7 papers
PatchIsland: Orchestration of LLM Agents for Continuous Vulnerability Repair
Wonyoung Kim, Seunggi Min, Minjae Gwon +11
Continuous fuzzing platforms such as OSS-Fuzz uncover large numbers of vulnerabilities, yet the subsequent repair process remains largely manual. Unfortunately, existing Automated…
Can We Stop Malicious AI? KILLBENCH: A Benchmark for External AI Kill Switch Feasibility
Sechan Lee, Hyounghun Kim, Sangdon Park
Malicious AI causing harm to humans is not just a Hollywood fantasy. Indeed, as highly capable models such as Claude Mythos emerge and agent systems like OpenClaw rapidly spread, t…
Aligning Large Language Models with Procedural Rules: An Autoregressive State-Tracking Prompting for In-Game Trading
Minkyung Kim, Junsik Kim, Woongcheol Yang +2
Large Language Models (LLMs) enable dynamic game interactions but fail to follow essential procedural flows in rule-governed trading systems, eroding player trust. This work resolv…
ATLANTIS: AI-driven Threat Localization, Analysis, and Triage Intelligence System
Taesoo Kim, HyungSeok Han, Soyeon Park +43
We present ATLANTIS, the cyber reasoning system developed by Team Atlanta that won 1st place in the Final Competition of DARPA's AI Cyber Challenge (AIxCC) at DEF CON 33 (August 20…
State-Inference-Based Prompting for Natural Language Trading with Game NPCs
Minkyung Kim, Junsik Kim, Hwidong Bae +3
Large Language Models enable dynamic game interactions but struggle with rule-governed trading systems. Current implementations suffer from rule violations, such as item hallucinat…
Towards Functional Correctness of Large Code Models with Selective Generation
Jaewoo Jeong, Taesoo Kim, Sangdon Park
The hallucination of code generation models hinders their applicability to systems requiring higher safety standards. One critical bottleneck in addressing code hallucination is th…
Unlearn to Relearn Backdoors: Deferred Backdoor Functionality Attacks on Deep Learning Models
Jeongjin Shin, Sangdon Park
Deep learning models are vulnerable to backdoor attacks, where adversaries inject malicious functionality during training that activates on trigger inputs at inference time. Extens…