CR^2 matches full-information routing performance for device-edge LLM inference using only device-side signals and cuts normalized deployment cost by up to 16.9% at matched accuracy.
Title resolution pending
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
years
2026 5verdicts
UNVERDICTED 5roles
background 1polarities
background 1representative citing papers
SURE-RAG aggregates pair-level claim-evidence relations into interpretable signals for selective RAG answering, reaching 0.9075 Macro-F1 on HotpotQA-RAG v3 while providing auditability and reducing unsafe answers by 37% at 30% coverage.
A data-driven tri-level adaptive robust optimization model with a scalable column-and-constraint generation algorithm jointly optimizes long-term grid configuration and short-term operational mitigation for wildfire ignition uncertainty, validated on synthetic data and a large utility distribution系统
Equivariant conformal prediction contracts non-conformity scores in increasing convex order via group averaging, yielding sharper sets with formal coverage guarantees.
Interval-POMDP shielding supplies runtime safety guarantees for agents whose perception error rates are estimated from finite labeled data, provided the true rates fall inside the learned intervals.
citing papers explorer
-
CR^2: Cost-Aware Risk-Controlled Routing for Wireless Device-Edge LLM Inference
CR^2 matches full-information routing performance for device-edge LLM inference using only device-side signals and cuts normalized deployment cost by up to 16.9% at matched accuracy.
-
SURE-RAG: Sufficiency and Uncertainty-Aware Evidence Verification for Selective Retrieval-Augmented Generation
SURE-RAG aggregates pair-level claim-evidence relations into interpretable signals for selective RAG answering, reaching 0.9075 Macro-F1 on HotpotQA-RAG v3 while providing auditability and reducing unsafe answers by 37% at 30% coverage.
-
Large-Scale Resilience Planning for Wildfire-Prone Electricity-System via Adaptive Robust Optimization
A data-driven tri-level adaptive robust optimization model with a scalable column-and-constraint generation algorithm jointly optimizes long-term grid configuration and short-term operational mitigation for wildfire ignition uncertainty, validated on synthetic data and a large utility distribution系统
-
eCP: Equivariant Conformal Prediction with pre-trained models
Equivariant conformal prediction contracts non-conformity scores in increasing convex order via group averaging, yielding sharper sets with formal coverage guarantees.
-
Interval POMDP Shielding for Imperfect-Perception Agents
Interval-POMDP shielding supplies runtime safety guarantees for agents whose perception error rates are estimated from finite labeled data, provided the true rates fall inside the learned intervals.