Papers with Pretrained Language Models
Can Pretrained Language Models Derive Correct Semantics from Corrupt Subwords under Noise? (2023.starsem-1)
Copied to clipboard
| Challenge: | Existing studies have shown that Pretrained Language Models (PLMs) perform poorly under noise due to subword segmentation. |
| Approach: | They propose a framework for subword segmentation that provides a systematic categorization of segmentation corruption under noise and evaluation protocols by generating contrastive datasets with canonical-noisy word pairs. |
| Outcome: | The proposed framework provides a systematic categorization of segmentation corruption under noise and evaluation protocols by generating contrastive datasets with canonical-noisy word pairs. |
BERT Goes Off-Topic: Investigating the Domain Transfer Challenge using Genre Classification (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Pretrained language models have improved performance of text classification tasks, but they still suffer from spurious domain-specific clues. |
| Approach: | They propose a method to augment pretrained language models by generating texts in any desired genre and on any desired topic. |
| Outcome: | The proposed method improves on genre classification tasks while showing no improvement for other topics. |
MAFIA: Multi-Adapter Fused Inclusive Language Models (2024.eacl-long)
Copied to clipboard
| Challenge: | Pretrained Language Models (PLMs) are widely used in NLP for various tasks. |
| Approach: | They propose to modularly debias a pre-trained language model across multiple bias dimensions using structured knowledge and a large generative model. |
| Outcome: | The proposed model is able to debias a pre-trained language model across multiple bias dimensions in a semi-automated way. |
GraSAME: Injecting Token-Level Structural Information to Pretrained Language Models via Graph-guided Self-Attention Mechanism (2024.findings-naacl)
Copied to clipboard
| Challenge: | Pretrained Language Models (PLMs) benefit from external knowledge stored in graph structures for various downstream tasks. |
| Approach: | They propose a graph-guided self-attention mechanism that integrates token-level structural information into PLMs without additional alignment or concatenation efforts. |
| Outcome: | The proposed model outperforms baseline models and achieves comparable results to state-of-the-art models on WebNLG datasets. |
Impact of Pretraining Term Frequencies on Few-Shot Numerical Reasoning (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Pretrained language models have demonstrated ability to perform numerical reasoning by extrapolating from a few examples in few-shot settings. |
| Approach: | They investigate how well pretrained language models reason with terms less frequent in pretraining data. |
| Outcome: | The models are more accurate on instances whose terms are more prevalent, in some cases above 70% more accurate than the bottom 10%. |
Measuring and Improving Consistency in Pretrained Language Models (2021.tacl-1)
Copied to clipboard
Yanai Elazar, Nora Kassner, Shauli Ravfogel, Abhilasha Ravichander, Eduard Hovy, Hinrich Schütze, Yoav Goldberg
| Challenge: | In this paper, we examine whether pretrained language models are consistent with factual knowledge. |
| Approach: | They propose a method to improve consistency of pretrained language models . consistency is a desirable property of a good language understanding model, they argue . |
| Outcome: | The proposed model improves consistency and shows that it is effective. |
From Generalist to Specialist: A Survey of Large Language Models for Chemistry (2025.coling-main)
Copied to clipboard
| Challenge: | Existing studies on pretraining of LLMs on extensive web-based texts are insufficient for advanced scientific discovery, especially in chemistry. |
| Approach: | They outline methodologies for incorporating domain-specific chemistry knowledge and multi-modal information into LLMs and conceptualize chemistry LLM agents using chemistry tools. |
| Outcome: | The proposed models are based on domain-specific chemistry knowledge and multi-modal information and are capable of accelerating scientific research. |
Analyzing Homonymy Disambiguation Capabilities of Pretrained Language Models (2024.lrec-main)
Copied to clipboard
Lorenzo Proietti, Stefano Perrella, Simone Tedeschi, Giulia Vulpis, Leonardo Lavalle, Andrea Sanchietti, Andrea Ferrari, Roberto Navigli
| Challenge: | Word Sense Disambiguation (WSD) is a key task in Natural Language Processing (NLP) but current pretrained language models lack the granularity to perform disambiguation . |
| Approach: | They propose a large-scale resource that leverages homonymy relations to cluster WordNet senses and train Homonymy Disambiguation systems. |
| Outcome: | The proposed model can distinguish homonyms with up to 95% accuracy even without fine-tuning the underlying PLM. |
Translate to Disambiguate: Zero-shot Multilingual Word Sense Disambiguation with Pretrained Language Models (2024.eacl-long)
Copied to clipboard
| Challenge: | Pretrained language models learn cross-lingual knowledge and perform well on diverse tasks when finetuned. |
| Approach: | They propose a zero-shot prompting approach that captures cross-lingual word sense with a contextual prompt. |
| Outcome: | The proposed approach outperforms baselines on recall in many evaluation languages without additional training or finetuning. |
POLITICS: Pretraining with Same-story Article Comparison for Ideology Prediction and Stance Detection (2022.findings-naacl)
Copied to clipboard
| Challenge: | a lack of general-purpose tools to characterize and predict ideology across genres of text remains a challenge . a recent study compared ideology-driven pretraining tasks with long or formal written texts . |
| Approach: | They propose to use a large-scale dataset to train pretraining models that compare political news articles on the same story written by different ideologies. |
| Outcome: | The proposed model outperforms baseline models and state-of-the-art models on ideology prediction and stance detection tasks. |
Understanding Politics via Contextualized Discourse Processing (2021.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in pretrained language models do not capture nuanced biases in political discourse . a new approach to represent political content is to use contextualized embeddings to create effective representations . |
| Approach: | They propose a model that captures and leverages political content to generate more effective representations . they use tweets, press releases, issues, news articles and participating entities to generate composed representations. |
| Outcome: | The proposed model generates representations for political entities over multiple issues or events . qualitative and quantitative analysis shows that the model is meaningful and effective . |
Made of Steel? Learning Plausible Materials for Components in the Vehicle Repair Domain (2023.eacl-main)
Copied to clipboard
| Challenge: | a novel approach to learn domain-specific plausible materials for components in the vehicle repair domain is proposed . connecting a symptom to an underlying cause is a crucial building block for natural language understanding across domains. |
| Approach: | They propose a method to aggregate salient predictions from a set of cloze task style templates and use a Wikipedia corpus to augment the model. |
| Outcome: | The proposed approach outperforms a traditional pattern-based approach by exploiting the compositionality assumption in a cloze task style setting. |
Can Pretrained Language Models (Yet) Reason Deductively? (2023.eacl-main)
Copied to clipboard
| Challenge: | Acquiring factual knowledge with Pretrained Language Models (PLMs) has attracted increasing attention, showing promising performance in many knowledge-intensive tasks. |
| Approach: | They conduct a comprehensive evaluation of the learnable deductive reasoning capability of pretrained language models and compare their performance against simple adversarial surface form edits. |
| Outcome: | The models are able to generalise learned logic rules and perform inconsistently against simple adversarial surface form edits, but catastrophically forget the previously learnt knowledge. |
Towards Understanding Task-agnostic Debiasing Through the Lenses of Intrinsic Bias and Forgetfulness (2024.findings-acl)
Copied to clipboard
Guangliang Liu, Milad Afshari, Xitong Zhang, Zhiyu Xue, Avrajit Ghosh, Bidhan Bashyal, Rongrong Wang, Kristen Johnson
| Challenge: | Debiasing Pretrained Language Models (PLMs) are task-agnostic and can be generalizable, but its impact on language modeling ability and the risk of relearning social biases remain as the two most significant challenges. |
| Approach: | They propose a framework which can Propagate Socially-fair Debiasing to Downstream Fine-tuning to alleviate the forgetting issue of PLMs by regularizing debiased attention heads based on the PLM’s bias levels from stages of pretraining and debiase. |
| Outcome: | The proposed framework can Propagate Socially-fair Debiasing to Downstream Fine-tuning, indicating that the ineffectiveness of debiase can be alleviated by overcoming the forgetting issue through regularizing successfully debiased attention heads based on the PLMs’ bias levels from stages of pretraining and debiases. |
Bigger Data or Fairer Data? Augmenting BERT via Active Sampling for Educational Text Classification (2022.coling-1)
Copied to clipboard
| Challenge: | Pretrained Language Models (PLMs) encode bias against protected groups in the representations they learn, which may harm the prediction fairness of downstream models. |
| Approach: | They propose to quantify the awareness that a pretrained language model (BERT) has regarding people’s protected attributes and augment it to enhance prediction fairness of downstream models. |
| Outcome: | The proposed method improves fairness and accuracy of models by inhibiting the awareness of protected attributes in the PLMs. |
How Much Do Encoder Models Know About Word Senses? (2025.acl-long)
Copied to clipboard
| Challenge: | Word Sense Disambiguation (WSD) is a key task in Natural Language Processing (NLP) however, how well these models inherently disambiguate word senses remains uncertain. |
| Approach: | They evaluate several encoder-only PLMs across WordNet and ODE sense inventories to evaluate their ability to separate word senses without any task-specific fine-tuning. |
| Outcome: | The proposed model outperforms output layer on WordNet and ODE sense inventories by 15 percentage points. |
Exploiting Abstract Meaning Representation for Open-Domain Question Answering (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing work attempts to address these challenges using Pretrained Language Models (PLMs) but the diversity of surface form expressions can hinder the model’s ability to capture accurate correlations, especially when the context is lengthy and complex. |
| Approach: | They propose a method known as Graph-as-Token (GST) to incorporate AMRs into PLMs to assist the model in understanding complex semantic information. |
| Outcome: | The proposed method outperforms existing methods and significantly improves performance on both Natural Questions and TriviaQA. |
Do PLMs Know and Understand Ontological Knowledge? (2023.acl-long)
Copied to clipboard
| Challenge: | Existing studies on pretrained language models focus mainly on factual knowledge, lacking a systematic probing of ontological knowledge. |
| Approach: | They investigate whether Pretrained Language Models store ontological knowledge and have a semantic un- derstanding of the knowledge rather than rote memorization of the surface form. |
| Outcome: | The proposed models can memorize certain ontological knowledge and perform logical reasoning with given knowledge according to ontological entailment rules. |
Generalizable and Stable Finetuning of Pretrained Language Models on Low-Resource Texts (2024.naacl-long)
Copied to clipboard
| Challenge: | Pretrained language models have advanced natural language processing tasks significantly, but finetuning them on low-resource datasets presents significant challenges such as instability and overfitting. |
| Approach: | They propose a regularization method based on attention-guided weight mixup for finetuning PLMs on low-resource datasets. |
| Outcome: | The proposed method improves generalization and combats overfitting on two splits of the training dataset. |
Addressing Healthcare-related Racial and LGBTQ+ Biases in Pretrained Language Models (2024.findings-naacl)
Copied to clipboard
| Challenge: | Pretrained language models (PLMs) propagate social stigmas and stereotypes, a critical concern given their widespread use. |
| Approach: | They adapt two intrinsic bias benchmarks to quantify racial and LGBTQ+ biases in prevalent PLMs and empirically evaluate the effectiveness of various debiasing methods in mitigating these biase. |
| Outcome: | The proposed methods reduce biases without compromising performance in downstream tasks. |
Integrating Task Specific Information into Pretrained Language Models for Low Resource Fine Tuning (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Existing pretrained language models are agnostic to downstream information and can overfit when fine-tuned with low resource datasets. |
| Approach: | They integrate label information as a task-specific prior into the self-attention component of pretrained BERT models. |
| Outcome: | Experiments on benchmarks and real-word datasets show that the proposed approach can improve the performance of pretrained models when fine-tuned with small datasets. |
Personalized Federated Learning for Text Classification with Gradient-Free Prompt Tuning (2024.findings-naacl)
Copied to clipboard
Rui Wang, Tong Yu, Ruiyi Zhang, Sungchul Kim, Ryan Rossi, Handong Zhao, Junda Wu, Subrata Mitra, Lina Yao, Ricardo Henao
| Challenge: | Pretrained language models (PLMs) are used for personalized federated learning . communication costs are high with large PLMs, and local training is expensive . |
| Approach: | They propose a framework for federated learning with pretrained language models . they propose 'discrete local search' and compression mechanism for local training . |
| Outcome: | The proposed framework achieves superior performance compared with baselines. |
A Survey on Patent Analysis: From NLP to Multimodal AI (2025.acl-long)
Copied to clipboard
| Challenge: | Recent advances in pretrained language models and large language models have demonstrated transformative capabilities across diverse domains. |
| Approach: | They propose a taxonomy for categorization based on tasks in the patent life cycle . they introduce a novel taxonomies for categorizing based upon tasks in patent life cycles . |
| Outcome: | The proposed method is based on tasks in the patent life cycle and provides a taxonomy for categorization based upon tasks in patent life cycles. |
Calibrating Factual Knowledge in Pretrained Language Models (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Existing studies show that Pretrained Language Models can store factual knowledge, but facts stored in PLMs are not always correct. |
| Approach: | They propose a lightweight method to calibrate factual knowledge in PLMs without re-training from scratch. |
| Outcome: | The proposed method can be used to calibrate factual knowledge in PLMs without re-training from scratch. |
Do Language Embeddings capture Scales? (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Pretrained Language Models possess significant linguistic, common sense and factual knowledge, but are short of the capability required for general common-sense reasoning. |
| Approach: | They propose to train pretrained language models with a method of canonicalizing numbers . they address a task which is also pre-requisite for general common-sense reasoning . |
| Outcome: | The proposed model can answer questions about common sense and linguistics, but lacks the capability to answer questions on scalar attributes. |
Enhancing Rumor Detection Methods with Propagation Structure Infused Language Model (2025.coling-main)
Copied to clipboard
| Challenge: | Pretrained Language Models excel in various Natural Language Processing tasks, but performance on social media applications like rumor detection remains suboptimal. |
| Approach: | They propose a pretraining strategy to infuse information from propagation structures into pretrained language models to capture interactions of stance and sentiment crucial for rumor detection. |
| Outcome: | The proposed model outperforms existing methods on social media applications and significantly improves rumor detection performance. |
Enhancing Court View Generation with Knowledge Injection and Guidance (2024.lrec-main)
Copied to clipboard
| Challenge: | Existing methods of natural language generation (NLG) rely on the extensive parameters of pretrained language models (PLMs) but their effectiveness may be compromised by insufficient domain-specific knowledge. |
| Approach: | They propose a knowledge-injected prompt encoder to incorporate domain knowledge during the training stage, thereby reducing computational overhead. |
| Outcome: | The proposed approach outperforms established baselines on real-world data in responsivity of claims and in the ability to transfer domain knowledge. |
Evaluating Unsupervised Dimensionality Reduction Methods for Pretrained Sentence Embeddings (2024.lrec-main)
Copied to clipboard
| Challenge: | Sentence embeddings produced by pretrained language models are high dimensional (ca. 1024-4096) this is problematic when representing large numbers of sentences in memory- or compute-constrained devices. |
| Approach: | They propose to use Principal Component Analysis to reduce the dimensionality of sentence embeddings produced by pretrained language models to reduce their complexity. |
| Outcome: | The proposed methods reduce the dimensionality of sentence embeddings by 50% without incurring significant loss in performance in multiple downstream tasks. |
Pre-trained Personalized Review Summarization with Effective Salience Estimation (2023.findings-acl)
Copied to clipboard
| Challenge: | Pretrained language models (PLMs) are a new paradigm in text generation for the strong ability of natural language comprehension. |
| Approach: | They propose a pre-trained personalized review summarization method that incorporates personalized information into the salience estimation of input reviews. |
| Outcome: | The proposed method performs better than the state-of-the-art methods on real-world datasets. |
What’s the Meaning of Superhuman Performance in Today’s NLU? (2023.acl-long)
Copied to clipboard
Simone Tedeschi, Johan Bos, Thierry Declerck, Jan Hajič, Daniel Hershcovich, Eduard Hovy, Alexander Koller, Simon Krek, Steven Schockaert, Rico Sennrich, Ekaterina Shutova, Roberto Navigli
| Challenge: | Recent research has focused on developing larger pretrained language models and introducing benchmarks such as SuperGLUE and SQuAD to measure their abilities. |
| Approach: | They propose to use benchmarks such as SuperGLUE and SQUAD to evaluate PLMs' abilities in language understanding, reasoning, and reading comprehension to assess their performance. |
| Outcome: | The proposed benchmarks have serious limitations affecting comparison between humans and PLMs and provide recommendations for fairer and more transparent benchmarks. |
Negated and Misprimed Probes for Pretrained Language Models: Birds Can Talk, But Cannot Fly (2020.acl-main)
Copied to clipboard
| Challenge: | Petroni et al. ( 2019) show that PLMs can recall factual knowledge that is part of their training corpus. |
| Approach: | They propose two new probing tasks that analyze factual knowledge stored in pretrained language models (PLMs) they introduce negated and non-negated cloze questions and add misprimes to clozing questions to mimic human priming methods. |
| Outcome: | The proposed tasks outperform automatically extracted knowledge bases on QA. |
Knowledge Unlearning for Mitigating Privacy Risks in Language Models (2023.acl-long)
Copied to clipboard
| Challenge: | Recent work shows that an adversary can extract training data from Pretrained Language Models including Personally Identifiable Information (PII) such as names, phone numbers, and email addresses. |
| Approach: | They propose to use knowledge unlearning to reduce privacy risks for LMs by performing gradient ascent on target token sequences instead of trying to unlearn all the data at once. |
| Outcome: | The proposed method can give a stronger empirical privacy guarantee in scenarios where the data vulnerable to extraction attacks are known a priori while being much more efficient and robust. |
Jargon: A Suite of Language Models and Evaluation Tasks for French Specialized Domains (2024.lrec-main)
Copied to clipboard
Vincent Segonne, Aidan Mannion, Laura Cristina Alonzo Canul, Alexandre Daniel Audibert, Xingyu Liu, Cécile Macaire, Adrien Pupier, Yongxin Zhou, Mathilde Aguiar, Felix E. Herron, Magali Norré, Massih R Amini, Pierrette Bouillon, Iris Eshkol-Taravella, Emmanuelle Esperança-Rodier, Thomas François, Lorraine Goeuriot, Jérôme Goulian, Mathieu Lafourcade, Benjamin Lecouteux, François Portet, Fabien Ringeval, Vincent Vandeghinste, Maximin Coavoux, Marco Dinarelli, Didier Schwab
| Challenge: | Pretrained language models are the de facto backbone of most state-of-the-art NLP systems. |
| Approach: | They propose a family of domain-specific pretrained PLMs for French focusing on three important domains: transcribed speech, medicine, and law. |
| Outcome: | The proposed models perform better on transcribed speech, medicine, and law domains than state-of-the-art models on a diverse set of tasks and datasets. |
Noise-Robust Fine-Tuning of Pretrained Language Models via External Guidance (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Pretrained Language Models (PLMs) are advanced but data labels are noisy due to the complex annotation process. |
| Approach: | They propose a framework for fine-tuning PLMs using noisy labels that incorporates guidance from Large Language Models like ChatGPT. |
| Outcome: | Experiments on synthetic and real-world noisy datasets show that the proposed framework outperforms the state-of-the-art framework. |