposals and the second one retrieving compact representations.
no code implementations • NeurIPS 2021 • Haoping Bai, Meng Cao, Ping Huang, Jiulong Shan
While single-shot quantized neural architecture search enjoys flexibility in both model architecture and quantization policy, the combined search space comes with many challenges, including instability when training the weight-sharing supernet and difficulty in navigating the exponentially growing search space.
Hardware Aware Neural Architecture Search Model Optimization +2
no code implementations • 11 May 2021 • Xi Li, Meng Cao, Yingying Tang, Scott Johnston, Zhendong Hong, Huimin Ma, Jiulong Shan
Inspired by the observation that audiences have different visual preferences on foreground and background objects, we for the first time propose to use saliency masks in the evaluation processes of the task of video frame interpolation.
no code implementations • 30 Apr 2021 • Dongming Yang, Yuexian Zou, Can Zhang, Meng Cao, Jie Chen
Upon the frame, an Interaction Intensifier Module and a Correlation Parsing Module are carefully designed, where: a) interactive semantics from humans can be exploited and passed to objects to intensify interactions, b) interactive correlations among humans, objects and interactions are integrated to promote predictions.
1 code implementation • CVPR 2021 • Can Zhang, Meng Cao, Dongming Yang, Jie Chen, Yuexian Zou
In this paper, we argue that learning by comparing helps identify these hard snippets and we propose to utilize snippet Contrastive learning to Localize Actions, CoLA for short.
no code implementations • 31 Dec 2020 • Meng Cao
The construction of matrix-product codes with certain self-orthogonality over finite fields is an effective way to obtain good $q$-ary quantum codes of large length.
Information Theory Information Theory Quantum Physics
1 code implementation • EMNLP 2020 • Meng Cao, Yue Dong, Jiapeng Wu, Jackie Chi Kit Cheung
Experimental results show that our model is able to correct factual errors in summaries generated by other neural summarization models and outperforms previous models on factual consistency evaluation on the CNN/DailyMail dataset.
1 code implementation • EMNLP 2020 • Jiapeng Wu, Meng Cao, Jackie Chi Kit Cheung, William L. Hamilton
Our analysis also reveals important sources of variability both within and across TKG datasets, and we introduce several simple but strong baselines that outperform the prior state of the art in certain settings.
1 code implementation • 4 Aug 2020 • Farah E. Shamout, Yiqiu Shen, Nan Wu, Aakash Kaku, Jungkyu Park, Taro Makino, Stanisław Jastrzębski, Duo Wang, Ben Zhang, Siddhant Dogra, Meng Cao, Narges Razavian, David Kudlowitz, Lea Azour, William Moore, Yvonne W. Lui, Yindalon Aphinyanaphongs, Carlos Fernandez-Granda, Krzysztof J. Geras
In order to verify performance in a real clinical setting, we silently deployed a preliminary version of the deep neural network at New York University Langone Health during the first wave of the pandemic, which produced accurate predictions in real-time.
no code implementations • 3 Jul 2020 • Meng Cao, Hao-Zhi Huang, Hao Wang, Xuan Wang, Li Shen, Sheng Wang, Linchao Bao, Zhifeng Li, Jiebo Luo
Compared with the state-of-the-art facial image editing methods, our framework generates video portraits that are more photo-realistic and temporally smooth.
no code implementations • 26 Apr 2020 • Meng Cao, Yuexian Zou
Specifically, \textit{NASK} consists of a Text Instance Segmentation network namely \textit{TIS} (\(1^{st}\) stage), a Text RoI Pooling module and a Fiducial pOint eXpression module termed as \textit{FOX} (\(2^{nd}\) stage).
no code implementations • 18 Mar 2020 • Qing Tian, Yanan Zhu, Chuang Ma, Meng Cao
Unsupervised domain adaptation (UDA) is an emerging research topic in the field of machine learning and pattern recognition, which aims to help the learning of unlabeled target domain by transferring knowledge from the source domain.
1 code implementation • IJCNLP 2019 • Meng Cao, Jackie Chi Kit Cheung
Referring Expression Generation (REG) is the task of generating contextually appropriate references to entities.