Three Roles, One Model: Role Orchestration at Inference Time to Close the Performance Gap Between Small and Large Agents

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	McClendon, S. Aaron, Gallego-Feliciano, Jorge, Zervoudakis, Stavros, Saravanos, Antonios
Format:	Preprint
Published:	2026
Subjects:	Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Hidden Dynamics of Massive Activations in Transformer Training
by: Gallego-Feliciano, Jorge, et al.
Published: (2025)

The Effect of Warm-Glow on User Behavioral Intention to Adopt Technology: Extending the UTAUT2 Model
by: Saravanos, Antonios, et al.
Published: (2022)

The Opaque Pointer Design Pattern in Python: Towards a Pythonic PIMPL for Modularity, Encapsulation, and Stability
by: Saravanos, Antonios, et al.
Published: (2026)

Reinforcement Learning for Machine Learning Model Deployment: Evaluating Multi-Armed Bandits in ML Ops Environments
by: McClendon, S. Aaron, et al.
Published: (2025)

VR as Library Technology: Early Faculty and Student Feedback on Educational Use of Immersive Technology
by: McClendon, V. J., et al.
Published: (2019)

Waterfall Model Simulation: A Systematic Mapping Study
by: Saravanos, Antonios
Published: (2025)

A Brief History of the Waterfall Model: Past, Present, and Future
by: Saravanos, Antonios
Published: (2025)

How Warm-Glow Alters the Usability of Technology
by: Saravanos, Antonios
Published: (2025)

Educational Media and Technology Yearbook, Volume 34
by: Orey, Michael, Ed., et al.
Published: (2009)

Monocyte‐Mimetic Contrast Agent Enables Targeted and Sensitive Magnetic Resonance Imaging of Atherosclerotic Lesions
by: Joshua Rousseau, et al.
Published: (2025)

Clamor Schürmann's Barngarla grammar: A commentary on the first section of A vocabulary of the Parnkalla language
by: Clendon, Mark
Published: (2015)

Worrorra: a language of the north-west Kimberley coast
by: Clendon, Mark
Published: (2015)

Pseudo Relevance Feedback is Enough to Close the Gap Between Small and Large Dense Retrieval Models
by: Li, Hang, et al.
Published: (2025)

Closing the Gap Between Directed Hopsets and Shortcut Sets
by: Bernstein, Aaron, et al.
Published: (2022)

Closing the Gap: The Role of the Librarian in Online Searching.
by: Dalrymple, Prudence W.
Published: (1984)

Gauging Public Acceptance of Conditionally Automated Vehicles in the United States
by: Saravanos, Antonios, et al.
Published: (2024)

Rosuvastatin: Role in Cardiovascular High-risk Patient
by: John E. Feliciano-Alfonso
Published: (2013)

Characterizing the Role of Similarity in the Property Inferences of Language Models
by: Rodriguez, Juan Diego, et al.
Published: (2024)

Distributed Stochastic Search for Multi-Agent Model Predictive Control
by: Yoon, Taehyun, et al.
Published: (2025)

Closing US Drinking Water Quality Gaps: The Role of Comprehensive Assessment
by: Gregory Pierce, et al.
Published: (2024)

Small Model as Master Orchestrator: Learning Unified Agent-Tool Orchestration with Parallel Subtask Decomposition
by: Yuan, Wenzhen, et al.
Published: (2026)

Tool-Augmented Agent for Closed-loop Optimization,Simulation,and Modeling Orchestration
by: Deng, Liyuan, et al.
Published: (2026)

Inference-Time Refinement Closes the Synthetic-Real Gap in Tabular Diffusion
by: Lomurno, Eugenio, et al.
Published: (2026)

COSMO-Agent: Tool-Augmented Agent for Closed-loop Optimization,Simulation,and Modeling Orchestration
by: Deng, Liyuan, et al.
Published: (2026)

Closing the Performance Gap Between AI and Radiologists in Chest X-Ray Reporting
by: Sharma, Harshita, et al.
Published: (2025)

Towards Closing the Performance Gap for Cryptographic Kernels Between CPUs and Specialized Hardware
by: Zhang, Naifeng, et al.
Published: (2025)

Lost in Inference: Rediscovering the Role of Natural Language Inference for Large Language Models
by: Madaan, Lovish, et al.
Published: (2024)

Enabling Adversarial Robustness in AI Models through Kubeflow MLOps
by: Bouras, Stavros, et al.
Published: (2026)

Intelligent Orchestration of Distributed Large Foundation Model Inference at the Edge
by: Koch, Fernando, et al.
Published: (2025)

Towards Generalized Routing: Model and Agent Orchestration for Adaptive and Efficient Inference
by: Guo, Xiyu, et al.
Published: (2025)

Understanding Performance Gap Between Parallel and Sequential Sampling in Large Reasoning Models
by: Gu, Xiangming, et al.
Published: (2026)

MemFlow: Intent-Driven Memory Orchestration for Small Language Model Agents
by: Chen, Jiayi, et al.
Published: (2026)

Closing the Confidence-Faithfulness Gap in Large Language Models
by: Miao, Miranda Muqing, et al.
Published: (2026)

Emissions and Performance Trade-off Between Small and Large Language Models
by: Garg, Anandita, et al.
Published: (2025)

David helps Goliath: Inference-Time Collaboration Between Small Specialized and Large General Diffusion LMs
by: Han, Xiaochuang, et al.
Published: (2023)

Closing the Gap Between Text and Speech Understanding in LLMs
by: Cuervo, Santiago, et al.
Published: (2025)

Closing the Gap Between Float and Posit Hardware Efficiency
by: Jonnalagadda, Aditya Anirudh, et al.
Published: (2026)

DOVA: Deliberation-First Multi-Agent Orchestration for Autonomous Research Automation
by: Shen, Aaron, et al.
Published: (2026)

Dialect Normalization using Large Language Models and Morphological Rules
by: Dimakis, Antonios, et al.
Published: (2025)

A Safety-Aware Role-Orchestrated Multi-Agent LLM Framework for Behavioral Health Communication Simulation
by: Cho, Ha Na
Published: (2026)