internalization

Tag

Cards List
#internalization

How to Train a Real-World Silicon Concierge? Internalizing Complex Business Workflow to Only OneModel

arXiv cs.CL · yesterday Cached

The paper proposes OneModel, a unified AI paradigm that consolidates complex business workflows into a single model's parameters, achieving over 50% latency reduction and improved accuracy in financial services through continual pre-training and logic compilation.

0 favorites 0 likes
#internalization

ICRL: Learning to Internalize Self-Critique with Reinforcement Learning

arXiv cs.AI · 2026-05-18 Cached

This paper introduces ICRL, a framework that jointly trains a solver and critic with reinforcement learning to internalize critique guidance, enabling the solver to improve without external critique. It uses distribution calibration and role-wise group advantage estimation, achieving 6-7 point gains over GRPO on agentic and mathematical reasoning tasks.

0 favorites 0 likes
← Back to home

Submit Feedback