TrustAgent‑RFF is a framework that enhances the reliability of radio frequency fingerprinting (RFF) models when faced with domain shifts such as changes in receiver, distance, or environment. It operates on a frozen I/Q classifier and gathers diagnostic evidence—counterfactual stability, prototype consistency, acquisition domain similarity, and signal quality—to estimate prediction reliability. A large language model planner selects diagnostic actions based on evidence, trust state, costs, and budget, while a calibrated decision layer classifies predictions as Accept, Reject, or Defer. Experiments on five RFF datasets show a 38.7% drop in source‑to‑target accuracy and highlight that many incorrect target‑domain predictions remain highly confident.
arXiv:2608. 14089v1 Announce Type: new Abstract: Safety classifiers deployed with large language models often fail for two reasons: their decisions reflect the policy learned during training rather than the deployer's desired policy, and their performance degrades as deployment traffic evolves.
The paper presents a calibrated radio‑frequency fingerprinting approach that handles co‑channel interference from multiple transmitters. By framing the task as a multi‑label classification problem, the authors use a 1D CNN and calibrate confidence thresholds to bound the average number of false negatives, ensuring reliable detection of spectrum violations. Experiments on the POWDER 5G testbed with Wi‑Fi, LTE, and 5G NR signals achieve up to 97% accuracy, with calibrated recall closely matching the specified false‑negative bounds even under out‑of‑distribution interference.
arXiv:2610.26999v1 Announce Type: new Abstract: Deep learning for radio frequency (RF) signal analysis relies on large annotated datasets, yet synthetic datasets trade physical realism for controllab...
arXiv:2606. 06871v1 Announce Type: new Abstract: Diagnosing 802.
arXiv:2508. 00042v2 Announce Type: replace-cross Abstract: Machine learning models deployed in non-stationary environments degrade silently, since as the input distribution drifts their accuracy decays without an error signal and without labels to reveal it.
arXiv:2607. 18278v1 Announce Type: cross Abstract: Calibration is usually evaluated in aggregate, but the most dangerous failures are often local: predictions that remain highly confident despite being wrong.
arXiv:2608. 07899v1 Announce Type: new Abstract: Agent systems increasingly expose execution traces, yet telemetry that reveals a failure may still be inadequate for identifying where that failure originated.
arXiv:2607. 07758v1 Announce Type: new Abstract: Foundation models (FMs) have transformed machine learning from isolated task-specific model development toward general-purpose models pretrained on broad data and adapted to multiple downstream tasks.
arXiv:2606. 01862v1 Announce Type: cross Abstract: Translating user intents into physical radio signals represents the critical yet notoriously tedious final step in wireless prototyping, as it requires intricate knowledge of physical layer details and presents immense implementation challenges.
The paper introduces PLSP (Pre-hoc Liminal Space Profiling), an anticipatory framework for predicting out-of-distribution (OOD) data before inference. It proposes a dataset‑independent metric called the CREDibility Score (CREDS) and introduces credibility curves and heat maps to analyze a model’s maximum credibility and behavior across datasets. Experiments on multiple datasets show that CREDS can improve model robustness to OOD prediction.
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%.
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.