paper-with-me

홈 › Papers

On the Prunability of Attention Heads in Multilingual BERT

2021-09-26 · Aakriti Budhraja, Madhura Pande, Pratyush Kumar, Mitesh M. Khapra

Large multilingual models, such as mBERT, have shown promise in crosslingual transfer. In this work, we employ pruning to quantify the robustness and interpret layer-wise importance of mBERT. On four GLUE tasks, the relative drops in accuracy due to pruning have almost identical results on mBERT and BERT suggesting that the reduced attention capacity of the multilingual models does not affect robustness to pruning. For the crosslingual task XNLI, we report higher drops in accuracy with pruning indicating lower robustness in crosslingual transfer. Also, the importance of the encoder layers sensitively depends on the language family and the pre-training corpus size. The top layers, which are relatively more influenced by fine-tuning, encode important information for languages similar to English (SVO) while the bottom layers, which are relatively less influenced by fine-tuning, are particularly important for agglutinative and low-resource languages.

📄 PDF Abstract BibTeX arXiv:2109.12683

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Refunds@Expedia|||How do I get a full refund from Expedia? “How do I get a full refund from Expedia? How do I get a full refund from Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Quick Help &…
Attention 설명 없음
Pruning 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
mBERT mBERT
Attention Dropout Attention Dropout is a type of dropout used in attention-based architectures, where elements are randomly dropped out of the…
Weight Decay 설명 없음
Linear Warmup With Linear Decay Linear Warmup With Linear Decay is a learning rate schedule in which we increase the learning rate linearly for $n$ updates and then linearly decay afterwards.

Similar Papers 제목 키워드 기반

On the weak link between importance and prunability of attention heads

2020-11-01 · EMNLP 2020 11 · Aakriti Budhraja, Madhura Pande, Preksha Nema, Pratyush Kumar 외

Given the success of Transformer-based models, two directions of study have emerged: interpreting role of individual attention heads and down-sizing the models for efficiency. Our work straddles these two streams: We ana…

Why Can You Lay Off Heads? Investigating How BERT Heads Transfer

2021-06-14 · Ting-Rui Chiang, Yun-Nung Chen

The huge size of the widely used BERT family models has led to recent efforts about model distillation. The main goal of distillation is to create a task-agnostic pre-trained model that can be fine-tuned on downstream ta…

Transfer Learning

Bridging Latent Reasoning and Target-Language Generation via Retrieval-Transition Heads

2026-02-25 · Shaswat Patel, Vishvesh Trivedi, Yue Han, Yihuai Hong 외 arxiv

Recent work has identified a subset of attention heads in Transformer as retrieval heads, which are responsible for retrieving information from the context. In this work, we first investigate retrieval heads in multiling…

Attention Can Reflect Syntactic Structure (If You Let It)

2021-01-26 · EACL 2021 2 · Vinit Ravishankar, Artur Kulmizev, Mostafa Abdou, Anders Søgaard 외

Since the popularization of the Transformer as a general-purpose feature encoder for NLP, many studies have attempted to decode linguistic structure from its novel multi-head attention mechanism. However, much of such wo…

Do Multilingual Neural Machine Translation Models Contain Language Pair Specific Attention Heads?

2021-05-31 · Findings (ACL) 2021 8 · Zae Myung Kim, Laurent Besacier, Vassilina Nikoulina, Didier Schwab

Recent studies on the analysis of the multilingual representations focus on identifying whether there is an emergence of language-independent representations, or whether a multilingual model partitions its weights among …

DecoderMachine TranslationNMTTranslation