AI Security · Privacy · Trustworthy Generative Models
I'm a first-year PhD student at Dartmouth in the JASPR Lab, advised by Shawn Shan. I work on the security, privacy, and trustworthiness of generative AI: what models reveal about their training data, whether we can verify what a model learned or unlearned, and how deployed systems fail under attack.
| Generative Models & Training Data | Provenance & Auditing | Security & Privacy of Deployed ML |
|---|---|---|
| Query-efficient black-box extraction of memorized training data from instruction-tuned models and production APIs. | Verifying what a model learned or unlearned. Concept erasure in diffusion models leaves traces of what was removed. | Jailbreaks and prompt injection in multi-agent systems, evaluation blind spots in LLMs, adversarial robustness. |
| Venue | Paper |
|---|---|
| ECCV 2026 U&ME Workshop |
Deletion Scars: Inferring Deleted Artist Styles from Paired Diffusion Models |
| ICML 2026 AI4Law Workshop |
Resistant to Lawyers, Defeated by Disagreement: Evaluation Blindspots in Legal Language Models |
| EACL 2026 LoResLM Workshop |
Making Large Language Models Speak Tulu: Structured Prompting for an Extremely Low-Resource Language |
| NeurIPS 2025 ML for Systems Workshop |
SLMs as Compiler Experts: Auto-Parallelization for Heterogeneous Systems |
| Applied Soft Computing Journal |
KASPER: Adversarially Robust PDF Malware Detection |
| arcane-ml | Distributed training across SSH clusters, Modal, and local multi-GPU setups | PyTorch Modal |
| Cerebrum | LLM training and serving with Mixture-of-Refusals safety routing | FSDP vLLM Kubernetes |
| JurisQwen | Qwen2.5-7B fine-tuned on Indian legal data | LoRA 4-bit Modal |
| KASPER | Adversarially robust PDF malware detection | PyTorch |
| goDFS | Content-addressable distributed file storage | Go |
| Raft3D | 3D printer management on Raft consensus | Go Raft |
| Dartmouth College | PhD, Computer Science · JASPR Lab | 2026 – |
| Google Summer of Code | Contributor, UC OSPO · Billion-scale ANN embedding benchmarks | 2025 |
| Lossfunk | AI Research Intern · Low-resource language modeling | 2025 |
| IIT Indore | Undergraduate Researcher · Adversarially robust malware detection | |
| PES University | B.Tech, Computer Science |