Pith. sign in

REVIEW 1 cited by

Efficient Document Re-Ranking for Transformers by Precomputing Term Representations

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 2004.14255 v2 pith:X3ATP4FX submitted 2020-04-29 cs.IR

classification cs.IR
keywords rankingdocumentrepresentationsnetworksquerytermtransformerapproach
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Deep pretrained transformer networks are effective at various ranking tasks, such as question answering and ad-hoc document ranking. However, their computational expenses deem them cost-prohibitive in practice. Our proposed approach, called PreTTR (Precomputing Transformer Term Representations), considerably reduces the query-time latency of deep transformer networks (up to a 42x speedup on web document ranking) making these networks more practical to use in a real-time ranking scenario. Specifically, we precompute part of the document term representations at indexing time (without a query), and merge them with the query representation at query time to compute the final ranking score. Due to the large size of the token representations, we also propose an effective approach to reduce the storage requirement by training a compression layer to match attention scores. Our compression technique reduces the storage required up to 95% and it can be applied without a substantial degradation in ranking performance.

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

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

  1. FullRecall: A Semantic Search-Based Ranking Approach for Maximizing Recall in Patent Retrieval

    cs.IR 2025-07 reject novelty 4.0 of 10

    A three-phase patent retrieval pipeline achieved 100% recall on five examiner-cited test queries, but the score is driven by post hoc cutoff choices and a candidate set that already contains the target patents.

Pith tools