mlsanityData to 14 Oct 2026

ActionPiece: Contextually Tokenizing Action Sequences for Generative Recommendation

arXiv 2502.13581 ↗First posted 19 Feb 2025cs.IR cs.LGcited by 18

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. Cited by 18 papers in the map (data through 14 Oct 2026). Builds on 62 of its 77 references that are in the map. First posted on arXiv 19 Feb 2025.

Builds on

62 of 77 references.

References · 62

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  2. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,599
  3. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 14,132
  4. 2014Neural Machine Translation by Jointly Learning to Align and Translatecited by 4,548
  5. 2014Sequence to Sequence Learning with Neural Networkscited by 2,986
  6. 2015Neural Machine Translation of Rare Words with Subword Unitscited by 2,918
  7. 2023A Survey of Large Language Modelscited by 2,576
  8. 2018SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processingcited by 1,989
  9. 2016Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translationcited by 1,475
  10. 2018Self-Attentive Sequential Recommendationcited by 1,184
  11. 2019BERT4Rec: Sequential Recommendation with Bidirectional Encoder Representations from Transformercited by 964
  12. 2021SoundStream: An End-to-End Neural Audio Codeccited by 771
38 more
  1. 2016Wide & Deep Learning for Recommender Systemscited by 747
  2. 2015Image-based Recommendations on Styles and Substitutescited by 706
  3. 2015Session-based Recommendations with Recurrent Neural Networkscited by 695
  4. 2018Subword Regularization: Improving Neural Network Translation Models with Multiple Subword Candidatescited by 639
  5. 2021Sentence-T5: Scalable Sentence Encoders from Pre-trained Text-to-Text Modelscited by 463
  6. 2022Recommendation as Language Processing (RLP): A Unified Pretrain, Personalized Prompt & Predict Paradigm (P5)cited by 402
  7. 2017Deep & Cross Network for Ad Click Predictionscited by 393
  8. 2024The Faiss librarycited by 376
  9. 2023TALLRec: An Effective and Efficient Tuning Framework to Align Large Language Model with Recommendationcited by 337
  10. 2020DCN V2: Improved Deep & Cross Network and Practical Lessons for Web-scale Learning to Rank Systemscited by 327
  11. 2020S^3-Rec: Self-Supervised Learning for Sequential Recommendation with Mutual Information Maximizationcited by 311
  12. 2023Large Language Models are Zero-Shot Rankers for Recommender Systemscited by 309
  13. 2022Transformer Memory as a Differentiable Search Indexcited by 247
  14. 2017Neural Attentive Session-based Recommendationcited by 237
  15. 2024Bridging Language and Items for Retrieval and Recommendation: Benchmarking LLMs as Semantic Encoderscited by 224
  16. 2023Recommender Systems with Generative Retrievalcited by 198
  17. 2019BPE-Dropout: Simple and Effective Subword Regularizationcited by 195
  18. 2024Actions Speak Louder than Words: Trillion-Parameter Sequential Transducers for Generative Recommendationscited by 188
  19. 2023Adapting Large Language Models by Integrating Collaborative Semantics for Recommendationcited by 179
  20. 2023Recommendation as Instruction Following: A Large Language Model Empowered Recommendation Approachcited by 168
  21. 2020RecBole: Towards a Unified, Comprehensive and Efficient Framework for Recommendation Algorithmscited by 138
  22. 2023Where to Go Next for Recommender Systems? ID- vs. Modality-based Recommender Models Revisitedcited by 124
  23. 2022Towards Universal Sequence Representation Learning for Recommender Systemscited by 121
  24. 2022Learning Vector-Quantized Item Representation for Transferable Sequential Recommenderscited by 112
  25. 2024Learnable Item Tokenization for Generative Recommendationcited by 110
  26. 2023How to Index Item IDs for Recommendation Foundation Modelscited by 102
  27. 2023CoLLM: Integrating Collaborative Embeddings into Large Language Models for Recommendationcited by 100
  28. 2024Tokenization Is More Than Compressioncited by 89
  29. 2024Getting the most out of your tokenizer for pre-training and domain adaptationcited by 66
  30. 2023Collaborative Large Language Model for Recommender Systemscited by 64
  31. 2024A Review of Modern Recommender Systems Using Generative Models (Gen-RecSys)cited by 58
  32. 2023LLaRA: Large Language-Recommendation Assistantcited by 56
  33. 2024IDGenRec: LLM-RecSys Alignment with Textual ID Learningcited by 53
  34. 2023Better Generalization with Semantic IDs: A Case Study in Ranking for Recommendationscited by 52
  35. 2024Large Language Models meet Collaborative Filtering: An Efficient All-round LLM-based Recommender Systemcited by 50
  36. 2020Dynamic Programming Encoding for Subword Segmentation in Neural Machine Translationcited by 38
  37. 2023A Formal Perspective on Byte-Pair Encodingcited by 36
  38. 2024CoST: Contrastive Quantization based Semantic Tokenization for Generative Recommendationcited by 26

50 most cited of 62 All 62 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 10; 2026: 8.

18 papers.

Sort:

By area:

  1. 2025RecoWorld: Building Simulated Environments for Agentic Recommender Systemscited by 19
  2. 2025CoFiRec: Coarse-to-Fine Tokenization for Generative Recommendationcited by 11
  3. 2025MMQ: Multimodal Mixture-of-Quantization Tokenization for Semantic ID Generation and User Behavioral Adaptationcited by 10
  4. 2025Pre-training Generative Recommender with Multi-Identifier Item Tokenizationcited by 4
  5. 2025Act-With-Think: Chunk Auto-Regressive Modeling for Generative Recommendationcited by 4
  6. 2025GRACE: Generative Recommendation via Journey-Aware Sparse Attention on Chain-of-Thought Tokenizationcited by 4
  7. 2025DiffGRM: Diffusion-based Generative Recommendation Modelcited by 4
  8. 2025Fading to Grow: Growing Preference Ratios via Preference Fading Discrete Diffusion for Recommendationcited by 3
  9. 2025Pctx: Tokenizing Personalized Context for Generative Recommendationcited by 3
  10. 2026Generative Recall, Dense Reranking: Learning Multi-View Semantic IDs for Efficient Text-to-Video Retrievalcited by 3
  11. 2026An Embarrassingly Simple Graph Heuristic Reveals Shortcut-Solvable Benchmarks for Sequential Recommendationcited by 1
  12. 2026ChronoID: Infusing Explicit Temporal Signals into Semantic IDs for Generative Recommendationcited by 1
6 more
  1. 2026T-RoPE: Time-Aware Rotary Position Embedding for Sequential Recommendationcited by 1
  2. 2025Token-Efficient Item Representation via Images for LLM Recommender Systemscited by 0
  3. 2026Position-Aware Drafting for Inference Acceleration in LLM-Based Generative List-Wise Recommendationcited by 0
  4. 2026On the Memorization Behavior of LLMs in Generative Recommendation: Observations, Implications, and Training Strategiescited by 0
  5. 2026SAGA: Structure-Attended Generative Action Embedding Model that encodes Multi-Surface User Action Sequencescited by 0
  6. 2026Towards a Densing Law for User Representation Learning at Billion-Scale Capacitycited by 0