René's URL Explorer Experiment


Title: Ayush Maheshwari

Open Graph Title: Ayush Maheshwari

Description: Personal site of ayush maheshwari.

Open Graph Description: Personal site of ayush maheshwari.

Mail addresses
.hakmngmail

Opengraph URL: /

X: @ayushmx

Generator: Wowchemy 5.9.0 for Hugo

direct link

Domain: ayushbits.github.io

Hey, it has json ld scripts: 
{
  "@context": "https://schema.org",
  "@type": "WebSite",
  "potentialAction": {
    "@type": "SearchAction",
    "target": "?q={search_term_string}",
    "query-input": "required name=search_term_string"
  },
  "url": ""
}
NoneIE=edge
authorAyush Maheshwari
theme-color#1565c0
twitter:cardsummary
twitter:creator@ayushmx
twitter:image/media/icon_hu0b7a4cb9992c9ac0e91bd28ffd38dd00_9727_512x512_fill_lanczos_center_3.png
profile:first_nameAyush
profile:last_nameMaheshwari
og:typeprofile
og:site_nameAyush Maheshwari
og:image/media/icon_hu0b7a4cb9992c9ac0e91bd28ffd38dd00_9727_512x512_fill_lanczos_center_3.png
og:localeen-us
og:updated_time2025-12-11T09:00:00+00:00

Links:

https://ayushbits.github.io
Ayush Maheshwarihttps://ayushbits.github.io/
Ayush Maheshwarihttps://ayushbits.github.io/
Homehttps://ayushbits.github.io/#about
Experiencehttps://ayushbits.github.io/#experience
Projectshttps://ayushbits.github.io/#projects
Publicationshttps://ayushbits.github.io/#featured
Contacthttps://ayushbits.github.io/#contact
https://ayushbits.github.io/uploads/cv_Ayush.pdf
https://topmate.io/ayush_iitb
https://ayushbits.github.io
NVIDIAhttps://www.nvidia.com
https://ayushbits.github.io/uploads/cv_Ayush.pdf
https://scholar.google.com/citations?user=7E4Vjm0AAAAJ
https://github.com/ayushbits
https://www.linkedin.com/in/ayushx/
https://twitter.com/ayushmx
https://topmate.io/ayush_iitb
NVIDIAhttps://www.nvidia.com/
NVAITC Indiahttps://resources.nvidia.com/en-us-gps-ai-capacity-building/nvaitc-research
CSE, IIT Bombayhttps://www.cse.iitb.ac.in
Prof. Ganesh Ramakrishnanhttps://www.cse.iitb.ac.in/~ganesh
Ekal Fellowshiphttps://www.ekal.org
UDAANhttps://udaanproject.org
resuméhttps://ayushbits.github.io/uploads/Resume_Ayush.pdf
IndicParam: Benchmark to evaluate LLMs on low-resource Indic langaugeshttps://arxiv.org/abs/2512.00333
LexGen: Domain aware multilingual lexicon generationhttps://arxiv.org/abs/2405.11200
Paperhttps://aclanthology.org/2025.findings-naacl.359/
Paperhttps://arxiv.org/abs/2210.06996
Click here for updates archivehttps://ayushbits.github.io/news
https://nvidia.com
NVIDIAhttps://nvidia.com
IIT Kharagpurhttps://www.iitkgp.ac.in
Slideshttps://github.com/ayushbits/large-scale-ai-lectures
Vizzhy Inchttps://vizzhy.com
https://research.adobe.com
Adobe Researchhttps://research.adobe.com
https://ayushbits.github.io/www.iitb.ac.in
IIT Bombayhttps://ayushbits.github.io/www.iitb.ac.in
https://ayushbits.github.io/tcs.com
Tata Consultancy Serviceshttps://ayushbits.github.io/tcs.com
UDAAN - An NMT pipeline + Post-editing tool to translate document (Best Paper Award at CODS-COMAD 2023)https://udaanproject.org
PDF https://ayushbits.github.io/https:/arxiv.org/pdf/2203.01644.pdf
Code https://ayushbits.github.io/https:/github.com/UDAAN-LEAP/leap-pe-tool
Project https://udaanproject.org
Websitehttps://udaanproject.org
https://udaanproject.org
SPEAR - Programmatically label and quickly build training datahttps://decile.org
PDF https://ayushbits.github.io/https:/arxiv.org/abs/2108.00373
Code https://ayushbits.github.io/https:/github.com/decile-team/spear
Websitehttps://decile.org
Followhttps://github.com/decile-team
https://decile.org
Temples of Indiahttps://templesofindia.org
PDF https://ayushbits.github.io/https:/www.aclweb.org/anthology/N18-5010/
Slides https://ayushbits.github.io/slides/example/
Websitehttps://templesofindia.org
Followhttps://twitter.com/tofi_official
Followhttps://instagram.com/tofi_in
https://templesofindia.org
IIT Bombayhttps://www.iitb.ac.in
Best Paper Awardhttps://ayushbits.github.io/publication/udaan-cods/
CODS-COMAD 2023https://cods-comad.in/2023/
Ekal Foundationhttps://www.ekal.org
📅 Book a Session on Topmate → https://topmate.io/ayush_iitb
See all publications →https://ayushbits.github.io/publication/
Google Scholarhttps://scholar.google.com/citations?user=7E4Vjm0AAAAJ
IndicParam: Benchmark to evaluate LLMs on low-resource Indic Languageshttps://ayushbits.github.io/publication/indicparam/
TL;DR: Comprehensive benchmark to evaluate LLMs on low-resource Indic languages. Dataset publicly available on HuggingFace. Addresses critical gap in multilingual NLP evaluation. https://ayushbits.github.io/publication/indicparam/
PDF https://ayushbits.github.io/https:/arxiv.org/abs/2512.00333
Cite https://ayushbits.github.io
Code https://ayushbits.github.io/https:/github.com/ayushbits/IndicParam
Dataset https://ayushbits.github.io/https:/huggingface.co/datasets/bharatgenai/IndicParam
LexGen: Domain-aware Multilingual Lexicon Generationhttps://ayushbits.github.io/publication/lexgen-acl/
TL;DR: Domain-aware multilingual lexicon generation for 6 Indian languages across 8 domains using routing-based architecture. Released benchmark with 75K+ translation pairs. Accepted at ACL Main Conference 2025. https://ayushbits.github.io/publication/lexgen-acl/
PDF https://ayushbits.github.io/https:/arxiv.org/pdf/2405.11200
Cite https://ayushbits.github.io
Code https://ayushbits.github.io/https:/github.com/Atulkmrsingh/lexgen
ARISE: Iterative Rule Induction and Synthetic Data Generation for Text Classificationhttps://ayushbits.github.io/publication/naacl-arise/
TL;DR: ARISE iteratively induces rules and generates synthetic data for text classification via bootstrapping. Outperforms complex methods like contrastive learning across diverse domains and languages. Published at NAACL 2025 Findings. https://ayushbits.github.io/publication/naacl-arise/
PDF https://ayushbits.github.io/https:/aclanthology.org/2025.findings-naacl.359/
Cite https://ayushbits.github.io
DictDis: Dictionary Constrained Disambiguation for Improved NMThttps://ayushbits.github.io/publication/emnlp-dictdis/
TL;DR: DictDis disambiguates between multiple dictionary candidate translations in lexically constrained NMT. Achieves 2-3 BLEU point improvements across regulatory, finance, engineering, and health domains. Published at EMNLP 2024 Findings. https://ayushbits.github.io/publication/emnlp-dictdis/
PDF https://ayushbits.github.io/https:/aclanthology.org/2024.findings-emnlp.643/
Cite https://ayushbits.github.io
Code https://ayushbits.github.io/https:/github.com/ayushbits/dictdis-multi
Sāmayik: A Benchmark and Dataset for English-Sanskrit Translationhttps://ayushbits.github.io/publication/saamayik/
TL;DR: First comprehensive benchmark and dataset for English-Sanskrit machine translation. Addresses critical gap in classical language NLP. Published at LREC-COLING 2024. https://ayushbits.github.io/publication/saamayik/
PDF https://ayushbits.github.io/https:/arxiv.org/pdf/2305.14004
Cite https://ayushbits.github.io
Code https://ayushbits.github.io/https:/github.com/ayushbits/saamayik
https://ayushbits.github.io/publication/udaan-cods/
UDAAN - Machine Learning based Post-Editing tool for Document Translationhttps://ayushbits.github.io/publication/udaan-cods/
TL;DR: A production-ready MT post-editing tool used by 100+ translators to translate technical content into Indian languages. Won Best Paper Award at CODS-COMAD 2023. Impact: First batch of engineering books translated using UDAAN were released by the President of India. https://ayushbits.github.io/publication/udaan-cods/
PDF https://ayushbits.github.io/https:/arxiv.org/abs/2203.01644
Cite https://ayushbits.github.io
Code https://ayushbits.github.io/https:/github.com/UDAAN-LEAP/leap-pe-tool
Project https://www.udaanproject.org
Reweighing auxiliary losses in supervised learninghttps://ayushbits.github.io/publication/reweighing-loss-aaai23/
TL;DR: AMAL learns instance-specific weights using meta-learning to optimally combine auxiliary losses in supervised learning. Provides significant gains in knowledge distillation and rule denoising. Published at AAAI 2023. https://ayushbits.github.io/publication/reweighing-loss-aaai23/
PDF https://ayushbits.github.io/https:/arxiv.org/abs/2202.03250
Cite https://ayushbits.github.io
Semi-Supervised Data Programming with Subset Selection,https://ayushbits.github.io/publication/semi-dp-acl/
TL;DR: SPEAR combines semi-supervised learning with data programming to improve noisy labeling functions. Significantly outperforms state-of-the-art on seven datasets by jointly learning from rules and labeled data. Published at ACL 2021 Findings. https://ayushbits.github.io/publication/semi-dp-acl/
PDF https://ayushbits.github.io/https:/arxiv.org/abs/2008.09887
Cite https://ayushbits.github.io
Code https://ayushbits.github.io/https:/github.com/decile-team/spear
Video https://ayushbits.github.io/https:/youtu.be/ZCm3kO7BT_k?t=3945
Joint Learning of Hyperbolic Label Embeddings for Hierarchical Multi-label Classificationhttps://ayushbits.github.io/publication/joint_learning_hyperbolic/
TL;DR: Jointly learns classifier parameters and hyperbolic label embeddings for hierarchical multi-label classification without assuming known label hierarchy. Achieves state-of-the-art results by capturing hierarchical structure in hyperbolic space. Published at EACL 2021. https://ayushbits.github.io/publication/joint_learning_hyperbolic/
PDF https://ayushbits.github.io/publication/joint_learning_hyperbolic/joint_learning_hyperbolic.pdf
Cite https://ayushbits.github.io
Code https://ayushbits.github.io/https:/github.com/soumyac1999/hyperbolic-label-emb-for-hmc
Poster https://ayushbits.github.io/publication/joint_learning_hyperbolic/Poster_EACL.pdf
Video https://ayushbits.github.io/https:/www.virtual2021.eacl.org/paper_main.1091.html
Tutorial at PReMI 2025 - Beyond Transformers - Deep Dive into Mamba and SSMshttps://ayushbits.github.io/talk/tutorial-at-premi-2025-beyond-transformers-deep-dive-into-mamba-and-ssms/
Tutorial session on state space models (SSMs) and Mamba architecture as alternatives to transformer models for sequence modeling. https://ayushbits.github.io/talk/tutorial-at-premi-2025-beyond-transformers-deep-dive-into-mamba-and-ssms/
Conferencehttps://premi25.iitd.ac.in
CASML 2025 - LLM Development Workshop at IISc Bangalorehttps://ayushbits.github.io/talk/casml-2025-llm-development-workshop-at-iisc-bangalore/
Day 2 workshop on mastering foundations of LLM development on NVIDIA platforms. IndiaAI Official Pre-Summit Event for AI Impact Summit 2026. https://ayushbits.github.io/talk/casml-2025-llm-development-workshop-at-iisc-bangalore/
C-DAC & NVIDIA Foundational Models Lecture Serieshttps://ayushbits.github.io/talk/c-dac-nvidia-foundational-models-lecture-series/
Multi-session hands-on lecture series on foundational models development lifecycle, powered by PARAM Siddhi-AI supercomputer. Delivered session on ‘Deploying Foundational Models - Challenges and Best Practices’. https://ayushbits.github.io/talk/c-dac-nvidia-foundational-models-lecture-series/
Workshop Presentation at NASSCOM NLTF 2024 - Building Indian Language Foundation Modelshttps://ayushbits.github.io/talk/workshop-presentation-at-nasscom-nltf-2024-building-indian-language-foundation-models/
Presented work on building large-scale foundation models for Indian languages, covering data collection, training architecture, and deployment challenges for multilingual LLMs. https://ayushbits.github.io/talk/workshop-presentation-at-nasscom-nltf-2024-building-indian-language-foundation-models/
Invited Tutorial at EDM 2023 - Data Efficient Machine Learning for Educational Content Creationhttps://ayushbits.github.io/talk/invited-tutorial-at-edm-2023-data-efficient-machine-learning-for-educational-content-creation/
Half-day invited tutorial on data-efficient machine learning for educational content creation, featuring the UDAAN translation ecosystem that has translated 100+ technical books across 11+ Indian languages. https://ayushbits.github.io/talk/invited-tutorial-at-edm-2023-data-efficient-machine-learning-for-educational-content-creation/
https://ayushbits.github.io/talk/invited-tutorial-at-edm-2023-data-efficient-machine-learning-for-educational-content-creation/
See all events https://ayushbits.github.io/event/
AIhttps://ayushbits.github.io/tag/ai/
best paperhttps://ayushbits.github.io/tag/best-paper/
C-DAChttps://ayushbits.github.io/tag/c-dac/
Data Efficiencyhttps://ayushbits.github.io/tag/data-efficiency/
Data Programminghttps://ayushbits.github.io/tag/data-programming/
Deep Learninghttps://ayushbits.github.io/tag/deep-learning/
Deploymenthttps://ayushbits.github.io/tag/deployment/
Digital Humanitieshttps://ayushbits.github.io/tag/digital-humanities/
Educational Data Mininghttps://ayushbits.github.io/tag/educational-data-mining/
Foundation Modelshttps://ayushbits.github.io/tag/foundation-models/
Foundational Modelshttps://ayushbits.github.io/tag/foundational-models/
GPUhttps://ayushbits.github.io/tag/gpu/
Indic Languageshttps://ayushbits.github.io/tag/indic-languages/
Information Retrievalhttps://ayushbits.github.io/tag/information-retrieval/
LLMhttps://ayushbits.github.io/tag/llm/
machine learninghttps://ayushbits.github.io/tag/machine-learning/
Machine Translationhttps://ayushbits.github.io/tag/machine-translation/
Multilingual NLPhttps://ayushbits.github.io/tag/multilingual-nlp/
NMThttps://ayushbits.github.io/tag/nmt/
NVIDIAhttps://ayushbits.github.io/tag/nvidia/
Twitter DMhttps://twitter.com/ayushmx
LinkedIn Messagehttps://www.linkedin.com/in/ayushx/
CC BY NC ND 4.0https://creativecommons.org/licenses/by-nc-nd/4.0
https://creativecommons.org/licenses/by-nc-nd/4.0
Wowchemyhttps://wowchemy.com/?utm_campaign=poweredby
open sourcehttps://github.com/wowchemy/wowchemy-hugo-themes
Copy https://ayushbits.github.io
Download https://ayushbits.github.io

Viewport: width=device-width, initial-scale=1

URLs of crawlers that visited me.