Tarun Suresh

I'm a third-year undergraduate student in the Computer Science Department at the University of Illinois, Urbana-Champaign advised by Prof. Sasa Misailovic, Prof. Gagandeep Singh, and Prof. Heng Ji.

Email  /  LinkedIn  /  Google Scholar

I am deeply interested in the intersection of deep learning, systems, and formal methods/programming languages. My research goal is to develop principled techniques that enhance the effectiveness, scalability, and reliability of AI systems, particularly in software development and safety-critical domains.

I'm currently researching:

- Structured Output Generation with Large Language Models (LLMs)

- Embedding Models, Re-Ranking, and Retrieval-Augmented Generation

- LLMs for Software Development

- Reinforcement Learning from Human Feedback (RLHF) and Preference-Based Reinforcement Learning (PBRL)

- Formal Verification and Adversarial Robustness of Deep Neural Networks

We introduce new gradient-based joint optimization algorithms and combine them into a scalable Branch-and-Bound-based verifier RABBit for precisely verifying relational properties defined over Deep Neural Networks, such as robustness against universal adversarial perturbations (UAP).

We present IRS, the first probabilistic approach for 5x faster robustness re-certification of Deep Neural Networks after model compression (pruning, quantization) or fine-tuning

We present the first study of the robustness of existing watermarking techniques on code generated by large language models and propose a parsing-based algorithm that easily removes these watermarks via semantic preserving transformations of the code.

We introduce CoRNStack, a large-scale, high-quality contrastive training dataset for code that spans multiple programming languages. We demonstrate that contrastive training of embedding models using CoRNStack leads to state-of-the-art performance across a variety of code retrieval tasks.

SynCode is a novel framework for the grammar-guided generation of Large Language Models (LLMs) that is scalable to general-purpose programming languages and has soundness and completeness guarantees. SynCode reduces syntax errors by 96-100% for various languages (JSON, Python, Go) and enables 1.5x-10x faster LLM inference than existing approaches.

IterGen is a framework that enables users to define and enforce semantic constraints into LLM output and efficiently backtrack and re-sample during LLM generation until these constraints are satisfied. IterGen improves LLM-generated SQL accuracy by 18% and eliminates LLM privacy leakage.

We develop a method, called TAR, for building tamper-resistant safeguards into open-weight LLMs such that adversaries cannot remove the safeguards even after thousands of steps of fine-tuning. In extensive evaluations and red teaming analyses, we find that our method greatly improves tamper-resistance while preserving benign capabilities.

To address challenges from the risk of reward hacking and the complexity of reinforcement learning during preference-based reinforcement learning, we develop a novel two-step learning method called PRC. The high-level idea is to limit the reinforcement learning agent to optimize over a constrained action space that excludes out-of-distribution state-actions, are unreliable and increase the complexity of the reinforcement learning problem at the second step.

We propose efficient deterministic formal verifiers to speed up DNN re-verification after pruning, quantization, or fine-tuning.

Template from here