eCoMAE aligns evidential pretraining with observed time-series content
On IMPTC weather-shift tests, OOD accuracy rose from 0.7971 to 0.8068 and OOD AURC fell from 0.0718 to 0.0648.
Stories per day · 30 days
3 RSS
Related3d-reconstruction1claude1computer-use-agents1computer-vision1evidential-learning1leash1masked-pretraining1motion-prediction1multi-robot1occupancy-forecasting1osworld1oversight1probing1qwen1reinforcement-learning1social-navigation1time-series1vggt1
18 stories, newest first · one card per story
Stories tagged #uncertainty
Linear probe predicts VGGT pose failures the model does not report
On HardMatch, where VGGT fails on 35% of 1,000 pairs, a frozen-feature probe reaches 93.5% ROC-AUC.
CoopUQ forecasts occupancy uncertainty for multi-robot navigation
Unity3D and real-robot tests report safer navigation than baselines under occlusion and out-of-distribution crowds.
LEASH: Uncertainty-Aware Weak-to-Strong Escalation and Oversight for Computer-Use Agents
LEASH escalates computer-use agents by latent uncertainty, matching Claude Opus 5 at 19% lower cost.
ProbabilityBench: Model Evaluation and Calibration with Exact Ground-Truth Uncertainty
ProbabilityBench gives exact class probabilities so researchers can measure model calibration and Bayes-error estimates.
In-Context Learning for System Identification: Probabilistic Formulation, Long-Sequence Handling, and Benchmark Validation
A probabilistic Transformer meta-model performs zero-shot system identification on long sequences with calibrated uncertainty.
Accurate but Not Humble: Evaluating Epistemic Humility in LLM Agents under Knowledge Conflict
High-accuracy LLM agents often detect knowledge conflicts but fail to communicate unresolved uncertainty.
Sparsifying Stochasticity, Not Capacity: Partial Stochasticity via Deep Weight Factorization of Prior Scales
A Bayesian network method sparsifies stochastic parameters via prior-scale factorization while matching fully stochastic accuracy.
Effective Resistance and Graph Neural Network Reliability in Tissue-Specific Interactomes
Effective resistance in 24 tissue interactomes mostly tracks inverse degree, limiting GNN reliability signals.
ReCIRC: Rectified Conformal Risk Control
ReCIRC reparameterizes conformal risk control so thresholds target conditional risk while keeping finite-sample marginal guarantees.
A Unified Uncertainty Representation for Graph Neural Networks via Doubly-Spectral Stochastic Expansion
DSS-GNN unifies GNN calibration, OOD detection, and shift robustness in one spectral representation.
Retrainable physics-integrated neural differentiable modeling of sintering across material systems
Sinter-PiNDiff predicts ceramic density and grain growth from sparse data across three material systems.
Pinocchio: Fast Uncertainty Estimates for Black-Box Language Models
Pinocchio introduces an external calibrator for black-box LLM API models with 0.862 AUROC and zero-shot transfer to thirteen unseen models.
People who know the most often sound the least certain
Opinion piece argues AI experts' hedged, qualified speech loses public influence to confident commentators, risking poorly informed AI policy decisions.
1
Moral Entropy: Auditing Bias and Uncertainty in Moral Judgment
Moral Entropy treats moral-label disagreement as a posterior and finds common vote rules badly misclassify items.
A Ranking Approach for Measuring Calibration
Researchers propose rankECE, a ranking-based calibration error measure with theoretical guarantees that outperforms binned ECE approximations.
Kalman Delta Networks: Uncertainty-aware Associative Memory
Researchers propose Kalman Delta Networks, adding Kalman-filter uncertainty tracking to delta-rule linear attention, improving perplexity and accuracy at 750M and 1.3B parameters.
Harnessing CLIP and DINO: An Uncertainty-Aware Cascaded Fusion Network for Generalizable Deepfake Image Detection
UCF-Net fuses CLIP and DINO features with entropy-based uncertainty weighting to improve generalizable deepfake image detection across generators.