Large Language Model-Guided Semantic Alignment for Human Activity Recognition Article Swipe

PDF

Yan Hua , Hongdong Tan , Yi Ding , Peng-Fei Zhou , Vinod Namboodiri , Yang Yu ·

YOU? · · 2024 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2410.00003

Human Activity Recognition (HAR) using Inertial Measurement Unit (IMU) sensors is critical for applications in healthcare, safety, and industrial production. However, variations in activity patterns, device types, and sensor placements create distribution gaps across datasets, reducing the performance of HAR models. To address this, we propose LanHAR, a novel system that leverages Large Language Models (LLMs) to generate semantic interpretations of sensor readings and activity labels for cross-dataset HAR. This approach not only mitigates cross-dataset heterogeneity but also enhances the recognition of new activities. LanHAR employs an iterative re-generation method to produce high-quality semantic interpretations with LLMs and a two-stage training framework that bridges the semantic interpretations of sensor readings and activity labels. This ultimately leads to a lightweight sensor encoder suitable for mobile deployment, enabling any sensor reading to be mapped into the semantic interpretation space. Experiments on five public datasets demonstrate that our approach significantly outperforms state-of-the-art methods in both cross-dataset HAR and new activity recognition. The source code is publicly available at https://github.com/DASHLab/LanHAR.

Related Topics

Computer Science

Philosophy

Concepts

Computer science Natural language processing Linguistics Philosophy

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2410.00003
PDF: https://arxiv.org/pdf/2410.00003
OA Status: green
Related Works: 10
OpenAlex ID: https://openalex.org/W4403853690

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4403853690

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2410.00003

Digital Object Identifier
Title: Large Language Model-Guided Semantic Alignment for Human Activity Recognition

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2024

Year of publication
Publication date: 2024-09-12

Full publication date if available
Authors: Yan Hua, Hongdong Tan, Yi Ding, Peng-Fei Zhou, Vinod Namboodiri, Yang Yu

List of authors in order
Landing page: https://arxiv.org/abs/2410.00003

Publisher landing page
PDF URL: https://arxiv.org/pdf/2410.00003

Direct link to full text PDF
Open access: Yes

Whether a free full text is available
OA status: green

Open access status per OpenAlex
OA URL: https://arxiv.org/pdf/2410.00003

Direct OA link when available
Concepts: Computer science, Natural language processing, Linguistics, Philosophy

Top concepts (fields/topics) attached by OpenAlex
Cited by: 0

Total citation count in OpenAlex
Related works (count): 10

Other works algorithmically related by OpenAlex

Full payload

id	https://openalex.org/W4403853690
doi	https://doi.org/10.48550/arxiv.2410.00003
ids.doi	https://doi.org/10.48550/arxiv.2410.00003
ids.openalex	https://openalex.org/W4403853690
fwci
type	preprint
title	Large Language Model-Guided Semantic Alignment for Human Activity Recognition
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
topics[0].id	https://openalex.org/T10812
topics[0].field.id	https://openalex.org/fields/17
topics[0].field.display_name	Computer Science
topics[0].score	0.8633999824523926
topics[0].domain.id	https://openalex.org/domains/3
topics[0].domain.display_name	Physical Sciences
topics[0].subfield.id	https://openalex.org/subfields/1707
topics[0].subfield.display_name	Computer Vision and Pattern Recognition
topics[0].display_name	Human Pose and Action Recognition
topics[1].id	https://openalex.org/T10444
topics[1].field.id	https://openalex.org/fields/17
topics[1].field.display_name	Computer Science
topics[1].score	0.7660999894142151
topics[1].domain.id	https://openalex.org/domains/3
topics[1].domain.display_name	Physical Sciences
topics[1].subfield.id	https://openalex.org/subfields/1707
topics[1].subfield.display_name	Computer Vision and Pattern Recognition
topics[1].display_name	Context-Aware Activity Recognition Systems
is_xpac	False
apc_list
apc_paid
concepts[0].id	https://openalex.org/C41008148
concepts[0].level	0
concepts[0].score	0.4804070293903351
concepts[0].wikidata	https://www.wikidata.org/wiki/Q21198
concepts[0].display_name	Computer science
concepts[1].id	https://openalex.org/C204321447
concepts[1].level	1
concepts[1].score	0.37645792961120605
concepts[1].wikidata	https://www.wikidata.org/wiki/Q30642
concepts[1].display_name	Natural language processing
concepts[2].id	https://openalex.org/C41895202
concepts[2].level	1
concepts[2].score	0.3362271785736084
concepts[2].wikidata	https://www.wikidata.org/wiki/Q8162
concepts[2].display_name	Linguistics
concepts[3].id	https://openalex.org/C138885662
concepts[3].level	0
concepts[3].score	0.0834871232509613
concepts[3].wikidata	https://www.wikidata.org/wiki/Q5891
concepts[3].display_name	Philosophy
keywords[0].id	https://openalex.org/keywords/computer-science
keywords[0].score	0.4804070293903351
keywords[0].display_name	Computer science
keywords[1].id	https://openalex.org/keywords/natural-language-processing
keywords[1].score	0.37645792961120605
keywords[1].display_name	Natural language processing
keywords[2].id	https://openalex.org/keywords/linguistics
keywords[2].score	0.3362271785736084
keywords[2].display_name	Linguistics
keywords[3].id	https://openalex.org/keywords/philosophy
keywords[3].score	0.0834871232509613
keywords[3].display_name	Philosophy
language	en
locations[0].id	pmh:oai:arXiv.org:2410.00003
locations[0].is_oa	True
locations[0].source.id	https://openalex.org/S4306400194
locations[0].source.issn
locations[0].source.type	repository
locations[0].source.is_oa	True
locations[0].source.issn_l
locations[0].source.is_core	False
locations[0].source.is_in_doaj	False
locations[0].source.display_name	arXiv (Cornell University)
locations[0].source.host_organization	https://openalex.org/I205783295
locations[0].source.host_organization_name	Cornell University
locations[0].source.host_organization_lineage	https://openalex.org/I205783295
locations[0].license
locations[0].pdf_url	https://arxiv.org/pdf/2410.00003
locations[0].version	submittedVersion
locations[0].raw_type
locations[0].license_id
locations[0].is_accepted	False
locations[0].is_published	False
locations[0].raw_source_name
locations[0].landing_page_url	http://arxiv.org/abs/2410.00003
locations[1].id	doi:10.48550/arxiv.2410.00003
locations[1].is_oa	True
locations[1].source.id	https://openalex.org/S4306400194
locations[1].source.issn
locations[1].source.type	repository
locations[1].source.is_oa	True
locations[1].source.issn_l
locations[1].source.is_core	False
locations[1].source.is_in_doaj	False
locations[1].source.display_name	arXiv (Cornell University)
locations[1].source.host_organization	https://openalex.org/I205783295
locations[1].source.host_organization_name	Cornell University
locations[1].source.host_organization_lineage	https://openalex.org/I205783295
locations[1].license	cc-by
locations[1].pdf_url
locations[1].version
locations[1].raw_type	article
locations[1].license_id	https://openalex.org/licenses/cc-by
locations[1].is_accepted	False
locations[1].is_published
locations[1].raw_source_name
locations[1].landing_page_url	https://doi.org/10.48550/arxiv.2410.00003
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5042638627
authorships[0].author.orcid	https://orcid.org/0000-0001-8316-3937
authorships[0].author.display_name	Yan Hua
authorships[0].author_position	first
authorships[0].raw_author_name	Yan, Hua
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5082469381
authorships[1].author.orcid	https://orcid.org/0000-0001-9706-2539
authorships[1].author.display_name	Hongdong Tan
authorships[1].author_position	middle
authorships[1].raw_author_name	Tan, Heng
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5101846047
authorships[2].author.orcid	https://orcid.org/0000-0002-4442-3393
authorships[2].author.display_name	Yi Ding
authorships[2].author_position	middle
authorships[2].raw_author_name	Ding, Yi
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5101569330
authorships[3].author.orcid
authorships[3].author.display_name	Peng-Fei Zhou
authorships[3].author_position	middle
authorships[3].raw_author_name	Zhou, Pengfei
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5114439623
authorships[4].author.orcid
authorships[4].author.display_name	Vinod Namboodiri
authorships[4].author_position	middle
authorships[4].raw_author_name	Namboodiri, Vinod
authorships[4].is_corresponding	False
authorships[5].author.id	https://openalex.org/A5056178220
authorships[5].author.orcid	https://orcid.org/0000-0002-7339-9126
authorships[5].author.display_name	Yang Yu
authorships[5].author_position	last
authorships[5].raw_author_name	Yang, Yu
authorships[5].is_corresponding	False
has_content.pdf	False
has_content.grobid_xml	False
is_paratext	False
open_access.is_oa	True
open_access.oa_url	https://arxiv.org/pdf/2410.00003
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2024-10-29T00:00:00
display_name	Large Language Model-Guided Semantic Alignment for Human Activity Recognition
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T06:51:31.235846
primary_topic.id	https://openalex.org/T10812
primary_topic.field.id	https://openalex.org/fields/17
primary_topic.field.display_name	Computer Science
primary_topic.score	0.8633999824523926
primary_topic.domain.id	https://openalex.org/domains/3
primary_topic.domain.display_name	Physical Sciences
primary_topic.subfield.id	https://openalex.org/subfields/1707
primary_topic.subfield.display_name	Computer Vision and Pattern Recognition
primary_topic.display_name	Human Pose and Action Recognition
related_works	https://openalex.org/W4391375266, https://openalex.org/W2899084033, https://openalex.org/W2748952813, https://openalex.org/W2390279801, https://openalex.org/W4391913857, https://openalex.org/W2358668433, https://openalex.org/W4396701345, https://openalex.org/W2376932109, https://openalex.org/W2001405890, https://openalex.org/W4396696052
cited_by_count	0
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2410.00003
best_oa_location.is_oa	True
best_oa_location.source.id	https://openalex.org/S4306400194
best_oa_location.source.issn
best_oa_location.source.type	repository
best_oa_location.source.is_oa	True
best_oa_location.source.issn_l
best_oa_location.source.is_core	False
best_oa_location.source.is_in_doaj	False
best_oa_location.source.display_name	arXiv (Cornell University)
best_oa_location.source.host_organization	https://openalex.org/I205783295
best_oa_location.source.host_organization_name	Cornell University
best_oa_location.source.host_organization_lineage	https://openalex.org/I205783295
best_oa_location.license
best_oa_location.pdf_url	https://arxiv.org/pdf/2410.00003
best_oa_location.version	submittedVersion
best_oa_location.raw_type
best_oa_location.license_id
best_oa_location.is_accepted	False
best_oa_location.is_published	False
best_oa_location.raw_source_name
best_oa_location.landing_page_url	http://arxiv.org/abs/2410.00003
primary_location.id	pmh:oai:arXiv.org:2410.00003
primary_location.is_oa	True
primary_location.source.id	https://openalex.org/S4306400194
primary_location.source.issn
primary_location.source.type	repository
primary_location.source.is_oa	True
primary_location.source.issn_l
primary_location.source.is_core	False
primary_location.source.is_in_doaj	False
primary_location.source.display_name	arXiv (Cornell University)
primary_location.source.host_organization	https://openalex.org/I205783295
primary_location.source.host_organization_name	Cornell University
primary_location.source.host_organization_lineage	https://openalex.org/I205783295
primary_location.license
primary_location.pdf_url	https://arxiv.org/pdf/2410.00003
primary_location.version	submittedVersion
primary_location.raw_type
primary_location.license_id
primary_location.is_accepted	False
primary_location.is_published	False
primary_location.raw_source_name
primary_location.landing_page_url	http://arxiv.org/abs/2410.00003
publication_date	2024-09-12
publication_year	2024
referenced_works_count	0
abstract_inverted_index.a	47, 98, 117
abstract_inverted_index.To	41
abstract_inverted_index.an	86
abstract_inverted_index.at	164
abstract_inverted_index.be	130
abstract_inverted_index.in	14, 22, 150
abstract_inverted_index.is	10, 161
abstract_inverted_index.of	38, 60, 81, 107
abstract_inverted_index.on	138
abstract_inverted_index.to	56, 90, 116, 129
abstract_inverted_index.we	44
abstract_inverted_index.HAR	39, 153
abstract_inverted_index.The	158
abstract_inverted_index.and	17, 27, 63, 97, 110, 154
abstract_inverted_index.any	126
abstract_inverted_index.but	76
abstract_inverted_index.for	12, 66, 122
abstract_inverted_index.new	82, 155
abstract_inverted_index.not	71
abstract_inverted_index.our	144
abstract_inverted_index.the	36, 79, 104, 133
abstract_inverted_index.HAR.	68
abstract_inverted_index.LLMs	96
abstract_inverted_index.This	69, 113
abstract_inverted_index.Unit	7
abstract_inverted_index.also	77
abstract_inverted_index.both	151
abstract_inverted_index.code	160
abstract_inverted_index.five	139
abstract_inverted_index.gaps	32
abstract_inverted_index.into	132
abstract_inverted_index.only	72
abstract_inverted_index.that	50, 102, 143
abstract_inverted_index.with	95
abstract_inverted_index.(HAR)	3
abstract_inverted_index.(IMU)	8
abstract_inverted_index.Human	0
abstract_inverted_index.Large	52
abstract_inverted_index.leads	115
abstract_inverted_index.novel	48
abstract_inverted_index.this,	43
abstract_inverted_index.using	4
abstract_inverted_index.(LLMs)	55
abstract_inverted_index.LanHAR	84
abstract_inverted_index.Models	54
abstract_inverted_index.across	33
abstract_inverted_index.create	30
abstract_inverted_index.device	25
abstract_inverted_index.labels	65
abstract_inverted_index.mapped	131
abstract_inverted_index.method	89
abstract_inverted_index.mobile	123
abstract_inverted_index.public	140
abstract_inverted_index.sensor	28, 61, 108, 119, 127
abstract_inverted_index.source	159
abstract_inverted_index.space.	136
abstract_inverted_index.system	49
abstract_inverted_index.types,	26
abstract_inverted_index.LanHAR,	46
abstract_inverted_index.address	42
abstract_inverted_index.bridges	103
abstract_inverted_index.employs	85
abstract_inverted_index.encoder	120
abstract_inverted_index.labels.	112
abstract_inverted_index.methods	149
abstract_inverted_index.models.	40
abstract_inverted_index.produce	91
abstract_inverted_index.propose	45
abstract_inverted_index.reading	128
abstract_inverted_index.safety,	16
abstract_inverted_index.sensors	9
abstract_inverted_index.Activity	1
abstract_inverted_index.However,	20
abstract_inverted_index.Inertial	5
abstract_inverted_index.Language	53
abstract_inverted_index.activity	23, 64, 111, 156
abstract_inverted_index.approach	70, 145
abstract_inverted_index.critical	11
abstract_inverted_index.datasets	141
abstract_inverted_index.enabling	125
abstract_inverted_index.enhances	78
abstract_inverted_index.generate	57
abstract_inverted_index.publicly	162
abstract_inverted_index.readings	62, 109
abstract_inverted_index.reducing	35
abstract_inverted_index.semantic	58, 93, 105, 134
abstract_inverted_index.suitable	121
abstract_inverted_index.training	100
abstract_inverted_index.available	163
abstract_inverted_index.datasets,	34
abstract_inverted_index.framework	101
abstract_inverted_index.iterative	87
abstract_inverted_index.leverages	51
abstract_inverted_index.mitigates	73
abstract_inverted_index.patterns,	24
abstract_inverted_index.two-stage	99
abstract_inverted_index.industrial	18
abstract_inverted_index.placements	29
abstract_inverted_index.ultimately	114
abstract_inverted_index.variations	21
abstract_inverted_index.Experiments	137
abstract_inverted_index.Measurement	6
abstract_inverted_index.Recognition	2
abstract_inverted_index.activities.	83
abstract_inverted_index.demonstrate	142
abstract_inverted_index.deployment,	124
abstract_inverted_index.healthcare,	15
abstract_inverted_index.lightweight	118
abstract_inverted_index.outperforms	147
abstract_inverted_index.performance	37
abstract_inverted_index.production.	19
abstract_inverted_index.recognition	80
abstract_inverted_index.applications	13
abstract_inverted_index.distribution	31
abstract_inverted_index.high-quality	92
abstract_inverted_index.recognition.	157
abstract_inverted_index.cross-dataset	67, 74, 152
abstract_inverted_index.heterogeneity	75
abstract_inverted_index.re-generation	88
abstract_inverted_index.significantly	146
abstract_inverted_index.interpretation	135
abstract_inverted_index.interpretations	59, 94, 106
abstract_inverted_index.state-of-the-art	148
abstract_inverted_index.https://github.com/DASHLab/LanHAR.	165
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	6
citation_normalized_percentile