no code implementations • 26 Feb 2024 • Khai Jiet Liong, Hongqiu Wu, Hai Zhao
(2) We introduce \textit{S-Attend}, a novel smoothing technique that effectively makes SA robust via structural perturbations.
1 code implementation • 17 Dec 2023 • Haoxin Lin, Hongqiu Wu, Jiaji Zhang, Yihao Sun, Junyin Ye, Yang Yu
Real-world decision-making problems are usually accompanied by delayed rewards, which affects the sample efficiency of Reinforcement Learning, especially in the extremely delayed case where the only feedback is the episodic reward obtained at the end of an episode.
1 code implementation • 9 Oct 2023 • Hongqiu Wu, Linfeng Liu, Hai Zhao, Min Zhang
Beyond the great cognitive powers showcased by language models, it is crucial to scrutinize whether their reasoning capabilities stem from strong generalization or merely exposure to relevant data.
2 code implementations • 17 Aug 2023 • Linfeng Liu, Hongqiu Wu, Hai Zhao
However, we note a critical flaw in the process of tagging one character to another, that the correction is excessively conditioned on the error.
1 code implementation • 28 May 2023 • Hongqiu Wu, Shaohua Zhang, Yuchen Zhang, Hai Zhao
In this paper, we study Chinese Spelling Correction (CSC) as a joint decision made by two separate models: a language model and an error model.
no code implementations • 9 May 2023 • Yifei Yang, Hongqiu Wu, Hai Zhao
This is due to the fine-grained nature of NER, as even minor word changes in the sentence can result in the emergence or mutation of any entities, resulting in invalid adversarial examples.
1 code implementation • 8 May 2023 • Hongqiu Wu, Yongxiang Liu, Hanwen Shi, Hai Zhao, Min Zhang
Based on the observation, we propose simple yet effective \textit{Contextualized representation-Adversarial Training} (CreAT), in which the attack is explicitly optimized to deviate the contextualized representation of the encoder.
2 code implementations • 19 Oct 2022 • Hongqiu Wu, Ruixue Ding, Hai Zhao, Boli Chen, Pengjun Xie, Fei Huang, Min Zhang
Multiple pre-training objectives fill the vacancy of the understanding capability of single-objective language modeling, which serves the ultimate purpose of pre-trained language models (PrLMs), generalizing well on a mass of scenarios.
1 code implementation • COLING 2022 • Yiyang Li, Hongqiu Wu, Hai Zhao
Based on the tremendous success of pre-trained language models (PrLMs) for source code comprehension tasks, current literature studies either ways to further improve the performance (generalization) of PrLMs, or their robustness against adversarial attacks.
1 code implementation • 25 Jun 2022 • Hongqiu Wu, Ruixue Ding, Hai Zhao, Pengjun Xie, Fei Huang, Min Zhang
Deep neural models (e. g. Transformer) naturally learn spurious features, which create a ``shortcut'' between the labels and inputs, thus impairing the generalization and robustness.
Ranked #1 on Machine Reading Comprehension on DREAM
Machine Reading Comprehension Named Entity Recognition (NER) +4
no code implementations • NeurIPS 2021 • Hongqiu Wu, Hai Zhao, Min Zhang
Beyond the success story of pre-trained language models (PrLMs) in recent natural language processing, they are susceptible to over-fitting due to unusual large model size.
1 code implementation • Findings (ACL) 2021 • Hongqiu Wu, Hai Zhao, Min Zhang
Code summarization (CS) is becoming a promising area in recent language understanding, which aims to generate sensible human language automatically for programming language in the format of source code, serving in the most convenience of programmer developing.