Copertina

Harness-Bench: Measuring Harness Effects across Models in Realistic Agent Workflows

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Yao, Yilun, Tan, Xinyu, Liu, Chao-Hsuan, Li, Yaoming, Wang, Zhengyang, Yu, Wenhan, Tan, Zhewen, Tian, Yuxuan, Zhao, Guangxiang, Sun, Lin, Zhang, Xiangzheng, Yang, Tong
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Artificial Intelligence
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

A Primer in Post-Training Reasoning Data: What We Know About How It Works
di: Li, Yaoming, et al.
Pubblicazione: (2026)

Efficient Switchable Safety Control in LLMs via Magic-Token-Guided Co-Training
di: Si, Jianfeng, et al.
Pubblicazione: (2025)

MemAudit: Post-hoc Auditing of Poisoned Agent Memory via Causal Attribution and Structural Anomaly Detection
di: Tan, Zhewen, et al.
Pubblicazione: (2026)

Beyond Parameter Arithmetic: Sparse Complementary Fusion for Distribution-Aware Model Merging
di: Lin, Weihong, et al.
Pubblicazione: (2026)

Training with Harnesses: On-Policy Harness Self-Distillation for Complex Reasoning
di: Zhao, Zhengyang, et al.
Pubblicazione: (2026)

Uncertainty Under the Curve: A Sequence-Level Entropy Area Metric for Reasoning LLM
di: Zhu, Yongfu, et al.
Pubblicazione: (2025)

Thinking with Reasoning Skills: Fewer Tokens, More Accuracy
di: Zhao, Guangxiang, et al.
Pubblicazione: (2026)

ARC: Active and Reflection-driven Context Management for Long-Horizon Information Seeking Agents
di: Yao, Yilun, et al.
Pubblicazione: (2026)

Code as Agent Harness
di: Ning, Xuying, et al.
Pubblicazione: (2026)

Large Language Models Badly Generalize across Option Length, Problem Types, and Irrelevant Noun Replacements
di: Zhao, Guangxiang, et al.
Pubblicazione: (2025)

TriPlay-RL: Tri-Role Self-Play Reinforcement Learning for LLM Safety Alignment
di: Tan, Zhewen, et al.
Pubblicazione: (2026)

Harness-1: Reinforcement Learning for Search Agents with State-Externalizing Harnesses
di: Jiang, Pengcheng, et al.
Pubblicazione: (2026)

Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses
di: Lin, Jiahang, et al.
Pubblicazione: (2026)

Transformer Meets Twicing: Harnessing Unattended Residual Information
di: Abdullaev, Laziz, et al.
Pubblicazione: (2025)

General Modular Harness for LLM Agents in Multi-Turn Gaming Environments
di: Zhang, Yuxuan, et al.
Pubblicazione: (2025)

HarnessForge: Joint Harness and Policy Evolution for Adaptive Agent Systems
di: Chen, Mingju, et al.
Pubblicazione: (2026)

FinHarness: An Inline Lifecycle Safety Harness for Finance LLM Agents
di: Jia, Haoxuan, et al.
Pubblicazione: (2026)

UserHarness: Harnessing User Minds for Stronger Agent Theory-of-Mind
di: Qian, Cheng, et al.
Pubblicazione: (2026)

Harness Updating Is Not Harness Benefit: Disentangling Evolution Capabilities in Self-Evolving LLM Agents
di: Lin, Minhua, et al.
Pubblicazione: (2026)

EpiBench: Benchmarking Multi-turn Research Workflows for Multimodal Agents
di: Dong, Xuan, et al.
Pubblicazione: (2026)

Auditing Agent Harness Safety
di: Liu, Chengzhi, et al.
Pubblicazione: (2026)

Natural-Language Agent Harnesses
di: Pan, Linyue, et al.
Pubblicazione: (2026)

Harnessing LLMs for Automated Video Content Analysis: An Exploratory Workflow of Short Videos on Depression
di: Liu, Jiaying Lizzy, et al.
Pubblicazione: (2024)

DIffSteISR: Harnessing Diffusion Prior for Superior Real-world Stereo Image Super-Resolution
di: Zhou, Yuanbo, et al.
Pubblicazione: (2024)

Harnessing RLHF for Robust Unanswerability Recognition and Trustworthy Response Generation in LLMs
di: Lin, Shuyuan, et al.
Pubblicazione: (2025)

Harnessing stimuli‐responsive biomaterials for advanced biomedical applications
di: Ziming Liao, et al.
Pubblicazione: (2024)

Formal Skill: Programmable Runtime Skills for Efficient and Accurate LLM Agents
di: Zhang, Xi, et al.
Pubblicazione: (2026)

HarnessAgent: Scaling Automatic Fuzzing Harness Construction with Tool-Augmented LLM Pipelines
di: Yang, Kang, et al.
Pubblicazione: (2025)

Symbiotic Cooperation for Web Agents: Harnessing Complementary Strengths of Large and Small LLMs
di: Zhang, Ruichen, et al.
Pubblicazione: (2025)

Harnessing Zn-Volatility for Compositional Tuning in PtZn Nanoalloy Catalysts
di: Yao, Bingqing, et al.
Pubblicazione: (2024)

Harnessing LLM Agents with Skill Programs
di: Liu, Hongjun, et al.
Pubblicazione: (2026)

Harnessing Croconaine Organic Photosensitizers for a Milder Surface‐Mediated Transfection
di: Ferdinandus, et al.
Pubblicazione: (2024)

CAX-Agent: A Lightweight Agent Harness for Reliable APDL Automation
di: Lin, Chenying, et al.
Pubblicazione: (2026)

Harnessing External Irradiation for Precise Activation of Metal‐Based Agents in Cancer Therapy
di: Yuting Wang, et al.
Pubblicazione: (2024)

Large-Scale Optimization Model Auto-Formulation: Harnessing LLM Flexibility via Structured Workflow
di: Liang, Kuo, et al.
Pubblicazione: (2026)

Denoise to Track: Harnessing Video Diffusion Priors for Robust Correspondence
di: Yuan, Tianyu, et al.
Pubblicazione: (2025)

VeRO: An Evaluation Harness for Agents to Optimize Agents
di: Ursekar, Varun, et al.
Pubblicazione: (2026)

Router Upcycling: Leveraging Mixture-of-Routers in Mixture-of-Experts Upcycling
di: Ran, Junfeng, et al.
Pubblicazione: (2025)

Harnessing the polar vortex motion in oxide heterostructures
di: Gupta, Pushpendra, et al.
Pubblicazione: (2025)

Harnessing the Hofmeister Effect for Dynamic Self‐Assembly of Supramolecular Hydrogels
di: Hongwang Tang, et al.
Pubblicazione: (2025)