arXiv AI
Jun 16
Auditing Reward Hackability in Code RL Training Environments
arXiv:2606. 16062v1 Announce Type: new Abstract: We measure the rate at which code RL environments accept incorrect solutions as correct.
Did We Actually Fix It? An Independent Adversarial Stress-Test of Post-Point-Adjustment Evaluation Metrics for Time-Series Anomaly Detection
arXiv:2607. 11969v1 Announce Type: cross Abstract: Point-adjustment (PA), long the default scoring protocol in time-series anomaly detection (TSAD), was shown by Kim et al.
Related stories
arXiv Machine Learning
Jun 10
What Demonstration Curation Metrics Do to Your Policy
arXiv:2606. 10229v1 Announce Type: cross Abstract: We study whether demonstration-curation metrics that detect defective training episodes also improve the downstream behavior-cloning policy that trains on the curated data.
arXiv AI
Jun 19
Calibration Without Comprehension: Diagnosing the Limits of Fine-Tuning LLMs for Vulnerability Detection in Systems Software
arXiv:2606. 20502v1 Announce Type: cross Abstract: Whether LLMs scoring well on vulnerability benchmarks genuinely reason about security or merely pattern-match on contaminated data remains unresolved.
arXiv AI
Aug 6
Just Repair: A Minimal Denoising Network for Time Series Anomaly Detection
arXiv:2604. 17388v3 Announce Type: replace-cross Abstract: Time series anomaly detectors have grown steadily more complex, incorporating attention mechanisms, adversarial training, and stochastic latent variables.
arXiv AI
Jul 23
Unlearning as Distribution Restoration: A Controlled Counterfactual Study, a Validated Selective Screen, and the Limits of Oracle-Free Certification
arXiv:2607. 19442v1 Announce Type: cross Abstract: Machine unlearning is commonly evaluated by matching a retrained oracle on trained probes.
arXiv Machine Learning
Jun 29
Test-Input Generation for Tensor Programs: What Actually Finds Kernel Bugs
arXiv:2606. 27396v1 Announce Type: cross Abstract: Test-input generation for tensor kernels is folkloric.
arXiv Machine Learning
Jul 14
When the Reward Suite Is Leaky: A Preregistered Causal Contrast of Natural Verifier False Positives in RLVR
arXiv:2607. 11022v1 Announce Type: new Abstract: The test suites used as RLVR rewards for code have natural false positives: per-task, persistent, asymmetric errors that accept the same wrong programs every time they appear, unlike the symmetric or resampled noise assumed by existing noise-robustness analyses.
arXiv Machine Learning
Jul 9
Prior-matched evaluation of operational Earth-observation classifiers: a three-number reporting method demonstrated on Sentinel-1 internal-wave detection
arXiv:2607. 07146v1 Announce Type: new Abstract: The Internal Waves Service screens the Sentinel-1 Wave-mode archive for internal solitary waves, routing detections to experts whose adjudication time is the resource the effort exists to conserve.
arXiv Machine Learning
Aug 11
Quality-Diversity Stress Tests for Process Reward Models:What Archive Coverage Can and Cannot Certify
arXiv:2608. 08008v1 Announce Type: new Abstract: Process reward models (PRMs) score intermediate reasoning steps and are widely used for search, ranking, and training, but optimization can exploit these learned proxies by increasing reward while turning correct reasoning into incorrect reasoning.
arXiv Machine Learning
Jul 21
When Benchmarks Lie: Evaluating Malicious Prompt Classifiers Under True Distribution Shift
arXiv:2602. 14161v2 Announce Type: replace Abstract: Detecting prompt injection, jailbreak attacks, and harmful requests is critical for deploying LLM-based agents safely, yet current evaluation practices in this literature overestimate generalization.
arXiv Machine Learning
Jun 15
Conformal calibration and look-elsewhere effect in anomaly detection for new-physics searches
arXiv:2606. 13780v1 Announce Type: cross Abstract: Machine-learned anomaly detection is reshaping searches for new physics, but it has outrun the statistics used to interpret it.
arXiv Machine Learning
Jun 17
Domain-Validity-Gated Metamorphic Testing of Scientific ML Surrogates
arXiv:2606. 17529v1 Announce Type: cross Abstract: Scientific machine-learning (SciML) surrogates approximate expensive simulations, but exact expected outputs for arbitrary inputs are unavailable (the oracle problem).