no code implementations • 27 Apr 2024 • Yuntao Shou, Tao Meng, FuChen Zhang, Nan Yin, Keqin Li
Specifically, on the one hand, in the feature disentanglement stage, we propose a Broad Mamba, which does not rely on a self-attention mechanism for sequence modeling, but uses state space models to compress emotional representation, and utilizes broad learning systems to explore the potential data distribution in broad space.
no code implementations • 27 Apr 2024 • Tao Meng, FuChen Zhang, Yuntao Shou, Wei Ai, Nan Yin, Keqin Li
Since consistency and complementarity information correspond to low-frequency and high-frequency information, respectively, this paper revisits the problem of multimodal emotion recognition in conversation from the perspective of the graph spectrum.
no code implementations • 24 Mar 2024 • Qin Liu, Fei Wang, Nan Xu, Tianyi Yan, Tao Meng, Muhao Chen
In this paper, we propose monotonic paraphrasing (MonoPara), an end-to-end decoding strategy that paraphrases given prompts or instructions into their lower perplexity counterparts based on an ensemble of a paraphrase LM for prompt (or instruction) rewriting, and a target LM (i. e. the prompt or instruction executor) that constrains the generation for lower perplexity.
no code implementations • 19 Jan 2024 • Wei Ai, CanHao Xie, Tao Meng, Yinghao Wu, Keqin Li
Community search is a derivative of community detection that enables online and personalized discovery of communities and has found extensive applications in massive real-world networks.
no code implementations • 19 Jan 2024 • JiaYi Du, Yinghao Wu, Wei Ai, Tao Meng, CanHao Xie, Keqin Li
Community Search (CS) aims to identify densely interconnected subgraphs corresponding to query vertices within a graph.
no code implementations • 3 Jan 2024 • Wei Ai, FuChen Zhang, Tao Meng, Yuntao Shou, HongEn Shao, Keqin Li
To address the above issues, we propose a two-stage emotion recognition model based on graph contrastive learning (TS-GCL).
no code implementations • 28 Dec 2023 • Yuntao Shou, Tao Meng, Wei Ai, Keqin Li
However, the existing feature fusion methods have usually mapped the features of different modalities into the same feature space for information fusion, which can not eliminate the heterogeneity between different modalities.
no code implementations • 17 Dec 2023 • Wei Ai, Yuntao Shou, Tao Meng, Keqin Li
Specifically, we construct a weighted multi-relationship graph to simultaneously capture the dependencies between speakers and event relations in a dialogue.
no code implementations • 11 Dec 2023 • Tao Meng, Yuntao Shou, Wei Ai, Nan Yin, Keqin Li
The main task of Multimodal Emotion Recognition in Conversations (MERC) is to identify the emotions in modalities, e. g., text, audio, image and video, which is a significant development direction for realizing machine intelligence.
no code implementations • 10 Dec 2023 • Yuntao Shou, Tao Meng, Wei Ai, Nan Yin, Keqin Li
Unlike the traditional single-utterance multi-modal emotion recognition or single-modal conversation emotion recognition, MCER is a more challenging problem that needs to deal with more complex emotional interaction relationships.
no code implementations • 5 Dec 2023 • Yuntao Shou, Wei Ai, Tao Meng
Furthermore, this paper innovatively introduces information bottleneck theory into graph contrastive learning to maximize task-related information while minimizing task-independent redundant information.
no code implementations • 4 Dec 2023 • Yuntao Shou, Wei Ai, Tao Meng, Keqin Li
Zero-shot age estimation aims to learn feature information about age from input images and make inferences about a given person's image or video frame without specific sample data.
no code implementations • 8 Jun 2023 • Kun Wang, Tao Meng, Jiakun Lei, Weijia Wang
In order to address this issue, we propose a control strategy based on control barrier functions, summarized as "safety check on kinematics" and "velocity tracking on dynamics" approach.
no code implementations • 31 May 2023 • Jiakun Lei, Tao Meng, Kun Wang, Weijia Wang, Shujian Sun
Further, the basic intermittent attitude controller is extended to a "constrained version" by introducing a strictly bounded virtual control law and an input saturation compensation auxiliary system.
no code implementations • 31 May 2023 • Jiakun Lei, Tao Meng, Yang Zhu, Kun Wang, Weijia Wang
To tackle this problem, we propose a modified framework called Compatible Performance Control (CPC), which integrates the Prescribed Performance Control (PPC) scheme with a contradiction detection and alleviation strategy.
no code implementations • 10 Nov 2022 • Jiakun Lei, Tao Meng, Kun Wang, Weijia Wang, Zhonghe Jin
The prescribed performance control (PPC) scheme is often employed for the control with guaranteed performance.
no code implementations • 13 Sep 2022 • Jiakun Lei, Tao Meng, Weijia Wang, Chengjin Yin, Zhonghe Jin
Based on the proposed structure, a backstepping controller is developed accordingly using a double-layer PPC framework.
no code implementations • 13 Sep 2022 • Jiakun Lei, Tao Meng, Weijia Wang, Shujian Sun, Heng Li, Zhonghe Jin
To resolve this problem, a switching controller structure is proposed in this paper based on the reduced-attitude representation, fusing the artificial potential field (APF) methodology and the Prescribed Performance Control (PPC) scheme together.
1 code implementation • 27 May 2022 • Tao Meng, Sidi Lu, Nanyun Peng, Kai-Wei Chang
We propose a general and efficient framework to control auto-regressive generation models with NeurAlly-Decomposed Oracle (NADO).
1 code implementation • 23 May 2022 • Honghua Zhang, Liunian Harold Li, Tao Meng, Kai-Wei Chang, Guy Van Den Broeck
Logical reasoning is needed in a wide range of NLP tasks.
no code implementations • NAACL 2021 • Tao Meng, Anjie Fang, Oleg Rokhlenko, Shervin Malmasi
We propose GEMNET, a novel approach for gazetteer knowledge integration, including (1) a flexible Contextual Gazetteer Representation (CGR) encoder that can be fused with any word-level model; and (2) a Mixture-of- Experts gating network that overcomes the feature overuse issue by learning to conditionally combine the context and gazetteer features, instead of assigning them fixed weights.
no code implementations • 12 Feb 2021 • Sidi Lu, Tao Meng, Nanyun Peng
We propose InsNet, an expressive insertion-based text generator with efficient training and flexible decoding (parallel or sequential).
1 code implementation • 18 Jun 2020 • Tao Meng, Kai-Wei Chang
This raises a question -- \emph{can we mine constraints and rules from data based on a learning algorithm?}
1 code implementation • ACL 2020 • Shengyu Jia, Tao Meng, Jieyu Zhao, Kai-Wei Chang
With little performance loss, our method can almost remove the bias amplification in the distribution.
1 code implementation • ACL 2020 • Fan Yin, Quanyu Long, Tao Meng, Kai-Wei Chang
We conduct a thorough study to diagnose the behaviors of pre-trained language encoders (ELMo, BERT, and RoBERTa) when confronted with natural grammatical errors.
2 code implementations • ACL 2020 • Da Yin, Tao Meng, Kai-Wei Chang
We propose SentiBERT, a variant of BERT that effectively captures compositional sentiment semantics.
1 code implementation • IJCNLP 2019 • Tao Meng, Nanyun Peng, Kai-Wei Chang
Experiments show that the Lagrangian relaxation and posterior regularization inference improve the performances on 15 and 17 out of 19 target languages, respectively.