Papers with contextual embeddings
ImaginE: An Imagination-Based Automatic Evaluation Metric for Natural Language Generation (2023.findings-eacl)
Copied to clipboard
| Challenge: | Existing evaluation methods for natural language generation rely on token-level or embedding-level comparisons with text references. |
| Approach: | They propose to use text-to-image generator to generate an image as the embodied imagination for the text snippet and compute the imagination similarity using contextual embeddings. |
| Outcome: | The proposed metric improves existing evaluation metrics’ correlations with human similarity judgments in both reference-based and reference-free scenarios. |
Jeff Da at COIN - Shared Task: BIG MOOD: Relating Transformers to Explicit Commonsense Knowledge (D19-60)
Copied to clipboard
| Challenge: | Recent studies show that large-scale pre-training models can be effective for large datasets. |
| Approach: | They propose a method of integrating contextual embeddings with commonsense graph embeddINGs by preprocessing knowledge bases and aligning tokens between misaligned tokenization methods. |
| Outcome: | The proposed method achieves higher accuracy than BERT and scores highest without pretraining. |
Detecting Primary Progressive Aphasia (PPA) from Text: A Benchmarking Study (2026.findings-eacl)
Copied to clipboard
Ghofrane Merhbene, Fabian Lecron, Philippe Fortemps, Bradford C. Dickerson, Mascha Kurpicz-Briki, Neguine Rezaii
| Challenge: | Primary progressive aphasia (PPA) is a neurodegenerative disorder characterized by progressive language deficits as the primary symptom. |
| Approach: | They benchmarked the performance of traditional machine learning models with various feature extraction techniques, transformer-based models, and large language models (LLMs) they found that transformer-Based models exceeded chance-level performance in terms of balanced accuracy, while MLP using MentalBert’s embeddings achieved the highest accuracy. |
| Outcome: | The proposed models outperform chance-level models in terms of balanced accuracy while using MentalBert’s embeddings achieve the highest accuracy. |
Substitution-based Semantic Change Detection using Contextual Embeddings (2023.acl-short)
Copied to clipboard
| Challenge: | a simplified approach to measuring semantic change using contextual embeddings is proposed . the static word vectors used for measuring semantic changes are difficult to interpret . |
| Approach: | They propose a simplified approach to measuring semantic change using contextual embeddings . they use the Jensen-Shannon Divergence between the distributions of most probable replacements for masked words in different time periods to measure semantic change. |
| Outcome: | The proposed approach is interpretable, efficient and much more efficient than static embeddings. |
Pre-training is a Hot Topic: Contextualized Document Embeddings Improve Topic Coherence (2021.acl-short)
Copied to clipboard
| Challenge: | Recent neural topic models extract words from documents, but they are not coherent . coherence is crucial for topic models, but many use bag-of-words document representations as input . pre-trained language models are becoming ubiquitous in natural language processing . |
| Approach: | They combine contextualized representations with neural topic models to produce more coherent topics . they say that future improvements in language models will translate into better topic models . |
| Outcome: | The proposed approach produces more meaningful and coherent topics than bag-of-words models and recent neural models. |
BERTTune: Fine-Tuning Neural Machine Translation with BERTScore (2021.acl-short)
Copied to clipboard
| Challenge: | Neural machine translation models are biased toward limited translation references . BERTScore is a scoring function based on contextual embeddings that overcomes the limitations of n-gram-based metrics. |
| Approach: | They propose to fine-tune models with a new evaluation metric based on contextual embeddings to overcome the limitations of n-gram-based metrics. |
| Outcome: | The proposed training objective improves translations that are different from the translations but close in the contextual embedding space. |
Cross-Lingual Alignment of Contextual Word Embeddings, with Applications to Zero-shot Dependency Parsing (N19-1)
Copied to clipboard
| Challenge: | Existing methods for multilingual transfer are limited by their dynamic nature. |
| Approach: | They propose a method that utilizes deep contextual embeddings, pretrained in an unsupervised fashion. |
| Outcome: | The proposed method outperforms the state-of-the-art on 6 languages, yielding an improvement of 6.8 LAS points on average. |
Event-Driven News Stream Clustering using Entity-Aware Contextual Embeddings (2021.eacl-main)
Copied to clipboard
| Challenge: | a novel method for online news stream clustering is proposed . a user can scour the many news sources multiple times a day to find news articles . |
| Approach: | They propose a method for online news stream clustering that is a variant of the streaming K-means algorithm. |
| Outcome: | The proposed model achieves state-of-the-art on a standard stream clustering dataset of English documents. |
Evaluation of Sentence Representations in Polish (2020.lrec-1)
Copied to clipboard
| Challenge: | Existing methods for learning sentence representations have been limited in low-resource languages such as Polish . |
| Approach: | They propose two new Polish datasets for evaluating sentence embeddings and evaluate eight different methods including Polish and multilingual models. |
| Outcome: | The proposed methods show strengths and weaknesses in Polish and multilingual models. |
Reinforcement Learning for Topic Models (2023.findings-acl)
Copied to clipboard
| Challenge: | Topic modeling is a method to extract information from documents by grouping topics into topics and linking them with words describing them. |
| Approach: | They propose to replace the variational autoencoder with a continuous action space reinforcement learning policy and modify the neural network architecture to weight the ELBO loss. |
| Outcome: | The proposed model outperforms all other unsupervised models and performs on par with or better than most models using supervised labeling and contrastive learning. |
The Devil is in the Details: Evaluating Limitations of Transformer-based Methods for Granular Tasks (2020.coling-main)
Copied to clipboard
| Challenge: | Contextual embeddings have shown state-of-the-art performance for various tasks such as question answering, sentiment analysis, and textual similarity. |
| Approach: | They propose to integrate transformer-based neural language models into their models to achieve relative improvements of up to 36% on granular tasks. |
| Outcome: | The proposed model outperforms baselines for more granular tasks while outperforming TF-IDF for more complex tasks. |
Attention weights accurately predict language representations in the brain (2022.findings-emnlp)
Copied to clipboard
| Challenge: | In Transformer-based language models, the attention mechanism converts token embeddings into contextual embeddables that incorporate information from neighboring words. |
| Approach: | They analyze fMRI recordings of English language learners and extract attention weights from them to determine how well they can predict brain responses. |
| Outcome: | The resulting hidden state embeddings are more accurate than lexical embeddngs or RNN-based models. |
Named Entity Recognition through Deep Representation Learning and Weak Supervision (2021.findings-acl)
Copied to clipboard
| Challenge: | Weakly supervised named entity recognition (NER) uses noisy labels to estimate the true labels of a dataset. |
| Approach: | They propose a model to learn optimal assignments of latent NER tags using observed tokens and weak labels provided by labeling functions. |
| Outcome: | The proposed model improves the quality of weak labels on four public datasets. |
More Embeddings, Better Sequence Labelers? (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Existing work suggests contextual embeddings improve sequence labeling accuracy . but, there is no definite conclusion on whether concatenating different kinds of embeddables is effective . |
| Approach: | They propose a family of contextual embeddings that improves sequence labeling accuracy . they conduct extensive experiments on 3 tasks over 18 datasets and 8 languages . |
| Outcome: | The proposed family of contextual embeddings improves the accuracy of sequence labelers over non-contextual embedders. |
Examining Cross-lingual Contextual Embeddings with Orthogonal Structural Probes (2021.emnlp-main)
Copied to clipboard
| Challenge: | Existing studies on whether multilingual embeddings can be aligned in a shared space across languages are lacking. |
| Approach: | They propose to learn a projection based on monolingual annotated datasets and evaluate syntactic and lexical information encoded in a shared cross-lingual embedding space. |
| Outcome: | The proposed model can be used to learn representations for languages with low resources. |
Towards Multi-Sense Cross-Lingual Alignment of Contextual Embeddings (2022.coling-1)
Copied to clipboard
| Challenge: | Existing approaches to learn cross-lingual word embeddings are sense agnostic . a novel framework to align contextual embeddables at the sense level is proposed . |
| Approach: | They propose a framework to align contextual embeddings at the sense level by leveraging cross-lingual signal from bilingual dictionaries only. |
| Outcome: | The proposed framework improves word sense disambiguation tasks by leveraging bilingual dictionaries . compared with baseline results, the proposed models achieve 0.52%, 2.09% and 1.29% performance improvements . |
Predicting Long-Term Citations from Short-Term Linguistic Influence (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods to quantify linguistic influence in timestamped documents are not informative about extent to which a paper affected subsequent publications. |
| Approach: | They propose to quantify linguistic influence in timestamped document collections by estimating a Hawkes process with a low-rank parameter matrix and identify lexical and semantic changes using contextual embeddings and word frequencies. |
| Outcome: | The proposed method is based on an online evaluation with incremental temporal training/test splits, in comparison with a strong baseline that includes predictors for initial citation counts, topics, and lexical features. |
Automatic Evaluation of Language Generation Technology Based on Structure Alignment (2025.coling-main)
Copied to clipboard
| Challenge: | Existing methods for automatic evaluation ignore syntax of sentences despite its importance in determining meaning. |
| Approach: | They propose an automatic evaluation metric that considers both the words in sentences and their syntactic structures. |
| Outcome: | The proposed method is comparable to baselines from two NLP tasks. |
Neural Architectures for Nested NER through Linearization (P19-1)
Copied to clipboard
| Challenge: | a nested named entity recognition (NER) is a set of entities that can overlap and be labeled with more than one label. |
| Approach: | They propose two neural network architectures for nested named entity recognition . they propose to model nesting entities as multilabels and predict a sequence-to-sequence problem . |
| Outcome: | The proposed methods outperform the state-of-the-art on four corpora . the proposed models also improve on the recently published contextual embeddings . |
High Quality ELMo Embeddings for Seven Less-Resourced Languages (2020.lrec-1)
Copied to clipboard
| Challenge: | Recent results show that deep neural networks using contextual embeddings outperform non-contextual embedders on a majority of text classification tasks. |
| Approach: | They propose to use contextual embeddings for seven languages to train new embeddables . they also show that existing embeddibles for listed languages shall be improved . |
| Outcome: | The proposed embeddings outperform non-contextual embeddables on a majority of text classification tasks. |
Merge and Label: A Novel Neural Network Architecture for Nested NER (P19-1)
Copied to clipboard
| Challenge: | Named entity recognition (NER) is one of the best studied tasks in natural language processing. |
| Approach: | They propose a neural network architecture that merges tokens and/or entities into nested entities and labels them independently. |
| Outcome: | The proposed approach achieves state-of-the-art F1 of 74.6 and improves with contextual embeddings to 82.4. |
Grounding Characters and Places in Narrative Text (2023.acl-long)
Copied to clipboard
| Challenge: | Prior work has analyzed characters and locations from text independently without grounding characters to their locations in narrative time. |
| Approach: | They propose a task to assign a spatial relationship category for every character and location co-mention within a window of text, taking into account linguistic context, narrative tense, and temporal scope. |
| Outcome: | The proposed model allows to test hypotheses on mobility and domestic space . women as characters tend to occupy more interior space than men, the model shows . |
Contextualized Embeddings based Transformer Encoder for Sentence Similarity Modeling in Answer Selection Task (2020.lrec-1)
Copied to clipboard
| Challenge: | Word embeddings that consider context have attracted great attention for natural language processing tasks in recent years. |
| Approach: | They propose two different approaches to integrate contextualized word embeddings with transformer encoders for sentence similarity modeling. |
| Outcome: | The proposed model outperforms the feature-based approach on six datasets. |
Multi-Sense Embeddings for Language Models and Knowledge Distillation (2025.findings-acl)
Copied to clipboard
| Challenge: | Transformer-based large language models generate different representations for the same token depending on context . however, words and tokens typically have only a limited number of senses . a knowledge distillation method can be used to learn a smaller student model . |
| Approach: | They propose a multi-sense embedding method that uses a clustering algorithm to generate a sense embeddable dictionary. |
| Outcome: | The proposed method offers significant space and inference time savings while maintaining competitive performance. |