MarkBERT: Marking Word Boundaries Improves Chinese BERT

Li, Linyang; Dai, Yong; Tang, Duyu; Feng, Zhangyin; Zhang, Cong; Qiu, Xipeng; Xu, Zenglin; Shi, Shuming

doi:10.48550/arxiv.2203.06378

Cited by 5 publications

(4 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(2) The need for data augmentation: they need to train a Doc2query model to provide the exact matching signal for improving the BERT re-ranker while our strategy does not need any extra overhead in terms of data augmentation. A few recent, but less related examples are Al-Hajj et al [23], who experiment with the use of different supervised signals into the input of the cross-encoder to emphasize target words in context and Li et al [24], who insert boundary markers into the input between contiguous words for Chinese named entity recognition. Additionally, there are various studies that show modifying cross-encoder re-ranker inputs by adding additional information represented as splitter tokens can improve their effectiveness [25][26][27][28].…”

Section: Modifying the Input Of Re-rankersmentioning

confidence: 99%

Injecting the score of the first-stage retriever as text improves BERT-based re-rankers

Askari,

Abolghasemi,

Pasi

et al. 2024

Discov Computing

View full text Add to dashboard Cite

In this paper we propose a novel approach for combining first-stage lexical retrieval models and Transformer-based re-rankers: we inject the relevance score of the lexical model as a token into the input of the cross-encoder re-ranker. It was shown in prior work that interpolation between the relevance score of lexical and Bidirectional Encoder Representations from Transformers (BERT) based re-rankers may not consistently result in higher effectiveness. Our idea is motivated by the finding that BERT models can capture numeric information. We compare several representations of the Best Match 25 (BM25) and Dense Passage Retrieval (DPR) scores and inject them as text in the input of four different cross-encoders. Since knowledge distillation, i.e., teacher-student training, proved to be highly effective for cross-encoder re-rankers, we additionally analyze the effect of injecting the relevance score into the student model while training the model by three larger teacher models. Evaluation on the MSMARCO Passage collection and the TREC DL collections shows that the proposed method significantly improves over all cross-encoder re-rankers as well as the common interpolation methods. We show that the improvement is consistent for all query types. We also find an improvement in exact matching capabilities over both the first-stage rankers and the cross-encoders. Our findings indicate that cross-encoder re-rankers can efficiently be improved without additional computational burden or extra steps in the pipeline by adding the output of the first-stage ranker to the model input. This effect is robust for different models and query types.

show abstract

Section: Modifying the Input Of Re-rankersmentioning

confidence: 99%

Injecting the score of the first-stage retriever as text improves BERT-based re-rankers

Askari,

Abolghasemi,

Pasi

et al. 2024

Discov Computing

View full text Add to dashboard Cite

show abstract

“…(2) The need for data augmentation: they need to train a Doc2query model to provide the exact matching signal for improving the BERT re-ranker while our strategy does not need any extra overhead in terms of data augmentation. A few recent, but less related examples are Al-Hajj et al [4], who experiment with the use of different supervised signals into the input of the cross-encoder to emphasize target words in context and Li et al [30], who insert boundary markers into the input between contiguous words for Chinese named entity recognition. Numerical information in Transformer models.…”

Section: Related Workmentioning

confidence: 99%

Injecting the BM25 Score as Text Improves BERT-Based Re-rankers

Askari¹,

Abolghasemi²,

Pasi³

et al. 2023

Preprint

View full text Add to dashboard Cite

In this paper we propose a novel approach for combining first-stage lexical retrieval models and Transformer-based re-rankers: we inject the relevance score of the lexical model as a token in the middle of the input of the cross-encoder re-ranker. It was shown in prior work that interpolation between the relevance score of lexical and BERT-based rerankers may not consistently result in higher effectiveness. Our idea is motivated by the finding that BERT models can capture numeric information. We compare several representations of the BM25 score and inject them as text in the input of four different cross-encoders. We additionally analyze the effect for different query types, and investigate the effectiveness of our method for capturing exact matching relevance. Evaluation on the MSMARCO Passage collection and the TREC DL collections shows that the proposed method significantly improves over all cross-encoder re-rankers as well as the common interpolation methods. We show that the improvement is consistent for all query types. We also find an improvement in exact matching capabilities over both BM25 and the cross-encoders. Our findings indicate that cross-encoder re-rankers can efficiently be improved without additional computational burden and extra steps in the pipeline by explicitly adding the output of the first-stage ranker to the model input, and this effect is robust for different models and query types.

show abstract

Section: Cross-encoder Cat (Ce Cat )mentioning

confidence: 99%

Injecting the Score of the First-stage Retriever as Text Improves BERT-Based Re-rankers

Askari,

Abolghasemi,

Pasi

et al. 2023

Preprint

View full text Add to dashboard Cite

In this paper we propose a novel approach for combining first-stage lexical retrieval models and Transformer-based re-rankers: we inject the relevance score of the lexical model as a token into the the input of the cross-encoder re-ranker. It was shown in prior work that interpolation between the relevance score of lexical and BERT-based re-rankers may not consistently result in higher effectiveness. Our idea is motivated by the finding that BERT models can capture numeric information. We compare several representations of the BM25 and Dense Passage Retrieval (DPR) scores and inject them as text in the input of four different cross-encoders. Since knowledge distillation, i.e., teacher-student training, proved to be highly effective for cross-encoder re-rankers, we additionally analyze the effect of injecting the relevance score into the student model while training the model by three larger teacher models. Evaluation on the MSMARCO Passage collection and the TREC DL collections shows that the proposed method significantly improves over all cross-encoder re-rankers as well as the common interpolation methods. We show that the improvement is consistent for all query types. We also find an improvement in exact matching capabilities over both the first-stage rankers and the cross-encoders. Our findings indicate that cross-encoder re-rankers can efficiently be improved without additional computational burden or extra steps in the pipeline by adding the output of the first-stage ranker to the model input. This effect is robust for different models and query types.

show abstract

MarkBERT: Marking Word Boundaries Improves Chinese BERT

Cited by 5 publications

References 13 publications

Injecting the score of the first-stage retriever as text improves BERT-based re-rankers

Injecting the score of the first-stage retriever as text improves BERT-based re-rankers

Injecting the BM25 Score as Text Improves BERT-Based Re-rankers

Injecting the Score of the First-stage Retriever as Text Improves BERT-Based Re-rankers

Contact Info

Product

Resources

About