no code implementations • 17 Apr 2024 • James Y. Huang, Wenxuan Zhou, Fei Wang, Fred Morstatter, Sheng Zhang, Hoifung Poon, Muhao Chen
Despite the strong capabilities of Large Language Models (LLMs) to acquire knowledge from their training corpora, the memorization of sensitive information in the corpora such as copyrighted, harmful, and private content has led to ethical and legal concerns.
1 code implementation • 17 Feb 2024 • Tianyi Yan, Fei Wang, James Y. Huang, Wenxuan Zhou, Fan Yin, Aram Galstyan, Wenpeng Yin, Muhao Chen
Instruction tuning has been used as a promising approach to improve the performance of large language models (LLMs) on unseen tasks.
1 code implementation • 23 Oct 2023 • Zekun Li, Wenxuan Zhou, Yao-Yi Chiang, Muhao Chen
This paper introduces GeoLM, a geospatially grounded language model that enhances the understanding of geo-entities in natural language.
no code implementations • 10 Oct 2023 • Fan Yang, Wenxuan Zhou, Zuxin Liu, Ding Zhao, David Held
This work introduces a novel approach that combines RL with trajectory optimization to manage this trade-off effectively.
no code implementations • 7 Aug 2023 • Wenxuan Zhou, Sheng Zhang, Yu Gu, Muhao Chen, Hoifung Poon
Instruction tuning has proven effective for distilling LLMs into more cost-efficient models such as Alpaca and Vicuna.
Ranked #1 on Named Entity Recognition (NER) on NCBI Disease
1 code implementation • 28 May 2023 • Fei Wang, James Y. Huang, Tianyi Yan, Wenxuan Zhou, Muhao Chen
However, previous ensemble-based debiasing methods typically apply debiasing on top-level logits without directly addressing biased attention patterns.
1 code implementation • 24 May 2023 • Fei Wang, Wenjie Mo, Yiwei Wang, Wenxuan Zhou, Muhao Chen
Building upon this SCM, we propose causal intervention techniques to mitigate entity bias for both white-box and black-box settings.
no code implementations • 24 May 2023 • Tianqing Fang, Zhaowei Wang, Wenxuan Zhou, Hongming Zhang, Yangqiu Song, Muhao Chen
However, knowledge conflicts arise when there is a mismatch between the actual temporal relations of events in the context and the prior knowledge or biases learned by the model.
1 code implementation • 22 May 2023 • Yiwei Wang, Bryan Hooi, Fei Wang, Yujun Cai, Yuxuan Liang, Wenxuan Zhou, Jing Tang, Manjuan Duan, Muhao Chen
In principle, textual context determines the ground-truth relation and the RE models should be able to correctly identify the relations reflected by the textual context.
no code implementations • 6 May 2023 • Wenxuan Zhou, Bowen Jiang, Fan Yang, Chris Paxton, David Held
In this work, we introduce Hybrid Actor-Critic Maps for Manipulation (HACMan), a reinforcement learning approach for 6D non-prehensile manipulation of objects using point cloud observations.
1 code implementation • 20 Mar 2023 • Wenxuan Zhou, Sheng Zhang, Hoifung Poon, Muhao Chen
However, their reliance on parametric knowledge may cause them to overlook contextual cues, leading to incorrect predictions in context-sensitive NLP tasks (e. g., knowledge acquisition tasks).
1 code implementation • 21 Dec 2022 • Keming Lu, I-Hung Hsu, Wenxuan Zhou, Mingyu Derek Ma, Muhao Chen
Relation Extraction (RE) has been extended to cross-document scenarios because many relations are not simply described in a single document.
no code implementations • 21 Dec 2022 • Wenxuan Zhou, Sheng Zhang, Tristan Naumann, Muhao Chen, Hoifung Poon
In this paper, we aim at bridging the gap and propose to pretrain and finetune the RE model using consistent objectives of contrastive learning.
no code implementations • 20 Dec 2022 • Tianqing Fang, Wenxuan Zhou, Fangyu Liu, Hongming Zhang, Yangqiu Song, Muhao Chen
However, data augmentation may introduce noisy data that impairs training.
no code implementations • 2 Nov 2022 • Wenxuan Zhou, David Held
Previous work in extrinsic dexterity usually has careful assumptions about contacts which impose restrictions on robot design, robot motions, and the variations of the physical parameters.
1 code implementation • 10 Oct 2022 • Xiaocong Yang, James Y. Huang, Wenxuan Zhou, Muhao Chen
Parameter-efficient tuning aims at updating only a small subset of parameters when adapting a pretrained model to downstream tasks.
1 code implementation • 19 May 2022 • Keming Lu, I-Hung Hsu, Wenxuan Zhou, Mingyu Derek Ma, Muhao Chen
Considering that summarization tasks aim at acquiring concise expressions of synoptical information from the longer context, these tasks naturally align with the objective of RE, i. e., extracting a kind of synoptical information that describes the relation of entity mentions.
Ranked #7 on Relation Extraction on TACRED
1 code implementation • NAACL 2022 • Yiwei Wang, Muhao Chen, Wenxuan Zhou, Yujun Cai, Yuxuan Liang, Dayiheng Liu, Baosong Yang, Juncheng Liu, Bryan Hooi
In this paper, we propose the CORE (Counterfactual Analysis based Relation Extraction) debiasing method that guides the RE models to focus on the main effects of textual context without losing the entity information.
no code implementations • Findings (NAACL) 2022 • Yiwei Wang, Muhao Chen, Wenxuan Zhou, Yujun Cai, Yuxuan Liang, Bryan Hooi
GRAPHCACHE aggregates the features from sentences in the whole dataset to learn global representations of properties, and use them to augment the local features within individual sentences.
1 code implementation • NAACL 2022 • Wenxuan Zhou, Qiang Ning, Heba Elfardy, Kevin Small, Muhao Chen
Current question answering (QA) systems primarily consider the single-answer scenario, where each question is assumed to be paired with one correct answer.
no code implementations • 12 Apr 2022 • Wenxuan Zhou, Steven Bohez, Jan Humplik, Abbas Abdolmaleki, Dushyant Rao, Markus Wulfmeier, Tuomas Haarnoja, Nicolas Heess
We propose the Offline Distillation Pipeline to break this trade-off by separating the training procedure into an online interaction phase and an offline distillation phase. Second, we find that training with the imbalanced off-policy data from multiple environments across the lifetime creates a significant performance drop.
no code implementations • 16 Dec 2021 • Wenxuan Zhou, Fangyu Liu, huan zhang, Muhao Chen
Deep neural networks are often overparameterized and may not easily achieve model generalization.
1 code implementation • ACL 2022 • Wenxuan Zhou, Fangyu Liu, Ivan Vulić, Nigel Collier, Muhao Chen
To achieve this, it is crucial to represent multilingual knowledge in a shared/unified space.
1 code implementation • EMNLP 2021 • Wenxuan Zhou, Fangyu Liu, Muhao Chen
Pretrained Transformers achieve remarkable performance when training and test data are from the same distribution.
1 code implementation • EMNLP 2021 • Wenxuan Zhou, Muhao Chen
Recent information extraction approaches have relied on training deep neural models.
Ranked #1 on Named Entity Recognition (NER) on CoNLL++
1 code implementation • 16 Mar 2021 • Harshit Sikchi, Wenxuan Zhou, David Held
Current RL agents explore the environment without considering these constraints, which can lead to damage to the hardware or even other agents in the environment.
1 code implementation • 2 Feb 2021 • Wenxuan Zhou, Muhao Chen
Sentence-level relation extraction (RE) aims at identifying the relationship between two entities in a sentence.
Ranked #2 on Relation Extraction on Re-TACRED
2 code implementations • 14 Nov 2020 • Wenxuan Zhou, Sujay Bajracharya, David Held
The goal of offline reinforcement learning is to learn a policy from a fixed dataset, without further interactions with the environment.
1 code implementation • 21 Oct 2020 • Wenxuan Zhou, Kevin Huang, Tengyu Ma, Jing Huang
In this paper, we propose two novel techniques, adaptive thresholding and localized context pooling, to solve the multi-label and multi-entity problems.
Ranked #6 on Relation Extraction on ReDocRED
Document-level Relation Extraction Multi-Label Classification +2
1 code implementation • 23 Aug 2020 • Harshit Sikchi, Wenxuan Zhou, David Held
In this work, we investigate a novel instantiation of H-step lookahead with a learned model and a terminal value function learned by a model-free off-policy algorithm, named Learning Off-Policy with Online Planning (LOOP).
no code implementations • 21 Aug 2020 • Ziqian Zeng, Wenxuan Zhou, Xin Liu, Zizheng Lin, Yangqin Song, Michael David Kuo, Wan Hang Keith Chiu
Our objective function is to predict an opinion word given a target word while our ultimate goal is to learn a sentiment classifier.
1 code implementation • 2 May 2020 • Wenxuan Zhou, Bill Yuchen Lin, Xiang Ren
Fine-tuning pre-trained language models (PTLMs), such as BERT and its better variant RoBERTa, has been a common practice for advancing performance in natural language understanding (NLU) tasks.
no code implementations • 10 Nov 2019 • Wenxuan Zhou, Junyi Du, Xiang Ren
Large pre-trained sentence encoders like BERT start a new chapter in natural language processing.
1 code implementation • ICLR 2020 • Ziqi Wang, Yujia Qin, Wenxuan Zhou, Jun Yan, Qinyuan Ye, Leonardo Neves, Zhiyuan Liu, Xiang Ren
While deep neural networks have achieved impressive performance on a range of NLP tasks, these data-hungry models heavily rely on labeled data, which restricts their applications in scenarios where data annotation is expensive.
2 code implementations • 5 Sep 2019 • Wenxuan Zhou, Hongtao Lin, Bill Yuchen Lin, Ziqi Wang, Junyi Du, Leonardo Neves, Xiang Ren
The soft matching module learns to match rules with semantically similar sentences such that raw corpora can be automatically labeled and leveraged by the RE module (in a much better coverage) as augmented supervision, in addition to the exactly matched sentences.
1 code implementation • ICLR 2019 • Wenxuan Zhou, Lerrel Pinto, Abhinav Gupta
A key challenge in reinforcement learning (RL) is environment generalization: a policy trained to solve a task in one environment often fails to solve the same task in a slightly different test environment.
1 code implementation • NAACL 2019 • Ziqian Zeng, Wenxuan Zhou, Xin Liu, Yangqiu Song
These word pairs can be extracted by using dependency parsers and simple rules.
1 code implementation • ACL 2018 • Yu Hong, Wenxuan Zhou, Jingli Zhang, Guodong Zhou, Qiaoming Zhu
Due to the ability of encoding and mapping semantic information into a high-dimensional latent feature space, neural networks have been successfully used for detecting events to a certain extent.
no code implementations • WS 2016 • Yu Hong, Liang Yao, Mengyi Liu, Tongtao Zhang, Wenxuan Zhou, Jianmin Yao, Heng Ji
We present a novel method of comparable corpora construction.
1 code implementation • 14 Nov 2012 • Amir Houmansadr, Wenxuan Zhou, Matthew Caesar, Nikita Borisov
As the operation of SWEET is not bound to specific email providers we argue that a censor will need to block all email communications in order to disrupt SWEET, which is infeasible as email constitutes an important part of today's Internet.
Cryptography and Security