paper-with-me

Papers

Enhancing Context Through Contrast

2024-01-06 · Kshitij Ambilduke, Aneesh Shetye, Diksha Bagade, Rishika Bhagwatkar, Khurshed Fitter, Prasad Vagdargi, Shital Chiddarwar

Neural machine translation benefits from semantically rich representations. Considerable progress in learning such representations has been achieved by language modelling and mutual information maximization objectives using contrastive learning. The language-dependent nature of language modelling introduces a trade-off between the universality of the learned representations and the model's performance on the language modelling tasks. Although contrastive learning improves performance, its success cannot be attributed to mutual information alone. We propose a novel Context Enhancement step to improve performance on neural machine translation by maximizing mutual information using the Barlow Twins loss. Unlike other approaches, we do not explicitly augment the data but view languages as implicit augmentations, eradicating the risk of disrupting semantic information. Further, our method does not learn embeddings from scratch and can be generalised to any set of pre-trained embeddings. Finally, we evaluate the language-agnosticism of our embeddings through language classification and use them for neural machine translation to compare with state-of-the-art approaches.

📄 PDF Abstract BibTeX arXiv:2401.03314

Code (0)

등록된 구현이 없습니다.

Tasks

Contrastive LearningLanguage ModellingMachine TranslationTranslation

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically
Barlow Twins Barlow Twins is a self-supervised learning method that applies redundancy-reduction — a principle first proposed in neuroscience — to self supervised learning. The objective…
Contrastive Learning 설명 없음

Similar Papers 제목 키워드 기반

Enhancing Input-Label Mapping in In-Context Learning with Contrastive Decoding

2025-02-19 · Keqin Peng, Liang Ding, Yuanxin Ouyang, Meng Fang 외

Large language models (LLMs) excel at a range of tasks through in-context learning (ICL), where only a few task examples guide their predictions. However, prior research highlights that LLMs often overlook input-label ma…

In-Context LearningNatural Language Understanding

Enhancing Contextual Understanding in Large Language Models through Contrastive Decoding

2024-05-04 · Zheng Zhao, Emilio Monti, Jens Lehmann, Haytham Assem

Large language models (LLMs) tend to inadequately integrate input context during text generation, relying excessively on encoded prior knowledge in model parameters, potentially resulting in generated text with factual i…

Open-Domain Question AnsweringQuestion AnsweringText Generation

Enhancing Retrieval-Augmented Generation: A Study of Best Practices

2025-01-13 · Siran Li, Linus Stenzel, Carsten Eickhoff, Seyed Ali Bahrainian

Retrieval-Augmented Generation (RAG) systems have recently shown remarkable advancements by integrating retrieval mechanisms into language models, enhancing their ability to produce more accurate and contextually relevan…

In-Context LearningRAGRetrievalRetrieval-augmented Generation+1

FusionMind -- Improving question and answering with external context fusion

2023-12-31 · Shreyas Verma, Manoj Parmar, Palash Choudhary, Sanchita Porwal

Answering questions using pre-trained language models (LMs) and knowledge graphs (KGs) presents challenges in identifying relevant knowledge and performing joint reasoning.We compared LMs (fine-tuned for the task) with t…

Knowledge GraphsMultiple-choiceQuestion Answering

HAD-Net: A Hierarchical Adversarial Knowledge Distillation Network for Improved Enhanced Tumour Segmentation Without Post-Contrast Images

2021-03-30 · Saverio Vadacchino, Raghav Mehta, Nazanin Mohammadi Sepahvand, Brennan Nichyporuk 외

Segmentation of enhancing tumours or lesions from MRI is important for detecting new disease activity in many clinical contexts. However, accurate segmentation requires the inclusion of medical images (e.g., T1 post cont…

Knowledge DistillationSegmentationUncertainty Quantification