Work

06/2026Pre-printEvery Eval Ever: A Unifying Schema and Community Repository for AI Evaluation ResultsJan Batzner*, Sree Harsha Nelaturu*, Damian Stachura, Anastassia Kornilova, et al.Preprint @ arXiv 2026 [* = equal contribution]
Paper link
06/2026Pre-printEvaluation Cards: An Interpretive Layer for AI Evaluation ReportingAvijit Ghosh, Anka Reuel, ..., Sree Harsha Nelaturu, et al.Preprint @ arXiv 2026
Paper link
05/2026Pre-printWhat Do Evolutionary Coding Agents Evolve?Nico Pelleriti, Sree Harsha Nelaturu, Zhanke Zhou, Zongze Li, Max Zimmer, Bo Han, Sebastian PokuttaPreprint @ arXiv 2026
Paper link
04/2025ConferenceINCLUDE: Evaluating Multilingual Language Understanding with Regional KnowledgeAngelika Romanou, Negar Foroutan, ..., Sree Harsha Nelaturu, et al.Spotlight @ ICLR 2025
Paper link
05/2024ConferenceOn The Fairness Impacts of Hardware Selection in Machine LearningSree Harsha Nelaturu, Nishaanth Kanna Ravichandran, Cuong Tran, Sara Hooker, Ferdinando FiorettoPoster @ ICML 2024
Paper link
2023ConferenceEnd-to-End Learnable Masks With Differentiable IndexingDibyanshu Shekhar, Sree Harsha Nelaturu, Ashwath Shetty, Ilia SucholutskyTiny Papers @ ICLR 2023
Paper link
2019WorkshopAccelerated CNN Training through Gradient ApproximationZiheng Wang, Sree Harsha Nelaturu, Saman AmarsingheWorkshop Paper @ EMC^2 / ISCA 2019
Paper link

2017HackathonPsych.aiMedia enhancement with art and deep learning.
Combining art and deep learning to enable media enhancement of entertainment content. First place winner at Microsoft GAINS AI Hackathon in 2017.
Project link
2021HackathonEndangered TalesComic generation for climate-change awareness.
Automatic comic generation for spreading awareness about climate change and sustainability. Best use of OpenAI API at Stanford Treehacks in 2021.
Project link
AcademicStoryBoxVRVR party game using language modelling.
Party game using VR and language modelling.
Project link
AcademicFair + Efficiency Federated LearningSparse learning for federated settings.
Gradient-based learning leveraging sparsity for centralized and federated learning. The project studies whether sparsity can help address communication bottlenecks and device fairness issues in non-IID federated settings.
AcademicLips sometimes lie: Towards understanding lip reading networksLip-reading robustness analysis with AV-HuBERT.
Class project for High Level Computer Vision, exploring the limitations of AV-HuBERT for lip-reading tasks.
AcademicStudy of effect of Optimizers on Accelerated Backpropagation TechniquesOptimizer behavior under approximate and sparse backpropagation.
Investigation of optimizers on networks trained with approximate and sparse operations for a class on Optimization for Machine Learning.
Project link
AcademicDeep GreetingsGreeting card generation with LSTMs and style transfer.
Using LSTMs and Neural Style Transfer for generating greeting cards.
Project link

Archive 8 items

InvitedArtistic Expression in AI, SRM Universityslides
- slides
InvitedOpen Source in AI, Sahajsoft
No public materials linked.
CommunityAutomatic Mixed Precision (AMP) as a source of stochasticity
No public materials linked.
Cohere LabsAccurate Pruning, Cohere For AI (ML Efficiency Group)
No public materials linked.
CommunitySelf Supervised Learning (Computer Vision Group)
No public materials linked.
CommunityEfficient Attention Mechanism (NLP Group)
No public materials linked.
CommunityLottery Ticket Hypothesis (ML Theory Group)
No public materials linked.
CommunityPresentation on Synthetic Petri Dishslides
- slides