Pith. sign in

REVIEW 4 cited by

Wacky Weights in Learned Sparse Representations and the Revenge of Score-at-a-Time Query Evaluation

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 2110.11540 v2 pith:OSWOTFXL submitted 2021-10-22 cs.IR

classification cs.IR
keywords querylatencyscore-at-a-timeapproachevaluationretrievalappearapproaches
verification ladder T0 review T1 audit T2 compute T3 formal

Signed reviews

No signed human review yet.

0 comments
read the original abstract

Recent advances in retrieval models based on learned sparse representations generated by transformers have led us to, once again, consider score-at-a-time query evaluation techniques for the top-k retrieval problem. Previous studies comparing document-at-a-time and score-at-a-time approaches have consistently found that the former approach yields lower mean query latency, although the latter approach has more predictable query latency. In our experiments with four different retrieval models that exploit representational learning with bags of words, we find that transformers generate "wacky weights" that appear to greatly reduce the opportunities for skipping and early exiting optimizations that lie at the core of standard document-at-a-time techniques. As a result, score-at-a-time approaches appear to be more competitive in terms of query evaluation latency than in previous studies. We find that, if an effectiveness loss of up to three percent can be tolerated, a score-at-a-time approach can yield substantial gains in mean query latency while at the same time dramatically reducing tail latency.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

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

  1. An Alternative to FLOPS Regularization to Effectively Productionize SPLADE-Doc

    cs.IR 2025-05 conditional novelty 6.0 of 10

    DF-FLOPS, a document-frequency-weighted variant of FLOPS regularization, cuts SPLADE-Doc latency in production Solr from 922 ms to 88-161 ms per query with only a 2.2-point MRR@10 loss in-domain.

  2. Effective Inference-Free Retrieval for Learned Sparse Representations

    cs.IR 2025-04 conditional novelty 6.0 of 10

    Li-LSR learns a static per-token query score from word embeddings, turning query encoding into a table lookup, and reports higher mRR@10 and nDCG@10 than Splade-v3-Doc.

  3. Beyond Quantile Methods: Improved Top-K Threshold Estimation for Traditional and Learned Sparse Indexes

    cs.IR 2024-12 conditional novelty 6.0 of 10

    New threshold estimation methods combining quantile precomputation with early termination and index lookups improve top-k score estimates, cutting the gap to the ideal MUF by large fractions on standard and learned sp...

  4. Rational Retrieval Acts: Leveraging Pragmatic Reasoning to Improve Sparse Retrieval

    cs.IR 2025-05 conditional novelty 5.0 of 10

    Rational Retrieval Acts applies Rational Speech Acts to sparse IR models, reweighting token-document associations by corpus context and improving retrieval accuracy on BEIR datasets.

Pith tools