SD
Sandipan Dandapat
cs.CLcs.AIcs.LGcs.CRcs.CYcs.HCcs.MA
On Valency
published · living versionsW_3b8zvz6v·v1 · currentpublished
VIAS: Value-Isolating Activation Steering for Multi-Value Alignment and Compliance for Organisations
with Avni Mittal, Avinash Anand, Dikshant Kukreja, Erik Cambria +1
1 version
Preprints & journals
19 papers in the corpus · 2020–2026Litmus (Re)Agent: A Benchmark and Agentic System for Predictive Evaluation of Multilingual Models2604.08970v1 · Avni Mittal, Shanu Kumar, Sandipan Dandapat et al.2026 · 0 citationsarXiv
Rethinking Tokenization for Rich Morphology: The Dominance of Unigram over BPE and Morphological Alignment2508.08424v3 · Saketh Reddy Vemula, Sandipan Dandapat, Dipti Misra Sharma et al.2025 · 1 citationarXiv
SAGE: A Generic Framework for LLM Safety Evaluation2504.19674v2 · Madhur Jindal, Hari Shrawgi, Parag Agrawal et al.2025 · 3 citationsarXiv
LLM Safety for Children2502.12552v1 · Prasanjit Rath, Hari Shrawgi, Parag Agrawal et al.2025 · 0 citationsarXiv
Socio-Culturally Aware Evaluation Framework for LLM-Based Content Moderation2412.13578v1 · Shanu Kumar, Gauri Kholkar, Saish Mendke et al.2024 · 0 citationsarXiv
Enhancing Zero-shot Chain of Thought Prompting via Uncertainty-Guided Strategy Selection2412.00353v2 · Shanu Kumar, Saish Mendke, Karody Lubna Abdul Rahman et al.2024 · 0 citationsarXiv
DiTTO: A Feature Representation Imitation Approach for Improving Cross-Lingual Transfer2303.02357v1 · Shanu Kumar, Abbaraju Soujanya, Sandipan Dandapat et al.2023 · 1 citationarXiv
Too Brittle To Touch: Comparing the Stability of Quantization and Distillation Towards Developing Lightweight Low-Resource MT Models2210.15184v2 · Harshita Diddee, Sandipan Dandapat, Monojit Choudhury et al.2022 · 2 citationsarXiv
On the Calibration of Massively Multilingual Language Models2210.12265v1 · Kabir Ahuja, Sunayana Sitaram, Sandipan Dandapat et al.2022 · 11 citationsarXiv
"Diversity and Uncertainty in Moderation" are the Key to Data Selection for Multilingual Few-shot Transfer2206.15010v1 · Shanu Kumar, Sandipan Dandapat, Monojit Choudhury2022 · 3 citationsarXiv
Multi Task Learning For Zero Shot Performance Prediction of Multilingual Models2205.06130v1 · Kabir Ahuja, Shanu Kumar, Sandipan Dandapat et al.2022 · 13 citationsarXiv
On the Economics of Multilingual Few-shot Learning: Modeling the Cost-Performance Trade-offs of Machine Translated and Manual Data2205.06350v2 · Kabir Ahuja, Monojit Choudhury, Sandipan Dandapat2022 · 4 citationsarXiv
Beyond Static Models and Test Sets: Benchmarking the Potential of Pre-trained Models Across Tasks and Languages2205.06356v2 · Kabir Ahuja, Sandipan Dandapat, Sunayana Sitaram et al.2022 · 7 citationsarXiv
Multilingual CheckList: Generation and Evaluation2203.12865v3 · Karthikeyan K, Shaily Bhatt, Pankaj Singh et al.2022 · 1 citationarXiv
Predicting the Performance of Multilingual NLP Models2110.08875v1 · Anirudh Srinivasan, Sunayana Sitaram, Tanuja Ganu et al.2021 · 17 citationsarXiv
On the Universality of Deep Contextual Language Models2109.07140v2 · Shaily Bhatt, Poonam Goyal, Sandipan Dandapat et al.2021 · 2 citationsarXiv
GLUECoS : An Evaluation Benchmark for Code-Switched NLP2004.12376v2 · Simran Khanuja, Sandipan Dandapat, Anirudh Srinivasan et al.2020 · 132 citationsarXiv
A New Dataset for Natural Language Inference from Code-mixed Conversations2004.05051v2 · Simran Khanuja, Sandipan Dandapat, Sunayana Sitaram et al.2020 · 26 citationsarXiv
Translating Web Search Queries into Natural Language Questions2002.02631v1 · Adarsh Kumar, Sandipan Dandapat, Sushil Chordia2020 · 9 citationsarXiv
Career total: 96 works. 19 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.