Papers by Abhilasha Sancheti
On the Influence of Gender and Race in Romantic Relationship Prediction from Large Language Models (2024.emnlp-main)
Copied to clipboard
| Challenge: | We show that models are less likely to predict romantic relationships for same-gender character pairs than different-grace character pairs. |
| Approach: | They perform name-replacement experiments to examine gender biases in large language models . they hypothesize that models mirror heteronormative biase and prejudice against interracial romantic relationships . |
| Outcome: | The results suggest that models may mirror heteronormative biases and prejudice against interracial romantic relationships in human and society. |
What do Large Language Models Learn about Scripts? (2022.starsem-1)
Copied to clipboard
| Challenge: | Script Knowledge is important for language understanding but expensive to produce manually and difficult to induce from text due to reporting bias. |
| Approach: | They propose a pipeline-based script induction framework which can generate good quality ESDs for unseen scenarios. |
| Outcome: | The proposed framework produces good quality ESDs for unseen scenarios, but manual evaluation shows there is room for improvement. |
Multi-Style Transfer with Discriminative Feedback on Disjoint Corpus (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing methods for style transfer require joint annotations across all stylistic dimensions, limiting their application to multiple styles. |
| Approach: | They initialize an encoder-decoder setup with transformer-based language model pre-trained on a generic corpus and enhance its re-writing capability to multiple target style dimensions by employing multiple style-aware language models as discriminators. |
| Outcome: | The proposed model can control styles across multiple style dimensions while preserving content of the input text. |
On the Mutual Influence of Gender and Occupation in LLM Representations (2025.acl-long)
Copied to clipboard
| Challenge: | We examine LLM representations of gender for first names in various occupational contexts to study how occupations and the gender perception of first names influence each other mutually. |
| Approach: | They examine LLM representations of gender for first names in various occupational contexts and examine how occupations and the gender perception of first names influence each other mutually. |
| Outcome: | The representations shift with the occupational context and are influenced by stereotypically feminine or masculine occupations. |
Post-Hoc Answer Attribution for Grounded and Trustworthy Long Document Comprehension: Task, Insights, and Challenges (2024.starsem-1)
Copied to clipboard
| Challenge: | Existing work on attribution of answer text to source document is limited . existing systems are prone to generating answers lacking sufficient grounding to knowledge sources . |
| Approach: | They propose to use existing datasets to assess the strengths and weaknesses of existing systems for this task. |
| Outcome: | The proposed system is based on retrieval-based and textual entailment-based optimal selection attribution systems. |
Less Mature is More Adaptable for Sentence-level Language Modeling (2025.acl-long)
Copied to clipboard
| Challenge: | Existing studies fine-tune encoders or contrastive learning approaches to learn sentence representations. |
| Approach: | They propose to use sentence-level models to study how sentence representations influence downstream task performance. |
| Outcome: | The proposed models outperform token-level models in terms of time and data efficiency. |
CaM-Gen: Causally Aware Metric-Guided Text Generation (2022.findings-acl)
Copied to clipboard
| Challenge: | Content is created for a well-defined purpose, often described by a metric or signal . external metrics and content tend to have inherent relationships and not all of them may be of consequence. |
| Approach: | They propose a mechanism to guide generative models by user-defined target metrics . authors propose generative networks guided by causally significant aspects of text . |
| Outcome: | The proposed models beat baselines in terms of the target metric control while maintaining fluency and language quality of the generated text. |
What to Read in a Contract? Party-Specific Summarization of Legal Obligations, Entitlements, and Prohibitions (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing systems that generate section-wise summaries of contracts can be tedious due to length and complexity of legalese. |
| Approach: | They propose a task of party-specific extractive summarization for legal contracts . they train a pairwise importance ranker and propose incorporating domain-specific notions of importance . |
| Outcome: | The proposed system generates a party-specific contract summary using a dataset of lease agreements and lease agreements. |
Agent-Specific Deontic Modality Detection in Legal Language (2022.emnlp-main)
Copied to clipboard
| Challenge: | a contract is a legal document executed by two or more parties. |
| Approach: | They propose a corpus of English contracts annotated with deontic modalities . they benchmark this corpus on agent-specific deontics and trigger span detection . |
| Outcome: | The proposed dataset can detect red flags with high recall. |