English
Related papers

Related papers: TRACE: A Metrologically-Grounded Engineering Frame…

200 papers

Evaluating the real-world capabilities of AI systems requires grounding benchmark performance in human-interpretable measures of task difficulty. Existing approaches that rely on direct human task completion time annotations are costly,…

Artificial Intelligence · Computer Science 2026-02-10 Fengyuan Liu , Jay Gala , Nilaksh , Dzmitry Bahdanau , Siva Reddy , Hugo Larochelle

Foundation models excel in stable environments, yet often fail where reliability matters most: medicine, finance, and policy. This Fidelity Paradox is not just a data problem; it is structural. In domains where rules change over time, extra…

Machine Learning · Computer Science 2026-03-27 Steffen Lukas

This paper presents a configuration-first framework for evaluating cross-backend compatibility in deep learning systems deployed on CPU, GPU, and compiled runtimes. The framework decouples experiments from code using YAML, supports both…

Machine Learning · Computer Science 2025-09-10 Zehua Li

We deployed an LLM agent with ReAct reasoning and full data access. It executed flawlessly, yet when asked "Why is completion rate 80%?", it returned metrics instead of causal explanation. The agent knew how to reason but we had not…

Software Engineering · Computer Science 2026-01-09 Bo Yu , Lei Zhao

The rapid advancement of Artificial Intelligence (AI) has created unprecedented demands for computational power, yet methods for evaluating the performance, efficiency, and environmental impact of deployed models remain fragmented. Current…

Performance · Computer Science 2025-10-22 Hongyuan Liu , Xinyang Liu , Guosheng Hu

Legacy systems concentrate business rules, architectural decisions, and operational exceptions that often remain implicit in code, data, configuration, and maintenance practices. At the same time, language-model-based coding agents depend…

Software Engineering · Computer Science 2026-05-19 Sanderson Oliveira de Macedo , Ronaldo Martins da Costa

Large language models (LLMs) are increasingly used in scientific domains. While they can produce reasoning-like content via methods such as chain-of-thought prompting, these outputs are typically unstructured and informal, obscuring whether…

Artificial Intelligence · Computer Science 2025-11-18 Pengze Li , Jiaqi Liu , Junchi Yu , Lihao Liu , Mingyu Ding , Wanli Ouyang , Shixiang Tang , Xi Chen

Large Language Models (LLMs) are increasingly used in agentic systems, where their interactions with diverse tools and environments create complex, multi-stage safety challenges. However, existing benchmarks mostly rely on static,…

Cryptography and Security · Computer Science 2026-02-03 Liming Lu , Xiang Gu , Junyu Huang , Jiawei Du , Xu Zheng , Yunhuai Liu , Yongbin Zhou , Shuchao Pang

The alignment of Large Language Models (LLMs) for multi-turn conversations typically relies on reward signals derived from the content of the text. This approach, however, overlooks a rich, complementary source of signal: the dynamics of…

Computation and Language · Computer Science 2025-11-12 Sian Gooding , Edward Grefenstette

The rapid deployment of large language model (LLM)-based agents introduces a new class of risks, driven by their capacity for autonomous planning, multi-step tool integration, and emergent interactions. It raises some risk factors for…

Multiagent Systems · Computer Science 2025-12-04 Rafflesia Khan , Declan Joyce , Mansura Habiba

Multi-hop Knowledge Graph Question Answering (KGQA) requires coherent reasoning across relational paths, yet existing methods often treat each reasoning step independently and fail to effectively leverage experience from prior explorations,…

Computation and Language · Computer Science 2026-04-14 Yingxu Wang , Jiaxin Huang , Mengzhu Wang , Nan Yin

Complex Verilog Design Problems (CVDP) challenge hardware LLM agents because solving them requires localizing verifier-relevant RTL, testbenches, include paths, and build dependencies inside large repository snapshots, making precise edits,…

Artificial Intelligence · Computer Science 2026-05-22 Zijian Du , Nathaniel Pinckney

In this short paper we address issues related to building multimodal AI systems for human performance support in manufacturing domains. We make two contributions: we first identify challenges of participatory design and training of such…

Human-Computer Interaction · Computer Science 2025-03-24 Elizabeth Anne Watkins , Emanuel Moss , Ramesh Manuvinakurike , Meng Shi , Richard Beckwith , Giuseppe Raffa

Artificial intelligence (AI) in its various forms finds more and more its way into complex distributed systems. For instance, it is used locally, as part of a sensor system, on the edge for low-latency high-performance inference, or in the…

Software Engineering · Computer Science 2022-12-29 Hans-Martin Heyn , Eric Knauss , Patrizio Pelliccione

When used in high-stakes settings, AI systems are expected to produce decisions that are transparent, interpretable and auditable, a requirement increasingly expected by regulations. Decision trees such as CART provide clear and verifiable…

Machine Learning · Computer Science 2026-04-07 Vincent Grari , Tim Arni , Thibault Laugel , Sylvain Lamprier , James Zou , Marcin Detyniecki

Optimizing the performance of large language models (LLMs) on large-scale AI training and inference systems requires a scalable and expressive mechanism to model distributed workload execution. Such modeling is essential for pre-deployment…

Distributed, Parallel, and Cluster Computing · Computer Science 2025-11-17 Changhai Man , Joongun Park , Hanjiang Wu , Huan Xu , Srinivas Sridharan , Tushar Krishna

Emerging deployments of Generative AI increasingly execute inference across decentralized and heterogeneous edge devices rather than on a single trusted server. In such environments, a single device failure or misbehavior can disrupt the…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-03-31 Chanh Nguyen , Erik Elmroth

Assessing the reliability of Large Language Models (LLMs) by confidence elicitation is a prominent approach to AI safety in high-stakes applications, such as healthcare and finance. Existing methods either require expensive computational…

Computation and Language · Computer Science 2026-04-08 Zhaohan Zhang , Ziquan Liu , Ioannis Patras

Multi-step reasoning tasks like mathematical problem solving are vulnerable to cascading failures, where a single incorrect step leads to complete solution breakdown. Current LLM routing methods assign entire queries to one model, treating…

Artificial Intelligence · Computer Science 2026-04-16 Vansh Kapoor , Aman Gupta , Hao Chen , Anurag Beniwal , Jing Huang , Aviral Kumar

Test-time scaling improves the reasoning performance of large language models but often results in token-inefficient overthinking, where models continue reasoning beyond what is necessary for a correct answer. Existing dynamic early-exit…

Artificial Intelligence · Computer Science 2026-04-21 Jiakun Li , Xingwei He , Kefan Li , Hongzheng Chai , Hongyue Yu , Yuan Yuan