An edge-cloud-expert LLM cascade for telecom knowledge systems minimizes processing cost subject to misalignment-risk bounds via multiple hypothesis testing on knowledge and confidence scores.
Overconfidence in LLM-as-a-judge: Diagnosis and confidence- driven solution.arXiv preprint arXiv:2508.06225
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
representative citing papers
VERDI derives three structural confidence signals from decomposed LLM verification traces and calibrates them with Platt-scaled logistic regression to achieve AUROC 0.72-0.91 on GPT models and 0.56-0.70 on Qwen models where log-probabilities fail.
Calibrating the full set of LLM judges with labeled data halves calibration error versus top-5 accuracy selection on RewardBench2 and outperforms on four benchmarks.
A single legitimate request can cause LLM orchestrators to output plans that violate security policies through the composition of benign subtasks, bypassing subtask-level checks.
citing papers explorer
-
Reliable LLM-Based Edge-Cloud-Expert Cascades for Telecom Knowledge Systems
An edge-cloud-expert LLM cascade for telecom knowledge systems minimizes processing cost subject to misalignment-risk bounds via multiple hypothesis testing on knowledge and confidence scores.
-
VERDI: Single-Call Confidence Estimation for Verification-Based LLM Judges via Decomposed Inference
VERDI derives three structural confidence signals from decomposed LLM verification traces and calibrates them with Platt-scaled logistic regression to achieve AUROC 0.72-0.91 on GPT models and 0.56-0.70 on Qwen models where log-probabilities fail.
-
Calibrate, Don't Curate: Label-Efficient Estimation from Noisy LLM Judges
Calibrating the full set of LLM judges with labeled data halves calibration error versus top-5 accuracy selection on RewardBench2 and outperforms on four benchmarks.
-
Semantic Intent Fragmentation: A Single-Shot Compositional Attack on Multi-Agent AI Pipelines
A single legitimate request can cause LLM orchestrators to output plans that violate security policies through the composition of benign subtasks, bypassing subtask-level checks.