Publications

Nandan Kumar Jha, Brandon Reagen (2026). AERO: Entropy-Guided Attention for Private LLM Inference. Under review.

Cite Code Poster Video Earlier arXiv Press release

Nandan Kumar Jha, Brandon Reagen (2026). Same Architecture, Different Capacity: Optimizer-Induced Spectral Scaling Laws. Under review.

PDF Cite Code Project Blog

Nandan Kumar Jha, Brandon Reagen (2026). NerVE: Nonlinear Eigenspectrum Dynamics in LLM Feed-Forward Networks. In ICLR 2026.

PDF Cite Code Project

Nandan Kumar Jha, Brandon Reagen (2025). Regularizing the Entropy Landscape of Self-Attention: Towards a Soft Inductive Bias in LLMs. In OPT Workshop at NeurIPS 2025.

Cite Workshop OpenReview

Nandan Kumar Jha, Brandon Reagen (2025). Spectral Scaling Laws in Language Models: How Effectively Do Feed-Forward Networks Use Their Latent Space?. In EMNLP 2025.

PDF Cite ICML 2025 AIW Related code

Nandan Kumar Jha, Brandon Reagen (2025). A Random Matrix Theory Perspective on the Learning Dynamics of Multi-head Latent Attention. In HiLD Workshop at ICML 2025.

Nandan Kumar Jha, Brandon Reagen (2025). Entropy-Guided Attention for Private LLMs. In PPAI (AAAI) Workshop 2025.

PDF Cite Code Poster Long video Press release

Nandan Kumar Jha, Brandon Reagen (2024). ReLU's Revival: On the Entropic Overload in Normalization-Free Large Language Models. In ATTRIB (NeurIPS) Workshop.

PDF Cite Code Poster Workshop

Nandan Kumar Jha, Brandon Reagen (2024). DeepReShape: Redesigning Neural Networks for Efficient Private Inference. In TMLR 2024.

PDF Cite Slides Video

Karthik Garimella, Zahra Ghodsi, Nandan Kumar Jha, Siddharth Garg, Brandon Reagen (2023). Characterizing and Optimizing End-to-End Systems for Private Inference. In ASPLOS 2023.

PDF Cite Code Poster

Karthik Garimella, Nandan Kumar Jha, Zahra Ghodsi, Siddharth Garg, Brandon Reagen (2021). CryptoNite: Revealing the Pitfalls of End-to-End Private Inference at Scale. In Arxiv Preprint.

Zahra Ghodsi, Nandan Kumar Jha, Brandon Reagen, Siddharth Garg (2021). Circa: Stochastic ReLUs for Private Deep Learning. In NeurIPS 2021.

PDF Cite Poster

Nandan Kumar Jha, Zahra Ghodsi, Siddharth Garg, Brandon Reagen (2021). DeepReDuce: ReLU Reduction for Fast Private Inference. In ICML 2021 (Spotlight).

PDF Cite Poster Slides ICML video Long video Spotlight slides Press release

Karthik Garimella, Nandan Kumar Jha, Brandon Reagen (2021). Sisyphus: A Cautionary Tale of Using Low-Degree Polynomial Activations in Privacy-Preserving Deep Learning. In (ACM CCS) PPML 2021.

PDF Cite Code Poster PPML Proceeding

Nandan Kumar Jha, Sparsh Mittal, Binod Kumar, Govardhan Mattela (2020). DeepPeep: Exploiting Design Ramifications to Decipher the Architecture of Compact DNNs. In ACM JETC 2020.

Nandan Kumar Jha, Sparsh Mittal (2020). Modeling Data Reuse in Deep Neural Networks by Taking Data-Types into Cognizance. In IEEE TC 2020.

Nandan Kumar Jha, Shreyas Ravishankar, Sparsh Mittal, Arvind Kaushik, Dipan Mandal, Mahesh Chandra (2020). DRACO: Co-Optimizing Hardware Utilization and Performance of DNNs on Systolic Accelerator. In ISVLSI 2020.

PDF Cite Slides Video DOI

Rajat Saini, Nandan Kumar Jha, Bedanta Das, Sparsh Mittal, C Krishna Mohan (2020). ULSAM: Ultra-lightweight subspace attention module for compact convolutional neural networks. In WACV 2020.

PDF Cite Code Poster Slides Video DOI

Nandan Kumar Jha, Rajat Saini, Subhrajit Nag, Sparsh Mittal (2020). E2GC: Energy-efficient group convolution in deep neural networks. In VLSID 2020.

PDF Cite Slides DOI

Nandan Kumar Jha, Sparsh Mittal, Sasikanth Avancha (2019). Data-type Aware Arithmetic Intensity for Deep Neural Networks. In ICCD 2019 (Poster).

PDF Cite Poster DOI

Nandan Kumar Jha, Sparsh Mittal, Govardhan Mattela (2019). The Ramifications of Making Deep Neural Networks Compact. In VLSID 2019.

PDF Cite Slides DOI