arXiv AI
Jun 30
When Models Know When They Do Not Know: Calibration, Cascading, and Cleaning
arXiv:2601. 07965v2 Announce Type: replace Abstract: When a model knows when it does not know, many possibilities emerge.
Calibration-Aware Uncertainty Cascades for Efficient Heterogeneous Model Collaboration
The paper introduces Calibration-Aware Uncertainty Cascades (CAUC), a post‑hoc framework that calibrates each model’s confidence independently and uses these calibrated scores to decide when to accept an early prediction, invoke a stronger model, or combine outputs. CAUC establishes a common reliability scale across heterogeneous models, decoupling deployment policies from specific model pools or budgets. Experiments on six language benchmarks show a 1.9% relative accuracy gain over strong‑model‑only inference while cutting strong‑model calls by about 47%, and on image classification it maintains or improves performance while reducing GFLOPs by up to 57%.
Related stories
arXiv AI
Jul 24
Routing Without Training: Controllable-Ratio LLM Offloading via Reliability Gating
arXiv:2607. 20481v1 Announce Type: new Abstract: Local-cloud collaboration is a practical way to deploy large language models under resource constraints, but existing methods often rely on trained routers or collaboration-aware finetuning that tie routing behavior to a particular operating regime.
arXiv Machine Learning
Aug 11
ReliableNet: A Chance-Constrained Approach to Trustworthy Classification in Deep Learning
arXiv:2608. 09768v1 Announce Type: new Abstract: A prediction that is both confident and wrong is a critical reliability failure because it can bypass abstention and human review precisely when the model is mistaken.
arXiv AI
Jun 12
Cross-Model Disagreement as a Label-Free Correctness Signal
arXiv:2603. 25450v2 Announce Type: replace Abstract: Detecting when a language model is wrong without ground truth labels is a fundamental challenge for safe deployment.
arXiv AI
Jun 2
CalArena: A Large-Scale Post-Hoc Calibration Benchmark
arXiv:2605. 30188v2 Announce Type: replace-cross Abstract: Reliable probability estimates are critical in many machine learning applications, yet modern classifiers are often poorly calibrated.
arXiv AI
Jun 19
Toward Calibrated Mixture-of-Experts Under Distribution Shift
arXiv:2606. 20544v1 Announce Type: new Abstract: Calibration aligns a model's predictive uncertainty with the frequencies of its empirical outcomes and is important for understanding and trusting reported probabilities.
Hugging Face Trending Papers
Jun 16
Uncertainty Quantification for Flow-Based Vision-Language-Action Models
Vision-language-action models (VLAs) combine vision-language backbones with expressive generative action heads trained via flow matching on large-scale robotic datasets. Despite their strong empirical performance in robotic manipulation, VLAs lack mechanisms to quantify confidence in their predictions and to detect when their actions may be unreliable.
arXiv Machine Learning
Jun 17
Uncertainty Quantification for Flow-Based Vision-Language-Action Models
arXiv:2606. 18043v1 Announce Type: cross Abstract: Vision-language-action models (VLAs) combine vision-language backbones with expressive generative action heads trained via flow matching on large-scale robotic datasets.
Hugging Face Trending Papers
Jul 2
Rethinking Post-Hoc Calibration in Semantic Segmentation
Reliable confidence estimates are essential in semantic segmentation, especially in safety-critical settings where overconfident errors can mislead downstream decisions. Yet modern segmentation models often remain miscalibrated.
arXiv AI
Aug 6
Provable Limits and Certified Deferral for Verbalized Uncertainty in Small Language Models
arXiv:2608. 05064v1 Announce Type: cross Abstract: Small open-weight language models increasingly run in private, offline, and cost-sensitive settings, where the key deployment question is not only what a model answers but when it should defer to a human.
arXiv Machine Learning
Aug 3
MARGIN: Runtime Confidence Calibration for Multi-Agent Foundation Model Coordination
arXiv:2605. 22949v3 Announce Type: replace Abstract: Foundation-model pools are increasingly used as black-box responders in coordinated systems where a coordinator must decide which response to trust.
arXiv Machine Learning
Aug 21
Answer-Level Trust Selection for Physical Vision-Language Reasoning
arXiv:2608. 19807v1 Announce Type: new Abstract: Vision-language models (VLMs) can estimate physical quantities such as duration, speed, and acceleration from visual observations, but existing benchmarks primarily assess overall model performance against annotated ground truth.