This program is tentative and subject to change.
Zero-knowledge proof (ZKP) circuits implemented in programming languages like Circom are fundamental to blockchain and privacy-preserving applications. These codes often suffer from constraint-related issues where constraints fail to accurately specify intended computations. While existing analysis tools have been proposed, they struggle with large-scale circuits containing complex template embeddings. We present ScaleCirc, a novel framework that addresses such limitations through: 1) systematic management of analysis redundancy via circuit deduplication strategies; 2) constrainedness propagation methods leveraging source code semantic information; and 3) a generalizable framework for different circuit analysis tasks. Evaluation on 691 real-world circuits shows ScaleCirc demonstrates higher efficiency, and successfully analyzes many Circom programs that existing works failed on. We also identified 3 previously unreported under-constrained bugs in the Circomlib library.
This program is tentative and subject to change.
Mon 17 NovDisplayed time zone: Seoul change
14:00 - 15:30 | |||
14:00 10mTalk | ScaleCirc: Scaling the Analysis over Circom Circuits Research Papers Jinan Jiang The Hong Kong Polytechnic University, Haoran Qin The Hong Kong Polytechnic University, Xiapu Luo Hong Kong Polytechnic University | ||
14:10 10mTalk | Improving NLSAT for Nonlinear Real Arithmetic Research Papers Zhonghan Wang Institute of Software, Chinese Academy of Sciences Pre-print | ||
14:20 10mTalk | Bridging Natural Language and Formal Specification - Automated Translation of Software Requirements to LTL via Hierarchical Semantics Decomposition Using LLMs Research Papers Zhi Ma Xidian University, Xin-Cheng Wen Harbin Institute of Technology, Zhexin Su Xidian University, Xiao Liang Yu National University of Singapore, Cong Tian Xidian University, Shengchao Qin Xidian University, Mengfei Yang China Academy of Space Technology | ||
14:30 10mTalk | Diagnosing Performance Differences in Model Checkers via Runtime-Guided Problem Generation Research Papers Yibo Dong National University of Singapore, Yicong Xu East China Normal University, Wenjing Deng East China Normal University, Yu Chen Chuzhou University, Xiaoyu Zhang East China Normal University, Jianwen Li East China Normal University, China, Chengyu Zhang Loughborough University, Geguang Pu East China Normal University, China | ||
14:40 10mTalk | VERT: Polyglot Verified Equivalent Rust Transpilation with Large Language Models Research Papers Aidan Z.H. Yang Carnegie Mellon University, Yoshiki Takashima Yale Law School, Brandon Paulsen Amazon, Joey Dodds Amazon, Daniel Kroening Amazon | ||
14:50 10mTalk | Agentic Specification Generator for Move Programs Research Papers Yu-Fu Fu Georgia Institute of Technology, Meng Xu University of Waterloo, Taesoo Kim Georgia Institute of Technology Pre-print | ||
15:00 10mTalk | How Big is the Automaton? Certified Lower Bounds on the Size of Presburger DFAs Research Papers Nicolas Amat ONERA - The French Aerospace Lab, Pierre Ganty IMDEA Software Institute, Spain, Alessio Mansutti IMDEA Software Institute | ||
15:10 10mTalk | Non-termination Witnesses and their Validation Research Papers Zsófia Ádám Department of Measurement and Information Systems, Budapest University of Technology and Economics, Paulína Ayaziová Masaryk University, Czechia, Levente Bajczi Budapest University of Technology and Economics, Dirk Beyer LMU Munich, Marek Jankola LMU Munich, Marian Lingsch-Rosenfeld LMU Munich, Jan Strejcek Masaryk University | ||
15:20 10mTalk | PAT-Agent: Autoformalization for Model Checking Research Papers Xinyue Zuo National University of Singapore, Yifan Zhang National University of Singapore, Hongshu Wang National University of Singapore, Yufan Cai National University of Singapore, Zhe Hou Griffith University, Jing Sun School of Computer Science, University of Auckland, Jin Song Dong National University of Singapore | ||