中文
相关论文

相关论文: AuditRepairBench: A Paired-Execution Trace Corpus …

200 篇论文

We introduce AuditBench, an alignment auditing benchmark. AuditBench consists of 56 language models with implanted hidden behaviors. Each model has one of 14 concerning behaviors--such as sycophantic deference, opposition to AI regulation,…

计算与语言 · 计算机科学 2026-03-11 Abhay Sheshadri , Aidan Ewart , Kai Fronsdal , Isha Gupta , Samuel R. Bowman , Sara Price , Samuel Marks , Rowan Wang

AI-driven program repair uses AI models to repair buggy software by producing patches. Rapid advancements in AI surely impact state-of-the-art performance of program repair. Yet, grasping this progress requires frequent and standardized…

软件工程 · 计算机科学 2025-10-28 André Silva , Martin Monperrus

Multi-agent systems achieve state-of-the-art outcomes through peer collaboration. However, when an agent in the pipeline silently drops a constraint, the system's final output may look correct even though the reasoning chain was quietly…

Automated Program Repair (APR) struggles with complex logic errors and silent failures. Current LLM-based APR methods are mostly static, relying on source code and basic test outputs, which fail to accurately capture complex runtime…

软件工程 · 计算机科学 2026-04-06 Jiaqing Wu , Tong Wu , Manqing Zhang , Yunwei Dong , Bo Shen

As AI systems are increasingly used to conduct research autonomously, misaligned systems could introduce subtle flaws that produce misleading results while evading detection. We introduce Auditing Sabotage Bench, a benchmark for evaluating…

人工智能 · 计算机科学 2026-04-28 Eric Gan , Aryan Bhatt , Buck Shlegeris , Julian Stastny , Vivek Hebbar

We present ActuBench, a multi-agent LLM pipeline for the automated generation and evaluation of advanced actuarial assessment items aligned with the International Actuarial Association (IAA) Education Syllabus. The pipeline separates four…

人工智能 · 计算机科学 2026-04-23 Jan-Philipp Schmidt

Production deployment of AI coding agents requires fast, reproducible evaluation signals. Existing industrial practices trade off speed and fidelity: online A/B testing takes weeks and risks user experience, shadow deployment yields signals…

软件工程 · 计算机科学 2026-05-12 Smriti Jha , Matteo Paltenghi , Chandra Maddila , Vijayaraghavan Murali , Shubham Ugare , Satish Chandra

Task success can hide process anomalies in real-world agent executions. An agent may pass the final task oracle while still accumulating unresolved ambiguity, unsafe external writes, ignored errors, weakly grounded commitments, or…

人工智能 · 计算机科学 2026-05-29 Yibing Liu , Yangze Liu , Xiaolong Yin , Bin Wang , Chong Zhang , Hao Yin , Zhongyi Han

Empirical claims about autonomous Kubernetes operations agents are largely unfalsifiable. Published work reports observational results without controlled comparisons against an agent-disabled baseline, selection bias is endemic,…

软件工程 · 计算机科学 2026-05-25 Joshua Odmark , Gideon Rubin , Deon van der Vyver

Autonomous driving stacks must pick one trajectory from a multi-modal candidate set; choosing by model confidence ignores safety, traffic-law, and comfort constraints. We present \textsc{RECTOR} (Rule-Enforced Constrained Trajectory…

人工智能 · 计算机科学 2026-05-26 Hadi Hajieghrary , Benedikt Walter , Chaitanya Shinde , Paul Schmitt , Miguel Hurtado

Agent systems often decompose a task across multiple roles, but these roles are typically specified by prompts rather than enforced by access controls. Without enforcement, a team pass rate can mask whether agents actually coordinated or…

We introduce a red-teaming methodology that exposes harder-to-catch attacks for coding-agent monitors, suggesting that current practices may under-elicit attacks and overstate monitor performance. We identify three challenges with current…

密码学与安全 · 计算机科学 2026-05-12 Monika Jotautaitė , Maria Angelica Martinez , Ollie Matthews , Tyler Tracy

Automated Program Repair (APR) agents leverage Large Language Models (LLMs) to autonomously diagnose and fix software bugs through reasoning, planning, and tool use. Despite impressive leaderboard gains on benchmarks such as SWE-bench,…

软件工程 · 计算机科学 2026-05-28 Ira Ceka , Hailie Mitchell , Saurabh Pujar , Luca Buratti , Shyam Ramji , Junfeng Yang , Gail Kaiser , Baishakhi Ray

Modern coding scaffolds turn LLMs into capable software agents, but their ability to follow scaffold-specified instructions remains under-examined, especially when constraints are heterogeneous and persist across interactions. To fill this…

Reproducibility problems that have long affected machine learning and reinforcement learning are now surfacing in agent research: papers compare systems by reported scores while leaving the rollout records behind those scores difficult to…

人工智能 · 计算机科学 2026-05-13 Charlie Masters , Ziyuan Liu , Stefano V. Albrecht

Tool using agents often fail for operational reasons even when language understanding is strong. Common causes include invalid arguments, interface drift, weak recovery, and inefficient retry behavior. We introduce ToolMisuseBench, an…

软件工程 · 计算机科学 2026-04-03 Akshey Sigdel , Rista Baral

Existing Agent benchmarks suffer from two critical limitations: high environment interaction overhead (up to 41\% of total evaluation time) and imbalanced task horizon and difficulty distributions that make aggregate scores unreliable. To…

As large-scale multi-agent systems evolve, the communication protocol layer has become a critical yet under-evaluated factor shaping performance and reliability. Despite the existence of diverse protocols (A2A, ACP, ANP, Agora, etc.),…

人工智能 · 计算机科学 2025-10-28 Hongyi Du , Jiaqi Su , Jisen Li , Lijie Ding , Yingxuan Yang , Peixuan Han , Xiangru Tang , Kunlun Zhu , Jiaxuan You

The rapid progress in Automated Program Repair (APR) has been fueled by advances in AI, particularly large language models (LLMs) and agent-based systems. SWE-Bench is a benchmark designed to evaluate repair systems using real issues mined…

软件工程 · 计算机科学 2026-02-05 Matias Martinez , Xavier Franch

Retrieval-Augmented Generation (RAG) shows promise for enterprise knowledge work, yet it often underperforms in high-stakes decision settings that require deep synthesis, strict traceability, and recovery from underspecified prompts.…

信息检索 · 计算机科学 2026-01-27 Xincheng You , Qi Sun , Neha Bora , Huayi Li , Shubham Goel , Kang Li , Sean Culatana
‹ 上一页 1 2 3 10 下一页 ›