distribution-shift

Tag

Cards List
#distribution-shift

RankShift: In-Database Detection and Explanation of Categorical Shifts

arXiv cs.LG · 3d ago Cached

RankShift is a novel in-database method for detecting and explaining categorical shifts in data streams, using Pearson scores to identify responsible categories, and it shows competitive performance against autoencoders on log datasets.

0 favorites 0 likes
#distribution-shift

When Does Dynamic Ensembling Pay Off? Diagnosing Regionwise Gains in Regression under Distribution Shift

arXiv cs.LG · 2026-08-20 Cached

The paper introduces D^CF5, a diagnostic to predict regionwise gains in dynamic ensembling for regression tasks under distribution shift, validated with high correlation across datasets.

0 favorites 0 likes
#distribution-shift

Dynamic Regime-Aware Conformal Calibration for Reliable Economic Forecast Intervals under Multiple Distribution Shifts

arXiv cs.LG · 2026-08-19 Cached

The paper introduces DRACP, a conformal prediction method that unifies multiple adaptation mechanisms for reliable economic forecast intervals under distribution shifts, achieving calibration with wider intervals.

0 favorites 0 likes
#distribution-shift

Accuracy and Robustness of Model Cascades Under Data Perturbations

arXiv cs.AI · 2026-08-19 Cached

This research paper investigates how data perturbations impact the accuracy and robustness of model cascades for efficient AI inference, identifying key failure modes and stressing the need for evaluation under distribution shift.

0 favorites 0 likes
#distribution-shift

Regime-Conditional Verification: Correctness Estimation for Adapting and Monitoring Safety Classifiers

arXiv cs.AI · 2026-08-17 Cached

This paper introduces Regime-Conditional Verification (RCV), a lightweight wrapper that adapts off-the-shelf safety classifiers for large language models by estimating prediction correctness and detecting distribution shift without retraining.

0 favorites 0 likes
#distribution-shift

From Uncertainty to Failure Attribution: Self-Diagnosing Models for Failure Attribution under Distribution Shift

arXiv cs.LG · 2026-08-11 Cached

The paper introduces self-diagnosing models that attribute model failures under distribution shift, linking uncertainty estimation with failure attribution.

0 favorites 0 likes
#distribution-shift

Invisible Shortcuts: Why Vision Encoders Know Your Camera

Hugging Face Daily Papers · 2026-08-05 Cached

This paper identifies invisible metadata traces at the pixel level as shortcuts that vision encoders exploit, leading to performance degradation under metadata distribution shifts. Mitigation strategies during and after pretraining reduce sensitivity to both targeted and unseen metadata without sacrificing downstream performance.

0 favorites 0 likes
#distribution-shift

Learning the Pareto Frontier of Predictive Models under Distribution Shift

arXiv cs.LG · 2026-08-04 Cached

This paper proposes Frontier Learning, a framework that combines representations and predictions from multiple black-box and white-box pretrained models to construct a unified target-domain representation, guaranteeing performance no worse than any individual reuse baseline under distribution shift. Evaluations on visual domain adaptation and clinical mortality prediction show consistent gains over strong baselines.

0 favorites 0 likes
#distribution-shift

Mitigating Class-Tail Undercoverage in Medical Vision-Language Models under Clinical Shift

arXiv cs.LG · 2026-08-03 Cached

Introduces CALCoDe, a post-hoc reliability layer for frozen medical vision-language models that mitigates class-tail undercoverage under clinical shift, achieving strong worst-class accepted coverage across multiple dermatology shifts and VLM backbones.

0 favorites 0 likes
#distribution-shift

OPERA: Offline Policy-guided Expert Routing and Adaptation for Universal Biomedical Image Analysis

Hugging Face Daily Papers · 2026-07-27 Cached

OPERA proposes a multi-agent ensemble framework that treats expert weight assignment as an offline policy learning problem for universal biomedical image analysis, enabling test-time adaptation without retraining and consistently improving performance across 9 datasets and 30+ baselines.

0 favorites 0 likes
#distribution-shift

Token-Level Off-Policy Learning for Faithful Generation Under Distribution Shift

Hugging Face Daily Papers · 2026-07-20 Cached

Proposes Token-Level Off-Policy Labeling (TOPL), an off-policy training paradigm for faithful generation that reframes post-training as token-level correctness prediction, achieving strong out-of-distribution generalization across summarization and machine translation tasks.

0 favorites 0 likes
#distribution-shift

Scoped Verification for Reliable Long-Horizon Agentic Context Evolution under Distribution Shift

arXiv cs.AI · 2026-07-13 Cached

GRACE uses a typed semantic graph to represent persistent instructions for LLM agents, enabling scoped verification of updates to improve reliability under distribution shift. Experiments on a telecom agent harness show significant improvements in strict reliability over baselines.

0 favorites 0 likes
#distribution-shift

PARA-PV: Physics-Aware Retrieval-Augmented PV Prediction Based on Frozen Foundation Model and Distribution Shift Correction

arXiv cs.AI · 2026-07-10 Cached

Introduces PARA-PV, a physics-aware retrieval-augmented framework for photovoltaic power forecasting that uses a frozen Chronos time-series foundation model and distribution shift correction to improve accuracy and handle peak, ramping, and low-power conditions.

0 favorites 0 likes
#distribution-shift

NEST: Tackling Dataset-Level Distribution Shifts via Regime-Oriented Mixture-of-Experts

arXiv cs.LG · 2026-07-09 Cached

This paper introduces NEST, a framework using a regime-oriented mixture-of-experts to handle dataset-level distribution shifts in time series forecasting, achieving state-of-the-art on various benchmarks.

0 favorites 0 likes
#distribution-shift

Drift Happens: An Empirical Study of Neural Architecture Robustness to Temporal Distribution Shift

arXiv cs.LG · 2026-07-08 Cached

This paper presents an empirical study comparing how different neural architectures (MLPs, CNNs, RNNs, pretrained transformers) degrade under temporal distribution shift across image and text domains, finding that models exploiting localized features degrade fastest while pretrained encoders drift more gradually.

0 favorites 0 likes
#distribution-shift

Domain-Adaptive Climate Downscaling Under Temporal Distribution Shift

arXiv cs.LG · 2026-07-08 Cached

This paper investigates temporal out-of-distribution shift in deep-learning-based climate downscaling and proposes a domain-adaptive framework that combines supervised reconstruction with domain alignment to improve high-resolution climate projections under non-stationary conditions.

0 favorites 0 likes
#distribution-shift

Demonstrating Generalization Failures via Mixtures of Conditional Policies

arXiv cs.AI · 2026-07-07 Cached

This paper proposes a method to construct language models that exhibit controllable generalization failures when trained with reinforcement learning, demonstrating that training success can diverge from generalization in structured ways.

0 favorites 0 likes
#distribution-shift

Loss Smoothing for Stable Adaptation Under Distribution Shift

arXiv cs.LG · 2026-07-02 Cached

Loss smoothing interpolates between source and target objectives during adaptation, preserving useful features while enabling specialization. Experiments across supervised shifts, RL, and language model fine-tuning show consistent improvements.

0 favorites 0 likes
#distribution-shift

ComMem: Complementary Memory Systems for Test-Time Adaptation of Vision-Language Models

arXiv cs.AI · 2026-06-30 Cached

ComMem proposes complementary memory systems inspired by biological memory to improve test-time adaptation of vision-language models, outperforming state-of-the-art on 15 benchmarks.

0 favorites 0 likes
#distribution-shift

When Can Conformal Risk Control Certify LLM Outputs? Bounds, Impossibility, and Adaptation for Structured Generation

arXiv cs.LG · 2026-06-30 Cached

This paper characterizes when conformal risk control can certify structured LLM outputs, proving impossibility bounds and analyzing certification hierarchies across different bounds. Empirical validation on six open-weight models shows that hard configurations are uncertifiable at low risk levels but practical certification is achievable at relaxed targets.

0 favorites 0 likes
Next →
← Back to home

Submit Feedback