Pith. sign in

REVIEW 11 cited by

Autoregressive Entity Retrieval

Not yet reviewed by Pith; the record is open.

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 2010.00904 v3 pith:XZACYZ2H submitted 2020-10-02 cs.CL cs.IRcs.LGstat.ML

Autoregressive Entity Retrieval

classification cs.CL cs.IRcs.LGstat.ML
keywords entityentitiesautoregressivefootprintmemorycontextdatadense
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

Entities are at the center of how we represent and aggregate knowledge. For instance, Encyclopedias such as Wikipedia are structured by entities (e.g., one per Wikipedia article). The ability to retrieve such entities given a query is fundamental for knowledge-intensive tasks such as entity linking and open-domain question answering. Current approaches can be understood as classifiers among atomic labels, one for each entity. Their weight vectors are dense entity representations produced by encoding entity meta information such as their descriptions. This approach has several shortcomings: (i) context and entity affinity is mainly captured through a vector dot product, potentially missing fine-grained interactions; (ii) a large memory footprint is needed to store dense representations when considering large entity sets; (iii) an appropriately hard set of negative data has to be subsampled at training time. In this work, we propose GENRE, the first system that retrieves entities by generating their unique names, left to right, token-by-token in an autoregressive fashion. This mitigates the aforementioned technical issues since: (i) the autoregressive formulation directly captures relations between context and entity name, effectively cross encoding both; (ii) the memory footprint is greatly reduced because the parameters of our encoder-decoder architecture scale with vocabulary size, not entity count; (iii) the softmax loss is computed without subsampling negative data. We experiment with more than 20 datasets on entity disambiguation, end-to-end entity linking and document retrieval tasks, achieving new state-of-the-art or very competitive results while using a tiny fraction of the memory footprint of competing systems. Finally, we demonstrate that new entities can be added by simply specifying their names. Code and pre-trained models at https://github.com/facebookresearch/GENRE.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 11 Pith papers

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

  1. ICICLE: Expanding Retrieval with In-Context Documents

    cs.IR 2026-05 unverdicted novelty 7.0

    ICICLE is an in-context indexing method for generative retrieval that uses source-aware docid generation with [COPY] routing and calibration to handle new documents without retraining.

  2. Beyond Semantic IDs: Encoding Business-Value Ranking into Document Identifiers for Generative Retrieval

    cs.IR 2026-07 conditional novelty 6.0

    Encoding business-value ranks inside semantic clusters yields collision-free DocIDs that improve generative-retrieval Hitrate and deliver +1.06% GMV online.

  3. Generative Chinese Statute Retrieval

    cs.IR 2026-07 conditional novelty 6.0

    A generative retriever with multi-granularity structured statute IDs and multi-task training outperforms strong sparse, dense, and legal baselines on the STARD Chinese statute benchmark.

  4. DeSQ: Decomposition-based SPARQL Query Generation

    cs.CL 2026-05 unverdicted novelty 6.0

    DeSQ decomposes questions into atomic constraints, maps them to SPARQL fragments with placeholders, grounds the placeholders, and assembles complete queries, outperforming prior methods on four of five benchmarks.

  5. TubiFM: Unified Item, Carousel, and Search Ranking for Streaming Discovery

    cs.IR 2026-05 unverdicted novelty 6.0

    A Llama-based model trained on serialized user stories unifies item, carousel, and search ranking and outperforms specialist baselines offline while improving some online metrics and reducing latency.

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

    cs.IR 2026-05 unverdicted novelty 6.0

    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.

  7. Cram Less to Fit More: Training Data Pruning Improves Memorization of Facts

    cs.CL 2026-04 conditional novelty 6.0

    Loss-based pruning of training data to limit facts and flatten their frequency distribution enables a 110M-parameter GPT-2 model to memorize 1.3 times more entity facts than standard training, matching a 1.3B-paramete...

  8. MVIGER: Multi-View Variational Integration of Complementary Knowledge for Generative Recommender

    cs.IR 2024-08 unverdicted novelty 6.0

    MVIGER integrates complementary knowledge from diverse prompts and indices in generative recommenders via a variational model with learnable prior over latent sources, showing superior performance on three datasets.

  9. G-DRAGON: Geospatial Reasoning and Dynamic Planning for Retrieval-Augmented Outdoor Navigation

    cs.RO 2026-05 unverdicted novelty 5.0

    G-DRAGON framework maps language commands to OSM coordinates via lightweight LLM for global planning and uses frontier exploration for local targets, outperforming baselines in simulation and completing real UGV perso...

  10. Multi-Perspective Evidence Synthesis and Reasoning for Unsupervised Multimodal Entity Linking

    cs.CL 2026-04 unverdicted novelty 5.0

    MSR-MEL synthesizes instance-centric, group-level, lexical, and statistical evidence with LLMs and asymmetric teacher-student GNNs to outperform prior unsupervised methods on multimodal entity linking benchmarks.

  11. OneRec: Unifying Retrieve and Rank with Generative Recommender and Iterative Preference Alignment

    cs.IR 2025-02 unverdicted novelty 5.0

    OneRec unifies retrieval and ranking in a generative recommender using session-wise decoding and iterative DPO-based preference alignment, achieving real-world gains on Kuaishou.