Papers by Alakananda Vempala

7 papers
Right for the Right Reason: Evidence Extraction for Trustworthy Tabular Reasoning (2022.acl-long)

Copied to clipboard

Challenge: Recent studies show that tabular reasoning models use spurious correlations and focus on false evidence or ignore it altogether.
Approach: They propose a task where models need to extract evidence and then inference labels . they crowdsource evidence row labels and develop unsupervised evidence extraction strategies .
Outcome: The proposed approach outperforms baseline models on the inference task using only the automatically extracted evidence as the premise.
Annotating If the Authors of a Tweet are Located at the Locations They Tweet About (L18-1)

Copied to clipboard

Challenge: a tweet's locations do not always indicate spatial information involving the author of the tweet . a corpus of 1,062 tweets contains 1,200 location named entities .
Approach: They propose a corpus annotating whether tweet authors are located in locations . they use temporal tags centered around tweet timestamps to temporally anchor this information .
Outcome: The proposed method annotates whether authors are located in tweet locations . it shows that no spatial relationship can be inferred in 21% of instances .
Determining Event Durations: Models and Error Analysis (N18-2)

Copied to clipboard

Challenge: a crucial piece of information regarding events is their duration, a rarely mentioned attribute . core tasks such as temporal understanding and reasoning would benefit from knowing the expected duration of events.
Approach: They introduce aspectual features that capture deeper linguistic information . they also experiment with neural networks to predict event durations .
Outcome: The proposed models capture deeper linguistic information than previous work and provide useful clues.
Categorizing and Inferring the Relationship between the Text and Image of Twitter Posts (P19-1)

Copied to clipboard

Challenge: Social media posts often contain images to provide content, provide context, or express feelings.
Approach: They build and release a dataset of image tweets annotated with four different classes which express whether the text or the image provides additional information to the other modality.
Outcome: The proposed method can be used in several downstream applications including pre-training image tagging models and collecting distantly supervised data for image captioning.
Don’t Retrain, Just Rewrite: Countering Adversarial Perturbations by Rewriting Text (2023.acl-long)

Copied to clipboard

Challenge: ATINTER model can be used to rewrite adversarial inputs to make them non-adversarial . if undefended, model should maintain good task performance and effectively mitigate adversarials .
Approach: They propose a model that intercepts adversarial inputs and learns to rewrite them . they show that it provides better adversarial robustness than existing defense approaches .
Outcome: The proposed model improves adversarial robustness without compromising task accuracy on a sentiment classification dataset.
ULTRA: Unleash LLMs’ Potential for Event Argument Extraction through Hierarchical Modeling and Pair-wise Self-Refinement (2024.findings-acl)

Copied to clipboard

Challenge: Structural extraction of events within discourse is critical for event-centric understanding . document-level EAE focuses on arguments that are scattered across an entire document . ULTRA is a hierarchical framework that extracts event arguments more cost-effectively .
Approach: They propose a hierarchical framework that extracts event arguments more cost-effectively . ULTRA sequentially reads text chunks of a document to generate a candidate argument set . they propose to use a supervised model to find the exact boundary of an argument .
Outcome: The proposed framework outperforms strong models and ChatGPT by 9.8% when evaluated by Exact Match (EM).
Annotating Temporally-Anchored Spatial Knowledge by Leveraging Syntactic Dependencies (L18-1)

Copied to clipboard

Challenge: Existing approaches to extract spatial knowledge focus on extracting locations of events, someone or something.
Approach: They propose a method to annotate temporally-anchored spatial knowledge on top of OntoNotes by crowdsourcing annotations.
Outcome: The proposed method can be automated and validated using syntactic dependencies and crowdsourced annotations.

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