mlsanityData to 14 Oct 2026

Examining and Adapting Time for Multilingual Classification via Mixture of Temporal Experts

arXiv 2502.08825 ↗First posted 12 Feb 2025cs.CLcited by 9

Not on the map. Why?

Abstract (arXiv)

Time is implicitly embedded in classification process: classifiers are usually built on existing data while to be applied on future data whose distributions (e.g., label and token) may change. However, existing state-of-the-art classification models merely consider the temporal variations and primarily focus on English corpora, which leaves temporal studies less explored, let alone under multilingual settings. In this study, we fill the gap by treating time as domains (e.g., 2024 vs. 2025), examining temporal effects, and developing a domain adaptation framework to generalize classifiers over time on multiple languages. Our framework proposes Mixture of Temporal Experts (MoTE) to leverage both semantic and data distributional shifts to learn and adapt temporal trends into classification models. Our analysis shows classification performance varies over time across different languages, and we experimentally demonstrate that MoTE can enhance classifier generalizability over temporal data shifts. Our study provides analytic insights and addresses the need for time-aware models that perform robustly in multilingual scenarios.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Builds on 34 of its 55 references that are in the map. First posted on arXiv 12 Feb 2025.

Builds on

34 of 55 references.

References · 34

  1. 2017Attention Is All You Needcited by 60,361
  2. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,305
  3. 2019Unsupervised Cross-lingual Representation Learning at Scalecited by 3,935
  4. 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,509
  5. 2014Unsupervised Domain Adaptation by Backpropagationcited by 1,917
  6. 2017Gradient Episodic Memory for Continual Learningcited by 1,898
  7. 2020mT5: A massively multilingual pre-trained text-to-text transformercited by 1,818
  8. 2018Experience Replay for Continual Learningcited by 1,095
  9. 2021LexGLUE: A Benchmark Dataset for Legal Language Understanding in Englishcited by 310
  10. 2019Benchmarking Zero-shot Text Classification: Datasets, Evaluation and Entailment Approachcited by 308
  11. 2021Time-Aware Language Models as Temporal Knowledge Basescited by 296
  12. 2020A Brief Review of Domain Adaptationcited by 288
22 more
  1. 2023ChatGPT Beyond English: Towards a Comprehensive Evaluation of Large Language Models in Multilingual Learningcited by 206
  2. 2022mGPT: Few-Shot Learners Go Multilingualcited by 142
  3. 2024Large Language Models Can Learn Temporal Reasoningcited by 111
  4. 2023Text Classification via Large Language Modelscited by 106
  5. 2021MultiEURLEX -- A multi-lingual and multi-label legal document classification dataset for zero-shot cross-lingual transfercited by 90
  6. 2023Domain Adaptation for Time Series Under Feature and Label Shiftscited by 60
  7. 2021Temporal Adaptation of BERT and Performance on Downstream Document Classification: Insights from Social Mediacited by 51
  8. 2021Temporal Effects on Pre-trained Models for Language Processing Taskscited by 46
  9. 2023ChatGPT may Pass the Bar Exam soon, but has a Long Way to Go for the LexGLUE benchmarkcited by 33
  10. 2022Domain Adaptation for Time-Series Classification to Mitigate Covariate Shiftcited by 24
  11. 2023TEILP: Time Prediction over Knowledge Graphs via Logical Reasoningcited by 23
  12. 2021Time-Stamped Language Model: Teaching Language Models to Understand the Flow of Eventscited by 22
  13. 2021Contributions of Transformer Attention Heads in Multi- and Cross-lingual Taskscited by 16
  14. 2024ChronosLex: Time-aware Incremental Training for Temporal Generalization of Legal Classification Taskscited by 11
  15. 2022Prototypical Fine-tuning: Towards Robust Performance Under Varying Data Sizescited by 10
  16. 2023Comparison between parameter-efficient techniques and full fine-tuning: A case study on multilingual news article classificationcited by 9
  17. 2017Measuring, Predicting and Visualizing Short-Term Change in Word Representation and Usage in VKontakte Social Networkcited by 7
  18. 2024Time Matters: Examine Temporal Effects on Biomedical Language Modelscited by 7
  19. 2024Adaptive Cross-lingual Text Classification through In-Context One-Shot Demonstrationscited by 6
  20. 2022Temporal Inductive Logic Reasoning over Hypergraphscited by 4
  21. 2024Examining Imbalance Effects on Performance and Demographic Fairness of Clinical Language Modelscited by 4
  22. 2024Using Machine Translation to Augment Multilingual Classificationcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 6.

9 papers.

  1. 2026Safeguarding LLM Agents against Long-Horizon Threats via Shadow Memorycited by 7
  2. 2025Attributes as Textual Genes: Leveraging LLMs as Genetic Algorithm Simulators for Conditional Synthetic Data Generationcited by 5
  3. 2025Deliberate Planning in Language Models with Symbolic Representationcited by 3
  4. 2026Model-Agnostic Meta Learning for Class Imbalance Adaptationcited by 1
  5. 2026Knowledge-driven Augmentation and Retrieval for Integrative Temporal Adaptationcited by 1
  6. 2025MELON: Multimodal Mixture-of-Experts with Spectral-Temporal Fusion for Long-Term Mobility Estimation in Critical Carecited by 0
  7. 2026StepGap: A Hybrid NLI-LLM Checker for Step-Level Evidence-Gap Detectionin Multi-Hop Question Answeringcited by 0
  8. 2026What Makes a Medical Checker Trainable? Diagnosing Signal Collapse and Reward Hacking in Checker-Guided RAG for Biomedical QAcited by 0
  9. 2026AIA2: Attribute-Agnostic Imbalance Augmentation for Subgroup Robustnesscited by 0