1 code implementation • 29 May 2025 • Ziyin Zhang, Jiahao Xu, Zhiwei He, Tian Liang, Qiuzhi Liu, Yansi Li, Linfeng Song, Zhengwen Liang, Zhuosheng Zhang, Rui Wang, Zhaopeng Tu, Haitao Mi, Dong Yu
In this work, we propose DeepTheorem, a comprehensive informal theorem-proving framework exploiting natural language to enhance LLM mathematical reasoning.
no code implementations • 20 May 2025 • Yijie Lu, Tianjie Ju, Manman Zhao, Xinbei Ma, Yuan Guo, Zhuosheng Zhang
As multimodal agents are increasingly trained to operate graphical user interfaces (GUIs) to complete user tasks, they face a growing threat from indirect prompt injection, attacks in which misleading instructions are embedded into the agent's visual environment, such as popups or chat messages, and misinterpreted as part of the intended task.
no code implementations • 20 May 2025 • Pengzhou Cheng, Haowen Hu, Zheng Wu, Zongru Wu, Tianjie Ju, Daizong Ding, Zhuosheng Zhang, Gongshen Liu
Graphical user interface (GUI) agents powered by multimodal large language models (MLLMs) have shown greater promise for human-interaction.
1 code implementation • 19 May 2025 • Zheng Wu, Pengzhou Cheng, Zongru Wu, Lingzhong Dong, Zhuosheng Zhang
Traditional OOD detection methods perform suboptimally in this domain due to the complex embedding space and evolving GUI environments.
no code implementations • 16 May 2025 • Lingxiao Diao, Xinyue Xu, Wanxuan Sun, Cheng Yang, Zhuosheng Zhang
Previous studies have made notable progress in benchmarking the instruction following capabilities of LLMs in general domains, with a primary focus on their inherent commonsense knowledge.
no code implementations • 25 Apr 2025 • Hongyu Zhu, Sichu Liang, Wenwen Wang, Boheng Li, Tongxin Yuan, Fangqi Li, Shilin Wang, Zhuosheng Zhang
In this work, we revisit these advances and uncover a critical issue: current MI benchmarks suffer from distribution shifts between member and non-member images, introducing shortcut cues that inflate MI performance.
1 code implementation • 15 Apr 2025 • Zhiwei He, Tian Liang, Jiahao Xu, Qiuzhi Liu, Xingyu Chen, Yue Wang, Linfeng Song, Dian Yu, Zhenwen Liang, Wenxuan Wang, Zhuosheng Zhang, Rui Wang, Zhaopeng Tu, Haitao Mi, Dong Yu
The capacity for complex mathematical reasoning is a key benchmark for artificial intelligence.
1 code implementation • 14 Apr 2025 • Tianjie Ju, Zhenyu Shao, Bowen Wang, Yujia Chen, Zhuosheng Zhang, Hao Fei, Mong-Li Lee, Wynne Hsu, Sufeng Duan, Gongshen Liu
We conduct probing experiments on 11 open-source LLMs over the PersonalityEdit benchmark and find that LLMs predominantly encode personality for responding in their middle and upper layers, with instruction-tuned models demonstrating a slightly clearer separation of personality traits.
no code implementations • 21 Mar 2025 • Yansi Li, Jiahao Xu, Tian Liang, Xingyu Chen, Zhiwei He, Qiuzhi Liu, Rui Wang, Zhuosheng Zhang, Zhaopeng Tu, Haitao Mi, Dong Yu
Traditional inference time scaling methods utilize scalar reward signals from process reward models to evaluate candidate reasoning steps, but these scalar rewards lack the nuanced qualitative information essential for understanding and justifying each step.
1 code implementation • 3 Mar 2025 • Tianjie Ju, Yi Hua, Hao Fei, Zhenyu Shao, Yubin Zheng, Haodong Zhao, Mong-Li Lee, Wynne Hsu, Zhuosheng Zhang, Gongshen Liu
Multi-Modal Large Language Models (MLLMs) have exhibited remarkable performance on various vision-language tasks such as Visual Question Answering (VQA).
no code implementations • 1 Mar 2025 • Zongru Wu, Pengzhou Cheng, Zheng Wu, Tianjie Ju, Zhuosheng Zhang, Gongshen Liu
Perception-enhanced pre-training, particularly through grounding techniques, is widely adopted to enhance the performance of graphical user interface (GUI) agents.
1 code implementation • 1 Mar 2025 • Yao Yao, Yifei Yang, Xinbei Ma, Dongjie Yang, Zhuosheng Zhang, Zuchao Li, Hai Zhao
Lastly, we examine whether LLMs are influenced by attention-grabbing headlines with larger semantic sizes in a real-world web shopping scenario.
1 code implementation • 26 Feb 2025 • Pengzhou Cheng, Zheng Wu, Zongru Wu, Aston Zhang, Zhuosheng Zhang, Gongshen Liu
To address the issue, we introduce OS-Kairos, an adaptive GUI agent capable of predicting confidence levels at each interaction step and efficiently deciding whether to act autonomously or seek human intervention.
1 code implementation • 21 Feb 2025 • Tianjie Ju, Bowen Wang, Hao Fei, Mong-Li Lee, Wynne Hsu, Yun Li, Qianren Wang, Pengzhou Cheng, Zongru Wu, Zhuosheng Zhang, Gongshen Liu
Recent advances in Large Language Models (LLMs) have upgraded them from sophisticated text generators to autonomous agents capable of corporation and tool use in multi-agent systems (MASs).
1 code implementation • 20 Feb 2025 • Shiqi Zhang, Xinbei Ma, Zouying Cao, Zhuosheng Zhang, Hai Zhao
The model then understands this task graph as input and generates a plan for parallel execution.
no code implementations • 30 Jan 2025 • Yue Wang, Qiuzhi Liu, Jiahao Xu, Tian Liang, Xingyu Chen, Zhiwei He, Linfeng Song, Dian Yu, Juntao Li, Zhuosheng Zhang, Rui Wang, Zhaopeng Tu, Haitao Mi, Dong Yu
To address underthinking, we propose a decoding strategy with thought switching penalty TIP that discourages premature transitions between thoughts, encouraging deeper exploration of each reasoning path.
1 code implementation • 11 Jan 2025 • Xiangru Tang, Tianyu Hu, Muyang Ye, Yanjun Shao, Xunjian Yin, Siru Ouyang, Wangchunshu Zhou, Pan Lu, Zhuosheng Zhang, Yilun Zhao, Arman Cohan, Mark Gerstein
To address these challenges, we present ChemAgent, a novel framework designed to improve the performance of LLMs through a dynamic, self-updating library.
no code implementations • 30 Dec 2024 • Xingyu Chen, Jiahao Xu, Tian Liang, Zhiwei He, Jianhui Pang, Dian Yu, Linfeng Song, Qiuzhi Liu, Mengfei Zhou, Zhuosheng Zhang, Rui Wang, Zhaopeng Tu, Haitao Mi, Dong Yu
The remarkable performance of models like the OpenAI o1 can be attributed to their ability to emulate human-like long-time thinking during inference.
1 code implementation • 10 Dec 2024 • Shaoqing Zhang, Zhuosheng Zhang, Kehai Chen, Rongxiang Weng, Muyun Yang, Tiejun Zhao, Min Zhang
This vulnerability poses significant risks to the real-world applications.
1 code implementation • 3 Dec 2024 • Zongru Wu, Pengzhou Cheng, Lingyong Fang, Zhuosheng Zhang, Gongshen Liu
Backdoor attacks remain significant security threats to generative large language models (LLMs).
1 code implementation • 16 Oct 2024 • Haodong Zhao, Jinming Hu, Peixuan Li, Fangqi Li, Jinrui Sha, Tianjie Ju, Peixuan Chen, Zhuosheng Zhang, Gongshen Liu
Language models (LMs) have emerged as critical intellectual property (IP) assets that necessitate protection.
1 code implementation • 1 Oct 2024 • Shaoqing Zhang, Zhuosheng Zhang, Kehai Chen, Xinbei Ma, Muyun Yang, Tiejun Zhao, Min Zhang
However, a key challenge lies in devising effective plans to guide action prediction in GUI tasks, though planning have been widely recognized as effective for decomposing complex tasks into a series of steps.
no code implementations • 20 Aug 2024 • Jiyang Qiu, Xinbei Ma, Zhuosheng Zhang, Hai Zhao
In our approach, we first leverage a language model to insert a trigger selected on fixed metrics into the input, then design a pipeline of model editing to directly embed a backdoor into an LLM.
1 code implementation • 5 Aug 2024 • Xinbei Ma, Yiting Wang, Yao Yao, Tongxin Yuan, Aston Zhang, Zhuosheng Zhang, Hai Zhao
This paper investigates the faithfulness of multimodal large language model (MLLM) agents in the graphical user interface (GUI) environment, aiming to address the research question of whether multimodal GUI agents can be distracted by environmental context.
1 code implementation • 15 Jul 2024 • Anni Zou, Wenhao Yu, Hongming Zhang, Kaixin Ma, Deng Cai, Zhuosheng Zhang, Hai Zhao, Dong Yu
In this paper, we introduce DocBench, a new benchmark designed to evaluate LLM-based document reading systems.
1 code implementation • 10 Jul 2024 • Tianjie Ju, Yiting Wang, Xinbei Ma, Pengzhou Cheng, Haodong Zhao, Yulong Wang, Lifeng Liu, Jian Xie, Zhuosheng Zhang, Gongshen Liu
The rapid adoption of large language models (LLMs) in multi-agent systems has highlighted their impressive capabilities in various applications, such as collaborative problem-solving and autonomous negotiation.
1 code implementation • 22 May 2024 • Pengzhou Cheng, Yidong Ding, Tianjie Ju, Zongru Wu, Wei Du, Ping Yi, Zhuosheng Zhang, Gongshen Liu
To improve the recall of the RAG for the target contexts, we introduce a knowledge graph to construct structured data to achieve hard matching at a fine-grained level.
1 code implementation • 22 May 2024 • Yiming Wang, Pei Zhang, Baosong Yang, Derek F. Wong, Zhuosheng Zhang, Rui Wang
Real-world data deviating from the independent and identically distributed (i. i. d.)
1 code implementation • 28 Mar 2024 • Yexin Wu, Zhuosheng Zhang, Hai Zhao
However, flawless CoT reasoning cannot be guaranteed due to the presence of indecomposable questions and the potential for erroneous reasoning chains, particularly in the case of small-scale language models.
1 code implementation • 29 Feb 2024 • Pengzhou Cheng, Wei Du, Zongru Wu, Fengwei Zhang, Libo Chen, Zhuosheng Zhang, Gongshen Liu
Although pre-training achieves remarkable performance, it suffers from task-agnostic backdoor attacks due to vulnerabilities in data and training mechanisms.
1 code implementation • 24 Feb 2024 • Tian Xia, Zhiwei He, Tong Ren, Yibo Miao, Zhuosheng Zhang, Yang Yang, Rui Wang
Bargaining is an important and unique part of negotiation between humans.
1 code implementation • 21 Feb 2024 • Zhiwei He, Binglin Zhou, Hongkun Hao, Aiwei Liu, Xing Wang, Zhaopeng Tu, Zhuosheng Zhang, Rui Wang
Furthermore, we analyze two key factors that contribute to the cross-lingual consistency in text watermarking and propose X-SIR as a defense method against CWRA.
1 code implementation • 19 Feb 2024 • Zongru Wu, Zhuosheng Zhang, Pengzhou Cheng, Gongshen Liu
In this paper, we investigate the learning mechanisms of backdoor LMs in the frequency space by Fourier analysis.
2 code implementations • 19 Feb 2024 • Xinbei Ma, Zhuosheng Zhang, Hai Zhao
We propose a Comprehensive Cognitive LLM Agent, CoCo-Agent, with two novel approaches, comprehensive environment perception (CEP) and conditional action prediction (CAP), to systematically improve the GUI automation performance.
1 code implementation • 19 Feb 2024 • Tianjie Ju, Yijin Chen, Xinwei Yuan, Zhuosheng Zhang, Wei Du, Yubin Zheng, Gongshen Liu
Recent work has showcased the powerful capability of large language models (LLMs) in recalling knowledge and reasoning.
1 code implementation • 8 Feb 2024 • Xinbei Ma, Tianjie Ju, Jiyang Qiu, Zhuosheng Zhang, Hai Zhao, Lifeng Liu, Yulong Wang
RQ1: Can edited LLMs behave consistently resembling communicative AI in realistic situations?
no code implementations • 6 Feb 2024 • Xiangru Tang, Qiao Jin, Kunlun Zhu, Tongxin Yuan, Yichi Zhang, Wangchunshu Zhou, Meng Qu, Yilun Zhao, Jian Tang, Zhuosheng Zhang, Arman Cohan, Zhiyong Lu, Mark Gerstein
Intelligent agents powered by large language models (LLMs) have demonstrated substantial promise in autonomously conducting experiments and facilitating scientific discoveries across various disciplines.
1 code implementation • 4 Feb 2024 • Xuanchang Zhang, Zhuosheng Zhang, Hai Zhao
Despite the rapid progress of large language models (LLMs), their task performance remains sensitive to prompt design.
1 code implementation • 23 Jan 2024 • Zhiwei He, Xing Wang, Wenxiang Jiao, Zhuosheng Zhang, Rui Wang, Shuming Shi, Zhaopeng Tu
In this work, we investigate the potential of employing the QE model as the reward model to predict human preferences for feedback training.
1 code implementation • 18 Jan 2024 • Tongxin Yuan, Zhiwei He, Lingzhong Dong, Yiming Wang, Ruijie Zhao, Tian Xia, Lizhen Xu, Binglin Zhou, Fangqi Li, Zhuosheng Zhang, Rui Wang, Gongshen Liu
We introduce R-Judge, a benchmark crafted to evaluate the proficiency of LLMs in judging and identifying safety risks given agent interaction records.
1 code implementation • 20 Nov 2023 • Zhuosheng Zhang, Yao Yao, Aston Zhang, Xiangru Tang, Xinbei Ma, Zhiwei He, Yiming Wang, Mark Gerstein, Rui Wang, Gongshen Liu, Hai Zhao
Large language models (LLMs) have dramatically enhanced the field of language intelligence, as demonstrably evidenced by their formidable empirical performance across a spectrum of complex reasoning tasks.
1 code implementation • 16 Nov 2023 • Siru Ouyang, Zhuosheng Zhang, Bing Yan, Xuan Liu, Yejin Choi, Jiawei Han, Lianhui Qin
Large Language Models (LLMs) excel in diverse areas, yet struggle with complex scientific reasoning, especially in the field of chemistry.
1 code implementation • 16 Nov 2023 • Xiangru Tang, Anni Zou, Zhuosheng Zhang, Ziming Li, Yilun Zhao, Xingyao Zhang, Arman Cohan, Mark Gerstein
Large language models (LLMs), despite their remarkable progress across various general domains, encounter significant barriers in medicine and healthcare.
1 code implementation • 10 Oct 2023 • Anni Zou, Zhuosheng Zhang, Hai Zhao, Xiangru Tang
Large language models (LLMs) have unveiled remarkable reasoning capabilities by exploiting chain-of-thought (CoT) prompting, which generates intermediate reasoning chains to serve as the rationale for deriving the answer.
7 code implementations • 20 Sep 2023 • Zhuosheng Zhang, Aston Zhang
Autonomous graphical user interface (GUI) agents aim to facilitate task automation by interacting with the user interface without manual intervention.
no code implementations • 18 Sep 2023 • Xinbei Ma, Yi Xu, Hai Zhao, Zhuosheng Zhang
On the other hand, the split segments are an appropriate element of multi-turn dialogue response selection.
1 code implementation • 30 Jun 2023 • Yiming Wang, Zhuosheng Zhang, Pei Zhang, Baosong Yang, Rui Wang
Neural-symbolic methods have demonstrated efficiency in enhancing the reasoning abilities of large language models (LLMs).
1 code implementation • COLING 2022 • Jialin Chen, Zhuosheng Zhang, Hai Zhao
Machine reading comprehension (MRC) poses new challenges over logical reasoning, which aims to understand the implicit logical relations entailed in the given contexts and perform inference over them.
1 code implementation • 22 May 2023 • Yiming Wang, Zhuosheng Zhang, Rui Wang
Further, we propose a Summary Chain-of-Thought (SumCoT) technique to elicit LLMs to generate summaries step by step, which helps them integrate more fine-grained details of source documents into the final summaries that correlate with the human writing mindset.
1 code implementation • 10 May 2023 • Anni Zou, Zhuosheng Zhang, Hai Zhao
Commonsense fact verification, as a challenging branch of commonsense question-answering (QA), aims to verify through facts whether a given commonsense claim is correct or not.
2 code implementations • 6 May 2023 • Zhiwei He, Tian Liang, Wenxiang Jiao, Zhuosheng Zhang, Yujiu Yang, Rui Wang, Zhaopeng Tu, Shuming Shi, Xing Wang
Compared to typical machine translation that focuses solely on source-to-target mapping, LLM-based translation can potentially mimic the human translation process which might take preparatory steps to ensure high-quality translation.
1 code implementation • 8 Feb 2023 • Chengwei Qin, Aston Zhang, Zhuosheng Zhang, Jiaao Chen, Michihiro Yasunaga, Diyi Yang
Spurred by advancements in scale, large language models (LLMs) have demonstrated the ability to perform a variety of natural language processing (NLP) tasks zero-shot -- i. e., without adaptation on downstream data.
3 code implementations • 2 Feb 2023 • Zhuosheng Zhang, Aston Zhang, Mu Li, Hai Zhao, George Karypis, Alex Smola
Experimental results on ScienceQA and A-OKVQA benchmark datasets show the effectiveness of our proposed approach.
Ranked #4 on
Science Question Answering
on ScienceQA
no code implementations • 10 Jan 2023 • Zhuosheng Zhang, Hai Zhao, Longxiang Liu
We decouple the contextualized word representations by masking mechanisms in Transformer-based PrLM, making each word only focus on the words in current utterance, other utterances, and two speaker roles (i. e., utterances of sender and utterances of the receiver), respectively.
no code implementations • 9 Jan 2023 • Zhuosheng Zhang, Kehai Chen, Rui Wang, Masao Utiyama, Eiichiro Sumita, Zuchao Li, Hai Zhao
Representation learning is the foundation of natural language processing (NLP).
1 code implementation • 16 Dec 2022 • Junlong Li, JinYuan Wang, Zhuosheng Zhang, Hai Zhao
This task becomes notably challenging in a zero-shot setting where no data is available to train tailored retrieval-reader models.
no code implementations • 1 Dec 2022 • Zhuosheng Zhang, Hai Zhao, Masao Utiyama, Eiichiro Sumita
Discriminative pre-trained language models (PLMs) learn to predict original texts from intentionally corrupted ones.
1 code implementation • 23 Oct 2022 • Wenhao Yu, Chenguang Zhu, Zhihan Zhang, Shuohang Wang, Zhuosheng Zhang, Yuwei Fang, Meng Jiang
However, applying such methods to commonsense reasoning tasks faces two unique challenges, i. e., the lack of a general large-scale corpus for retrieval and a corresponding effective commonsense retriever.
1 code implementation • 13 Oct 2022 • Sizhe Zhou, Siru Ouyang, Zhuosheng Zhang, Hai Zhao
In open-retrieval conversational machine reading (OR-CMR) task, machines are required to do multi-turn question answering given dialogue history and a textual knowledge base.
1 code implementation • 12 Oct 2022 • Zhuosheng Zhang, Shuohang Wang, Yichong Xu, Yuwei Fang, Wenhao Yu, Yang Liu, Hai Zhao, Chenguang Zhu, Michael Zeng
Leveraging task-aware annotated data as supervised signals to assist with self-supervised learning on large-scale unlabeled data has become a new trend in pre-training language models.
Ranked #1 on
Sentence Completion
on HellaSwag
1 code implementation • 11 Oct 2022 • Zhuosheng Zhang, Hai Zhao, Ming Zhou
They treat training instances equally throughout the training process, with little attention on the individual contribution of those instances.
5 code implementations • 7 Oct 2022 • Zhuosheng Zhang, Aston Zhang, Mu Li, Alex Smola
Providing these steps for prompting demonstrations is called chain-of-thought (CoT) prompting.
Ranked #4 on
Question Answering
on TruthfulQA
(EM metric)
1 code implementation • 23 Aug 2022 • Dongjie Yang, Zhuosheng Zhang, Hai Zhao
Masked Language Modeling (MLM) has been widely used as the denoising objective in pre-training language models (PrLMs).
no code implementations • 21 Jul 2022 • Jiayi Wang, Rongzhou Bao, Zhuosheng Zhang, Hai Zhao
However, we find that most existing textual adversarial examples are unnatural, which can be easily distinguished by both human and machine.
1 code implementation • 18 Apr 2022 • Yiyang Li, Hai Zhao, Zhuosheng Zhang
Multi-turn dialogue modeling as a challenging branch of natural language understanding (NLU), aims to build representations for machines to understand human dialogues, which provides a solid foundation for multiple downstream tasks.
1 code implementation • Findings (ACL) 2022 • Jiayi Wang, Rongzhou Bao, Zhuosheng Zhang, Hai Zhao
We question the validity of current evaluation of robustness of PrLMs based on these non-natural adversarial samples and propose an anomaly detector to evaluate the robustness of PrLMs with more natural adversarial samples.
no code implementations • 23 Nov 2021 • Xiangxiang Zhu, Kunde Yang, Zhuosheng Zhang
By the analysis of the properties of the IF equation, we prove that a good IF equation can unify the well-known IF and group delay estimators and provides an effective way to characterize the mixture of time-varying and frequency-varying signals.
1 code implementation • ACL 2022 • Baorong Huang, Zhuosheng Zhang, Hai Zhao
In this paper, we imitate the human reading process in connecting the anaphoric expressions and explicitly leverage the coreference information of the entities to enhance the word embeddings from the pre-trained language model, in order to highlight the coreference mentions of the entities that must be identified for coreference-intensive question answering in QUOREF, a relatively new dataset that is specifically designed to evaluate the coreference-related performance of a model.
1 code implementation • ACL 2022 • Xinbei Ma, Zhuosheng Zhang, Hai Zhao
Tangled multi-party dialogue contexts lead to challenges for dialogue reading comprehension, where multiple dialogue threads flow simultaneously within a common dialogue record, increasing difficulties in understanding the dialogue history for both human and machine.
no code implementations • ACL 2022 • Bohong Wu, Zhuosheng Zhang, JinYuan Wang, Hai Zhao
In detail, we introduce an in-passage negative sampling strategy to encourage a diverse generation of sentence representations within the same passage.
1 code implementation • 13 Oct 2021 • Zhuosheng Zhang, Hanqing Zhang, Keming Chen, Yuhang Guo, Jingyun Hua, Yulong Wang, Ming Zhou
Although pre-trained models (PLMs) have achieved remarkable improvements in a wide range of NLP tasks, they are expensive in terms of time and resources.
no code implementations • 11 Oct 2021 • Zhuosheng Zhang, Hai Zhao
In this paper, we review the previous methods from the technical perspective of dialogue modeling for the dialogue comprehension task.
1 code implementation • PACLIC 2021 • Yuchen He, Zhuosheng Zhang, Hai Zhao
Multi-party dialogue machine reading comprehension (MRC) raises an even more challenging understanding goal on dialogue with more than two involved speakers, compared with the traditional plain passage style MRC.
Ranked #1 on
Discourse Parsing
on Molweni
no code implementations • 29 Sep 2021 • Siru Ouyang, Zhuosheng Zhang, Hai Zhao
Pre-trained language models (PrLMs) have been shown useful for enhancing a broad range of natural language understanding (NLU) tasks.
no code implementations • 9 Sep 2021 • Xinbei Ma, Zhuosheng Zhang, Hai Zhao
Multi-party multi-turn dialogue comprehension brings unprecedented challenges on handling the complicated scenarios from multiple speakers and criss-crossed discourse relationship among speaker-aware utterances.
Ranked #1 on
Question Answering
on Molweni
no code implementations • Findings (EMNLP) 2021 • Rongzhou Bao, Zhuosheng Zhang, Hai Zhao
Pre-trained language models (PrLM) have to carefully manage input units when training on a very large text with a vocabulary consisting of millions of words.
1 code implementation • EMNLP 2021 • Zhuosheng Zhang, Siru Ouyang, Hai Zhao, Masao Utiyama, Eiichiro Sumita
In this work, we propose an effective gating strategy by smoothing the two dialogue states in only one decoder and bridge decision making and question generation to provide a richer dialogue state reference.
no code implementations • 2 Aug 2021 • Xiangxiang Zhu, Bei Li, Kunde Yang, Zhuosheng Zhang, Wenting Li
The standard chirplet transform (CT) with a chirp-modulated Gaussian window provides a valuable tool for analyzing linear chirp signals.
no code implementations • 27 Jul 2021 • Zuchao Li, Kevin Parnow, Hai Zhao, Zhuosheng Zhang, Rui Wang, Masao Utiyama, Eiichiro Sumita
Though the pre-trained contextualized language model (PrLM) has made a significant impact on NLP, training PrLMs in languages other than English can be impractical for two reasons: other languages often lack corpora sufficient for training powerful PrLMs, and because of the commonalities among human languages, computationally expensive PrLM training for different languages is somewhat redundant.
no code implementations • 25 Jul 2021 • Bohong Wu, Zhuosheng Zhang, Hai Zhao
Multi-hop reading comprehension (MHRC) requires not only to predict the correct answer span in the given passage, but also to provide a chain of supporting evidences for reasoning interpretability.
no code implementations • 4 Jun 2021 • Zhuosheng Zhang, Shucheng Yu
However, extending BO to the setting of DBA is nontrivial because in DBA only output labels instead of real-valued scores, as needed by BO, are available to attackers.
no code implementations • ACL 2021 • Zhuosheng Zhang, Hai Zhao
Pre-trained language models (PrLMs) have demonstrated superior performance due to their strong ability to learn universal language representations from self-supervised pre-training.
2 code implementations • NeurIPS 2021 • Siru Ouyang, Zhuosheng Zhang, Hai Zhao
Recent years have witnessed an increasing interest in training machines with reasoning ability, which deeply relies on accurately and clearly presented clue forms.
Ranked #23 on
Reading Comprehension
on ReClor
no code implementations • 4 Mar 2021 • Zhuosheng Zhang, Hai Zhao
In this paper, we review the previous methods from the technical perspective of dialogue modeling for the dialogue comprehension task.
no code implementations • 11 Feb 2021 • Zuchao Li, Zhuosheng Zhang, Hai Zhao, Rui Wang, Kehai Chen, Masao Utiyama, Eiichiro Sumita
In this paper, we propose explicit and implicit text compression approaches to enhance the Transformer encoding and evaluate models using this approach on several typical downstream tasks that rely on the encoding heavily.
no code implementations • 10 Feb 2021 • Zhuosheng Zhang, Junlong Li, Hai Zhao
Experimental results on four dialogue comprehension benchmark tasks show that our proposed model achieves great improvements on baselines.
no code implementations • 4 Feb 2021 • Zhuosheng Zhang, Jiarui Li, Shucheng Yu, Christian Makaya
For model privacy, local model parameters in federated learning shall be obfuscated before sent to the remote aggregator.
no code implementations • 1 Jan 2021 • Rongzhou Bao, Zhuosheng Zhang, Hai Zhao
Instead of too early fixing the linguistic unit input as nearly all previous work did, we propose a novel method that combines span-level information into the representations generated by PrLMs during fine-tuning phase for better flexibility.
no code implementations • 1 Jan 2021 • Zuchao Li, Kevin Barry Parnow, Hai Zhao, Zhuosheng Zhang, Rui Wang, Masao Utiyama, Eiichiro Sumita
Though the pre-trained contextualized language model (PrLM) has made a significant impact on NLP, training PrLMs in languages other than English can be impractical for two reasons: other languages often lack corpora sufficient for training powerful PrLMs, and because of the commonalities among human languages, computationally expensive PrLM training for different languages is somewhat redundant.
no code implementations • 30 Dec 2020 • Zhuosheng Zhang, Haojie Yu, Hai Zhao, Rui Wang, Masao Utiyama
Word representation is a fundamental component in neural language understanding models.
no code implementations • 30 Dec 2020 • Rongzhou Bao, Jiayi Wang, Zhuosheng Zhang, Hai Zhao
By substituting complex words with simple alternatives, lexical simplification (LS) is a recognized method to reduce such lexical diversity, and therefore to improve the understandability of sentences.
1 code implementation • Findings (ACL) 2021 • Siru Ouyang, Zhuosheng Zhang, Hai Zhao
Conversational Machine Reading (CMR) aims at answering questions in a complicated manner.
no code implementations • 27 Dec 2020 • Zhuosheng Zhang, Yuwei Wu, Junru Zhou, Sufeng Duan, Hai Zhao, Rui Wang
In detail, for self-attention network (SAN) sponsored Transformer-based encoder, we introduce syntactic dependency of interest (SDOI) design into the SAN to form an SDOI-SAN with syntax-guided self-attention.
1 code implementation • 7 Dec 2020 • Yilin Zhao, Zhuosheng Zhang, Hai Zhao
Thus we propose a novel reference-based knowledge enhancement model called Reference Knowledgeable Network (RekNet), which simulates human reading strategies to refine critical information from the passage and quote explicit knowledge in necessity.
1 code implementation • Findings of the Association for Computational Linguistics 2020 • Junru Zhou, Zhuosheng Zhang, Hai Zhao, Shuailiang Zhang
Besides, LIMIT-BERT takes a semi-supervised learning strategy to offer the same large amount of linguistics task data as that for the language model training.
1 code implementation • 26 Sep 2020 • Yi Xu, Hai Zhao, Zhuosheng Zhang
In the retrieval-based multi-turn dialogue modeling, it remains a challenge to select the most appropriate response according to extracting salient features in context utterances.
no code implementations • 15 Sep 2020 • Junjie Yang, Zhuosheng Zhang, Hai Zhao
Generative machine reading comprehension (MRC) requires a model to generate well-formed answers.
no code implementations • 14 Sep 2020 • Zhuosheng Zhang, Yiqing Zhang, Hai Zhao, Xi Zhou, Xiang Zhou
This paper presents a novel method to generate answers for non-extraction machine reading comprehension (MRC) tasks whose answers cannot be simply extracted as one span from the given passages.
1 code implementation • 14 Sep 2020 • Longxiang Liu, Zhuosheng Zhang, Hai Zhao, Xi Zhou, Xiang Zhou
A multi-turn dialogue is composed of multiple utterances from two or more different speaker roles.
1 code implementation • 10 Sep 2020 • Junlong Li, Zhuosheng Zhang, Hai Zhao
Pre-trained language models (PrLMs) have achieved great success on a wide range of natural language processing tasks by virtue of the universal language representation ability obtained by self-supervised learning on a large corpus.
1 code implementation • 13 May 2020 • Zhuosheng Zhang, Hai Zhao, Rui Wang
In this survey, we provide a comprehensive and comparative review on MRC covering overall research topics about 1) the origin and development of MRC and CLM, with a particular focus on the role of CLMs; 2) the impact of MRC and CLM to the NLP community; 3) the definition, datasets, and evaluation of MRC; 4) general MRC architecture and technical methods in the view of two-stage Encoder-Decoder solving architecture from the insights of the cognitive process of humans; 5) previous highlights, emerging topics, and our empirical analysis, among which we especially focus on what works in different periods of MRC researches.
no code implementations • ICLR 2020 • Zuchao Li, Rui Wang, Kehai Chen, Masso Utiyama, Eiichiro Sumita, Zhuosheng Zhang, Hai Zhao
However, MLE focuses on once-to-all matching between the predicted sequence and gold-standard, consequently treating all incorrect predictions as being equally incorrect.
1 code implementation • ICLR 2020 • Zhuosheng Zhang, Kehai Chen, Rui Wang, Masao Utiyama, Eiichiro Sumita, Zuchao Li, Hai Zhao
Though visual information has been introduced for enhancing neural machine translation (NMT), its effectiveness strongly relies on the availability of large amounts of bilingual parallel sentence pairs with manual image annotations.
no code implementations • 29 Apr 2020 • Junlong Li, Zhuosheng Zhang, Hai Zhao
In this paper, the relevance of each turn to the question are calculated to choose key turns.
2 code implementations • 27 Jan 2020 • Zhuosheng Zhang, Junjie Yang, Hai Zhao
Inspired by how humans solve reading comprehension questions, we proposed a retrospective reader (Retro-Reader) that integrates two stages of reading and verification strategies: 1) sketchy reading that briefly investigates the overall interactions of passage and question, and yield an initial judgment; 2) intensive reading that verifies the answer and gives the final prediction.
Ranked #7 on
Question Answering
on SQuAD2.0
1 code implementation • 27 Dec 2019 • Zuchao Li, Rui Wang, Kehai Chen, Masao Utiyama, Eiichiro Sumita, Zhuosheng Zhang, Hai Zhao
In this paper, we propose an explicit sentence compression method to enhance the source sentence representation for NMT.
no code implementations • 7 Nov 2019 • Zhuosheng Zhang, Rui Wang, Kehai Chen, Masao Utiyama, Eiichiro Sumita, Hai Zhao
We present a universal framework to model contextualized sentence representations with visual awareness that is motivated to overcome the shortcomings of the multimodal parallel data with manual annotations.
no code implementations • CONLL 2019 • Zuchao Li, Hai Zhao, Zhuosheng Zhang, Rui Wang, Masao Utiyama, Eiichiro Sumita
This paper describes our SJTU-NICT{'}s system for participating in the shared task on Cross-Framework Meaning Representation Parsing (MRP) at the 2019 Conference for Computational Language Learning (CoNLL).
no code implementations • 31 Oct 2019 • Junru Zhou, Zhuosheng Zhang, Hai Zhao, Shuailiang Zhang
In this paper, we present a Linguistic Informed Multi-Task BERT (LIMIT-BERT) for learning language representations across multiple linguistic tasks by Multi-Task Learning (MTL).
1 code implementation • 5 Sep 2019 • Zhuosheng Zhang, Yuwei Wu, Hai Zhao, Zuchao Li, Shuailiang Zhang, Xi Zhou, Xiang Zhou
The latest work on language representations carefully integrates contextualized features into language model training, which enables a series of success especially in various machine reading comprehension and natural language inference tasks.
Ranked #8 on
Natural Language Inference
on SNLI
no code implementations • 3 Sep 2019 • Zhuosheng Zhang, Bingjie Tang, Zuchao Li, Hai Zhao
This work models named entity distribution from a way of visualizing topological structure of embedding space, so that we make an assumption that most, if not all, named entities (NEs) for a language tend to aggregate together to be accommodated by a specific hypersphere in embedding space.
no code implementations • 3 Sep 2019 • Zhuosheng Zhang, Zhen Meng, Hai Zhao
This paper presents a smart sliding Chinese pinyin Input Method Editor (IME) for touchscreen devices which allows user finger sliding from one key to another on the touchscreen instead of tapping keys one by one, while the target Chinese character sequence will be predicted during the sliding process to help user input Chinese characters efficiently.
no code implementations • 31 Aug 2019 • Ying Luo, Hai Zhao, Zhuosheng Zhang, Bingjie Tang
For monolingual cases, the proposed named entity model gives an open description of diverse named entity types and different languages.
2 code implementations • 30 Aug 2019 • Shuailiang Zhang, Hai Zhao, Yuwei Wu, Zhuosheng Zhang, Xi Zhou, Xiang Zhou
Multi-choice reading comprehension is a challenging task to select an answer from a set of candidate options when given passage and question.
1 code implementation • 14 Aug 2019 • Zhuosheng Zhang, Yuwei Wu, Junru Zhou, Sufeng Duan, Hai Zhao, Rui Wang
In detail, for self-attention network (SAN) sponsored Transformer-based encoder, we introduce syntactic dependency of interest (SDOI) design into the SAN to form an SDOI-SAN with syntax-guided self-attention.
Ranked #5 on
Question Answering
on SQuAD2.0 dev
no code implementations • 22 Apr 2019 • Shu Jiang, Zhuosheng Zhang, Hai Zhao, Jiangtong Li, Yang Yang, Bao-liang Lu, Ning Xia
Chemical reaction practicality is the core task among all symbol intelligence based chemical information processing, for example, it provides indispensable clue for further automatic synthesis route inference.
no code implementations • 27 Jan 2019 • Shuailiang Zhang, Hai Zhao, Yuwei Wu, Zhuosheng Zhang, Xi Zhou, Xiang Zhou
Multi-choice reading comprehension is a challenging task that requires complex reasoning procedure.
Ranked #3 on
Question Answering
on RACE
1 code implementation • 16 Jan 2019 • Zuchao Li, Shexia He, Hai Zhao, Yiqing Zhang, Zhuosheng Zhang, Xi Zhou, Xiang Zhou
Semantic role labeling (SRL) aims to discover the predicateargument structure of a sentence.
Ranked #9 on
Semantic Role Labeling
on CoNLL 2005
1 code implementation • ACL 2019 • Zhuosheng Zhang, Yafang Huang, Hai Zhao
Pinyin-to-character (P2C) conversion is the core component of pinyin-based Chinese input method engine (IME).
1 code implementation • 6 Nov 2018 • Zhuosheng Zhang, Hai Zhao, Kangwei Ling, Jiangtong Li, Zuchao Li, Shexia He, Guohong Fu
Representation learning is the foundation of machine reading comprehension and inference.
1 code implementation • EMNLP 2018 • Zuchao Li, Shexia He, Jiaxun Cai, Zhuosheng Zhang, Hai Zhao, Gongshen Liu, Linlin Li, Luo Si
Semantic role labeling (SRL) aims to recognize the predicate-argument structure of a sentence.
1 code implementation • CONLL 2018 • Zuchao Li, Shexia He, Zhuosheng Zhang, Hai Zhao
This paper describes the system of team LeisureX in the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies.
no code implementations • 8 Sep 2018 • Zhuosheng Zhang, Yuwei Wu, Zuchao Li, Hai Zhao
Who did what to whom is a major focus in natural language understanding, which is right the aim of semantic role labeling (SRL) task.
Ranked #11 on
Natural Language Inference
on SNLI
Machine Reading Comprehension
Natural Language Understanding
+1
no code implementations • 8 Sep 2018 • Zhuosheng Zhang, Shexia He, Zuchao Li, Hai Zhao
The goal of semantic role labeling (SRL) is to discover the predicate-argument structure of a sentence, which plays a critical role in deep processing of natural language.
no code implementations • COLING 2018 • Pengfei Zhu, Zhuosheng Zhang, Jiangtong Li, Yafang Huang, Hai Zhao
Traditional chatbots usually need a mass of human dialogue data, especially when using supervised machine learning method.
no code implementations • 7 Aug 2018 • Zhuosheng Zhang, Yafang Huang, Pengfei Zhu, Hai Zhao
Machine reading comprehension is a task to model relationship between passage and query.
no code implementations • ACL 2018 • Yafang Huang, Zuchao Li, Zhuosheng Zhang, Hai Zhao
Chinese pinyin input method engine (IME) lets user conveniently input Chinese into a computer by typing pinyin through the common keyboard.
1 code implementation • COLING 2018 • Zhuosheng Zhang, Hai Zhao
Answering questions from university admission exams (Gaokao in Chinese) is a challenging AI task since it requires effective representation to capture complicated semantic relations between questions and answers.
1 code implementation • COLING 2018 • Zhuosheng Zhang, Jiangtong Li, Pengfei Zhu, Hai Zhao, Gongshen Liu
In this paper, we formulate previous utterances into context using a proposed deep utterance aggregation model to form a fine-grained context representation.
Ranked #14 on
Conversational Response Selection
on E-commerce
1 code implementation • COLING 2018 • Zhuosheng Zhang, Yafang Huang, Hai Zhao
Representation learning is the foundation of machine reading comprehension.
no code implementations • SEMEVAL 2018 • Zhuosheng Zhang, Jiangtong Li, Hai Zhao, Bingjie Tang
This paper describes a hypernym discovery system for our participation in the SemEval-2018 Task 9, which aims to discover the best (set of) candidate hypernyms for input concepts or entities, given the search space of a pre-defined vocabulary.
Ranked #5 on
Hypernym Discovery
on Music domain