Picture for Raj Dabre

Raj Dabre

Nilekani Centre at AI4Bharat, Indian Institute of Technology Madras, India, National Institute of Information and Communications Technology, Kyoto, Japan, Indian Institute of Technology Bombay, India

TikZero: Zero-Shot Text-Guided Graphics Program Synthesis

Add code
Mar 14, 2025
Viaarxiv icon

IteRABRe: Iterative Recovery-Aided Block Reduction

Add code
Mar 08, 2025
Viaarxiv icon

RomanLens: Latent Romanization and its role in Multilinguality in LLMs

Add code
Feb 11, 2025
Viaarxiv icon

Are Language Models Agnostic to Linguistically Grounded Perturbations? A Case Study of Indic Languages

Add code
Dec 14, 2024
Figure 1 for Are Language Models Agnostic to Linguistically Grounded Perturbations? A Case Study of Indic Languages
Figure 2 for Are Language Models Agnostic to Linguistically Grounded Perturbations? A Case Study of Indic Languages
Figure 3 for Are Language Models Agnostic to Linguistically Grounded Perturbations? A Case Study of Indic Languages
Figure 4 for Are Language Models Agnostic to Linguistically Grounded Perturbations? A Case Study of Indic Languages
Viaarxiv icon

Pralekha: An Indic Document Alignment Evaluation Benchmark

Add code
Nov 28, 2024
Viaarxiv icon

BhasaAnuvaad: A Speech Translation Dataset for 14 Indian Languages

Add code
Nov 07, 2024
Figure 1 for BhasaAnuvaad: A Speech Translation Dataset for 14 Indian Languages
Figure 2 for BhasaAnuvaad: A Speech Translation Dataset for 14 Indian Languages
Figure 3 for BhasaAnuvaad: A Speech Translation Dataset for 14 Indian Languages
Figure 4 for BhasaAnuvaad: A Speech Translation Dataset for 14 Indian Languages
Viaarxiv icon

Cross-Lingual Auto Evaluation for Assessing Multilingual LLMs

Add code
Oct 17, 2024
Viaarxiv icon

WorldCuisines: A Massive-Scale Benchmark for Multilingual and Multicultural Visual Question Answering on Global Cuisines

Add code
Oct 16, 2024
Figure 1 for WorldCuisines: A Massive-Scale Benchmark for Multilingual and Multicultural Visual Question Answering on Global Cuisines
Figure 2 for WorldCuisines: A Massive-Scale Benchmark for Multilingual and Multicultural Visual Question Answering on Global Cuisines
Figure 3 for WorldCuisines: A Massive-Scale Benchmark for Multilingual and Multicultural Visual Question Answering on Global Cuisines
Figure 4 for WorldCuisines: A Massive-Scale Benchmark for Multilingual and Multicultural Visual Question Answering on Global Cuisines
Viaarxiv icon

An Empirical Comparison of Vocabulary Expansion and Initialization Approaches for Language Models

Add code
Jul 08, 2024
Figure 1 for An Empirical Comparison of Vocabulary Expansion and Initialization Approaches for Language Models
Figure 2 for An Empirical Comparison of Vocabulary Expansion and Initialization Approaches for Language Models
Figure 3 for An Empirical Comparison of Vocabulary Expansion and Initialization Approaches for Language Models
Figure 4 for An Empirical Comparison of Vocabulary Expansion and Initialization Approaches for Language Models
Viaarxiv icon

How effective is Multi-source pivoting for Translation of Low Resource Indian Languages?

Add code
Jun 19, 2024
Viaarxiv icon