TR

Tiasa Singha Roy

cs.CLcs.AIcs.LGcs.CR

On Valency

published · living versions

W_xnwpaqt6·v1 · currentpublished

Quantifying Claim Smoothing by LLM Research Agents

with Manya Wadhwa, Greg Durrett

1 version

Preprints & journals

7 papers in the corpus · 2021–2026

Evaluating Risks in Weak-to-Strong Alignment: A Bias-Variance Perspective2604.25077v3 · Hamid Osooli, Kareema Batool, Rick Gentry et al.2026 · 0 citationsarXiv

CREATE: Testing LLMs for Associative Creativity2603.09970v2 · Manya Wadhwa, Tiasa Singha Roy, Harvey Lederman et al.2026 · 0 citationsarXiv

Interpreting and Mitigating Unwanted Uncertainty in LLMs2510.22866v1 · Tiasa Singha Roy, Ayush Rajesh Jhaveri, Ilias Triantafyllopoulos2025 · 0 citationsarXiv

RankAug: Augmented data ranking for text classification2311.04535v1 · Tiasa Singha Roy, Priyam Basu2023 · 1 citationarXiv

Interpretability of Fine-grained Classification of Sadness and Depression2203.10432v1 · Tiasa Singha Roy, Priyam Basu, Aman Priyanshu et al.2022 · 2 citationsarXiv

Privacy enabled Financial Text Classification using Differential Privacy and Federated Learning2110.01643v1 · Priyam Basu, Tiasa Singha Roy, Rakshit Naidu et al.2021 · 29 citationsarXiv

Benchmarking Differential Privacy and Federated Learning for BERT Models2106.13973v2 · Priyam Basu, Tiasa Singha Roy, Rakshit Naidu et al.2021 · 32 citationsarXiv

Career total: 9 works. 7 are in this corpus.

Profile built from the corpus for this byline.

Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.