activity
20242026
collaborators

5 papers

cs.CL2026

Learning Faster with Better Tokens: Parameter-Efficient Vocabulary Adaptation for Specialized Text Summarization

Gunjan Balde, Soumyadeep Roy, Mainack Mondal +1

Large language models pretrained on general-domain corpora often exhibit tokenization inefficiencies when applied to specialized domains. Although continual pretraining for domain…

cs.LG2026

Sample Efficient Active Algorithms for Offline Reinforcement Learning

Soumyadeep Roy, Shashwat Kushwaha, Ambedkar Dukkipati

Offline reinforcement learning (RL) enables policy learning from static data but often suffers from poor coverage of the state-action space and distributional shift problems. This…

cs.CL2025

"Where does it hurt?" -- Dataset and Study on Physician Intent Trajectories in Doctor Patient Dialogues

Tom Röhr, Soumyadeep Roy, Fares Al Mohamad +4

In a doctor-patient dialogue, the primary objective of physicians is to diagnose patients and propose a treatment plan. Medical doctors guide these conversations through targeted q…

cs.CL2025

Evaluation of LLMs in Medical Text Summarization: The Role of Vocabulary Adaptation in High OOV Settings

Gunjan Balde, Soumyadeep Roy, Mainack Mondal +1

Large Language Models (LLMs) recently achieved great success in medical text summarization by simply using in-context learning. However, these recent efforts do not perform fine-gr…

cs.CL2024

Adaptive BPE Tokenization for Enhanced Vocabulary Adaptation in Finetuning Pretrained Language Models

Gunjan Balde, Soumyadeep Roy, Mainack Mondal +1

In this work, we show a fundamental limitation in vocabulary adaptation approaches that use Byte-Pair Encoding (BPE) tokenization scheme for fine-tuning pretrained language models…