paper-with-me

홈 › Papers

Test-Time Training for Zero-Resource Dense Retrieval Reranking

2026-05-31 · Shiyan Liu, Yichen Li arxiv

Dense retrievers excel at first-stage candidate generation but lack effective reranking in zero-resource settings. Existing approaches face a fundamental dilemma: cross-encoders deliver strong reranking quality but require costly supervised training and incur high latency, while unsupervised BM25 reranking consistently degrades dense retrieval performance on most of BEIR benchmarks. We propose DART (Dense Adaptive Reranking at Test-time), which resolves this dilemma by adapting the scoring function at inference time. For each query, the top-ranked documents serve as pseudo-positive examples and the bottom-ranked as pseudo-negative examples, providing noisy but readily available supervision to adapt a bilinear scoring matrix $W$ via a small number of gradient updates. We further introduce a confidence-weighted margin loss and a cross-query momentum buffer that warm-starts adaptation across queries. On six BEIR benchmarks, DART achieves a mean per-dataset relative NDCG@10 gain of +2.1% over the dense retrieval baseline with under 10ms additional latency per query, demonstrating a powerful capability for zero-shot performance enhancement and cross-domain generalization.

📄 PDF Abstract BibTeX arXiv:2606.01070

Code (0)

등록된 구현이 없습니다.

Tasks

Domain Generalization

Similar Papers 제목 키워드 기반

RSC: Accelerating Graph Neural Networks Training via Randomized Sparse Computations

2022-10-19 · Zirui Liu, Shengyuan Chen, Kaixiong Zhou, Daochen Zha 외

The training of graph neural networks (GNNs) is extremely time consuming because sparse graph-based operations are hard to be accelerated by hardware. Prior art explores trading off the computational precision to reduce …

Cross-Lingual Training with Dense Retrieval for Document Retrieval

2021-09-03 · Peng Shi, Rui Zhang, He Bai, Jimmy Lin

Dense retrieval has shown great success in passage ranking in English. However, its effectiveness in document retrieval for non-English languages remains unexplored due to the limitation in training resources. In this wo…

Document RankingPassage RankingRetrieval

Multilingual acoustic word embedding models for processing zero-resource languages

2020-02-06 · Herman Kamper, Yevgen Matusevych, Sharon Goldwater

Acoustic word embeddings are fixed-dimensional representations of variable-length speech segments. In settings where unlabelled speech is the only available resource, such embeddings can be used in "zero-resource" speech…

Transfer LearningWord Embeddings

Back to Basics: A Simple Recipe for Improving Out-of-Domain Retrieval in Dense Encoders

2023-11-16 · Hyunji Lee, Luca Soldaini, Arman Cohan, Minjoon Seo 외

Prevailing research practice today often relies on training dense retrievers on existing large datasets such as MSMARCO and then experimenting with ways to improve zero-shot generalization capabilities to unseen domains.…

Data AugmentationDomain GeneralizationRetrievalZero-shot Generalization

Energy-efficient Dense DNN Acceleration with Signed Bit-slice Architecture

2022-03-15 · Dongseok Im, Gwangtae Park, Zhiyong Li, Junha Ryu 외

As the number of deep neural networks (DNNs) to be executed on a mobile system-on-chip (SoC) increases, the mobile SoC suffers from the real-time DNN acceleration within its limited hardware resources and power budget. A…