no code implementations • ECCV 2020 • Shuchen Weng, Wenbo Li, Dawei Li, Hongxia Jin, Boxin Shi
We study conditional image repainting where a model is trained to generate visual content conditioned on user inputs, and composite the generated content seamlessly onto a user provided image while preserving the semantics of users' inputs.
no code implementations • NAACL 2022 • Yu Wang, V.srinivasan@samsung.com V.srinivasan@samsung.com, Hongxia Jin
Knowledge based question answering (KBQA) is a complex task for natural language understanding.
no code implementations • 24 Jan 2025 • James Seale Smith, Chi-Heng Lin, Shikhar Tuli, Haris Jeelani, Shangqian Gao, Yilin Shen, Hongxia Jin, Yen-Chang Hsu
The rapid proliferation of large language models (LLMs) in natural language processing (NLP) has created a critical need for techniques that enable efficient deployment on memory-constrained devices without compromising performance.
1 code implementation • 15 Oct 2024 • Shangqian Gao, Chi-Heng Lin, Ting Hua, Tang Zheng, Yilin Shen, Hongxia Jin, Yen-Chang Hsu
We evaluate our method on various LLMs, including OPT, LLaMA, LLaMA-2, Phi-1. 5, and Phi-2.
no code implementations • 19 Aug 2024 • Chi-Heng Lin, Shangqian Gao, James Seale Smith, Abhishek Patel, Shikhar Tuli, Yilin Shen, Hongxia Jin, Yen-Chang Hsu
Large Language Models (LLMs) have reshaped the landscape of artificial intelligence by demonstrating exceptional performance across various tasks.
no code implementations • 26 Jun 2024 • Vikas Yadav, Hyuk Joon Kwon, Vijay Srinivasan, Hongxia Jin
Question Answer Generation (QAG) is an effective data augmentation technique to improve the accuracy of question answering systems, especially in low-resource domains.
no code implementations • 1 May 2024 • Shikhar Tuli, Chi-Heng Lin, Yen-Chang Hsu, Niraj K. Jha, Yilin Shen, Hongxia Jin
We also propose systematic qualitative and quantitative methods to rigorously test the quality of generated text for non-autoregressive generation.
no code implementations • 25 Dec 2023 • Avik Ray, Yilin Shen, Hongxia Jin
State-of-the-art spoken language understanding (SLU) models have shown tremendous success in benchmark SLU datasets, yet they still fail in many practical scenario due to the lack of model compositionality when trained on limited training data.
no code implementations • 2 Dec 2023 • Minchul Kim, Shangqian Gao, Yen-Chang Hsu, Yilin Shen, Hongxia Jin
In this paper, we introduce "Token Fusion" (ToFu), a method that amalgamates the benefits of both token pruning and token merging.
no code implementations • 2 Dec 2023 • Lingyu Zhang, Ting Hua, Yilin Shen, Hongxia Jin
In order to achieve this goal, a model has to be "smart" and "knowledgeable".
no code implementations • 30 Nov 2023 • James Seale Smith, Yen-Chang Hsu, Zsolt Kira, Yilin Shen, Hongxia Jin
We show that STAMINA outperforms the prior SOTA for the setting of text-to-image continual customization on a 50-concept benchmark composed of landmarks and human faces, with no stored replay data.
no code implementations • 25 Sep 2023 • Kalpa Gunaratna, Vijay Srinivasan, Hongxia Jin
Therefore to bridge this gap, we transform the full joint NLU model to be `inherently' explainable at granular levels without compromising on accuracy.
1 code implementation • 31 Jul 2023 • Jun Yan, Vikas Yadav, Shiyang Li, Lichang Chen, Zheng Tang, Hai Wang, Vijay Srinivasan, Xiang Ren, Hongxia Jin
To demonstrate the threat, we propose a simple method to perform VPI by poisoning the model's instruction tuning data, which proves highly effective in steering the LLM.
no code implementations • 20 Jul 2023 • Shiyang Li, Jun Yan, Hai Wang, Zheng Tang, Xiang Ren, Vijay Srinivasan, Hongxia Jin
We conduct a comprehensive evaluation of four major model families across nine datasets, employing twelve sets of verbalizers for each of them.
3 code implementations • 17 Jul 2023 • Lichang Chen, Shiyang Li, Jun Yan, Hai Wang, Kalpa Gunaratna, Vikas Yadav, Zheng Tang, Vijay Srinivasan, Tianyi Zhou, Heng Huang, Hongxia Jin
Large language models (LLMs) strengthen instruction-following capability through instruction-finetuning (IFT) on supervised instruction/response data.
no code implementations • 12 Apr 2023 • James Seale Smith, Yen-Chang Hsu, Lingyu Zhang, Ting Hua, Zsolt Kira, Yilin Shen, Hongxia Jin
We show that C-LoRA not only outperforms several baselines for our proposed setting of text-to-image continual customization, which we refer to as Continual Diffusion, but that we achieve a new state-of-the-art in the well-established rehearsal-free continual learning setting for image classification.
no code implementations • 6 Apr 2023 • Yashas Malur Saidutta, Rakshith Sharma Srinivasa, Ching-Hua Lee, Chouchang Yang, Yilin Shen, Hongxia Jin
We show that existing deep keyword spotting mechanisms can be improved by Successive Refinement, where the system first classifies whether the input audio is speech or not, followed by whether the input is keyword-like or not, and finally classifies which keyword was uttered.
no code implementations • 30 Jan 2023 • Kaiwen Zhou, Kaizhi Zheng, Connor Pryor, Yilin Shen, Hongxia Jin, Lise Getoor, Xin Eric Wang
Such object navigation tasks usually require large-scale training in visual environments with labeled objects, which generalizes poorly to novel objects in unknown environments.
no code implementations • 13 Jan 2023 • Miao Yin, Burak Uzkent, Yilin Shen, Hongxia Jin, Bo Yuan
We first develop a graph-based ranking for measuring the importance of attention heads, and the extracted importance information is further integrated to an optimization-based procedure to impose the heterogeneous structured sparsity patterns on the ViT models.
no code implementations • 20 Dec 2022 • Yu Wang, Hongxia Jin
A coreference resolution system is to cluster all mentions that refer to the same entity in a given context.
no code implementations • 18 Dec 2022 • Yu Wang, Hongxia Jin
In this paper, we introduce a robust semantic frame parsing pipeline that can handle both \emph{OOD} patterns and \emph{OOV} tokens in conjunction with a new complex Twitter dataset that contains long tweets with more \emph{OOD} patterns and \emph{OOV} tokens.
no code implementations • 18 Dec 2022 • Yu Wang, Hongxia Jin
The target of a coreference resolution system is to cluster all mentions that refer to the same entity in a given context.
no code implementations • 2 Nov 2022 • Ting Hua, Yen-Chang Hsu, Felicity Wang, Qian Lou, Yilin Shen, Hongxia Jin
However, standard SVD treats the parameters within the matrix with equal importance, which is a simple but unrealistic assumption.
no code implementations • 19 Oct 2022 • Kalpa Gunaratna, Vijay Srinivasan, Akhila Yerukola, Hongxia Jin
In this work, we propose a novel approach that: (i) learns to generate additional slot type specific features in order to improve accuracy and (ii) provides explanations for slot filling decisions for the first time in a joint NLU model.
no code implementations • ICLR 2022 • Yen-Chang Hsu, Ting Hua, SungEn Chang, Qian Lou, Yilin Shen, Hongxia Jin
In other words, the optimization objective of SVD is not aligned with the trained model's task accuracy.
no code implementations • 18 Mar 2022 • Yen-Chang Hsu, James Smith, Yilin Shen, Zsolt Kira, Hongxia Jin
Knowledge distillation (KD) is a substantial strategy for transferring learned knowledge from one neural network model to another.
no code implementations • 25 Jan 2022 • Peixi Xiong, Yilin Shen, Hongxia Jin
In contrast to previous works, our model splits alignment into different levels to achieve learning better correlations without needing additional data and annotations.
no code implementations • NAACL 2021 • Ting Hua, Yilin Shen, Changsheng Zhao, Yen-Chang Hsu, Hongxia Jin
Most existing continual learning approaches suffer from low accuracy and performance fluctuation, especially when the distributions of old and new data are significantly different.
no code implementations • CVPR 2022 • Qian Lou, Yen-Chang Hsu, Burak Uzkent, Ting Hua, Yilin Shen, Hongxia Jin
The key primitive is that Dictionary-Lookup-Transformormations (DLT) is proposed to replace Linear Transformation (LT) in multi-modal detectors where each weight in Linear Transformation (LT) is approximately factorized into a smaller dictionary, index, and coefficient.
no code implementations • 30 Dec 2021 • Changsheng Zhao, Ting Hua, Yilin Shen, Qian Lou, Hongxia Jin
Knowledge distillation, Weight pruning, and Quantization are known to be the main directions in model compression.
1 code implementation • 13 Dec 2021 • Manas Gaur, Kalpa Gunaratna, Vijay Srinivasan, Hongxia Jin
To address this open problem, we propose Information SEEking Question generator (ISEEQ), a novel approach for generating ISQs from just a short user query, given a large text corpus relevant to the user query.
no code implementations • 28 Oct 2021 • Junjiao Tian, Yen-Change Hsu, Yilin Shen, Hongxia Jin, Zsolt Kira
We are the first to propose a method that works well across both OOD detection and calibration and under different types of shifts.
no code implementations • 29 Sep 2021 • Junjiao Tian, Yen-Chang Hsu, Yilin Shen, Hongxia Jin, Zsolt Kira
To this end, we theoretically derive two score functions for OOD detection, the covariate shift score and concept shift score, based on the decomposition of KL-divergence for both scores, and propose a geometrically-inspired method (Geometric ODIN) to improve OOD detection under both shifts with only in-distribution data.
no code implementations • ICLR 2022 • Qian Lou, Ting Hua, Yen-Chang Hsu, Yilin Shen, Hongxia Jin
DictFormer significantly reduces the redundancy in the transformer's parameters by replacing the prior transformer's parameters with compact, shared dictionary, a few unshared coefficients, and indices.
no code implementations • 23 Aug 2021 • Kalpa Gunaratna, Vijay Srinivasan, Sandeep Nama, Hongxia Jin
Information Extraction from visual documents enables convenient and intelligent assistance to end users.
no code implementations • ACL 2021 • Yilin Shen, Yen-Chang Hsu, Avik Ray, Hongxia Jin
This paper proposes to train a model with only IND data while supporting both IND intent classification and OOD detection.
2 code implementations • ICCV 2021 • James Smith, Yen-Chang Hsu, Jonathan Balloch, Yilin Shen, Hongxia Jin, Zsolt Kira
Modern computer vision applications suffer from catastrophic forgetting when incrementally learning new concepts over time.
Ranked #5 on
Class Incremental Learning
on cifar100
no code implementations • 6 Jun 2021 • Yu Wang, Yilin Shen, Hongxia Jin
In this paper, we introduce a novel multi-step spoken language understanding system based on adversarial learning that can leverage the multiround user's feedback to update slot values.
no code implementations • 1 Jun 2021 • Yu Wang, Hongxia Jin
In this paper, we present a coarse to fine question answering (CFQA) system based on reinforcement learning which can efficiently processes documents with different lengths by choosing appropriate actions.
no code implementations • EACL 2021 • Akhila Yerukola, Mason Bretan, Hongxia Jin
We introduce a data augmentation technique based on byte pair encoding and a BERT-like self-attention model to boost performance on spoken language understanding tasks.
no code implementations • 29 Mar 2021 • Kalpa Gunaratna, Yu Wang, Hongxia Jin
Then we learn entity embeddings through this new type of triples.
2 code implementations • ICLR 2021 • Abhishek Sinha, Kumar Ayush, Jiaming Song, Burak Uzkent, Hongxia Jin, Stefano Ermon
Empirically, models trained with our method achieve improved conditional/unconditional image generation along with improved anomaly detection capabilities.
Ranked #8 on
Image Generation
on CIFAR-100
no code implementations • ICLR 2021 • Qian Lou, Yilin Shen, Hongxia Jin, Lei Jiang
A cryptographic neural network inference service is an efficient way to allow two parties to execute neural network inference without revealing either party’s data or model.
no code implementations • 16 Oct 2020 • Yilin Shen, Wenhu Chen, Hongxia Jin
We design a Dirichlet Prior RNN to model high-order uncertainty by degenerating as softmax layer for RNN model training.
no code implementations • EMNLP 2020 • Wei-Jen Ko, Avik Ray, Yilin Shen, Hongxia Jin
Existing open-domain dialogue generation models are usually trained to mimic the gold response in the training set using cross-entropy loss on the vocabulary.
no code implementations • 22 May 2020 • Kechen Qin, Yu Wang, Cheng Li, Kalpa Gunaratna, Hongxia Jin, Virgil Pavlu, Javed A. Aslam
Multi-hop knowledge based question answering (KBQA) is a complex task for natural language understanding.
3 code implementations • 4 May 2020 • Yang Cao, Yonghui Xiao, Shun Takagi, Li Xiong, Masatoshi Yoshikawa, Yilin Shen, Jinfei Liu, Hongxia Jin, Xiaofeng Xu
Third, we design a private location trace release framework that pipelines the detection of location exposure, policy graph repair, and private trajectory release with customizable and rigorous location privacy.
Cryptography and Security Computers and Society
no code implementations • 4 May 2020 • Ruiyi Zhang, Tong Yu, Yilin Shen, Hongxia Jin, Changyou Chen, Lawrence Carin
Text-based interactive recommendation provides richer user feedback and has demonstrated advantages over traditional interactive recommender systems.
2 code implementations • CVPR 2020 • Yen-Chang Hsu, Yilin Shen, Hongxia Jin, Zsolt Kira
Deep neural networks have attained remarkable performance when applied to data that comes from the same distribution as that of the training set, but can significantly degrade otherwise.
Out-of-Distribution Detection
Out of Distribution (OOD) Detection
no code implementations • NeurIPS 2019 • Ruiyi Zhang, Tong Yu, Yilin Shen, Hongxia Jin, Changyou Chen
Text-based interactive recommendation provides richer user preferences and has demonstrated advantages over traditional interactive recommender systems.
no code implementations • IJCNLP 2019 • Yilin Shen, Xiangyu Zeng, Hongxia Jin
ProgModel consists of a novel context gate that transfers previously learned knowledge to a small size expanded component; and meanwhile enables this new component to be fast trained to learn from new data.
no code implementations • WS 2019 • Avik Ray, Yilin Shen, Hongxia Jin
However, state-of-the art attention based neural parsers are slow to retrain which inhibits real time domain adaptation.
no code implementations • 15 Oct 2019 • Avik Ray, Yilin Shen, Hongxia Jin
Recurrent neural network (RNN) based joint intent classification and slot tagging models have achieved tremendous success in recent years for building spoken language understanding and dialog systems.
no code implementations • NAACL 2019 • Yilin Shen, Avik Ray, Hongxia Jin, S Nama, eep
We present SkillBot that takes the first step to enable end users to teach new skills in personal assistants (PA).
no code implementations • ICCV 2019 • Ramprasaath R. Selvaraju, Stefan Lee, Yilin Shen, Hongxia Jin, Shalini Ghosh, Larry Heck, Dhruv Batra, Devi Parikh
Many vision and language models suffer from poor visual grounding - often falling back on easy-to-learn language priors rather than basing their decisions on visual concepts in the image.
no code implementations • 26 Dec 2018 • Yu Wang, Abhishek Patel, Hongxia Jin
In this paper, a new deep reinforcement learning based augmented general sequence tagging system is proposed.
1 code implementation • NAACL 2018 • Yu Wang, Yilin Shen, Hongxia Jin
The most effective algorithms are based on the structures of sequence to sequence models (or "encoder-decoder" models), and generate the intents and semantic tags either using separate models or a joint model.
Ranked #1 on
Intent Detection
on ATIS
no code implementations • ICLR 2019 • Wenhu Chen, Yilin Shen, Hongxia Jin, William Wang
With the recently rapid development in deep learning, deep neural networks have been widely adopted in many real-life applications.
1 code implementation • EMNLP 2018 • Bailin Wang, Wei Lu, Yu Wang, Hongxia Jin
It is common that entity mentions can contain other mentions recursively.
Ranked #6 on
Nested Named Entity Recognition
on NNE
Nested Mention Recognition
Nested Named Entity Recognition
+1
no code implementations • 18 Sep 2018 • Yilin Shen, Xiangyu Zeng, Yu Wang, Hongxia Jin
The results show that our approach leverages such simple user information to outperform state-of-the-art approaches by 0. 25% for intent detection and 0. 31% for slot filling using standard training data.
no code implementations • 17 Sep 2018 • Avik Ray, Yilin Shen, Hongxia Jin
Learning intents and slot labels from user utterances is a fundamental step in all spoken language understanding (SLU) and dialog systems.
no code implementations • COLING 2018 • Yu Wang, Abhishek Patel, Hongxia Jin
In this paper, a new deep reinforcement learning based augmented general tagging system is proposed.
Deep Reinforcement Learning
Named Entity Recognition (NER)
+4
no code implementations • ACL 2018 • Yilin Shen, Avik Ray, Abhishek Patel, Hongxia Jin
We present a system, CRUISE, that guides ordinary software developers to build a high quality natural language understanding (NLU) engine from scratch.
no code implementations • 1 Jan 2018 • Walid Shalaby, Wlodek Zadrozny, Hongxia Jin
We evaluate our concept embedding models on two tasks: (1) analogical reasoning, where we achieve a state-of-the-art performance of 91% on semantic analogies, (2) concept categorization, where we achieve a state-of-the-art performance on two benchmark datasets achieving categorization accuracy of 100% on one and 98% on the other.
no code implementations • 21 Oct 2017 • Shiva Prasad Kasiviswanathan, Nina Narodytska, Hongxia Jin
Deep neural networks are powerful learning models that achieve state-of-the-art performance on many computer vision, speech, and language processing tasks.
no code implementations • 4 Jan 2017 • Shiva Prasad Kasiviswanathan, Kobbi Nissim, Hongxia Jin
Our first contribution is a generic transformation of private batch ERM mechanisms into private incremental ERM mechanisms, based on a simple idea of invoking the private batch ERM procedure at some regular time intervals.