ORACLE: Agentic AI Orchestrator Routing Via Adaptive Verifier Calibration Feedback
cs.AI, cs.LG, cs.MA
Submitted: 2026-07-24
Updated: 2026-10-01
Code: https://github.com/ORACLEorg/ORACLE
Terminology
Sources
- AutoMix: Automatically Mixing Language Models
- $\tau^2$-Bench: Evaluating Conversational Agents in a Dual-Control Environment
- Reinforcement Learning with Verifiable yet Noisy Rewards under Imperfect Verifiers
- FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performance
- CONCUR: High-Throughput Agentic Batch Inference of LLM via Congestion-Based Concurrency Control
- Robust KV Cache Management for LLM Serving under Output Token Length Uncertainty
- Aragog: Just-in-Time Model Routing for Scalable Serving of Agentic Workflows
- Hybrid LLM: Cost-Efficient and Quality-Aware Query Routing
- MuxServe: Flexible Spatial-Temporal Multiplexing for Multiple LLM Serving
- Gemini: A Family of Highly Capable Multimodal Models
- SWE-bench: Can Language Models Resolve Real-World GitHub Issues?
- ThunderAgent: A Simple, Fast and Program-Aware Agentic Inference System
- Near-Optimal Online Deployment and Routing for Streaming LLMs
- AlpaServe: Statistical Multiplexing with Model Parallelism for Deep Learning Serving
- Ministral 3
- vLLM Semantic Router: Signal Driven Decision Routing for Mixture-of-Modality Models
- Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces
- RouteLLM: Learning to Route LLMs with Preference Data
- Agentic Reward Modeling: Integrating Human Preferences with Verifiable Correctness Signals for Reliable Reward Systems
- Humanity's Last Exam
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems