paper-with-me

홈 › Papers

Contrastive Conditioning for Assessing Disambiguation in MT: A Case Study of Distilled Bias

2021-05-01 · ACL ARR May 2021 5 · Jannis Vamvas, Rico Sennrich

Lexical disambiguation is a major challenge for machine translation systems, especially if some senses of a word are trained less often than others. Identifying patterns of overgeneralization requires evaluation methods that are both reliable and scalable. We propose contrastive conditioning as a reference-free black-box method for detecting disambiguation errors. Specifically, we score the quality of a translation by conditioning on variants of the source that provide contrastive disambiguation cues. After validating our method, we apply it in a case study to perform a targeted evaluation of sequence-level knowledge distillation. By probing word sense disambiguation and translation of gendered occupation names, we show that distillation-trained models tend to overgeneralize more than other models with a comparable BLEU score. Contrastive conditioning thus highlights a side effect of distillation that is not fully captured by standard evaluation metrics. Code and data to reproduce our findings are publicly available.

📄 PDF Abstract BibTeX

Code (1)

zurichnlp/contrastive-conditioning 공식 구현 pytorch

Tasks

Knowledge DistillationMachine TranslationTranslationWord Sense Disambiguation

Similar Papers 제목 키워드 기반

Contrastive Entity Coreference and Disambiguation for Historical Texts

2024-06-21 · Abhishek Arora, Emily Silcock, Leander Heldring, Melissa Dell

Massive-scale historical document collections are crucial for social science research. Despite increasing digitization, these documents typically lack unique cross-document identifiers for individuals mentioned within th…

Articlescoreference-resolutionCoreference ResolutionCross Document Coreference Resolution+1

EMODIS: A Benchmark for Context-Dependent Emoji Disambiguation in Large Language Models

2025-11-10 · Jiacheng Huang, Ning Yu, Xiaoyin Yi arxiv

Large language models (LLMs) are increasingly deployed in real-world communication settings, yet their ability to resolve context-dependent ambiguity remains underexplored. In this work, we present EMODIS, a new benchmar…

Contrastive Self-Supervised Learning for Commonsense Reasoning

2020-05-02 · ACL 2020 6 · Tassilo Klein, Moin Nabi

We propose a self-supervised method to solve Pronoun Disambiguation and Winograd Schema Challenge problems. Our approach exploits the characteristic structure of training corpora related to so-called "trigger" words, whi…

Self-Supervised Learning

SimCLAD: A Simple Framework for Contrastive Learning of Acronym Disambiguation

2021-11-29 · Bin Li, Fei Xia, Yixuan Weng, Xiusheng Huang 외

Acronym disambiguation means finding the correct meaning of an ambiguous acronym from the dictionary in a given sentence, which is one of the key points for scientific document understanding (SDU@AAAI-22). Recently, many…

Contrastive Learningdocument understandingSentence

PiCO+: Contrastive Label Disambiguation for Robust Partial Label Learning

2022-01-22 · Haobo Wang, Ruixuan Xiao, Yixuan Li, Lei Feng 외

Partial label learning (PLL) is an important problem that allows each training example to be labeled with a coarse candidate set, which well suits many real-world data annotation scenarios with label ambiguity. Despite t…

Contrastive LearningPartial Label LearningPICORepresentation Learning