English
Related papers

Related papers: Tracking Capabilities for Safer Agents

200 papers

Recent advances in AI agents capable of solving complex, everyday tasks, from scheduling to customer service, have enabled deployment in real-world settings, but their possibilities for unsafe behavior demands rigorous evaluation. While…

Artificial Intelligence · Computer Science 2026-02-18 Sanidhya Vijayvargiya , Aditya Bharat Soni , Xuhui Zhou , Zora Zhiruo Wang , Nouha Dziri , Graham Neubig , Maarten Sap

Tracing the sequence of library and system calls that a program makes is very helpful in the characterization of its interactions with the surrounding environment and ultimately of its semantics. Due to entanglements of real-world software…

Cryptography and Security · Computer Science 2024-10-30 Daniele Cono D'Elia , Simone Nicchi , Matteo Mariani , Matteo Marini , Federico Palmaro

AI safety is still largely framed as alignment: training models to follow human preferences, safety policies, and normative constraints. That framing has improved the behavior of modern language models, but aligned behavior does not by…

Artificial Intelligence · Computer Science 2026-05-27 Yige Li , Yunhao Feng , Jun Sun

We propose a novel type system for verifying that programs correctly implement constant-resource behavior. Our type system extends recent work on automatic amortized resource analysis (AARA), a set of techniques that automatically derive…

Programming Languages · Computer Science 2018-01-09 Van Chan Ngo , Mario Dehesa-Azuara , Matthew Fredrikson , Jan Hoffmann

The performance of large language model (LLM) agents depends critically on the execution harness, the system layer that orchestrates tool use, context management, and state persistence. Yet this same architectural centrality makes the…

Cryptography and Security · Computer Science 2026-05-12 Xixun Lin , Yang Liu , Yancheng Chen , Yongxuan Wu , Yucheng Ning , Yilong Liu , Nan Sun , Shun Zhang , Bin Chong , Chuan Zhou , Yanan Cao

AI agents promise to serve as general-purpose personal assistants for their users, which requires them to have access to private user data (e.g., personal and financial information). This poses a serious risk to security and privacy.…

Cryptography and Security · Computer Science 2026-04-22 Robert Stanley , Avi Verma , Lillian Tsai , Konstantinos Kallas , Sam Kumar

When AI agents don't align their actions with human values they may cause serious harm. One way to solve the value alignment problem is by including a human operator who monitors all of the agent's actions. Despite the fact, that this…

Human-Computer Interaction · Computer Science 2023-06-13 Yitzhak Spielberg , Amos Azaria

Autonomous AI agents powered by large language models are being deployed in production with capabilities including shell execution, file system access, database queries, and multi-party communication. Recent red teaming research…

Cryptography and Security · Computer Science 2026-03-19 Saikat Maiti

Sequential multi-agent systems built with large language models (LLMs) can automate complex software tasks, but they are hard to trust because errors quietly pass from one stage to the next. We study a traceable and accountable pipeline,…

Artificial Intelligence · Computer Science 2025-10-10 Amine Barrak

Agentic AI systems present both significant opportunities and novel risks due to their capacity for autonomous action, encompassing tasks such as code execution, internet interaction, and file modification. This poses considerable…

Artificial Intelligence · Computer Science 2025-12-30 Shaun Khoo , Jessica Foo , Roy Ka-Wei Lee

Advances in large language models have enabled agentic AI systems that can reason, plan, and interact with external tools to execute multi-step workflows, while public blockchains have evolved into a programmable substrate for value…

Artificial Intelligence · Computer Science 2026-01-09 Saad Alqithami

Large Language Models produce a controllability gap in safety-critical engineering: even low rates of undetected constraint violations render a system undeployable. Current orchestration paradigms suffer from sycophantic compliance, context…

Artificial Intelligence · Computer Science 2026-05-05 Tianbao Zhang

Agent benchmarks typically report only final outcomes: pass or fail. This threatens evaluation credibility in three ways. First, scores may be inflated or deflated by shortcuts and benchmark artifacts, misrepresenting capability. Second,…

Capturing types in Scala unify static effect and resource tracking with object capabilities, enabling lightweight effect polymorphism with minimal notational overhead. However, their expressiveness has been insufficient for tracking…

Programming Languages · Computer Science 2025-09-10 Yichen Xu , Oliver Bračevac , Cao Nguyen Pham , Martin Odersky

Foundation models have transformed automated code generation, yet autonomous software-engineering agents remain unreliable in realistic development settings. The dominant explanation locates this gap in model capability. We propose a…

Software Engineering · Computer Science 2026-05-14 Hailin Zhong , Shengxin Zhu

Safety-critical system's failure or malfunction can cause loss of human lives or damage to the physical environment; therefore, continuous safety assessment is crucial for such systems. In many domains this includes the use of Safety…

Software Engineering · Computer Science 2023-07-17 Ankit Agrawal , Jane Cleland-Huang

The safety of autonomous AI agents is increasingly recognized as a critical open problem. As agents transition from passive text generators to active actors capable of executing shell commands, modifying files, calling APIs, and browsing…

Artificial Intelligence · Computer Science 2026-05-19 Ashwin Aravind

AI agents, predominantly powered by large language models (LLMs), are vulnerable to indirect prompt injection, in which malicious instructions embedded in untrusted data can trigger dangerous agent actions. This position paper discusses our…

Cryptography and Security · Computer Science 2026-04-01 Chong Xiang , Drew Zagieboylo , Shaona Ghosh , Sanjay Kariyappa , Kai Greshake , Hanshen Xiao , Chaowei Xiao , G. Edward Suh

Corrigibility is a safety property for artificially intelligent agents. A corrigible agent will not resist attempts by authorized parties to alter the goals and constraints that were encoded in the agent when it was first started. This…

Artificial Intelligence · Computer Science 2020-04-06 Koen Holtman

Large language model-based agents are rapidly evolving from simple conversational assistants into autonomous systems capable of performing complex, professional-level tasks in various domains. While these advancements promise significant…