DSE learns sentence embeddings from cross-attention models to speed up sentence-pair similarity computation.
problem Computing sentence-pair similarity is computationally expensive when candidate sentences are large.
method Distilled Sentence Embedding (DSE) based on knowledge distillation from cross-attentive models.
result DSE significantly outperforms ELMO variants and other sentence embedding methods, accelerating computation by several orders of magnitude.
ROTS improves sentence similarity by incorporating structural information.
problem Measuring sentence similarity with theoretical insights and structural awareness.
method Recursive Optimal Transport (ROT) framework to incorporate structural information.
result ROTS outperforms weakly supervised approaches in sentence similarity tasks.
Improves document summarization by combining word embeddings and n-grams.
problem Exact word matching fails to measure semantic similarity between sentences.
method Uses deep embedding features and tf-idf features to improve sentence similarity measure; builds an improved sentence similarity graph; employs a submodular objective function; develops a Transformer-based compression model.
result Outperforms tf-idf based approach and achieves state-of-the-art performance on DUC04 dataset.
Paper proposes continual learning for sentence encoders.
problem Optimize sentence encoders for new corpora while maintaining old corpus accuracy.
method Initialize encoders with corpus-independent features, update using Boolean operations of conceptor matrices.
result Proposed sentence encoder can continually learn features from new corpora.
This paper explores sentence vector properties for automatic summarization.
problem Understanding the internal structure and properties of sentence vectors.
method Compositional sentence vector representations using artificial neural networks.
result Cosine similarity correlates with sentence importance and can identify gaps in summaries.
Paper proposes a new method for sentence embeddings using weighted word vectors.
problem Improving sentence embeddings for natural language processing tasks.
method A simple sentence embedding method using weighted average of word vectors followed by soft projection.
result Demonstrates effectiveness on clinical semantic textual similarity task.
We train multi-task autoencoders on linguistic tasks and analyze the learned hidden sentence representations. The representations change significantly when translation and part-of-speech decoders are added. The more decoders a model employs, the better it clusters sentences according to their syntactic similarity, as t…
Method compares sentences by cosine similarity of vector projections.
problem Measuring semantic similarity of sentences.
method Cosine similarity of vector projections of sentence groups.
result Advantages over existing methods in preserving word order and syntactic connections.
SFBoW provides sentence embeddings with predefined dimensions.
problem Sentence embeddings problem at document-level.
method Refinement of Fuzzy Bag-of-Words, predefined dimension.
result Competitive performances in Semantic Textual Similarity benchmarks.
We propose a new generative model of sentences that first samples a prototype sentence from the training corpus and then edits it into a new sentence. Compared to traditional models that generate from scratch either left-to-right or by first sampling a latent sentence vector, our prototype-then-edit model improves perp…
The study examines how character and word-level representations improve sentence-level tasks.
problem Improving the quality of word and sentence representations using character-level information.
method Feature-wise sigmoid gating mechanism for combining character and word-level representations.
result Modeling characters improves final word and sentence representations, especially for less frequent words.
Deep learning models improved sentence similarity in medical records.
problem Improving sentence similarity in electronic medical records.
method Developed models using traditional machine learning and deep learning approaches, pre-trained sentence embeddings on biomedical corpora.
result Ensembled model achieved a Person correlation coefficient of 0.8528.
Method aligns multilingual news for better stock return prediction.
problem Losing nuances in multilingual news translations.
method Optimal transport for sentence alignment.
result Aligned sentences correlate better with stock returns.
This study improves sentence embeddings from BERT models.
problem Capturing the underlying meaning of sentences using BERT models.
method Comprehensive review and testing of various sentence embedding extraction and refinement methods.
result Representation-shaping techniques significantly improve sentence embeddings from BERT-based and simple baseline models.
Parallel sentences are a relatively scarce but extremely useful resource for many applications including cross-lingual retrieval and statistical machine translation. This research explores our methodology for mining such data from previously obtained comparable corpora. The task is highly practical since non-parallel m…
Improved autoencoders guide latent sentence representations for better text generation and manipulation.
problem Current autoencoders struggle to maintain coherent latent spaces for meaningful text manipulations.
method Adversarial autoencoders with a denoising objective (DAAE) to guide latent space geometry.
result DAAE provides the best trade-off between generation quality and reconstruction capacity.
New document embedding method finds more relevant tech docs.
problem Computing document similarities for technical content.
method Hybrid approach using graph techniques to extract keyphrases and score sentences.
result Proposed methods outperform baselines up to 27% in NDCG.
Optimizes LLM prompts using logged user feedback.
problem Naive approaches to optimizing LLM prompts suffer from high variance and bias.
method Kernel-based off-policy gradient method leveraging sentence similarity.
result Substantially reduces variance and suppresses bias in optimizing prompts.
Improved text generation using transferable rewards from related tasks.
problem Non-differentiable task-specific scores limit the use of policy gradient methods in text generation.
method Transferable Reward Learner that uses model-based rewards for sentence-level and phrase-level similarity.
result Improved performance on semantic evaluation measures in image captioning tasks.
WME generates document embeddings from word embeddings, outperforming state-of-the-art techniques.
problem Lack of unsupervised document embeddings from pre-trained word embeddings.
method Word Mover's Embedding (WME) approach.
result WME consistently matches or outperforms state-of-the-art techniques on various text classification and similarity tasks.
Generative model uses captions to generate images, improving semantic understanding.
problem Complex image generation models require large datasets and intricate learning.
method Adapts captioning models to generate images, using learned sentence and frame vectors.
result Images generated from multiple captions better capture semantic meaning.
CURE extracts relations without supervision by clustering similar entity pairs.
problem Extracting relations unsupervised without considering sentence correlations.
method CURE uses Encoder-Decoder architecture for self-supervised learning and clustering similar relations.
result CURE outperforms state-of-the-art models on NYT and UNPC datasets.
New approach detects fake news stance using deep learning and similarity features.
problem Detecting fake news through stance detection in news articles.
method Combines deep neural representations with string similarity features for headline and article analysis.
result Model outperforms previous state-of-the-art, especially with pre-training and combined representations.
Estimating the difficulty level of math word problems is an important task for many educational applications. Identification of relevant and irrelevant sentences in math word problems is an important step for calculating the difficulty levels of such problems. This paper addresses a novel application of text categoriza…
New metrics predict human sentence comprehension across languages.
problem Predicting human sentence comprehension using computational models.
method Developed sentence-level metrics using multilingual large language models.
result Achieved high accuracy in predicting human sentence reading speeds.
Existing models based on artificial neural networks (ANNs) for sentence classification often do not incorporate the context in which sentences appear, and classify sentences individually. However, traditional sentence classification approaches have been shown to greatly benefit from jointly classifying subsequent sente…
Paper proposes a novel model to improve n-ary cross-sentence relation extraction by addressing noisy data and non-consecutive sentences.
problem Noisy labeled data and non-consecutive sentences in n-ary cross-sentence relation extraction.
method Two-level agent reinforcement learning model and hybrid attention mechanism/PCNN approach.
result The model reduces the impact of noisy data and achieves better performance.
Prob2Vec embeds problems for adaptive tutoring, achieving high similarity accuracy.
problem Retrieve problems with similar mathematical concepts for adaptive tutoring.
method Hierarchical problem embedding algorithm (Prob2Vec) combining abstraction and embedding steps.
result 96.88% accuracy on problem similarity test, significantly outperforming state-of-the-art sentence embedding methods.
CGMH uses Metropolis-Hastings sampling to generate sentences with complex constraints.
problem Generating sentences with specific constraints while maintaining fluency and naturalness.
method CGMH employs Metropolis-Hastings sampling for constrained sentence generation.
result CGMH outperforms previous methods in various constrained sentence generation tasks.
New method improves sentence classification using context information.
problem Classifying sentences with limited context information.
method Context-LSTM-CNN method that considers large contexts and long-range dependencies.
result Consistently improves over previous methods on two datasets.
BERTopic improves financial text analysis with FinTextSim's contextual embeddings.
problem Analyzing financial text data for insights and predictions.
method Integrates BERTopic with FinTextSim for topic modeling and clustering.
result BERTopic performs better with FinTextSim's embeddings, improving topic clarity and reducing misclassification.
Framework quantifies semantic similarity between groups of embeddings.
problem Quantifying semantic similarity between groups of embeddings.
method Formulates model comparison task, contrasts generative models, uses information criteria.
result Achieves competitive results in Semantic Textual Similarity tasks.
Determinantal point processes (DPPs), which arise in random matrix theory and quantum physics, are natural models for subset selection problems where diversity is preferred. Among many remarkable properties, DPPs offer tractable algorithms for exact inference, including computing marginal probabilities and sampling; ho…
Framework for unsupervised entity representations using Optimal Transport.
problem Building representations of entities and their contexts.
method Optimal Transport, Wasserstein distance, barycenters.
result Significant gains in sentence similarity tasks (4.1% relative improvement).
Unsupervised scheme ranks sentences in text documents based on semantic importance.
problem Ranking sentences in text documents without labeled data.
method Extracts essential words and phrases, constructs semantic phrase and sentence graphs, applies PageRank, combines scores, and optimizes for topic diversity.
result SSR outperforms individual judges and compares favorably with combined rankings on benchmarks.
In this paper we explore the "vector semantics" problem from the perspective of "almost orthogonal" property of high-dimensional random vectors. We show that this intriguing property can be used to "memorize" random vectors by simply adding them, and we provide an efficient probabilistic solution to the set membership …
Improved text summarization using neural semantic encoders with hierarchical structure.
problem Capturing long-term dependencies in text summarization.
method Proposed a novel hierarchical Neural Semantic Encoder (NSE) model augmented with lemma and PoS tags.
result Significantly outperformed state-of-the-art models in ROUGE metric.
This study identifies sentence relationships in legal transcripts.
problem Improving understanding of legal case proceedings through sentence relationships.
method Combining machine learning and rule-based approach to classify sentence relationships.
result First study to use discourse relationships for legal court case transcripts.
PAC-Bayes analysis explains sentence vector learning from unlabeled data.
problem Understanding and improving sentence vector learning from unlabeled data.
method PAC-Bayes bound analysis for transfer learning.
result Simple heuristics and new algorithms derived from PAC-Bayes analysis.
New model improves image captioning's ability to describe unseen concepts.
problem Image captioning models struggle with describing unseen combinations of concepts.
method Proposes a multi-task model combining caption generation and image-sentence ranking, with a decoding mechanism to re-rank captions based on image similarity.
result The model significantly outperforms state-of-the-art models in compositional generalization.
Paper tackles noisy relation classification by sentence-level reinforcement learning.
problem Noisy distant supervision in relation classification.
method Two-module approach: instance selector using reinforcement learning, relation classifier making sentence-level predictions.
result Jointly trained model optimizes instance selection and relation classification, effectively handling noisy data.
BERT improved for propaganda detection with imbalanced, dissimilar data.
problem BERT struggles with dissimilar imbalanced datasets in propaganda detection.
method Cost-sensitive BERT with dissimilarity measure for imbalanced, dissimilar datasets.
result Achieved second-highest score on sentence-level propaganda classification.
NeuralWarp aligns time-series indices using deep learning for improved similarity.
problem Need for elastic time-series similarity measures.
method Proposes NeuralWarp, a deep learning model for aligning time-series indices.
result NeuralWarp outperforms non-parametric and un-warped deep models.
Bayesian EnKF improves sentence comprehension uncertainty modeling.
problem Uncertainty in human language comprehension, especially with ambiguous inputs.
method Bayesian framework using ensemble Kalman filter (EnKF) for uncertainty quantification.
result Enhanced model's ability to approximate human cognitive processing with linguistic ambiguities.
A graph model improves short text classification by integrating sentence relationships.
problem Sparse features in short text classification due to limited text length.
method PathWalk model combining graph networks and short sentences.
result PathWalk achieves state-of-the-art results on four datasets.
Framework learns sentence order from paragraphs using attention and transformer networks.
problem Learning to order sentences from a paragraph.
method Bidirectional sentence encoder and self-attention transformer network for ranking.
result Framework outperforms state-of-the-art methods on sentence ordering and discrimination tasks.
Method selects task-specific neurons for unsupervised transfer learning.
problem Challenges in fine-tuning large language models for specific tasks.
method Selects important neurons for a specific classification task and extends to multi-source transfer learning.
result Higher similarity between task-specific fingerprints leads to better transferability.
Paper tackles counterfactual sentence detection and evaluation.
problem Detect and evaluate counterfactual sentences in natural language.
method Used a BERT base model for classification and a hybrid BERT Multi-Layer Perceptron for sequence identification. Introduced cascaded linear inputs to improve performance.
result Achieved an F1 score of 85.00% in Task 1 and 83.90% in Task 2.