@inproceedings{zhou-etal-2023-fine,
title = "Fine-grained Artificial Neurons in Audio-transformers for Disentangling Neural Auditory Encoding",
author = "Zhou, Mengyue and
Liu, Xu and
Liu, David and
Wu, Zihao and
Liu, Zhengliang and
Zhao, Lin and
Zhu, Dajiang and
Guo, Lei and
Han, Junwei and
Liu, Tianming and
Hu, Xintao",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2023",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.findings-acl.503",
doi = "10.18653/v1/2023.findings-acl.503",
pages = "7943--7956",
abstract = "The Wav2Vec and its variants have achieved unprecedented success in computational auditory and speech processing. Meanwhile, neural encoding studies that integrate the superb representation capability of Wav2Vec and link those representations to brain activities have provided novel insights into a fundamental question of how auditory and speech processing unfold in the human brain. Without an explicit definition, most existing studies treat each transformer encoding layer in Wav2Vec as a single artificial neuron (AN). That is, the layer-level embeddings are used to predict neural responses. However, the comprehensive layer-level embedding aggregates multiple types of contextual attention captured by multi-head self-attention (MSA) modules. Thus, the layer-level ANs lack fine-granularity for neural encoding. To address this limitation, we define the elementary units, i.e., each hidden dimension, as neuron-level ANs in Wav2Vec2.0, quantify their temporal responses, and couple those ANs with their biological-neuron (BN) counterparts in the human brain. Our experimental results demonstrated that: 1) The proposed neuron-level ANs carry meaningful neurolinguistic information; 2) Those ANs anchor to their BN signatures; 3) The AN-BN anchoring patterns are interpretable from a neurolinguistic perspective. More importantly, our results suggest an intermediate stage in both the computational representation in Wav2Vec2.0 and the cortical representation in the brain. Our study validates the fine-grained ANs in Wav2Vec2.0, which may serve as a novel and general strategy to link transformer-based deep learning models to neural responses for probing the sensory processing in the brain.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhou-etal-2023-fine">
<titleInfo>
<title>Fine-grained Artificial Neurons in Audio-transformers for Disentangling Neural Auditory Encoding</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mengyue</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xu</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zihao</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhengliang</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lin</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dajiang</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lei</namePart>
<namePart type="family">Guo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Junwei</namePart>
<namePart type="family">Han</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tianming</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xintao</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2023</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Boyd-Graber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naoaki</namePart>
<namePart type="family">Okazaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>The Wav2Vec and its variants have achieved unprecedented success in computational auditory and speech processing. Meanwhile, neural encoding studies that integrate the superb representation capability of Wav2Vec and link those representations to brain activities have provided novel insights into a fundamental question of how auditory and speech processing unfold in the human brain. Without an explicit definition, most existing studies treat each transformer encoding layer in Wav2Vec as a single artificial neuron (AN). That is, the layer-level embeddings are used to predict neural responses. However, the comprehensive layer-level embedding aggregates multiple types of contextual attention captured by multi-head self-attention (MSA) modules. Thus, the layer-level ANs lack fine-granularity for neural encoding. To address this limitation, we define the elementary units, i.e., each hidden dimension, as neuron-level ANs in Wav2Vec2.0, quantify their temporal responses, and couple those ANs with their biological-neuron (BN) counterparts in the human brain. Our experimental results demonstrated that: 1) The proposed neuron-level ANs carry meaningful neurolinguistic information; 2) Those ANs anchor to their BN signatures; 3) The AN-BN anchoring patterns are interpretable from a neurolinguistic perspective. More importantly, our results suggest an intermediate stage in both the computational representation in Wav2Vec2.0 and the cortical representation in the brain. Our study validates the fine-grained ANs in Wav2Vec2.0, which may serve as a novel and general strategy to link transformer-based deep learning models to neural responses for probing the sensory processing in the brain.</abstract>
<identifier type="citekey">zhou-etal-2023-fine</identifier>
<identifier type="doi">10.18653/v1/2023.findings-acl.503</identifier>
<location>
<url>https://aclanthology.org/2023.findings-acl.503</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>7943</start>
<end>7956</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Fine-grained Artificial Neurons in Audio-transformers for Disentangling Neural Auditory Encoding
%A Zhou, Mengyue
%A Liu, Xu
%A Liu, David
%A Wu, Zihao
%A Liu, Zhengliang
%A Zhao, Lin
%A Zhu, Dajiang
%A Guo, Lei
%A Han, Junwei
%A Liu, Tianming
%A Hu, Xintao
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Findings of the Association for Computational Linguistics: ACL 2023
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F zhou-etal-2023-fine
%X The Wav2Vec and its variants have achieved unprecedented success in computational auditory and speech processing. Meanwhile, neural encoding studies that integrate the superb representation capability of Wav2Vec and link those representations to brain activities have provided novel insights into a fundamental question of how auditory and speech processing unfold in the human brain. Without an explicit definition, most existing studies treat each transformer encoding layer in Wav2Vec as a single artificial neuron (AN). That is, the layer-level embeddings are used to predict neural responses. However, the comprehensive layer-level embedding aggregates multiple types of contextual attention captured by multi-head self-attention (MSA) modules. Thus, the layer-level ANs lack fine-granularity for neural encoding. To address this limitation, we define the elementary units, i.e., each hidden dimension, as neuron-level ANs in Wav2Vec2.0, quantify their temporal responses, and couple those ANs with their biological-neuron (BN) counterparts in the human brain. Our experimental results demonstrated that: 1) The proposed neuron-level ANs carry meaningful neurolinguistic information; 2) Those ANs anchor to their BN signatures; 3) The AN-BN anchoring patterns are interpretable from a neurolinguistic perspective. More importantly, our results suggest an intermediate stage in both the computational representation in Wav2Vec2.0 and the cortical representation in the brain. Our study validates the fine-grained ANs in Wav2Vec2.0, which may serve as a novel and general strategy to link transformer-based deep learning models to neural responses for probing the sensory processing in the brain.
%R 10.18653/v1/2023.findings-acl.503
%U https://aclanthology.org/2023.findings-acl.503
%U https://doi.org/10.18653/v1/2023.findings-acl.503
%P 7943-7956
Markdown (Informal)
[Fine-grained Artificial Neurons in Audio-transformers for Disentangling Neural Auditory Encoding](https://aclanthology.org/2023.findings-acl.503) (Zhou et al., Findings 2023)
ACL
- Mengyue Zhou, Xu Liu, David Liu, Zihao Wu, Zhengliang Liu, Lin Zhao, Dajiang Zhu, Lei Guo, Junwei Han, Tianming Liu, and Xintao Hu. 2023. Fine-grained Artificial Neurons in Audio-transformers for Disentangling Neural Auditory Encoding. In Findings of the Association for Computational Linguistics: ACL 2023, pages 7943–7956, Toronto, Canada. Association for Computational Linguistics.