ActionPiece: Contextually Tokenizing Action Sequences for Generative Recommendation Article Swipe

PDF

Yupeng Hou , Jianmo Ni , Zhankui He , Noveen Sachdeva , Wang-Cheng Kang , Ed H. , Julian McAuley , Derek Zhiyuan Cheng ·

YOU? · · 2025 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2502.13581

Generative recommendation (GR) is an emerging paradigm where user actions are tokenized into discrete token patterns and autoregressively generated as predictions. However, existing GR models tokenize each action independently, assigning the same fixed tokens to identical actions across all sequences without considering contextual relationships. This lack of context-awareness can lead to suboptimal performance, as the same action may hold different meanings depending on its surrounding context. To address this issue, we propose ActionPiece to explicitly incorporate context when tokenizing action sequences. In ActionPiece, each action is represented as a set of item features. Given the action sequence corpora, we construct the vocabulary by merging feature patterns as new tokens, based on their co-occurrence frequency both within individual sets and across adjacent sets. Considering the unordered nature of feature sets, we further introduce set permutation regularization, which produces multiple segmentations of action sequences with the same semantics. Our code is available at: https://github.com/google-deepmind/action_piece.

Related Topics

Generative Grammar

Computer Science

Artificial Intelligence

Physics

Quantum Mechanics

Concepts

Generative grammar Action (physics) Computer science Artificial intelligence Physics Quantum mechanics

Metadata

Type: preprint
Language: en
Landing Page: http://arxiv.org/abs/2502.13581
PDF: https://arxiv.org/pdf/2502.13581
OA Status: green
Cited By: 1
Related Works: 10
OpenAlex ID: https://openalex.org/W4407764389

All OpenAlex metadata

Raw OpenAlex JSON

OpenAlex ID: https://openalex.org/W4407764389

Canonical identifier for this work in OpenAlex
DOI: https://doi.org/10.48550/arxiv.2502.13581

Digital Object Identifier
Title: ActionPiece: Contextually Tokenizing Action Sequences for Generative Recommendation

Work title
Type: preprint

OpenAlex work type
Language: en

Primary language
Publication year: 2025

Year of publication
Publication date: 2025-02-19

Full publication date if available
Authors: Yupeng Hou, Jianmo Ni, Zhankui He, Noveen Sachdeva, Wang-Cheng Kang, Ed H., Julian McAuley, Derek Zhiyuan Cheng

List of authors in order
Landing page: https://arxiv.org/abs/2502.13581

Publisher landing page
PDF URL: https://arxiv.org/pdf/2502.13581

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/2502.13581

Direct OA link when available
Concepts: Generative grammar, Action (physics), Computer science, Artificial intelligence, Physics, Quantum mechanics

Top concepts (fields/topics) attached by OpenAlex
Cited by: 1

Total citation count in OpenAlex
Citations by year (recent): 2025: 1

Per-year citation counts (last 5 years)
Related works (count): 10

Other works algorithmically related by OpenAlex

Full payload

id	https://openalex.org/W4407764389
doi	https://doi.org/10.48550/arxiv.2502.13581
ids.doi	https://doi.org/10.48550/arxiv.2502.13581
ids.openalex	https://openalex.org/W4407764389
fwci
type	preprint
title	ActionPiece: Contextually Tokenizing Action Sequences for Generative Recommendation
biblio.issue
biblio.volume
biblio.last_page
biblio.first_page
topics[0].id	https://openalex.org/T10028
topics[0].field.id	https://openalex.org/fields/17
topics[0].field.display_name	Computer Science
topics[0].score	0.991599977016449
topics[0].domain.id	https://openalex.org/domains/3
topics[0].domain.display_name	Physical Sciences
topics[0].subfield.id	https://openalex.org/subfields/1702
topics[0].subfield.display_name	Artificial Intelligence
topics[0].display_name	Topic Modeling
topics[1].id	https://openalex.org/T10203
topics[1].field.id	https://openalex.org/fields/17
topics[1].field.display_name	Computer Science
topics[1].score	0.9732999801635742
topics[1].domain.id	https://openalex.org/domains/3
topics[1].domain.display_name	Physical Sciences
topics[1].subfield.id	https://openalex.org/subfields/1710
topics[1].subfield.display_name	Information Systems
topics[1].display_name	Recommender Systems and Techniques
topics[2].id	https://openalex.org/T13702
topics[2].field.id	https://openalex.org/fields/17
topics[2].field.display_name	Computer Science
topics[2].score	0.9693999886512756
topics[2].domain.id	https://openalex.org/domains/3
topics[2].domain.display_name	Physical Sciences
topics[2].subfield.id	https://openalex.org/subfields/1702
topics[2].subfield.display_name	Artificial Intelligence
topics[2].display_name	Machine Learning in Healthcare
is_xpac	False
apc_list
apc_paid
concepts[0].id	https://openalex.org/C39890363
concepts[0].level	2
concepts[0].score	0.7647656798362732
concepts[0].wikidata	https://www.wikidata.org/wiki/Q36108
concepts[0].display_name	Generative grammar
concepts[1].id	https://openalex.org/C2780791683
concepts[1].level	2
concepts[1].score	0.6743399500846863
concepts[1].wikidata	https://www.wikidata.org/wiki/Q846785
concepts[1].display_name	Action (physics)
concepts[2].id	https://openalex.org/C41008148
concepts[2].level	0
concepts[2].score	0.32616350054740906
concepts[2].wikidata	https://www.wikidata.org/wiki/Q21198
concepts[2].display_name	Computer science
concepts[3].id	https://openalex.org/C154945302
concepts[3].level	1
concepts[3].score	0.306252658367157
concepts[3].wikidata	https://www.wikidata.org/wiki/Q11660
concepts[3].display_name	Artificial intelligence
concepts[4].id	https://openalex.org/C121332964
concepts[4].level	0
concepts[4].score	0.07562202215194702
concepts[4].wikidata	https://www.wikidata.org/wiki/Q413
concepts[4].display_name	Physics
concepts[5].id	https://openalex.org/C62520636
concepts[5].level	1
concepts[5].score	0.0
concepts[5].wikidata	https://www.wikidata.org/wiki/Q944
concepts[5].display_name	Quantum mechanics
keywords[0].id	https://openalex.org/keywords/generative-grammar
keywords[0].score	0.7647656798362732
keywords[0].display_name	Generative grammar
keywords[1].id	https://openalex.org/keywords/action
keywords[1].score	0.6743399500846863
keywords[1].display_name	Action (physics)
keywords[2].id	https://openalex.org/keywords/computer-science
keywords[2].score	0.32616350054740906
keywords[2].display_name	Computer science
keywords[3].id	https://openalex.org/keywords/artificial-intelligence
keywords[3].score	0.306252658367157
keywords[3].display_name	Artificial intelligence
keywords[4].id	https://openalex.org/keywords/physics
keywords[4].score	0.07562202215194702
keywords[4].display_name	Physics
language	en
locations[0].id	pmh:oai:arXiv.org:2502.13581
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/2502.13581
locations[0].version	submittedVersion
locations[0].raw_type	text
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/2502.13581
locations[1].id	doi:10.48550/arxiv.2502.13581
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.2502.13581
indexed_in	arxiv, datacite
authorships[0].author.id	https://openalex.org/A5052347581
authorships[0].author.orcid	https://orcid.org/0000-0002-0747-8010
authorships[0].author.display_name	Yupeng Hou
authorships[0].author_position	first
authorships[0].raw_author_name	Hou, Yupeng
authorships[0].is_corresponding	False
authorships[1].author.id	https://openalex.org/A5077817759
authorships[1].author.orcid	https://orcid.org/0000-0002-6863-8073
authorships[1].author.display_name	Jianmo Ni
authorships[1].author_position	middle
authorships[1].raw_author_name	Ni, Jianmo
authorships[1].is_corresponding	False
authorships[2].author.id	https://openalex.org/A5061585298
authorships[2].author.orcid	https://orcid.org/0000-0002-9139-8004
authorships[2].author.display_name	Zhankui He
authorships[2].author_position	middle
authorships[2].raw_author_name	He, Zhankui
authorships[2].is_corresponding	False
authorships[3].author.id	https://openalex.org/A5005200778
authorships[3].author.orcid	https://orcid.org/0000-0003-1228-746X
authorships[3].author.display_name	Noveen Sachdeva
authorships[3].author_position	middle
authorships[3].raw_author_name	Sachdeva, Noveen
authorships[3].is_corresponding	False
authorships[4].author.id	https://openalex.org/A5090597599
authorships[4].author.orcid	https://orcid.org/0009-0006-8795-3665
authorships[4].author.display_name	Wang-Cheng Kang
authorships[4].author_position	middle
authorships[4].raw_author_name	Kang, Wang-Cheng
authorships[4].is_corresponding	False
authorships[5].author.id	https://openalex.org/A5028125399
authorships[5].author.orcid	https://orcid.org/0000-0003-3230-5338
authorships[5].author.display_name	Ed H.
authorships[5].author_position	middle
authorships[5].raw_author_name	Chi, Ed H.
authorships[5].is_corresponding	False
authorships[6].author.id	https://openalex.org/A5021827617
authorships[6].author.orcid	https://orcid.org/0000-0003-0955-7588
authorships[6].author.display_name	Julian McAuley
authorships[6].author_position	middle
authorships[6].raw_author_name	McAuley, Julian
authorships[6].is_corresponding	False
authorships[7].author.id	https://openalex.org/A5055484401
authorships[7].author.orcid	https://orcid.org/0009-0000-7943-8328
authorships[7].author.display_name	Derek Zhiyuan Cheng
authorships[7].author_position	last
authorships[7].raw_author_name	Cheng, Derek Zhiyuan
authorships[7].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/2502.13581
open_access.oa_status	green
open_access.any_repository_has_fulltext	False
created_date	2025-10-10T00:00:00
display_name	ActionPiece: Contextually Tokenizing Action Sequences for Generative Recommendation
has_fulltext	False
is_retracted	False
updated_date	2025-11-06T06:51:31.235846
primary_topic.id	https://openalex.org/T10028
primary_topic.field.id	https://openalex.org/fields/17
primary_topic.field.display_name	Computer Science
primary_topic.score	0.991599977016449
primary_topic.domain.id	https://openalex.org/domains/3
primary_topic.domain.display_name	Physical Sciences
primary_topic.subfield.id	https://openalex.org/subfields/1702
primary_topic.subfield.display_name	Artificial Intelligence
primary_topic.display_name	Topic Modeling
related_works	https://openalex.org/W4391375266, https://openalex.org/W2899084033, https://openalex.org/W2748952813, https://openalex.org/W2380075625, https://openalex.org/W2390279801, https://openalex.org/W4391913857, https://openalex.org/W2358668433, https://openalex.org/W4396701345, https://openalex.org/W2376932109, https://openalex.org/W2001405890
cited_by_count	1
counts_by_year[0].year	2025
counts_by_year[0].cited_by_count	1
locations_count	2
best_oa_location.id	pmh:oai:arXiv.org:2502.13581
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/2502.13581
best_oa_location.version	submittedVersion
best_oa_location.raw_type	text
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/2502.13581
primary_location.id	pmh:oai:arXiv.org:2502.13581
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/2502.13581
primary_location.version	submittedVersion
primary_location.raw_type	text
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/2502.13581
publication_date	2025-02-19
publication_year	2025
referenced_works_count	0
abstract_inverted_index.a	88
abstract_inverted_index.GR	23
abstract_inverted_index.In	81
abstract_inverted_index.To	66
abstract_inverted_index.an	4
abstract_inverted_index.as	19, 53, 87, 106
abstract_inverted_index.by	102
abstract_inverted_index.is	3, 85, 148
abstract_inverted_index.of	46, 90, 126, 139
abstract_inverted_index.on	62, 110
abstract_inverted_index.to	34, 50, 73
abstract_inverted_index.we	70, 98, 129
abstract_inverted_index.Our	146
abstract_inverted_index.all	38
abstract_inverted_index.and	16, 118
abstract_inverted_index.are	10
abstract_inverted_index.at:	150
abstract_inverted_index.can	48
abstract_inverted_index.its	63
abstract_inverted_index.may	57
abstract_inverted_index.new	107
abstract_inverted_index.set	89, 132
abstract_inverted_index.the	30, 54, 94, 100, 123, 143
abstract_inverted_index.(GR)	2
abstract_inverted_index.This	44
abstract_inverted_index.both	114
abstract_inverted_index.code	147
abstract_inverted_index.each	26, 83
abstract_inverted_index.hold	58
abstract_inverted_index.into	12
abstract_inverted_index.item	91
abstract_inverted_index.lack	45
abstract_inverted_index.lead	49
abstract_inverted_index.same	31, 55, 144
abstract_inverted_index.sets	117
abstract_inverted_index.this	68
abstract_inverted_index.user	8
abstract_inverted_index.when	77
abstract_inverted_index.with	142
abstract_inverted_index.Given	93
abstract_inverted_index.based	109
abstract_inverted_index.fixed	32
abstract_inverted_index.sets,	128
abstract_inverted_index.sets.	121
abstract_inverted_index.their	111
abstract_inverted_index.token	14
abstract_inverted_index.where	7
abstract_inverted_index.which	135
abstract_inverted_index.across	37, 119
abstract_inverted_index.action	27, 56, 79, 84, 95, 140
abstract_inverted_index.issue,	69
abstract_inverted_index.models	24
abstract_inverted_index.nature	125
abstract_inverted_index.tokens	33
abstract_inverted_index.within	115
abstract_inverted_index.actions	9, 36
abstract_inverted_index.address	67
abstract_inverted_index.context	76
abstract_inverted_index.feature	104, 127
abstract_inverted_index.further	130
abstract_inverted_index.merging	103
abstract_inverted_index.propose	71
abstract_inverted_index.tokens,	108
abstract_inverted_index.without	40
abstract_inverted_index.However,	21
abstract_inverted_index.adjacent	120
abstract_inverted_index.context.	65
abstract_inverted_index.corpora,	97
abstract_inverted_index.discrete	13
abstract_inverted_index.emerging	5
abstract_inverted_index.existing	22
abstract_inverted_index.meanings	60
abstract_inverted_index.multiple	137
abstract_inverted_index.paradigm	6
abstract_inverted_index.patterns	15, 105
abstract_inverted_index.produces	136
abstract_inverted_index.sequence	96
abstract_inverted_index.tokenize	25
abstract_inverted_index.assigning	29
abstract_inverted_index.available	149
abstract_inverted_index.construct	99
abstract_inverted_index.depending	61
abstract_inverted_index.different	59
abstract_inverted_index.features.	92
abstract_inverted_index.frequency	113
abstract_inverted_index.generated	18
abstract_inverted_index.identical	35
abstract_inverted_index.introduce	131
abstract_inverted_index.sequences	39, 141
abstract_inverted_index.tokenized	11
abstract_inverted_index.unordered	124
abstract_inverted_index.Generative	0
abstract_inverted_index.contextual	42
abstract_inverted_index.explicitly	74
abstract_inverted_index.individual	116
abstract_inverted_index.semantics.	145
abstract_inverted_index.sequences.	80
abstract_inverted_index.suboptimal	51
abstract_inverted_index.tokenizing	78
abstract_inverted_index.vocabulary	101
abstract_inverted_index.ActionPiece	72
abstract_inverted_index.Considering	122
abstract_inverted_index.considering	41
abstract_inverted_index.incorporate	75
abstract_inverted_index.permutation	133
abstract_inverted_index.represented	86
abstract_inverted_index.surrounding	64
abstract_inverted_index.ActionPiece,	82
abstract_inverted_index.performance,	52
abstract_inverted_index.predictions.	20
abstract_inverted_index.co-occurrence	112
abstract_inverted_index.segmentations	138
abstract_inverted_index.independently,	28
abstract_inverted_index.recommendation	1
abstract_inverted_index.relationships.	43
abstract_inverted_index.regularization,	134
abstract_inverted_index.autoregressively	17
abstract_inverted_index.context-awareness	47
abstract_inverted_index.https://github.com/google-deepmind/action_piece.	151
cited_by_percentile_year
countries_distinct_count	0
institutions_distinct_count	8
citation_normalized_percentile