Papers by Elisa Bassignana

9 papers
CrossRE: A Cross-Domain Dataset for Relation Extraction (2022.findings-emnlp)

Copied to clipboard

Challenge: Relation Extraction (RE) evaluation is limited to in-domain setups . despite the drought of research on cross-domain RE, its practical importance remains .
Approach: They propose a cross-domain benchmark for relation extraction which includes multi-label annotations and meta-data to include explanations and flags of difficult instances.
Outcome: The proposed model includes explanations and flags of difficult instances.
What’s wrong with your model? A Quantitative Analysis of Relation Classification (2024.starsem-1)

Copied to clipboard

Challenge: A major trend in NLP research aims at designing more sophisticated setups to improve the state-of-the-art (SOTA) on a target task.
Approach: They propose an in-depth analysis suite for Relation Classification to be used for prediction tasks.
Outcome: The proposed model improves over the baseline by >3 Micro-F1 . the proposed model is based on a case study and a preliminary error-guided analysis .
How to Encode Domain Information in Relation Classification (2024.lrec-main)

Copied to clipboard

Challenge: Existing deep learning models require a lot of training data to obtain high performance.
Approach: They propose a multi-domain training setup for Relation Classification (RC) they compare different ways to enrich input instances with domain information .
Outcome: The proposed model improves > 2 Macro-F1 against the baseline setup.
Silver Syntax Pre-training for Cross-Domain Relation Extraction (2023.findings-acl)

Copied to clipboard

Challenge: Relation Extraction (RE) is the task of extracting structured knowledge from unstructured text.
Approach: They exploit the affinity between syntactic structure and semantic RE to obtain low-cost pre-training data.
Outcome: The proposed model outperforms baseline models in five out of six cross-domain setups without additional annotated data.
What Do You Mean by Relation Extraction? A Survey on Datasets and Study on Scientific Relation Classification (2022.acl-srw)

Copied to clipboard

Challenge: Existing RE surveys focus on modeling techniques, but there are few that are based on real-world scenarios.
Approach: They propose to survey RE datasets and revisit the task definition and its adoption by the community.
Outcome: The proposed approach improves the reliability of RE evaluations across multiple datasets and reveals significant discrepancies in annotations.
The AI Gap: How Socioeconomic Status Affects Language Technology Interactions (2025.acl-long)

Copied to clipboard

Challenge: Socioeconomic status (SES) fundamentally influences how people interact with technology, but it is limited by proxy metrics and synthetic data.
Approach: They collect 6,482 prompts from previous interactions of 1,000 individuals from ‘diverse socioeconomic backgrounds’ about their use of language technologies and generative AI.
Outcome: The findings show that higher SES groups have higher levels of abstraction, convey requests more concisely, and topics like ‘inclusivity’ and ‘travel’.
Experimental Standards for Deep Learning in Natural Language Processing Research (2022.findings-emnlp)

Copied to clipboard

Challenge: a lack of common experimental standards remains an open challenge to the field at large .
Approach: They propose to distill discussions on experimental standards into a single, widely-applicable methodology.
Outcome: Using best practices, we can strengthen experimental evidence, improve reproducibility and enable scientific progress.
Can Humans Identify Domains? (2024.lrec-main)

Copied to clipboard

Challenge: Textual domain is a crucial property within the Natural Language Processing community due to its effects on downstream model performance.
Approach: They examine the level of human disagreement and the relative difficulty of each annotation task by training classifiers to perform the same task.
Outcome: The authors show that human proficiency in identifying related intrinsic textual properties is low and that disagreements are high.
Evidence > Intuition: Transferability Estimation for Encoder Selection (2022.emnlp-main)

Copied to clipboard

Challenge: Existing studies on LM transferability have focused on a priori tuning of encoders . prior work has examined the different yet related tasks of performance prediction .
Approach: They propose to generate quantitative evidence to predict which LM will perform best on a target task without fine-tuning all candidates.
Outcome: The proposed model outperforms the standard of human practitioner ranking in 94% of the setups.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations