Publications

📚 Workshops

2026

Adrita Das, Peiran Jiang, Dantong Zhu, Barnabás Póczos, José Lugo-Martínez.
Understanding Deterministic Diffusion through Reverse Transition Kernels.
ICLR 2026 Workshop (DeLTa).

Paper

Peiran Jiang, Adrita Das, Weifeng Wu, Simran Sodhi, Huaiying Zhang, José Lugo-Martínez.
Design of Phase-Separating Biosystems via Joint Diffusion and Positive-Unlabeled Guidance.
ICLR 2026 Workshop on Generative and Experimental Methods (GEM).

Paper

2025

Adrita Das, Peiran Jiang, Dantong Zhu, Barnabás Póczos, José Lugo-Martínez.
State-Space Architectures for Scalable Diffusion-based 3D Molecule Generation.
NeurIPS 2025 Workshop on Structured Probabilistic Inference & Generative Modeling

Paper

✍️ Blog & Articles

2026

Adrita Das, Dantong Zhu.
Understanding and Fixing Bottlenecks in State Space Models: What Recency and Over-Smoothing Tell Us.
ICLR 2026 Blogpost Track (Poster).

Read the article

📄 Preprints

2026

Adrita Das, Peiran Jiang, Dantong Zhu, Barnabás Póczos, José Lugo-Martínez.
Breaking the Bottlenecks: Scalable Diffusion Models for 3D Molecular Generation Preprint

📑 arXiv• 💻 Code

Under Review

Peiran Jiang, Adrita Das, Weifeng Wu, Dantong Zhu, Huaiying Zhang, José Lugo-Martínez.
OpenPhase: Condition-Aware Exploration of Multicomponent Biosystem Phase-Separating Behavior.

📄 Paper
🧾 Preprint
💻 Code

2023

Chakradhar Guntuboina†, Adrita Das†, Parisa Mollaei, Seongwon Kim, Amir Barati Farimani.
PeptideBERT: A Transformer Language Model for Peptide Property Prediction.
Journal of Physical Chemistry Letters (JPCL), 2023.

📄 Paper

Protein Architects: Generative AI for De Novo Protein Design

GenAI Hackathon – Carnegie Mellon University School of Computer Science (2023)

Developed a generative AI platform for de novo protein engineering using transformer-based protein language models. Implemented iterative masked-token sampling to generate novel protein variants for Alzheimer’s disease drug discovery and built an interactive web application for AI-assisted protein sequence generation.

Technologies: PyTorch • HuggingFace Transformers • Python

GitHub • Demo • Poster

Please do reach out to me through email if you have any questions about my research.
I’d love to talk more!

Return to Main Page