Rocktim Das — accepted AI-conference papers

2024

EXAMS-V: A Multi-Discipline Multilingual Multimodal Exam Benchmark for Evaluating Vision Language Models

Rocktim Das, Simeon Hristov, Haonan Li, Dimitar Dimitrov, Ivan Koychev, Preslav Nakov

ACL 2024long

We introduce EXAMS-V, a new challenging multi-discipline multimodal multilingual exam benchmark for evaluating vision language models. It consists of 20,932 multiple-choice questions across 20 school disciplines covering natural science, social science, and other miscellaneous studies, e.g., religio…

Cited by 19 Source ↗PDF ↗code ↗Scholar ↗

2023

DKAF: KB Arbitration for Learning Task-Oriented Dialog Systems with Dialog-KB Inconsistencies

Vishal Saley, Rocktim Das, Dinesh Raghu, Mausam

ACL 2023findings

Task-oriented dialog (TOD) agents often ground their responses on external knowledge bases (KBs). These KBs can be dynamic and may be updated frequently. Existing approaches for learning TOD agents assume the KB snapshot contemporary to each individual dialog is available during training. However, i…

Cited by 0 Source ↗PDF ↗code ↗Scholar ↗