mlsanityData to 14 Oct 2026

Weakly-Supervised Scientific Document Classification via Retrieval-Augmented Multi-Stage Training

arXiv 2306.07193 ↗First posted 12 Jun 2023cs.CL cs.IRcited by 8

On the map: Retrieval-augmented generation → Generation and context integration

Abstract (arXiv)

Scientific document classification is a critical task for a wide range of applications, but the cost of obtaining massive amounts of human-labeled data can be prohibitive. To address this challenge, we propose a weakly-supervised approach for scientific document classification using label names only. In scientific domains, label names often include domain-specific concepts that may not appear in the document corpus, making it difficult to match labels and documents precisely. To tackle this issue, we propose WANDER, which leverages dense retrieval to perform matching in the embedding space to capture the semantics of label names. We further design the label name expansion module to enrich the label name representations. Lastly, a self-training step is used to refine the predictions. The experiments on three datasets show that WANDER outperforms the best baseline by 11.9% on average. Our code will be published at https://github.com/ritaranx/wander.

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Placed in Generation and context integration, one of 15 directions in Retrieval-augmented generation. Builds on 30 of its 48 references that are in the map. First posted on arXiv 12 Jun 2023.

Builds on

30 of 48 references.

References · 30

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  2. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  3. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,824
  4. 2015Character-level Convolutional Networks for Text Classificationcited by 2,330
  5. 2019SciBERT: A Pretrained Language Model for Scientific Textcited by 1,414
  6. 2020Understanding Contrastive Representation Learning through Alignment and Uniformity on the Hyperspherecited by 1,405
  7. 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,216
  8. 2022BERTopic: Neural topic modeling with a class-based TF-IDF procedurecited by 801
  9. 2020SPECTER: Document-level Representation Learning using Citation-informed Transformerscited by 430
  10. 2021Unsupervised Corpus Aware Language Model Pre-training for Dense Passage Retrievalcited by 203
  11. 2019On the Use of ArXiv as a Datasetcited by 103
  12. 2021WRENCH: A Comprehensive Benchmark for Weak Supervisioncited by 96
18 more
  1. 2020BOND: BERT-Assisted Open-Domain Named Entity Recognition with Distant Supervisioncited by 91
  2. 2020Fine-Tuning Pre-trained Language Model with Weak Supervision: A Contrastive-Regularized Self-Training Approachcited by 83
  3. 2022FBNETGEN: Task-aware GNN-based fMRI Analysis via Functional Brain Network Generationcited by 67
  4. 2018Weakly-Supervised Neural Text Classificationcited by 66
  5. 2020X-Class: Text Classification with Extremely Weak Supervisioncited by 46
  6. 2022COCO-DR: Combating Distribution Shifts in Zero-Shot Dense Retrieval with Contrastive and Distributionally Robust Learningcited by 20
  7. 2016Query Expansion with Locally-Trained Word Embeddingscited by 17
  8. 2023The Effect of Metadata on Scientific Literature Tagging: A Cross-Field Cross-Model Studycited by 17
  9. 2023Pre-training Multi-task Contrastive Learning Models for Scientific Literature Understandingcited by 16
  10. 2023Neighborhood-Regularized Self-Training for Learning with Few Labelscited by 14
  11. 2022ReSel: N-ary Relation Extraction from Scientific Text and Tables by Learning to Retrieve and Selectcited by 10
  12. 2023PTGB: Pre-Train Graph Neural Networks for Brain Network Analysiscited by 10
  13. 2021MotifClass: Weakly Supervised Text Classification with Higher-order Metadata Informationcited by 9
  14. 2022How Can Graph Neural Networks Help Document Retrieval: A Case Study on CORD19 with Concept Map Generationcited by 7
  15. 2022Seed-Guided Topic Discovery with Out-of-Vocabulary Seedscited by 7
  16. 2023A Benchmark on Extremely Weakly Supervised Text Classification: Reconcile Seed Matching and Prompting Approachescited by 6
  17. 2022Adaptive Multi-view Rule Discovery for Weakly-Supervised Compatible Products Predictioncited by 4
  18. 2022FastClass: A Time-Efficient Approach to Weakly-Supervised Text Classificationcited by 1

Built on by

Cited by year

Citing papers per year: 2023: 3; 2024: 5.

8 papers.

  1. 2023ToolQA: A Dataset for LLM Question Answering with External Toolscited by 274
  2. 2024RAM-EHR: Retrieval Augmentation Meets Clinical Predictions on Electronic Health Recordscited by 21
  3. 2023Knowledge-Infused Prompting: Assessing and Advancing Clinical Text Data Generation with Large Language Modelscited by 20
  4. 2024ARL2: Aligning Retrievers for Black-box Large Language Models via Self-guided Adaptive Relevance Labelingcited by 10
  5. 2024HiGen: Hierarchy-Aware Sequence Generation for Hierarchical Text Classificationcited by 6
  6. 2023Open Visual Knowledge Extraction via Relation-Oriented Multimodality Model Promptingcited by 5
  7. 2024Efficient Few-shot Learning for Multi-label Classification of Scientific Documents with Many Classescited by 3
  8. 2024Artificial Intuition: Efficient Classification of Scientific Abstractscited by 2

Similar, not cited

Close in wording, no citation either way

  1. 2026Label-Centric Knowledge Retrieval for Zero-Shot Text Classification
  2. 2023Retrieval-augmented Multi-label Text Classification
  3. 2024textttMixGR: Enhancing Retriever Generalization for Scientific Domain through Complementary Granularity
  4. 2025Automatic Labelling with Open-source LLMs using Dynamic Label Schema Integration
  5. 2024DocGenome: An Open Large-scale Scientific Document Benchmark for Training and Testing Multi-modal Large Language Models
  6. 2023ATLANTIC: Structure-Aware Retrieval-Augmented Language Model for Interdisciplinary Science
  7. 2024Clustering Algorithms and RAG Enhancing Semi-Supervised Text Classification with Large LLMs
  8. 2026RAPT: Retrieval-Augmented Post-hoc Thresholding for Multi-Label Classification