Challenge: Character sets change over time and character frequency distributions shift across historical periods or regions . character distribution alignment can improve existing models at inference time without requiring retraining .
Approach: They propose a loss function that incorporates the Wasserstein distance between predicted and target distributions.
Outcome: The proposed method improves accuracy and robustness under temporal and contextual shifts.

Similar Papers

Deep Generative Model for Joint Alignment and Word Representation (N18-1)

Copied to clipboard

Challenge: EmbedAlign model embeds words in their complete observed context and learns by marginalisation of latent lexical alignments.
Approach: They exploit translation as a distributional context and embed words as posterior probability densities, rather than point estimates, which allows them to compare words in context using a measure of overlap between distributions.
Outcome: The proposed model performs on a range of lexical semantics tasks and achieves competitive results on benchmarks including natural language inference, paraphrasing, and text similarity.
Text Classification Under Class Distribution Shift: A Survey (2026.eacl-long)

Copied to clipboard

Challenge: ML models assume that training and test data are sampled from the same distribution, but in daily practice, this assumption is often broken.
Approach: They survey articles studying open-set text classification to understand the distribution shifts and mitigation approaches for each problem setup.
Outcome: The proposed methods can solve problems caused by the shifting class distribution in open-set text classification and related tasks.
Looking Beyond Label Noise: Shifted Label Distribution Matters in Distantly Supervised Relation Extraction (D19-1)

Copied to clipboard

Challenge: Existing studies on DS-based relation extraction (RE) methods focus on handling label noise, but other factors may have been overlooked.
Approach: They propose a method to automatically adjust DS-RE models to a shifted label distribution problem . they find this problem exists in real-world DS datasets and can be overcome .
Outcome: The proposed method achieves consistent performance gains on DS-trained models with an up to 23% relative F1 improvement, which verifies their assumptions.
Neural Temporality Adaptation for Document Classification: Diachronic Word Embeddings and Domain Adaptation Models (P19-1)

Copied to clipboard

Challenge: Recent studies show that document classifiers can become more stable over time when trained in ways that account for temporal variations.
Approach: They propose a method for embedding diachronic word embedds into document classification models . they propose 'time-driven neural classification model' that accounts for temporal variations .
Outcome: The proposed model can be trained on six corpora and make it more robust over time.
Diachronic word embeddings and semantic shifts: a survey (C18-1)

Copied to clipboard

Challenge: Existing methods for tracing time-related semantic shifts with word embedding models lack the cohesion, common terminology and shared practices of more established areas of natural language processing.
Approach: They propose several axes along which these methods can be compared and propose a framework for comparison.
Outcome: The proposed methods are compared with existing methods and outline their main challenges and potential applications.
Understanding Cross-Lingual Alignment—A Survey (2024.findings-acl)

Copied to clipboard

Challenge: Cross-lingual alignment is the meaningful similarity of representations across languages in multilingual language models.
Approach: They propose a taxonomy of methods to improve cross-lingual alignment . they argue that an effective trade-off between language-neutral and language-specific information is key .
Outcome: The proposed methods can be applied to encoder models and encoder-decoder-only models . they show that language-neutral and language-specific information is key .
Letters From the Past: Modeling Historical Sound Change Through Diachronic Character Embeddings (2022.acl-long)

Copied to clipboard

Challenge: a great deal of work has been done on NLP approaches to lexical semantic change detection, but other aspects of language change have received less attention from the NLP community.
Approach: They propose to compare the relative distance through time between the distributions of the characters involved before and after a sound change has taken place.
Outcome: The proposed method can trace the well-known historical change of lenition of plosives in Danish historical sources and identify several of the changes under consideration and uncover meaningful contexts in which they appeared.
Function Words as Statistical Cues for Language Learning (2026.acl-long)

Copied to clipboard

Challenge: Existing studies have argued that function words aid learning abstract grammatical knowledge from linear input.
Approach: They examine the statistical distribution of function words and their properties . they show that function words are reliable, diverse, and informative .
Outcome: The results show that function words preserve high frequency, reliable syntactic association, phrase-boundary alignment and are informative to structural dependency.
Predict the Future from the Past? On the Temporal Data Distribution Shift in Financial Sentiment Classifications (2023.emnlp-main)

Copied to clipboard

Challenge: Existing methods for financial sentiment analysis use random splits of a dataset into training and testing to ensure there is no distribution shift between training and deployment.
Approach: They propose a method that combines out-of-distribution detection with time series modeling for temporal financial sentiment analysis.
Outcome: The proposed method improves the model’s ability to adapt to evolving temporal shifts in a volatile financial market.
WSPAlign: Word Alignment Pre-training via Large-Scale Weakly Supervised Span Prediction (2023.acl-long)

Copied to clipboard

Challenge: Existing word alignment methods rely on manual data and lack generalization ability.
Approach: They propose to use a weakly-supervised large-scale weakly supervised dataset for word alignment pre-training via span prediction to reduce the need for manual data.
Outcome: The proposed method improves upon the best supervised baseline by 3.3 6.1 points in F1 and 1.5 6.1 point in AER.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations