3 papers
cs.LG2026
Pay Only for Disagreement: Certified No-Regression Verdicts for Model Updates with Matching Label-Complexity Bounds
Vishnu Bindu Balachandran
Every production model is updated, by retraining, fine-tuning, quantization, or a silent vendor swap, and each update risks being worse than what it replaced. We formalize update p…
cs.LG2026
Self-Poisoning in Adaptive Out-of-Distribution Detection: A Sharp-Threshold Theory and Certified Label-Free Calibration
Vishnu Bindu Balachandran
Test-time adaptive out-of-distribution (OOD) detectors update a memory bank from the unlabelled stream. We show this adaptation obeys a provable dynamical law. Modelling bank impur…
cs.LG2026
Decodable but Not Detectable: A Leakage Fingerprint for Near-OOD Benchmarks
Vishnu Bindu Balachandran
While auditing a perturbation-based OOD detector on a document benchmark, we recorded an AUROC of 0.326 -- well below the 0.5 chance level. The cause is a benchmark leak: the desig…