Papers by Daniel Preotiuc-Pietro
Improving Multimodal Classification of Social Media Posts by Leveraging Image-Text Auxiliary Tasks (2024.findings-eacl)
Copied to clipboard
| Challenge: | Prior work on multimodal content classification has not addressed these challenges. |
| Approach: | They propose to use two auxiliary tasks to fine-tune multimodal models to address hidden cross-modal semantics and weak image-text relationships when modeling text and images. |
| Outcome: | The proposed model improves by up to 2.6 F1 score across five diverse social media datasets. |
Identifying Named Entities as they are Typed (2021.eacl-main)
Copied to clipboard
| Challenge: | Named Entity Recognition (NER) systems are not applicable to systems that process text in real time as the text is typed. |
| Approach: | They propose a new experimental setup for evaluating Named Entity Recognition systems that evaluates named entities as they are typed on a sentence level . they propose to adapt existing evaluation setups to suit the new setup . |
| Outcome: | The proposed setup shows that the best systems that are evaluated on each token after its typed reach performance within 1–5 F1 points of systems that were evaluated at the end of the sentence. |
STARQA: A Question Answering Dataset for Complex Analytical Reasoning over Structured Databases (2025.emnlp-main)
Copied to clipboard
| Challenge: | Existing Text2SQL systems do not support SQL analysts in their primary work of performing complex analytics on specialized databases. |
| Approach: | They propose to decompose STARQA questions using SQL and Python to perform reasoning on specialized relational databases. |
| Outcome: | The proposed approach decomposes the task through a combination of SQL and Python, and achieves better performance on the more difficult questions. |
EntSUM: A Data Set for Entity-Centric Extractive Summarization (2022.acl-long)
Copied to clipboard
| Challenge: | Existing methods for controllable summarization fail to generate entity-centric summaries. |
| Approach: | They propose to use a human-annotated data set EntSUM to generate controllable summarization with a focus on named entities as the aspects to control. |
| Outcome: | The proposed data set shows that existing methods fail to generate entity-centric summaries. |
Extractive Entity-Centric Summarization as Sentence Selection using Bi-Encoders (2022.aacl-short)
Copied to clipboard
| Challenge: | Entity-centric summarization is a type of controllable summarizing that aims to produce a summary specific to a given target entity. |
| Approach: | They propose to recast a sentence selection task as a controllable summarization using a dataset supported by EntSUM. |
| Outcome: | The proposed framework outperforms the current state-of-the-art in the sentence selection task and outperformed the competitive entity-centric Lead 3 heuristic by 1.1 F1. |
Towards a Unified Multi-Domain Multilingual Named Entity Recognition Model (2023.eacl-main)
Copied to clipboard
Mayank Kulkarni, Daniel Preotiuc-Pietro, Karthik Radhakrishnan, Genta Indra Winata, Shijie Wu, Lingjue Xie, Shaohua Yang
| Challenge: | Named Entity Recognition is a key task whose performance is sensitive to genre and language. |
| Approach: | They propose a setup for Named Entity Recognition which includes multi-domain and multilingual training and evaluation across 13 domains and 4 languages. |
| Outcome: | The proposed model improves on 13 domains and 4 languages across 13 domain and 4 language domains. |
Multi-Domain Named Entity Recognition with Genre-Aware and Agnostic Inference (2020.acl-main)
Copied to clipboard
| Challenge: | Named entity recognition (NER) is a key component of many text processing pipelines. |
| Approach: | They propose a new architecture tailored to the task of identifying named entities with data from multiple genres. |
| Outcome: | The proposed architecture outperforms baseline and competitive methods on all three setups with differences ranging between +1.95 to +3.11 average F1 across multiple genres when compared to standard approaches. |
Calibrating LLMs for Text-to-SQL Parsing by Leveraging Sub-clause Frequencies (2025.emnlp-main)
Copied to clipboard
| Challenge: | Large language models (LLMs) exhibit unexpected failures in which they are confidently incorrect. |
| Approach: | They propose a method for calibrating SQL outputs that leverages structured nature to provide more granular signals of correctness. |
| Outcome: | The proposed method improves on two popular text-to-SQL datasets and provides a confidence score that is calibrated. |
Overcoming Catastrophic Forgetting in Massively Multilingual Continual Learning (2023.findings-acl)
Copied to clipboard
Genta Winata, Lingjue Xie, Karthik Radhakrishnan, Shijie Wu, Xisen Jin, Pengxiang Cheng, Mayank Kulkarni, Daniel Preotiuc-Pietro
| Challenge: | Existing methods to handle catastrophic forgetting fail to retain knowledge learnt in the past when sudden shifts occur in training data distributions. |
| Approach: | They propose a learning rate scheduling method that preserves new information without strongly overwriting past knowledge. |
| Outcome: | The proposed method preserves new information without overwriting past knowledge in a multilingual continuous learning framework. |
Point-of-Interest Type Inference from Social Media Text (2020.aacl-main)
Copied to clipboard
| Challenge: | Using a dataset of 200,000 English tweets, we can predict the type of the place from which a tweet was sent from. |
| Approach: | They propose to analyze a dataset of 200,000 tweets from 2,761 points-of-interest in the U.S. and train classifiers to predict the type of the location a tweet was sent from. |
| Outcome: | The proposed method can predict the type of the location a tweet was sent from and reach a macro F1 of 43.67 across eight classes. |
Combining Humor and Sarcasm for Improving Political Parody Detection (2022.naacl-main)
Copied to clipboard
| Challenge: | Parody is a figurative device used for mimicking entities for comedic or critical purposes. |
| Approach: | They propose a multi-encoder model that combines three parallel encoders to enrich parody-specific representations with humor and sarcasm information. |
| Outcome: | The proposed model outperforms state-of-the-art methods on a dataset of political parody tweets. |
Unsupervised Contrast-Consistent Ranking with Language Models (2024.eacl-long)
Copied to clipboard
| Challenge: | Language models contain ranking-based knowledge and are powerful solvers of in-context ranking tasks. |
| Approach: | They propose to use a model to elicit language models' ranking knowledge without supervision by using a pairwise, pointwise and listwise prompting method. |
| Outcome: | The proposed method is inspired by an unsupervised probing method called Contrast-Consistent Search (CCS). |
Fact vs. Opinion: the Role of Argumentation Features in News Classification (2020.coling-main)
Copied to clipboard
| Challenge: | A 2018 study led by the Media Insight Project showed that most journalists think that their news organizations should clearly mark what is news reporting and what is commentary or opinion in order to combat fake news and gain public trust. |
| Approach: | They propose to classify news articles into newsstories and opinion pieces using models that aim to sup-plement the article content representation with argumentation features. |
| Outcome: | The proposed model outperforms linguistic features and improves on fine-tuned transformer-based models on data from publishers. |
Cross-lingual Few-Shot Learning on Unseen Languages (2022.aacl-main)
Copied to clipboard
| Challenge: | Large pre-trained language models have demonstrated the ability to obtain good performance on downstream tasks with limited examples in resource-rich languages. |
| Approach: | They propose to use a downstream sentiment analysis task to analyze the effectiveness of several few-shot learning strategies across 12 languages, including 8 unseen languages, to compare results. |
| Outcome: | The proposed model, XLM-R, gives the best performance on a task with few examples in resource-rich languages. |
EntSUMv2: Dataset, Models and Evaluation for More Abstractive Entity-Centric Summarization (2023.emnlp-main)
Copied to clipboard
| Challenge: | Entity-centric summarization is a form of controllable summarizing that aims to generate a summary for a specific entity given a document. |
| Approach: | They propose to use a more abstract version of the original entity-centric ENTSUM summarization dataset to generate a shorter annotated summary for downstream users. |
| Outcome: | The proposed method is more abstract and uses supervised fine-tuning and large-scale instruction tuning to provide more specific and useful summaries for downstream users. |
Temporally-Informed Analysis of Named Entity Recognition (2020.acl-main)
Copied to clipboard
| Challenge: | Existing methods to evaluate text data are rarely reported by taking the timestamp of the document into account. |
| Approach: | They propose methods that make better use of temporally-diverse training data with a focus on named entity recognition. |
| Outcome: | The proposed models make better use of temporally-diverse training data, with a focus on named entity recognition. |
Improving Instruct Models for Free: A Study on Partial Adaptation (2025.emnlp-main)
Copied to clipboard
Ozan Irsoy, Pengxiang Cheng, Jennifer L Chen, Daniel Preotiuc-Pietro, Shiyue Zhang, Duccio Pappadopulo
| Challenge: | Instruct models are deemed superior and more usable but can be eroded by instruction tuning . a recent study shows that instruct models are better at following instructions than base models . |
| Approach: | They scale down the strength of instruction tuning to improve model performance . they show that reducing instruction tuning results in material improvement . |
| Outcome: | The proposed model improves on a few-shot in-context learning benchmark . but it loses some degree of its in-training ability . |
Analyzing Political Parody in Social Media (2020.acl-main)
Copied to clipboard
| Challenge: | Parody is a figurative device used to imitate an entity for comedic or critical purposes. |
| Approach: | They propose a dataset of tweets from real politicians and their corresponding parody accounts to run supervised machine learning models for automatic classification. |
| Outcome: | The proposed models predict political parody tweets with 90% accuracy . they also identify the markers of parody through a linguistic analysis . |
Automatic Identification and Classification of Bragging in Social Media (2022.acl-long)
Copied to clipboard
| Challenge: | Bragging is a speech act employed to build a favorable self-image through positive statements about oneself. |
| Approach: | They propose to use tweets annotated for bragging to build a model that can predict bragging with macro F1 up to 72.42 and 35.95 for binary and multi-class bragging classification tasks respectively. |
| Outcome: | The proposed models predict bragging with macro F1 up to 72.42 and 35.95 in binary and multi-class classification tasks respectively. |
Who Is Bragging More Online? A Large Scale Analysis of Bragging in Social Media (2024.lrec-main)
Copied to clipboard
| Challenge: | Social media is a natural platform for users to use bragging to gain admiration, respect, attention and followers from their audiences. |
| Approach: | They employ computational sociolinguistics methods to conduct the first large scale study of bragging behavior on Twitter by focusing on its overall prevalence, temporal dynamics and impact of demographic factors. |
| Outcome: | The proposed study shows that the prevalence of bragging decreases over time within the same population of users and younger, more educated and popular users in the U.S. are more likely to brag. |
Leveraging Contextual Information for Effective Entity Salience Detection (2024.findings-naacl)
Copied to clipboard
Rajarshi Bhowmik, Marco Ponza, Atharva Tendle, Anant Gupta, Rebecca Jiang, Xingyu Lu, Qian Zhao, Daniel Preotiuc-Pietro
| Challenge: | Prior work on salient entity detection focused on machine learning models that require heavy feature engineering. |
| Approach: | They propose to fine-tune medium-sized language models with a cross-encoder style architecture to achieve significant performance gains over feature engineering approaches. |
| Outcome: | The proposed model fine-tunes medium-sized pre-trained language models with a cross-encoder style architecture yields substantial performance gains over feature engineering approaches. |