paper-with-me

Papers

Diverse Pretrained Context Encodings Improve Document Translation

2021-06-07 · ACL 2021 5 · Domenic Donato, Lei Yu, Chris Dyer

We propose a new architecture for adapting a sentence-level sequence-to-sequence transformer by incorporating multiple pretrained document context signals and assess the impact on translation performance of (1) different pretraining approaches for generating these signals, (2) the quantity of parallel data for which document context is available, and (3) conditioning on source, target, or source and target contexts. Experiments on the NIST Chinese-English, and IWSLT and WMT English-German tasks support four general conclusions: that using pretrained context representations markedly improves sample efficiency, that adequate parallel data resources are crucial for learning to use document context, that jointly conditioning on multiple context representations outperforms any single representation, and that source context is more valuable for translation performance than target side context. Our best multi-context model consistently outperforms the best existing context-aware transformers.

📄 PDF Abstract BibTeX arXiv:2106.03717

Code (0)

등록된 구현이 없습니다.

Tasks

Document TranslationSentenceTranslation

Similar Papers 제목 키워드 기반

Graph-tree Fusion Model with Bidirectional Information Propagation for Long Document Classification

2024-10-03 · Sudipta Singha Roy, Xindi Wang, Robert E. Mercer, Frank Rudzicz

Long document classification presents challenges in capturing both local and global dependencies due to their extensive content and complex structure. Existing methods often struggle with token limits and fail to adequat…

Document ClassificationGraph AttentionSentence

Deep Relevance Ranking Using Enhanced Document-Query Interactions

2018-09-05 · EMNLP 2018 10 · Ryan McDonald, Georgios-Ioannis Brokos, Ion Androutsopoulos

We explore several new models for document relevance ranking, building upon the Deep Relevance Matching Model (DRMM) of Guo et al. (2016). Unlike DRMM, which uses context-insensitive encodings of terms and query-document…

Ad-Hoc Information RetrievalQuestion Answering

DIALKI: Knowledge Identification in Conversational Systems through Dialogue-Document Contextualization

2021-09-10 · EMNLP 2021 11 · Zeqiu Wu, Bo-Ru Lu, Hannaneh Hajishirzi, Mari Ostendorf

Identifying relevant knowledge to be used in conversational systems that are grounded in long documents is critical to effective response generation. We introduce a knowledge identification model that leverages the docum…

Response Generation

Perceptually Aligning Representations of Music via Noise-Augmented Autoencoders

2025-11-07 · Mathias Rose Bjare, Giorgia Cantisani, Marco Pasini, Stefan Lattner 외 arxiv

We argue that training autoencoders to reconstruct inputs from noised versions of their encodings, when combined with perceptually motivated losses, yields encodings that are structured according to a perceptual hierarch…

Task Matrices: Linear Maps for Cross-Model Finetuning Transfer

2025-12-16 · Darrin O' Brien, Dhikshith Gajulapalli, Eric Xia arxiv

Results in interpretability suggest that large vision and language models learn implicit linear encodings when models are biased by in-context prompting. However, the existence of similar linear representations in more g…