Papers by Mitesh Khapra

8 papers
Finding Blind Spots in Evaluator LLMs with Interpretable Checklists (2024.emnlp-main)

Copied to clipboard

Challenge: Large Language Models are increasingly relied upon to evaluate text outputs of other LLMs . however, concerns persist over the accuracy of these assessments and the potential for misleading conclusions.
Approach: They propose a framework to assess the reliability of Large Language Models (LLMs) they propose ' FBI' framework to examine the proficiency of Evaluator LLMs in assessing four critical abilities .
Outcome: The proposed framework assesses the performance of LLMs in text generation tasks.
IndicVoices: Towards building an Inclusive Multilingual Speech Dataset for Indian Languages (2024.findings-acl)

Copied to clipboard

Challenge: Using INDICVOICES, we build the first ASR model to support all 22 languages listed in the 8th Schedule of the Constitution of India.
Approach: They propose a dataset of natural and spontaneous speech from 16237 speakers covering 145 Indian districts and 22 languages.
Outcome: The proposed dataset contains 7348 hours of read, extempore and conversational audio from 16237 speakers covering 145 Indian districts and 22 languages.
How Good is Zero-Shot MT Evaluation for Low Resource Indian Languages? (2024.acl-short)

Copied to clipboard

Challenge: a recent study focused on machine translation evaluation for low-resource languages . linguistic aspects that vary across languages are factors that will exacerbate the problem in low-source languages due to the reliance on extensive data resources.
Approach: They propose to use multi-dimensional quality metrics and DA annotations to meta-evaluate MT evaluation metrics for low-resource languages.
Outcome: The proposed evaluation metrics are based on human scores on the candidate translations of assamese, maithili, and Punjabi.
Active Evaluation: Efficient NLG Evaluation with Few Pairwise Comparisons (2022.acl-long)

Copied to clipboard

Challenge: Recent studies show that evaluating NLG systems using pairwise comparisons is expensive as the number of human annotations grows linearly with k.
Approach: They propose a framework to efficiently identify the top-ranked system by actively choosing system pairs for comparison using dueling bandit algorithms.
Outcome: The proposed framework reduces human annotations by 80% on 13 NLG evaluation datasets spanning 5 tasks .
Input-specific Attention Subnetworks for Adversarial Detection (2022.findings-acl)

Copied to clipboard

Challenge: a new method to prune attention heads is proposed for adversarial detection . attention heads in models such as BERT are over-provisioned and can be pruned .
Approach: They propose a method to construct input-specific attention subnetworks from which three features are extracted to discriminate between authentic and adversarial inputs.
Outcome: The proposed method significantly improves state-of-the-art adversarial detection accuracy on 10 NLU datasets with 11 different adversarials.
Aksharantar: Open Indic-language Transliteration datasets and models for the Next Billion Users (2023.findings-emnlp)

Copied to clipboard

Challenge: Indian subcontinent is home to diverse languages written in multiple scripts . widespread use of romanization and lack of standardization means accurate transliteration models form a critical component in the NLP stack for Indian languages used by over 735 million Internet users.
Approach: They propose to build a transliteration dataset using monolingual and parallel corpora and human annotators.
Outcome: The proposed model improves accuracy by 15% on the Dakshina test set and establishes strong baselines on the Aksharantar test set.
OpenHands: Making Sign Language Recognition Accessible with Pose-based Pretrained Models across Languages (2022.acl-long)

Copied to clipboard

Challenge: a new study examines the performance of pretraining for sign language recognition in low-resource settings.
Approach: They propose using pose extracted through pretrained models as the standard modality of data to reduce training time and enable efficient inference.
Outcome: The proposed model reduces training time and allows efficient inference in sign languages.
IndicBART: A Pre-trained Model for Indic Natural Language Generation (2022.findings-acl)

Copied to clipboard

Challenge: IndicBART is a multilingual, sequence-to-sequence pre-trained model focusing on 11 Indic languages and English.
Approach: They present a multilingual sequence-to-sequence pre-trained model for Indic languages . they evaluate it on two NLG tasks: Neural Machine Translation and extreme summarization .
Outcome: The proposed model performs well on low-resource translation scenarios . Script sharing, multilingual training, and better utilization contribute to the performance.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations