Pith. sign in

REVIEW 17 cited by

Recommender Systems with Generative Retrieval

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2305.05065 v3 pith:A4NCPS2R submitted 2023-05-08 cs.IR cs.LG

Shashank Rajput , Nikhil Mehta , Anima Singh , Raghunandan H. Keshavan , Trung Vu , Lukasz Heldt , Lichan Hong , Yi Tay
show 5 more authors
Vinh Q. Tran Jonah Samost Maciej Kula Ed H. Chi Maheswaran Sathiamoorthy
This is my paper
classification cs.IRcs.LG
keywords retrievalsemanticmodelcandidatesgenerativeitemrecommendersystems
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Modern recommender systems perform large-scale retrieval by first embedding queries and item candidates in the same unified space, followed by approximate nearest neighbor search to select top candidates given a query embedding. In this paper, we propose a novel generative retrieval approach, where the retrieval model autoregressively decodes the identifiers of the target candidates. To that end, we create semantically meaningful tuple of codewords to serve as a Semantic ID for each item. Given Semantic IDs for items in a user session, a Transformer-based sequence-to-sequence model is trained to predict the Semantic ID of the next item that the user will interact with. To the best of our knowledge, this is the first Semantic ID-based generative model for recommendation tasks. We show that recommender systems trained with the proposed paradigm significantly outperform the current SOTA models on various datasets. In addition, we show that incorporating Semantic IDs into the sequence-to-sequence model enhances its ability to generalize, as evidenced by the improved retrieval performance observed for items with no prior interaction history.

Discussion (0). Sign in to comment.

Forward citations

Cited by 17 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Learning Variable-Length Tokenization for Generative Recommendation

    cs.LG 2026-05 unverdicted novelty 7.0 of 10

    VarLenRec learns variable-length semantic IDs for generative recommendation by allocating longer codes to tail items via popularity-weighted information budget allocation, hyperbolic residual quantization, and a diffe...

  2. Hierarchical Residual Policy Optimization for Generative Recommendations

    cs.IR 2026-08 conditional novelty 6.0 of 10

    HRPO decomposes item-level rewards into token-level 'residual credits' along semantic identifier hierarchies and optimizes the generator with a PPO-style objective, improving session utility in KuaiSim and production ...

  3. Mosaic: A Fleet of User Embedding Specialists for Recommendation at Meta

    cs.IR 2026-07 conditional novelty 6.0 of 10

    Mosaic shows that a fleet of four heterogeneous user-embedding specialists, trained with redundancy-reduction and composite-label losses, improves downstream recommendation quality at Meta.

  4. EGR: Embedding-Native Generative Retrieval with a Shared LLM

    cs.IR 2026-07 conditional novelty 6.0 of 10

    EGR jointly trains a single LLM to embed both item metadata and user histories in one dense-vector space, and reports superior retrieval and a +2.91% production CVR lift.

  5. An LLM-powered Agentic Recommendation System for Connected TV Content Discovery

    cs.IR 2026-07 conditional novelty 6.0 of 10

    A hybrid system where an LLM agent generates CTV topic channels and traditional ML ranks the media inside them is engineered to run at production latency via async caching, but only its retrieval component is quality-...

  6. Gated Bidirectional Linear Attention for Generative Retrieval

    cs.IR 2026-06 unverdicted novelty 6.0 of 10

    GBLA extends kernelized linear attention with local causal mixing, key gating, and gated RMSNorm; a 1:2 hybrid with self-attention matches full bidirectional self-attention quality on Yandex Music data while deliverin...

  7. CapsID: Soft-Routed Variable-Length Semantic IDs for Generative Recommendation

    cs.IR 2026-05 unverdicted novelty 6.0 of 10

    CapsID uses probabilistic capsule routing and confidence-based termination to generate variable-length semantic IDs, improving recall by 9.6% over strong baselines with half the latency of dual-representation systems.

  8. Vectorizing the Trie: Efficient Constrained Decoding for LLM-based Generative Retrieval on Accelerators

    cs.IR 2026-02 conditional novelty 6.0 of 10

    Constrained decoding for generative retrieval can be made accelerator-friendly by flattening the trie of valid items into a CSR sparse matrix and doing branch-free vectorized lookups.

  9. Session-Level Optimization for Large-Scale Retrieval using REINFORCE with Multi-Step Off-Policy Correction

    cs.IR 2026-07 conditional novelty 5.5 of 10

    Autoregressive multi-step off-policy REINFORCE plus a user-feedback model improves offline cumulative session reward for generative retrieval on Yambda-5B without large retrieval degradation.

  10. Serving the Long Tail: Training-Free LLM Candidate Generation for Vacation Rental Marketplaces

    cs.LG 2026-07 conditional novelty 5.0 of 10

    Union fusion of LLM metadata queries with IBKNN extends candidate coverage to cold-start and long-tail Vrbo listings while matching or beating IBKNN recall at every K and collapsing small-vs-frontier LLM gaps under 1%.

  11. Session-Level Optimization for Large-Scale Retrieval using REINFORCE with Multi-Step Off-Policy Correction

    cs.IR 2026-07 conditional novelty 5.0 of 10

    Off-policy REINFORCE with up to 10 importance-weight factors raises estimated discounted session reward over next-item and positive-only baselines in offline evaluation on the Yambda-5B dataset.

  12. VirtualMLE: A Virtual ML Engineer that Optimizes Sequential Recommenders

    cs.IR 2026-06 unverdicted novelty 5.0 of 10

    VirtualMLE deploys an LLM agent with execution-reflection-memory to tune sequential recommenders, reaching competitive quality on Amazon benchmarks with fewer trials and transferring heuristics across datasets.

  13. TwiSTAR:Think Fast, Think Slow, Then Act,Generative Recommendation with Adaptive Reasoning

    cs.IR 2026-05 unverdicted novelty 5.0 of 10

    TwiSTAR learns to switch between fast SID retrieval and slow rationale-generating reasoning in generative recommendation, yielding better accuracy-latency trade-offs on three datasets.

  14. SID-Coord: Coordinating Semantic IDs for ID-based Ranking in Short-Video Search

    cs.IR 2026-04 unverdicted novelty 5.0 of 10

    SID-Coord coordinates semantic IDs with hashed item IDs via attention fusion, adaptive gating, and interest alignment, yielding +0.664% long-play rate and +0.369% playback duration gains in production search ranking.

  15. An LLM-powered Agentic Recommendation System for Connected TV Content Discovery

    cs.IR 2026-07 conditional novelty 4.5 of 10

    A hybrid agentic CTV recommender uses LLMs for contextual topic generation/ranking and traditional ML for media ranking, made production-viable via async caching and MediaBrain generative retrieval.

  16. Fine-Tuned LLM as a Complementary Predictor Improving Ads System

    cs.IR 2026-05 unverdicted novelty 4.0 of 10

    Fine-tuned LLM acts as ancillary advertiser predictor in production ads RecSys, augmenting retrieval and ranking with measurable offline and online gains.

  17. Retrieval-Augmented Generation for Large Language Models: A Survey

    cs.CL 2023-12 unverdicted novelty 3.0 of 10

    A survey of RAG paradigms, components, benchmarks, and challenges for improving LLMs on knowledge-intensive tasks.

Pith tools