Alert button
Picture for Kai-Wei Chang

Kai-Wei Chang

Alert button

Understanding ME? Multimodal Evaluation for Fine-grained Visual Commonsense

Nov 10, 2022
Zhecan Wang, Haoxuan You, Yicheng He, Wenhao Li, Kai-Wei Chang, Shih-Fu Chang

Figure 1 for Understanding ME? Multimodal Evaluation for Fine-grained Visual Commonsense
Figure 2 for Understanding ME? Multimodal Evaluation for Fine-grained Visual Commonsense
Figure 3 for Understanding ME? Multimodal Evaluation for Fine-grained Visual Commonsense
Figure 4 for Understanding ME? Multimodal Evaluation for Fine-grained Visual Commonsense
Viaarxiv icon

Unsupervised Syntactically Controlled Paraphrase Generation with Abstract Meaning Representations

Nov 02, 2022
Kuan-Hao Huang, Varun Iyer, Anoop Kumar, Sriram Venkatapathy, Kai-Wei Chang, Aram Galstyan

Figure 1 for Unsupervised Syntactically Controlled Paraphrase Generation with Abstract Meaning Representations
Figure 2 for Unsupervised Syntactically Controlled Paraphrase Generation with Abstract Meaning Representations
Figure 3 for Unsupervised Syntactically Controlled Paraphrase Generation with Abstract Meaning Representations
Figure 4 for Unsupervised Syntactically Controlled Paraphrase Generation with Abstract Meaning Representations
Viaarxiv icon

Investigating Ensemble Methods for Model Robustness Improvement of Text Classifiers

Oct 28, 2022
Jieyu Zhao, Xuezhi Wang, Yao Qin, Jilin Chen, Kai-Wei Chang

Figure 1 for Investigating Ensemble Methods for Model Robustness Improvement of Text Classifiers
Figure 2 for Investigating Ensemble Methods for Model Robustness Improvement of Text Classifiers
Figure 3 for Investigating Ensemble Methods for Model Robustness Improvement of Text Classifiers
Figure 4 for Investigating Ensemble Methods for Model Robustness Improvement of Text Classifiers
Viaarxiv icon

How well can Text-to-Image Generative Models understand Ethical Natural Language Interventions?

Oct 27, 2022
Hritik Bansal, Da Yin, Masoud Monajatipoor, Kai-Wei Chang

Figure 1 for How well can Text-to-Image Generative Models understand Ethical Natural Language Interventions?
Figure 2 for How well can Text-to-Image Generative Models understand Ethical Natural Language Interventions?
Figure 3 for How well can Text-to-Image Generative Models understand Ethical Natural Language Interventions?
Figure 4 for How well can Text-to-Image Generative Models understand Ethical Natural Language Interventions?
Viaarxiv icon

The Tail Wagging the Dog: Dataset Construction Biases of Social Bias Benchmarks

Oct 18, 2022
Nikil Roashan Selvam, Sunipa Dev, Daniel Khashabi, Tushar Khot, Kai-Wei Chang

Figure 1 for The Tail Wagging the Dog: Dataset Construction Biases of Social Bias Benchmarks
Figure 2 for The Tail Wagging the Dog: Dataset Construction Biases of Social Bias Benchmarks
Figure 3 for The Tail Wagging the Dog: Dataset Construction Biases of Social Bias Benchmarks
Figure 4 for The Tail Wagging the Dog: Dataset Construction Biases of Social Bias Benchmarks
Viaarxiv icon

SUPERB @ SLT 2022: Challenge on Generalization and Efficiency of Self-Supervised Speech Representation Learning

Oct 16, 2022
Tzu-hsun Feng, Annie Dong, Ching-Feng Yeh, Shu-wen Yang, Tzu-Quan Lin, Jiatong Shi, Kai-Wei Chang, Zili Huang, Haibin Wu, Xuankai Chang, Shinji Watanabe, Abdelrahman Mohamed, Shang-Wen Li, Hung-yi Lee

Figure 1 for SUPERB @ SLT 2022: Challenge on Generalization and Efficiency of Self-Supervised Speech Representation Learning
Figure 2 for SUPERB @ SLT 2022: Challenge on Generalization and Efficiency of Self-Supervised Speech Representation Learning
Figure 3 for SUPERB @ SLT 2022: Challenge on Generalization and Efficiency of Self-Supervised Speech Representation Learning
Figure 4 for SUPERB @ SLT 2022: Challenge on Generalization and Efficiency of Self-Supervised Speech Representation Learning
Viaarxiv icon

Watermarking Pre-trained Language Models with Backdooring

Oct 14, 2022
Chenxi Gu, Chengsong Huang, Xiaoqing Zheng, Kai-Wei Chang, Cho-Jui Hsieh

Figure 1 for Watermarking Pre-trained Language Models with Backdooring
Figure 2 for Watermarking Pre-trained Language Models with Backdooring
Figure 3 for Watermarking Pre-trained Language Models with Backdooring
Figure 4 for Watermarking Pre-trained Language Models with Backdooring
Viaarxiv icon

An Analysis of the Effects of Decoding Algorithms on Fairness in Open-Ended Language Generation

Oct 07, 2022
Jwala Dhamala, Varun Kumar, Rahul Gupta, Kai-Wei Chang, Aram Galstyan

Figure 1 for An Analysis of the Effects of Decoding Algorithms on Fairness in Open-Ended Language Generation
Figure 2 for An Analysis of the Effects of Decoding Algorithms on Fairness in Open-Ended Language Generation
Figure 3 for An Analysis of the Effects of Decoding Algorithms on Fairness in Open-Ended Language Generation
Figure 4 for An Analysis of the Effects of Decoding Algorithms on Fairness in Open-Ended Language Generation
Viaarxiv icon

Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoning

Sep 29, 2022
Pan Lu, Liang Qiu, Kai-Wei Chang, Ying Nian Wu, Song-Chun Zhu, Tanmay Rajpurohit, Peter Clark, Ashwin Kalyan

Figure 1 for Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoning
Figure 2 for Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoning
Figure 3 for Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoning
Figure 4 for Dynamic Prompt Learning via Policy Gradient for Semi-structured Mathematical Reasoning
Viaarxiv icon

MAGIC: Mask-Guided Image Synthesis by Inverting a Quasi-Robust Classifier

Sep 23, 2022
Mozhdeh Rouhsedaghat, Masoud Monajatipoor, Kai-Wei Chang, C. -C. Jay Kuo, Iacopo Masi

Figure 1 for MAGIC: Mask-Guided Image Synthesis by Inverting a Quasi-Robust Classifier
Figure 2 for MAGIC: Mask-Guided Image Synthesis by Inverting a Quasi-Robust Classifier
Figure 3 for MAGIC: Mask-Guided Image Synthesis by Inverting a Quasi-Robust Classifier
Figure 4 for MAGIC: Mask-Guided Image Synthesis by Inverting a Quasi-Robust Classifier
Viaarxiv icon