Papers with paraphrase detection
Recovering Lexically and Semantically Reused Texts (2021.starsem-1)
Copied to clipboard
| Challenge: | Writers often repurpose material from existing texts when composing new documents. |
| Approach: | They propose to use local text reuse detection to detect localized regions of lexically or semantically similar text embedded in otherwise unrelated material. |
| Outcome: | The proposed methods perform better on three LTRD tasks, detecting plagiarism, modeling journalists’ use of press releases, and identifying scientists’ citation of earlier papers. |
Calibration of Pre-trained Transformers (2020.emnlp-main)
Copied to clipboard
| Challenge: | Pre-trained Transformers dominate benchmark tasks but use a large number of self-attention heads across many layers in a way that is difficult to unpack. |
| Approach: | They analyze pre-trained Transformer models' posterior probabilities to determine whether they are calibrated for three tasks: natural language inference, paraphrase detection, and commonsense reasoning. |
| Outcome: | The models are calibrated in-domain and out-of-domain, and their calibration error out-domain can be as much as 3.5x lower. |
Neural Relation Extraction for Knowledge Base Enrichment (P19-1)
Copied to clipboard
| Challenge: | Existing studies focus on the extraction itself and rely on Named Entity Disambiguation (NED) to map triples into knowledge base (KB) enrichment. |
| Approach: | They propose an end-to-end relation extraction model for knowledge base enrichment based on a neural encoder-decoder model . they propose to extract entities and their relationships from sentences in the form of triples and map the elements of the extracted triples to an existing KB in an end to end manner. |
| Outcome: | The proposed model outperforms state-of-the-art baselines by 15.51% and 8.38% on two real-world datasets. |
Semi-supervised Adversarial Text Generation based on Seq2Seq models (2022.emnlp-industry)
Copied to clipboard
| Challenge: | In contrast, adversarial training has been used in computer vision to improve models’ robustness due to the discrete nature of text. |
| Approach: | They propose a way to generate adversarial samples by using pseudo-labeled in-domain text data to train a seq2seq model for adversarials and combine it with paraphrase detection. |
| Outcome: | The proposed model generates realistic and relevant adversarial samples compared to other state-of-the-art models and recovers up to 70% of errors. |
Abstract Meaning Representation for Paraphrase Detection (N18-1)
Copied to clipboard
| Challenge: | Abstract Meaning Representation (AMR) parsing is ideal for paraphrase detection . it abstracts away from the syntactic realization of a sentence, and denotes only its meaning in a canonical form. |
| Approach: | They propose a technique that uses latent semantic analysis to translate sentences into AMR graphs . they show that the technique can be used to detect whether two sentences have the same meaning . |
| Outcome: | The proposed technique significantly advances state-of-the-art paraphrase detection for the Microsoft Research Paraphrase Corpus. |
GenDLN: Evolutionary Algorithm-Based Stacked LLM Framework for Joint Prompt Optimization (2025.acl-srw)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) are increasingly replacing traditional classification and inference models due to their generality, ability to perform a wide range of tasks, and seemingly advanced "reasoning" prompt optimization is a promising alternative to manual/human prompt engineering, but the cost of using LLMs for prompt optimization via commercial APIs remains high. |
| Approach: | They propose an open-source, efficient genetic algorithm-based prompt pair optimization framework that leverages commercial APIs. |
| Outcome: | The proposed approach allows teams with limited resources to efficiently use commercial LLMs for prompt optimization. |
Exploring Multitask Learning for Low-Resource Abstractive Summarization (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Recent work shows that training text encoders using data from multiple tasks helps to produce an encoder that can be used in numerous downstream tasks with minimal fine-tuning. |
| Approach: | They incorporate four different tasks to improve abstractive summarization performance . they use a pretrained BERT model and train all tasks using a small-scale training corpus . |
| Outcome: | The proposed model outperforms a model trained in a multitask setting with no additional summarization data. |
Striking a Balance: Alleviating Inconsistency in Pre-trained Models for Symmetric Classification Tasks (2022.findings-acl)
Copied to clipboard
| Challenge: | Inconsistency is observed in symmetric classification tasks that take two inputs and require the output to be invariant of the order of the inputs. |
| Approach: | They propose a consistency loss function to alleviate inconsistency in symmetric classification tasks that take two inputs and require the output to be invariant of the order of the inputs. |
| Outcome: | The proposed model improves consistency in predictions for three paraphrase detection datasets without significant drop in accuracy scores. |
To Attend or not to Attend: A Case Study on Syntactic Structures for Semantic Relatedness (P18-1)
Copied to clipboard
| Challenge: | Recent success of Recurrent Neural Networks (RNNs) in Machine Translation (MT) has prompted attention mechanisms to be used in machine translation. |
| Approach: | They propose a tree-structured attention model on Tree Long Short-Term Memory Networks . they also experiment with three LSTM variants: bidirectional-LSTMs, Constituency Tree-LSTS, and Dependency Tree LSTS. |
| Outcome: | The proposed model is based on tree-LSTMs, constituency trees, and dependencies trees. |
MetaICL: Learning to Learn In Context (2022.naacl-main)
Copied to clipboard
| Challenge: | Large language models can do in-context learning by conditioning on a few training examples with no parameter updates or task-specific templates. |
| Approach: | They propose a meta-training framework where a pretrained language model is tuned to do in-context learning on a large set of training tasks. |
| Outcome: | The proposed framework outperforms baseline models on 142 NLP datasets and a range of target tasks with domain shifts. |
Measuring and Improving Attentiveness to Partial Inputs with Counterfactuals (2024.findings-emnlp)
Copied to clipboard
Yanai Elazar, Bhargavi Paranjape, Hao Peng, Sarah Wiegreffe, Khyathi Chandu, Vivek Srikumar, Sameer Singh, Noah Smith
| Challenge: | Existing studies have found that datasets with paired inputs are prone to spurious correlations, resulting in models trained only on those outperform chance. |
| Approach: | They propose a counterfactual attentiveness test to measure reliance on spurious correlations by replacing part of the input with its counterpart from a different example. |
| Outcome: | The proposed method improves models' attentiveness on ten datasets spanning four tasks: natural language inference, reading comprehension, paraphrase detection, and visual & language reasoning. |
EENLP: Cross-lingual Eastern European NLP Index (2022.lrec-1)
Copied to clipboard
Alexey Tikhonov, Alex Malkhasov, Andrey Manoshin, George-Andrei Dima, Réka Cserháti, Md.Sadek Hossain Asif, Matt Sárdi
| Challenge: | Existing NLP resources for Eastern European languages are sparse. |
| Approach: | They propose to use existing Eastern European language resources to build cross-lingual datasets for five different semantic tasks to support commonsense reasoning. |
| Outcome: | The proposed model trains on 104 languages and shows impressive results on text analysis tasks. |
On the Importance of Adaptive Data Collection for Extremely Imbalanced Pairwise Tasks (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Recent datasets heuristically choose examples to ensure label balance . state-of-the-art models trained on QQP and WikiQA have only 2.4% average precision . |
| Approach: | They show that recent datasets heuristically choose examples to ensure label balance . they instead use active learning to retrieve uncertain points from a large pool of unlabeled utterance pairs . |
| Outcome: | The proposed model improves on QQP and WikiQA by using more informative negative examples. |
Evaluating Multilingual Sentence Representation Models in a Real Case Scenario (2022.lrec-1)
Copied to clipboard
| Challenge: | a recent study has shown that the infamous Protocols are actually plagiarized . a convoluted task with no standard benchmarks for paraphrase detection and sentence similarity is a problem . |
| Approach: | They evaluate sentence representation models on the paraphrase detection task . they use a forged text from the so-called "Protocols of the Elders of Zion" scholars have demonstrated that the first text plagiarizes from the second . |
| Outcome: | The proposed model is based on the forged “Protocols of the Elders of Zion” . the model is similar to the standard model but has some problems . |
Lexical Generalization Improves with Larger Models and Longer Training (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Excessive utilization of lexical overlap heuristics can lead to failure on challenging inputs. |
| Approach: | They analyze the use of lexical overlap heuristics in natural language inference, paraphrase detection, and reading comprehension using a contrastive dataset. |
| Outcome: | The proposed model can be used to improve performance on a wide range of tasks, but it is often wrong. |
ESCRITO - An NLP-Enhanced Educational Scoring Toolkit (L18-1)
Copied to clipboard
| Challenge: | Existing implementations are very specific to specific use cases and datasets. |
| Approach: | ESCRITO is a toolkit for scoring student writings using NLP techniques . authors propose teachers and NLP researchers to use APIs for scoring pipelines . |
| Outcome: | ESCRITO is a toolkit for scoring student writings using NLP techniques . it addresses two main user groups: teachers and NLP researchers . |
On the Calibration of Pre-trained Language Models using Mixup Guided by Area Under the Margin and Saliency (2022.acl-long)
Copied to clipboard
| Challenge: | Existing studies have shown that mixing up can improve model calibration on image classification tasks, but little is known about using it on natural language understanding (NLU) tasks. |
| Approach: | They propose a mixup strategy for pre-trained language models that improves model calibration further by using the AUM statistic and saliency map. |
| Outcome: | The proposed mixup improves model calibration on natural language understanding tasks while maintaining competitive accuracy. |
Semantic Equivalence Detection: Are Interrogatives Harder than Declaratives? (L18-1)
Copied to clipboard
| Challenge: | Semantic Text Similarity (STS) tasks are often not seen as similar to semantic equivalence detection tasks. |
| Approach: | They propose to assess the performance of different approaches to STS over different types of textual segments. |
| Outcome: | The proposed methods differ in performance over different types of textual segments, including declaratives and interrogatives, under conditions of comparability. |
Similarity-Based Reconstruction Loss for Meaning Representation (D18-1)
Copied to clipboard
| Challenge: | a new paper proposes and evaluates a set of loss functions that can be used to train models for representation learning . cross-entropy loss penalizes models when they fail to generate the exact word from ground truth data . |
| Approach: | They propose and evaluate loss functions that can be used to train any neural model for representation learning. |
| Outcome: | The proposed loss functions amplify semantic diversity while preserving original meaning . they show performance improvement on paraphrase detection and language inference tasks . |
Improving Paraphrase Detection with the Adversarial Paraphrasing Task (2021.acl-long)
Copied to clipboard
| Challenge: | a new adversarial method of paraphrase identification is being used to identify paraphrases based on word overlap and syntax . authors propose a dataset that generates semantically equivalent but lexically and syntactically disparate paraphrase pairs . |
| Approach: | They propose an adversarial method for paraphrase identification that uses word overlap and syntax to identify paraphrases. |
| Outcome: | The proposed method improves paraphrase detection accuracy and speed of generation of datasets. |
Zero-shot and Few-shot Learning with Instruction-following LLMs for Claim Matching in Automated Fact-checking (2025.coling-main)
Copied to clipboard
| Challenge: | Claim matching (CM) is a binary classification task that can be used to determine if two claims can be verified using the same piece of evidence or fact-check. |
| Approach: | They propose a claim matching task that uses binary classification and large language models to test out learning approaches to the task. |
| Outcome: | The proposed task can be tackled by leveraging mature tasks such as natural language inference or paraphrase detection. |
LuxEmbedder: A Cross-Lingual Approach to Enhanced Luxembourgish Sentence Embeddings (2025.coling-main)
Copied to clipboard
| Challenge: | Sentence embedding models are limited for many low-resource languages, including Luxembourgish. |
| Approach: | They propose to use Luxembourgish as an enhanced sentence embedding model with strong cross-lingual capabilities to address this issue. |
| Outcome: | The proposed model can embed Luxembourgish sentences better than high-resource languages. |
AraReasoner: Evaluating Reasoning-Based LLMs for Arabic NLP (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large language models have shown remarkable progress in reasoning abilities and general natural language processing tasks, yet their performance on Arabic data remains underexplored. |
| Approach: | They compare reasoning-focused LLMs with deepSeek models across 15 Arabic NLP tasks . they use zero-shot, few-shot and fine-tuning to evaluate their capacity for linguistic reasoning . |
| Outcome: | The proposed models outperform strong models on Arabic datasets and are compared with other models. |
ALVIN: Active Learning Via INterpolation (2024.emnlp-main)
Copied to clipboard
| Challenge: | Experimental results show that Active Learning methods ignore example groups whose prevalence may vary . supervised fine-tuning remains a critical component of model development, authors say . |
| Approach: | They propose an approach that uses interpolations to create anchors between examples . they propose to use the model to identify informative examples that counteract shortcuts . |
| Outcome: | The proposed model outperforms state-of-the-art active learning methods on six datasets . it prioritizes high-certainty instances that integrate representations from different example groups . |