1 code implementation • 20 Jan 2024 • Guangyuan Ma, Xing Wu, Zijia Lin, Songlin Hu
In this study, we aim to shed light on this issue by revealing that masked auto-encoder (MAE) pre-training with enhanced decoding significantly improves the term coverage of input tokens in dense representations, compared to vanilla BERT checkpoints.
1 code implementation • 30 Oct 2023 • Zhenpeng Su, Xing Wu, Xue Bai, Zijia Lin, Hui Chen, Guiguang Ding, Wei Zhou, Songlin Hu
Experiments reveal that models incorporating the proposed MiLe Loss can gain consistent performance improvement on downstream benchmarks.
1 code implementation • 6 Sep 2023 • Zhenpeng Su, Xing Wu, Wei Zhou, Guangyuan Ma, Songlin Hu
ChatGPT has gained significant interest due to its impressive performance, but people are increasingly concerned about its potential risks, particularly around the detection of AI-generated content (AIGC), which is often difficult for untrained humans to identify.
no code implementations • 16 Aug 2023 • Guangyuan Ma, Xing Wu, Peng Wang, Zijia Lin, Songlin Hu
Concretely, we leverage the capabilities of LLMs for document expansion, i. e. query generation, and effectively transfer expanded knowledge to retrievers using pre-training strategies tailored for passage retrieval.
1 code implementation • 7 Jun 2023 • Zhenpeng Su, Xing Wu, Wei Zhou, Guangyuan Ma, Songlin Hu
Dialogue response selection aims to select an appropriate response from several candidates based on a given user and system utterance history.
Ranked #1 on Conversational Response Selection on E-commerce
1 code implementation • 25 Apr 2023 • Guangyuan Ma, Hongtao Liu, Xing Wu, Wanhui Qian, Zhepeng Lv, Qing Yang, Songlin Hu
Firstly, we introduce the user behavior masking pre-training task to recover the masked user behaviors based on their contextual behaviors.
no code implementations • 20 Apr 2023 • Guangyuan Ma, Xing Wu, Peng Wang, Songlin Hu
Siamese or fully separated dual-encoders are often adopted as basic retrieval architecture in the pre-training and fine-tuning stages for encoding queries and passages into their latent embedding spaces.
no code implementations • 5 Apr 2023 • Xing Wu, Guangyuan Ma, Peng Wang, Meng Lin, Zijia Lin, Fuzheng Zhang, Songlin Hu
As an effective representation bottleneck pretraining technique, the contextual masked auto-encoder utilizes contextual embedding to assist in the reconstruction of passages.
2 code implementations • 19 Dec 2022 • Xing Wu, Guangyuan Ma, Wanhui Qian, Zijia Lin, Songlin Hu
Recently, methods have been developed to improve the performance of dense passage retrieval by using context-supervised pre-training.
1 code implementation • 13 Oct 2022 • Xing Wu, Chaochen Gao, Zijia Lin, Zhongyuan Wang, Jizhong Han, Songlin Hu
Sparse sampling is also likely to miss important frames corresponding to some text portions, resulting in textual redundancy.
2 code implementations • 8 Oct 2022 • Xing Wu, Chaochen Gao, Zijia Lin, Jizhong Han, Zhongyuan Wang, Songlin Hu
Contrastive learning has been extensively studied in sentence embedding learning, which assumes that the embeddings of different views of the same sentence are closer.
no code implementations • 18 Aug 2022 • Yike Guo, Qifeng Liu, Jie Chen, Wei Xue, Jie Fu, Henrik Jensen, Fernando Rosas, Jeffrey Shaw, Xing Wu, Jiji Zhang, Jianliang Xu
This report presents a comprehensive view of our vision on the development path of the human-machine symbiotic art creation.
2 code implementations • 16 Aug 2022 • Xing Wu, Guangyuan Ma, Meng Lin, Zijia Lin, Zhongyuan Wang, Songlin Hu
Dense passage retrieval aims to retrieve the relevant passages of a query from a large corpus based on dense representations (i. e., vectors) of the query and the passages.
1 code implementation • 8 Jul 2022 • Xing Wu, Qiulian Fang
In the cancer survival prediction for TCGA cases, SAEsurv-net addresses the curse of dimensionality with a two-stage dimensionality reduction strategy and handles multi-omics heterogeneity with a stacked autoencoder model.
1 code implementation • ACL 2022 • Xing Wu, Chaochen Gao, Meng Lin, Liangjun Zang, Zhongyuan Wang, Songlin Hu
Before entering the neural network, a token is generally converted to the corresponding one-hot representation, which is a discrete distribution of the vocabulary.
1 code implementation • 10 Dec 2021 • Chaochen Gao, Xing Wu, Peng Wang, Jue Wang, Liangjun Zang, Zhongyuan Wang, Songlin Hu
To tackle that, we propose an effective knowledge distillation framework for contrastive sentence embeddings, termed DistilCSE.
no code implementations • 30 Oct 2021 • Jue Wang, Haofan Wang, Xing Wu, Chaochen Gao, Debing Zhang
In this paper, we present TransAug (Translate as Augmentation), which provide the first exploration of utilizing translated sentence pairs as data augmentation for text, and introduce a two-stage paradigm to advances the state-of-the-art sentence embeddings.
2 code implementations • COLING 2022 • Xing Wu, Chaochen Gao, Yipeng Su, Jizhong Han, Zhongyuan Wang, Songlin Hu
Contrastive learning has been gradually applied to learn high-quality unsupervised sentence embedding.
2 code implementations • COLING 2022 • Xing Wu, Chaochen Gao, Liangjun Zang, Jizhong Han, Zhongyuan Wang, Songlin Hu
Unsup-SimCSE takes dropout as a minimal data augmentation method, and passes the same input sentence to a pre-trained Transformer encoder (with dropout turned on) twice to obtain the two corresponding embeddings to build a positive pair.
no code implementations • 8 May 2020 • Xing Wu, Yibing Liu, Xiangyang Zhou, dianhai yu
As an alternative, we propose a new method for BERT distillation, i. e., asking the teacher to generate smoothed word ids, rather than labels, for teaching the student model in knowledge distillation.
no code implementations • 22 Feb 2020 • Xiaohui Song, Liangjun Zang, Yipeng Su, Xing Wu, Jizhong Han, Songlin Hu
While several state-of-the-art approaches to dialogue state tracking (DST) have shown promising performances on several benchmarks, there is still a significant performance gap between seen slot values (i. e., values that occur in both training set and test set) and unseen ones (values that occur in training set but not in test set).
no code implementations • 5 Sep 2019 • Xing Wu, Dongjun Wei, Liangjun Zang, Jizhong Han, Songlin Hu
Automatic and human evaluation results show that TransSent can generate structured sentences with high quality, and has certain scalability in different tasks.
no code implementations • 21 Aug 2019 • Xing Wu, Tao Zhang, Liangjun Zang, Jizhong Han, Songlin Hu
So we propose a two step approach "Mask and Infill".
no code implementations • 28 Mar 2019 • Tao Zhang, Xing Wu, Meng Lin, Jizhong Han, Songlin Hu
Imbalanced data commonly exists in real world, espacially in sentiment-related corpus, making it difficult to train a classifier to distinguish latent sentiment in text data.
5 code implementations • 17 Dec 2018 • Xing Wu, Shangwen Lv, Liangjun Zang, Jizhong Han, Songlin Hu
BERT demonstrates that a deep bidirectional language model is more powerful than either an unidirectional language model or the shallow concatenation of a forward and backward model.
no code implementations • 30 Aug 2018 • Changqing Zou, Haoran Mo, Ruofei Du, Xing Wu, Chengying Gao, Hongbo Fu
We introduce LUCSS, a language-based system for interactive col- orization of scene sketches, based on their semantic understanding.