11 papers
Did the LLM Answer the Question? A Semantic Reconstruction Game for Jailbreak Evaluation
Liam Davis, Romain Emil-Popescu, Suman Jana
Jailbreak attacks against large language models (LLMs) are evaluated using attack success rate (ASR), which depends on correctly identifying responses that fulfill prohibited reque…
Cost-Efficient Theorem Proving via Agent Orchestration in Program Verification
Shuangjie Yao, Nikolaus Holzer, Mark Paul Santolucito +3
Program verification establishes software correctness through machine-checkable proofs constructed in theorem provers. It's a guarantee especially valuable for code generated by la…
Efficient Branch-and-Bound Testing and Verification of zkVMs
Hideaki Takahashi, Suman Jana, Junfeng Yang
Zero-knowledge virtual machines (zkVMs) enable verifiable execution of general-purpose programs by translating virtual machine semantics into algebraic constraints over execution t…
GIF: Locally Sound Geometric Information Flow Control for LLMs
Adam Storek, Nikolaus Holzer, Zhuo Zhang +1
Large language models increasingly mediate interactions between sensitive data, untrusted inputs, and privileged actions in agentic systems, creating security and privacy risks. Th…
EffiPair: Improving the Efficiency of LLM-generated Code with Differential Execution Feedback
Samira Hajizadeh, Suman Jana
Large language models (LLMs) can generate functionally correct programs that differ substantially in execution efficiency. Existing inference-time optimization methods typically re…
Cross-Model Disagreement as a Label-Free Correctness Signal
Matt Gorbett, Suman Jana
Detecting when a language model is wrong without ground truth labels is a fundamental challenge for safe deployment. Existing approaches rely on a model's own uncertainty -- such a…
Red Teaming Program Repair Agents: When Correct Patches can Hide Vulnerabilities
Simin Chen, Yixin He, Suman Jana +1
LLM-based agents are increasingly deployed for software maintenance tasks such as automated program repair (APR). APR agents automatically fetch GitHub issues and use backend LLMs…
zkFuzz: Foundation and Framework for Effective Fuzzing of Zero-Knowledge Circuits
Hideaki Takahashi, Jihwan Kim, Suman Jana +1
Zero-knowledge (ZK) circuits enable privacy-preserving computations and are central to many cryptographic protocols. Systems like Circom simplify ZK development by combining witnes…
Comment on Revisiting Neural Program Smoothing for Fuzzing
Dongdong She, Kexin Pei, Junfeng Yang +2
MLFuzz, a work accepted at ACM FSE 2023, revisits the performance of a machine learning-based fuzzer, NEUZZ. We demonstrate that its main conclusion is entirely wrong due to severa…
Beyond Accuracy: Evaluating Self-Consistency of Code Large Language Models with IdentityChain
Marcus J. Min, Yangruibo Ding, Luca Buratti +4
Code Large Language Models (Code LLMs) are being increasingly employed in real-life applications, so evaluating them is critical. While the conventional accuracy evaluates the perf…
PatchCURE: Improving Certifiable Robustness, Model Utility, and Computation Efficiency of Adversarial Patch Defenses
Chong Xiang, Tong Wu, Sihui Dai +3
State-of-the-art defenses against adversarial patch attacks can now achieve strong certifiable robustness with a marginal drop in model utility. However, this impressive performanc…