Structural uncertainty from self-preference-induced rankings of LLM reasoning paths complements answer dispersion for identifying unreliable instances on logical tasks while collapsing on factual retrieval.
hub
What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
15 Pith papers cite this work. Polarity classification is still indexing.
15
Pith papers citing it
abstract
There are two major types of uncertainty one can model. Aleatoric uncertainty captures noise inherent in the observations. On the other hand, epistemic uncertainty accounts for uncertainty in the model -- uncertainty which can be explained away given enough data. Traditionally it has been difficult to model epistemic uncertainty in computer vision, but with new Bayesian deep learning tools this is now possible. We study the benefits of modeling epistemic vs. aleatoric uncertainty in Bayesian deep learning models for vision tasks. For this we present a Bayesian deep learning framework combining input-dependent aleatoric uncertainty together with epistemic uncertainty. We study models under the framework with per-pixel semantic segmentation and depth regression tasks. Further, our explicit uncertainty formulation leads to new loss functions for these tasks, which can be interpreted as learned attenuation. This makes the loss more robust to noisy data, also giving new state-of-the-art results on segmentation and depth regression benchmarks.
hub tools
citation-role summary
background 1
citation-polarity summary
fields
cs.LG 4 astro-ph.GA 2 cs.RO 2 stat.ML 2 cond-mat.mtrl-sci 1 cs.AI 1 cs.CV 1 cs.SE 1 physics.chem-ph 1roles
background 1polarities
background 1representative citing papers
Face-Feature Tuning is a label-free logit remapping method that reduces FPR/TPR gaps across groups in deepfake detection while preserving overall accuracy.
A neural network trained on simulations infers stripping times for Sagittarius stream stars from phase-space data, measuring a 0.3 dex/Gyr metallicity gradient and estimating ages for globular clusters such as Pal 12 and NGC 2419.
ElemeNet is a unified ML software package for molecular property prediction across elements 1-100 with built-in uncertainty quantification and competitive benchmarks on diverse chemistry datasets.
MoPe propagates historical dynamic posteriors via SE(3) warping and bounded Bayesian fusion to maintain persistent motion state in monocular Gaussian SLAM.
Ensemble Semantic Entropy improves correlation with code correctness over single-model methods and powers a cascading scaling system that cuts FLOPs by 64.9% while preserving performance on LiveCodeBench.
Derives optimal low-rank subspace for Laplace approx in BNNs, provides scalable outperforming version, and new comparison metric.
A deep RL vulnerability-prediction policy trained in semantic embedding space finds up to 23% more unique robot manipulation failures than vision-language baselines and enables more efficient fine-tuning.
TwinSpecNet uses empirical paired learning on spectral twins to denoise low-S/N APOGEE spectra and predict stellar parameters and abundances with lower scatter than the standard pipeline.
A graph-structured framework fuses 3D perception with rule-based, LLM, and memory reasoning to raise hazard coverage from 57% to 93% across 115 simulated underground mine scenarios.
Proposes the VER framework as a diagnostic sequence for identifying explanatory insufficiency in learned representations, distinguishing it from standard errors and shifts.
Mechanistic learning from ML is generically underdetermined in high-dimensional proxy regimes, with LLMs worsening the problem by collapsing many possible explanations into one fluent narrative.
Co-training an SDC and pedestrians with MAPPO yields 78% goal success and 14% collisions versus 35%/33% for rule-based baselines, with jaywalking causing 62% of collisions and evidence of poor anticipation via speed differentials.
A unified taxonomy of uncertainty in ML for physics is introduced together with validation tools such as coverage, calibration, and proper scoring rules, illustrated on regression and classification tasks.
Uncertainty-aware neural networks using Gaussian negative log-likelihood and dropout are applied to predict intrinsic magnetic properties and coercivity via graph neural networks in permanent magnet research.
citing papers explorer
-
Quantifying Consistency in LLM Logical Reasoning via Structural Uncertainty
Structural uncertainty from self-preference-induced rankings of LLM reasoning paths complements answer dispersion for identifying unreliable instances on logical tasks while collapsing on factual retrieval.
-
Toward Calibrated, Fair, and accurate Deepfake Detection
Face-Feature Tuning is a label-free logit remapping method that reduces FPR/TPR gaps across groups in deepfake detection while preserving overall accuracy.
-
Reconstructing the Stripping History of the Sagittarius Stream with Neural Networks
A neural network trained on simulations infers stripping times for Sagittarius stream stars from phase-space data, measuring a 0.3 dex/Gyr metallicity gradient and estimating ages for globular clusters such as Pal 12 and NGC 2419.
-
ElemeNet: Multiscale Molecular Machine Learning with Uncertainty Quantification Across the Periodic Table
ElemeNet is a unified ML software package for molecular property prediction across elements 1-100 with built-in uncertainty quantification and competitive benchmarks on diverse chemistry datasets.
-
MoPe: Motion Permanence for Robust Monocular Gaussian Mapping in Dynamic Environments
MoPe propagates historical dynamic posteriors via SE(3) warping and bounded Bayesian fusion to maintain persistent motion state in monocular Gaussian SLAM.
-
Ensemble-Based Uncertainty Estimation for Code Correctness Estimation
Ensemble Semantic Entropy improves correlation with code correctness over single-model methods and powers a cascading scaling system that cuts FLOPs by 64.9% while preserving performance on LiveCodeBench.
-
Low Rank Based Subspace Inference for the Laplace Approximation of Bayesian Neural Networks
Derives optimal low-rank subspace for Laplace approx in BNNs, provides scalable outperforming version, and new comparison metric.
-
RoboMD: Uncovering Robot Vulnerabilities through Semantic Potential Fields
A deep RL vulnerability-prediction policy trained in semantic embedding space finds up to 23% more unique robot manipulation failures than vision-language baselines and enables more efficient fine-tuning.
-
TwinSpecNet: Extending APOGEE's chemical reach to low-S/N spectra via empirical paired learning
TwinSpecNet uses empirical paired learning on spectral twins to denoise low-S/N APOGEE spectra and predict stellar parameters and abundances with lower scatter than the standard pipeline.
-
From 3D Perception to Safety Reasoning: A Graph-Based Framework for Real-Time Underground Mine Monitoring
A graph-structured framework fuses 3D perception with rule-based, LLM, and memory reasoning to raise hazard coverage from 57% to 93% across 115 simulated underground mine scenarios.
-
Detecting Explanatory Insufficiency in Learned Representations: A Framework for Representational Vigilance
Proposes the VER framework as a diagnostic sequence for identifying explanatory insufficiency in learned representations, distinguishing it from standard errors and shifts.
-
Position: Prioritize Identifying Structure, Not Complex Models, for Scientific Discovery
Mechanistic learning from ML is generically underdetermined in high-dimensional proxy regimes, with LLMs worsening the problem by collapsing many possible explanations into one fluent narrative.
-
Multi-Agent Reinforcement Learning for Safe Autonomous Driving Under Pedestrian Behavioral Uncertainty
Co-training an SDC and pedestrians with MAPPO yields 78% goal success and 14% collisions versus 35%/33% for rule-based baselines, with jaywalking causing 62% of collisions and evidence of poor anticipation via speed differentials.
-
Modelling magnetic material properties with uncertainty-aware neural networks
Uncertainty-aware neural networks using Gaussian negative log-likelihood and dropout are applied to predict intrinsic magnetic properties and coercivity via graph neural networks in permanent magnet research.