Counter-fitting Word Vectors to Linguistic Constraints

In this work, we present a novel counter-fitting method which injects antonymy and synonymy constraints into vector space representations in order to improve the vectors' capability for judging semantic similarity. Applying this method to publicly available pre-trained word vectors leads to a new state of the art performance on the SimLex-999 dataset. We also show how the method can be used to tailor the word vector space for the downstream task of dialogue state tracking, resulting in robust improvements across different dialogue domains.

• Publication year

  2016

• Venue

  North American Chapter of the Association for Computational Linguistics

• Publication date

  2016-03-02

• Fields of study

  Linguistics, Computer Science

• Identifiers
  DOI 10.18653/v1/N16-1018arXiv 1603.00892
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Learning Semantic Word Embeddings based on Ordinal Knowledge Constraints

  2015cited by this paper
• From Paraphrase Database to Compositional Paraphrase Model and Back

  2015cited by this paper
• Symmetric Pattern Based Word Embeddings for Improved Word Similarity Prediction

  2015cited by this paper
• PPDB 2.0: Better paraphrase ranking, fine-grained entailment relations, word embeddings, and style classification

  2015cited by this paper
• Multi-domain Dialog State Tracking using Recurrent Neural Networks

  2015cited by this paper
• Non-distributional Word Vector Representations

  2015cited by this paper
• Word Embedding-based Antonym Detection using Thesauri and Distributional Information

  2015cited by this paper
• Specializing Word Embeddings for Similarity or Relatedness

  2015cited by this paper
• Tailoring Continuous Word Representations for Dependency Parsing

  2014cited by this paper
• The Second Dialog State Tracking Challenge

  2014cited by this paper
• The third Dialog State Tracking Challenge

  2014cited by this paper
• Knowledge-Powered Deep Learning for Word Embedding

  2014cited by this paper
• GloVe: Global Vectors for Word Representation

  2014cited by this paper
• Robust dialog state tracking using delexicalised recurrent neural networks and unsupervised adaptation

  2014cited by this paper
• Word-Based Dialog State Tracking with Recurrent Neural Networks

  2014cited by this paper
• SimLex-999: Evaluating Semantic Models With (Genuine) Similarity Estimation

  2014cited by this paper
• Improving Lexical Embeddings with Semantic Knowledge

  2014cited by this paper
• Embedding Word Similarity with Neural Machine Translation

  2014cited by this paper
• Computing Lexical Contrast

  2013cited by this paper
• Bilingual Word Embeddings for Phrase-Based Machine Translation

  2013cited by this paper
• PPDB: The Paraphrase Database

  2013influential reference
• Distributed Representations of Words and Phrases and their Compositionality

  2013cited by this paper
• Polarity Inducing Latent Semantic Analysis

  2012cited by this paper
• Excitatory or Inhibitory: A New Semantic Orientation Extracts Contradiction and Causality from the Web

  2012cited by this paper
• A machine learning approach to textual entailment recognition

  2009cited by this paper
• A Uniform Approach to Analogies, Synonyms, Antonyms, and Associations

  2008cited by this paper
• Finding Contradictions in Text

  2008cited by this paper
• Computing Word-Pair Antonymy

  2008cited by this paper
• From distributional to semantic similarity

  2004cited by this paper
• Identifying Synonyms among Distributionally Similar Words

  2003cited by this paper
• An Unsupervised Approach to Recognizing Discourse Relations

  2002cited by this paper
• WordNet: A Lexical Database for English

  1995cited by this paper

• Beyond Cosine Similarity: Taming Semantic Drift and Antonym Intrusion in a 15-Million Node Turkish Synonym Graph

  2026cites this paper
• A Hybrid Protocol for Large-Scale Semantic Dataset Generation in Low-Resource Languages: The Turkish Semantic Relations Corpus

  2026cites this paper
• Embedding Retrofitting: Data Engineering for better RAG

  2026cites this paper
• Sentences Based Adversarial Attack on AI-Generated Text Detectors

  2026cites this paper
• One Word is Enough: Minimal Adversarial Perturbations for Neural Text Ranking

  2026cites this paper
• F2Attack: Two-Factors Scoring Method for Query-Efficient Hard-Label Black-Box Textual Adversarial Attacks

  2025cites this paper
• Semantically Improved Adversarial Attack Based on Masked Language Model via Context Preservation

  2025cites this paper
• From Benchmark to Better Embeddings: Leveraging Synonym Substitution to Enhance Multimodal Models in Ukrainian

  2025cites this paper
• Efficient Gradient-Oriented Optimization for Black-box Hard-label Textual Adversarial Attack

  2025cites this paper
• Beyond Weaponization: NLP Security for Medium and Lower-Resourced Languages in Their Own Right

  2025cites this paper
• MVGText: Momentum and Variance Guided Hard-Label Text Test Case Generation

  2025cites this paper
• Hard label adversarial attack with high query efficiency against NLP models

  2025cites this paper
• Confidence Elicitation: A New Attack Vector for Large Language Models

  2025cites this paper
• GSSA-AT: a textual adversarial defense method based on synonym substitution

  2025cites this paper
• TextGuard: Identifying and neutralizing adversarial threats in textual data

  2025cites this paper
• Semantic enrichment of neural word embeddings: Leveraging taxonomic similarity for enhanced distributional semantics

  2025influential citation
• Harnessing the Universal Geometry of Embeddings

  2025cites this paper
• SCALA: Toward Imperceptible and Efficient Black-Box Textual Adversarial Perturbations

  2025cites this paper
• HyGloadAttack: Hard-label black-box textual adversarial attacks via hybrid optimization

  2024cites this paper
• Evaluating the Validity of Word-level Adversarial Attacks with Large Language Models

  2024influential citation
• Query-Efficient Textual Adversarial Example Generation for Black-Box Attacks

  2024cites this paper
• Automated Trustworthiness Oracle Generation for Machine Learning Text Classifiers

  2024cites this paper
• SHAPAttack: Shapley-Guided Multigranularity Adversarial Attack Against Text Transformers

  2024cites this paper
• Implications of Minimum Description Length for Adversarial Attack in Natural Language Processing

  2024cites this paper
• Data augmentation and adversary attack on limit resources text classification

  2024influential citation
• On the Robustness of Generative Information Retrieval Models

  2024cites this paper
• Rethinking Word-level Adversarial Attack: The Trade-off between Efficiency, Effectiveness, and Imperceptibility

  2024cites this paper
• Overview of the CLEF-2024 CheckThat! Lab Task 6 on Robustness of Credibility Assessment with Adversarial Examples (InCrediblAE)

  2024cites this paper
• A Constraint-Enforcing Reward for Adversarial Attacks on Text Classifiers

  2024cites this paper
• Explaining the role of Intrinsic Dimensionality in Adversarial Training

  2024cites this paper
• Generating Diverse Translation with Perturbed kNN-MT

  2024cites this paper
• Towards a unified framework for imperceptible textual attacks

  2024cites this paper
• Improvements to adversarial training for text classification tasks

  2024cites this paper
• FastTextDodger: Decision-Based Adversarial Attack Against Black-Box NLP Models With Extremely High Efficiency

  2024cites this paper
• A More Context-Aware Approach for Textual Adversarial Attacks Using Probability Difference-Guided Beam Search

  2024cites this paper
• Flexible margins and multiple samples learning to enhance lexical semantic similarity

  2024cites this paper
• ROIC-DM: Robust Text Inference and Classification via Diffusion Model

  2024cites this paper
• Linguistically Informed Language Generation: A Multi-faceted Approach

  2024influential citation
• Improving Example Quality in Black Box Hard Label Text Adversarial Attacks

  2024cites this paper
• SemRoDe: Macro Adversarial Training to Learn Representations that are Robust to Word-Level Attacks

  2024cites this paper
• HQA-Attack: Toward High Quality Black-Box Hard-Label Adversarial Attack on Text

  2024cites this paper
• TextCheater: A Query-Efficient Textual Adversarial Attack in the Hard-Label Setting

  2024cites this paper
• Unitary Multi-Margin BERT for Robust Natural Language Processing

  2024cites this paper
• A Survey on Data Augmentation in Large Model Era

  2024influential citation
• Emergent Visual-Semantic Hierarchies in Image-Text Representations

  2024cites this paper
• Reversible jump attack to textual classifiers with modification reduction

  2024cites this paper
• CROWD: Certified Robustness via Weight Distribution for Smoothed Classifiers against Backdoor Attack

  2024cites this paper
• TF-Attack: Transferable and Fast Adversarial Attacks on Large Language Models

  2024cites this paper
• Synergy of Human Language Processing and Artificial Intelligence

  2024cites this paper
• Assessing biomedical knowledge robustness in large language models by query-efficient sampling attacks

  2024cites this paper
• Adversarial Attacks and Dimensionality in Text Classifiers

  2024cites this paper
• Humanizing Machine-Generated Content: Evading AI-Text Detection through Adversarial Attack

  2024cites this paper
• OpenFact at CheckThat! 2024: Combining Multiple Attack Methods for Effective Adversarial Text Generation

  2024cites this paper
• Defending Text-to-image Diffusion Models: Surprising Efficacy of Textual Perturbations Against Backdoor Attacks

  2024cites this paper
• A Fingerprint Scheme for Deep Neural Network Models Based on Adversarial Samples

  2024cites this paper
• Semantic-Preserving Adversarial Example Attack against BERT

  2024cites this paper
• Analyzing Robustness of Automatic Scientific Claim Verification Tools against Adversarial Rephrasing Attacks

  2024cites this paper
• UCTT: universal and low-cost adversarial example generation for tendency classification

  2024cites this paper
• Re-appraising the Schema Linking for Text-to-SQL

  2023cites this paper
• Robustness-Aware Word Embedding Improves Certified Robustness to Adversarial Word Substitutions

  2023cites this paper
• SSPAttack: A Simple and Sweet Paradigm for Black-Box Hard-Label Textual Adversarial Attack

  2023cites this paper
• RMLM: A Flexible Defense Framework for Proactively Mitigating Word-level Adversarial Attacks

  2023cites this paper
• Evaluating the Robustness of Deep learning Models for NLP

  2023influential citation
• SCAT: Robust Self-supervised Contrastive Learning via Adversarial Training for Text Classification

  2023cites this paper
• TextVerifier: Robustness Verification for Textual Classifiers with Certifiable Guarantees

  2023cites this paper
• Robustifying Language Models with Test-Time Adaptation

  2023cites this paper
• Novel Data Augmentation for resource constrained Image captioning

  2023cites this paper
• Can Pretrained Language Models Derive Correct Semantics from Corrupt Subwords under Noise?

  2023cites this paper
• Interpretability and Transparency-Driven Detection and Transformation of Textual Adversarial Examples (IT-DT)

  2023cites this paper
• Adversary for Social Good: Leveraging Adversarial Attacks to Protect Personal Attribute Privacy

  2023cites this paper
• Enhancing robustness of AI offensive code generators via data augmentation

  2023cites this paper
• Generation-based parallel particle swarm optimization for adversarial text attacks

  2023cites this paper
• DA^3: A Distribution-Aware Adversarial Attack against Language Models

  2023cites this paper
• On the Robustness of Generative Retrieval Models: An Out-of-Distribution Perspective

  2023cites this paper
• Do-GOOD: Towards Distribution Shift Evaluation for Pre-Trained Visual Document Understanding Models

  2023influential citation
• DARE: Towards Robust Text Explanations in Biomedical and Healthcare Applications

  2023cites this paper
• Randomized Smoothing with Masked Inference for Adversarially Robust Text Classifications

  2023cites this paper
• Lost In Translation: Generating Adversarial Examples Robust to Round-Trip Translation

  2023influential citation
• Bert is Robust! A Case Against Word Substitution-Based Adversarial Attacks

  2023cites this paper
• Sparse Black-Box Multimodal Attack for Vision-Language Adversary Generation

  2023cites this paper
• Don’t Retrain, Just Rewrite: Countering Adversarial Perturbations by Rewriting Text

  2023cites this paper
• Knowledge Distillation on Cross-Modal Adversarial Reprogramming for Data-Limited Attribute Inference

  2023cites this paper
• Robustness Tests for Automatic Machine Translation Metrics with Adversarial Attacks

  2023cites this paper
• BERT Lost Patience Won't Be Robust to Adversarial Slowdown

  2023influential citation
• The Importance of Context in the Evaluation of Word Embeddings: The Effects of Antonymy and Polysemy

  2023cites this paper
• On generalization of the sense retrofitting model

  2023cites this paper
• Large Language Models for Healthcare Data Augmentation: An Example on Patient-Trial Matching.

  2023cites this paper
• Retrofitting Light-weight Language Models for Emotions using Supervised Contrastive Learning

  2023cites this paper
• Topic-oriented Adversarial Attacks against Black-box Neural Ranking Models

  2023cites this paper
• Self-supervised Post-processing Method to Enrich Pretrained Word Vectors

  2023influential citation
• A Comparative Study of Different Dimensionality Reduction Techniques for Arabic Machine Translation

  2023cites this paper
• BufferSearch: Generating Black-Box Adversarial Texts With Lower Queries

  2023cites this paper
• Towards Effective Paraphrasing for Information Disguise

  2023cites this paper
• Learning Custom Experience Ontologies via Embedding-based Feedback Loops

  2023cites this paper
• Adversarial Robustness of Phishing Email Detection Models

  2023cites this paper
• Text-Defend: Detecting Adversarial Examples using Local Outlier Factor

  2023cites this paper
• Modeling Adversarial Attack on Pre-trained Language Models as Sequential Decision Making

  2023influential citation
• LimeAttack: Local Explainable Method for Textual Hard-Label Adversarial Attack

  2023cites this paper
• Black-Box Adversarial Attacks Against Language Model Detector

  2023influential citation
• Al-ways Bad: Retrofitting Cross-Lingual Word Embeddings to Dictionaries . Association for Computational Linguistics , 2020

  2023cites this paper