Pith. sign in

REVIEW 23 cited by

In Search of Lost Domain Generalization

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2007.01434 v1 pith:SZ6CYXB7 submitted 2020-07-02 cs.LG stat.ML

Ishaan Gulrajani , David Lopez-Paz This is my paper
classification cs.LGstat.ML
keywords domaingeneralizationalgorithmsmodelselectiondatasetsdomainbedcriteria
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

The goal of domain generalization algorithms is to predict well on distributions different from those seen during training. While a myriad of domain generalization algorithms exist, inconsistencies in experimental conditions -- datasets, architectures, and model selection criteria -- render fair and realistic comparisons difficult. In this paper, we are interested in understanding how useful domain generalization algorithms are in realistic settings. As a first step, we realize that model selection is non-trivial for domain generalization tasks. Contrary to prior work, we argue that domain generalization algorithms without a model selection strategy should be regarded as incomplete. Next, we implement DomainBed, a testbed for domain generalization including seven multi-domain datasets, nine baseline algorithms, and three model selection criteria. We conduct extensive experiments using DomainBed and find that, when carefully implemented, empirical risk minimization shows state-of-the-art performance across all datasets. Looking forward, we hope that the release of DomainBed, along with contributions from fellow researchers, will streamline reproducible and rigorous research in domain generalization.

Discussion (0). Sign in to comment.

Forward citations

Cited by 23 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Quantization Meets OOD: Generalizable Quantization-aware Training from a Flatness Perspective

    cs.CV 2025-08 conditional novelty 7.0 of 10

    Quantization-aware training degrades out-of-distribution accuracy, and a flatness-aware method with gradient-disorder freezing, FQAT, partially recovers it.

  2. Face4FairShifts: A Large Image Benchmark for Fairness and Robust Learning across Visual Domains

    cs.CV 2025-08 conditional novelty 7.0 of 10

    A new face benchmark with four visual domains and fairness-sensitive labels provides larger measured distribution shifts and lower baseline performance than existing fairness datasets.

  3. OpFlow: Learning Opportunity-Conditioned Choice Potentials for Robust OD Flow Prediction

    cs.LG 2026-07 conditional novelty 6.5 of 10

    Robust OD flow prediction comes from learning row-centered exposure-to-choice potentials and reconstructing counts as scale times allocation, not from raw-count supervision.

  4. Exposure is not manifestation: measurement target and output resolution jointly determine which behavioural-faithfulness evaluator wins

    cs.CL 2026-07 conditional novelty 6.0 of 10

    Small hyperbolic models (146M–3B) report 100% creative-seed preference, 90.7% compliance-gap detection, and a selective-gating skeleton–wallpaper memory pilot as a companion-AI stack.

  5. Rethink Domain Generalization in Heterogeneous Sequence MRI Segmentation

    eess.IV 2025-07 conditional novelty 6.0 of 10

    A semi-supervised pretraining method improves cross-sequence pancreas segmentation Dice from 43.55% to 70.39% (NU) and from 35.62% to 66.61% (IH) on the new PancreasDG benchmark.

  6. A Supervised Machine Learning Framework for Multipactor Breakdown Prediction in High-Power Radio Frequency Devices and Accelerator Components: A Case Study in Planar Geometry

    physics.acc-ph 2025-07 conditional novelty 6.0 of 10

    Random Forest and Extra Trees trained on PIC data predict planar multipactor susceptibility maps about as well as a Monte Carlo benchmark, while neural networks generalize more poorly to unseen materials.

  7. Simulate, Refocus and Ensemble: An Attention-Refocusing Scheme for Domain Generalization

    cs.CV 2025-07 conditional novelty 6.0 of 10

    SRE improves CLIP's domain generalization by training an attention-refocuser on simulated target domains and ensembling the most attention-consistent checkpoints.

  8. Subgraph Generation for Generalizing on Out-of-Distribution Links

    cs.LG 2025-07 conditional novelty 6.0 of 10

    FLEX is a generative framework that synthesizes counterfactual subgraphs with a semi-implicit graph VAE and adversarially co-trains a GNN to improve out-of-distribution link prediction.

  9. Harmonizing and Merging Source Models for CLIP-based Domain Generalization

    cs.CV 2025-06 conditional novelty 6.0 of 10

    HAM trains per-domain CLIP encoders, enriches them with confident cross-domain samples, aligns their update directions, and merges them with redundancy trimming, reaching 79.0% average accuracy on five DG benchmarks w...

  10. Moment Alignment: Unifying Gradient and Hessian Matching for Domain Generalization

    cs.LG 2025-06 reject novelty 6.0 of 10

    A unified moment-alignment theory bounds target-domain error by cross-domain differences in loss derivatives, and the new CMA algorithm implements exact gradient and Hessian matching in closed form.

  11. Hidden-Domain Routing for All-Type Audio Deepfake Detection

    cs.SD 2026-08 accept novelty 5.0 of 10

    A router-then-specialist audio deepfake detector, which classifies audio type first and then applies type-specific models and thresholds, achieved 96.10% Macro-F1 and first place on AT-ADD Track2.

  12. OpenEvoShield: Dual Non-Stationary Continual Defense for Open-World Multi-Agent System Attacks

    cs.AI 2026-05 reject novelty 5.0 of 10

    OpenEvoShield claims to defend LLM multi-agent systems against evolving and novel attacks by combining asymmetric-rate continual learning with energy-based OOD detection.

  13. Towards Domain-Generalized Open-Vocabulary Object Detection: A Progressive Domain-invariant Cross-modal Alignment Method

    cs.CV 2026-03 conditional novelty 5.0 of 10

    A progressive curriculum that trains open-vocabulary detectors on low-ambiguity, high-signal cross-modal alignments first improves robustness to visual domain shifts, with modest, test-tuned gains.

  14. Causal Transfer in Medical Image Analysis

    cs.CV 2026-03 accept novelty 5.0 of 10

    Causal Transfer Learning unifies structural causal models, invariant risk minimisation and counterfactuals with transfer learning to produce domain-robust medical image models.

  15. $\Delta \mathrm{Energy}$: Optimizing Energy Change During Vision-Language Alignment Improves both OOD Detection and OOD Generalization

    cs.CV 2025-10 reject novelty 5.0 of 10

    ΔEnergy, an energy-change OOD score for CLIP, and its EBM fine-tuning loss simultaneously improve OOD detection and covariate-shift generalization.

  16. Domain-Shift-Aware Conformal Prediction for Large Language Models

    stat.ML 2025-10 reject novelty 5.0 of 10

    DS-CP reweights calibration scores via embedding-based density ratios to improve conformal coverage under domain shift, but its stated guarantee requires a weight condition that the default configuration violates.

  17. MorphGen: Morphology-Guided Representation Learning for Robust Single-Domain Generalization in Histopathological Cancer Classification

    cs.CV 2025-08 conditional novelty 5.0 of 10

    MorphGen uses supervised contrastive learning to align histopathology images with nuclear masks and applies SWA, reporting improved out-of-domain cancer classification accuracy on CAMELYON17, BCSS, and OCELOT.

  18. Robust Invariant Representation Learning by Distribution Extrapolation

    cs.LG 2025-05 reject novelty 5.0 of 10

    A new IRMv1 penalty based on extrapolated per-sample gradient magnitudes is proposed, with reported gains over IRM baselines on synthetic and vision benchmarks.

  19. Simple Domain Generalization for Strong Pixel-Level Image Tampering Detection in Modern VLMs

    cs.CV 2026-07 reject novelty 4.0 of 10

    A simple domain-generalization training recipe (balanced real/tampered batches, late injection of a companion VLM domain, low learning rate) yields large pixel-level tampering-localization gains on out-of-distribution...

  20. Saving for the future: Enhancing generalization via partial logic regularization

    cs.LG 2025-08 reject novelty 4.0 of 10

    PL-Reg adds a trainable mask and a defined/undefined classification loss to logic-based regularization, improving unknown-class accuracy across GCD, mDG+GCD, and CIL benchmarks.

  21. Calibrated and Robust Foundation Models for Vision-Language and Medical Image Tasks Under Distribution Shift

    cs.CV 2025-07 reject novelty 4.0 of 10

    StaRFM reuses the authors' earlier CalShift penalties, extends them to 3D medical segmentation with patch-wise and voxel-wise variants, and claims large gains that are not consistently supported by the paper's own tables.

  22. Generalizing vision-language models to novel domains: A comprehensive survey

    cs.CV 2025-06 conditional novelty 3.0 of 10

    A survey of VLM generalization literature organized by transferred module, with benchmark tables and a review of multimodal LLMs.

  23. Data Heterogeneity Modeling for Trustworthy Machine Learning

    cs.LG 2025-06 conditional novelty 3.0 of 10

    A survey that frames heterogeneity-aware machine learning as a paradigm spanning data collection, training, evaluation, and deployment, drawing mostly on the authors' prior results.

Pith tools