Skip to content
AI Atlas
PaperActive

Calibration-Aware Uncertainty Cascades for Efficient Heterogeneous Model Collaboration

arxiv.org/abs/2609.11446

quality89

Updated 57 min ago · first seen 12 Sept 2026

paper_01M29X34MZH23K38TQJSYTW835

Published
12 Sept 2026
T1 · 57 min ago
arXiv
2609.11446
T1 · 57 min ago
Category
cs.AI
T1 · 57 min ago

Abstract

Heterogeneous model collaboration seeks to exploit the complementary strengths of different models to balance predictive performance and inference cost. Existing approaches typically rely either on trained routers, which tie routing decisions to a fixed task and model pool, or on raw-confidence cascades, whose thresholds lack consistent reliability semantics across heterogeneous models. Consequently, these approaches adapt poorly to changing model pools and deployment budgets. We propose Calibration-Aware Uncertainty Cascades (CAUC), a simple post-hoc framework that independently calibrates each model's confidence and selects deployment policies using validation data. The resulting calibrated confidence scores establish a common reliability scale for accepting an early prediction, invoking a stronger model, or selectively combining model outputs. This unified decision criterion decouples deployment policies from any particular model pool or operating budget. We further show theoretically that calibration gives confidence thresholds an explicit selective-risk interpretation, whereas uncalibrated scores offer no comparable reliability guarantee. Extensive experiments demonstrate that, across six language benchmarks, CAUC achieves an average relative accuracy improvement of 1.9% over strong-model-only inference while avoiding approximately 47% of strong-model calls. On image classification benchmarks, it maintains or improves predictive performance while reducing measured GFLOPs by up to 57%.

Authors 5

Cai Xu, Han Jiang, Wei Zhao, Yilin Zhang, Ying Liu

Specification

Official page

Source:arXiv (Atom API + RSS)T1observed 57 min agohigh

Arxiv announce type
new

Source:arXiv (Atom API + RSS)T1observed 57 min agohigh

arXiv id
2609.11446

Source:arXiv (Atom API + RSS)T1observed 57 min agohigh

Categories
cs.AI

Source:arXiv (Atom API + RSS)T1observed 57 min agohigh

PDF

Source:arXiv (Atom API + RSS)T1observed 57 min agohigh

Primary category
cs.AI

Source:arXiv (Atom API + RSS)T1observed 57 min agohigh

Published
12 Sept 2026

Source:arXiv (Atom API + RSS)T1observed 57 min agohigh

Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →

Provenance

Attributed facts

9

Source tiers

T19

Freshest observation

57 min ago

Conflicts

None