Browse State-of-the-Art › Domain Adaptation

Domain Adaptation

2,400 papers with code · 58 benchmarks · 96 datasets archive 2025-07-28

Computer VisionMethodology

Domain Adaptation is the task of adapting models across domains. This is motivated by the challenge where the test and training datasets fall from different data distributions due to some factor. Domain adaptation aims to build machine learning models that can be generalized into a target domain and dealing with the discrepancy across domain distributions.

Further readings:

  • A Brief Review of Domain Adaptation

( Image credit: Unsupervised Image-to-Image Translation Networks )

Description from the archive archive 2025-07-28.

Benchmarks archive 2025-07-28

58 leaderboard tables shown for this task, 58 with rows (a “benchmark” on this site is a table with at least one row, as on /sota), ordered by row count. “Best model” is the first row in the archive's own order at snapshot; nothing is re-ranked here and metric direction is not recorded in the archive. PwC's Trend sparklines are not in the archive, so that column is omitted. 10 shown of 58 until expanded.

Syntology 7 newer papers were judged to report on this task's benchmarks so far (Claude Sonnet 4.5, reading their tables: a model's judgement, not a result; 0 placed): SYNTHIA-to-Cityscapes (1 judged, 0 placed), Office-Home (2 judged, 0 placed), GTA5 to Cityscapes (1 judged, 0 placed), VisDA2017 (1 judged, 0 placed), Office-Caltech (1 judged, 0 placed), PACS (1 judged, 0 placed). Drawn from 8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read).

DatasetBest model (first row in archive order)PaperCodeSyntologyCompare
Office-31 (40 rows) FFTAT Feature Fusion Transferability Aware Transformer for Unsupervised... code — Compare
SYNTHIA-to-Cityscapes (33 rows) HALO Hyperbolic Active Learning for Semantic Segmentation under Domain Shift code Syntology official (archive's flag): 3 ran · 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 5 unverified Compare
Office-Home (29 rows) SWG Combining inherent knowledge of vision-language models with... code — Compare
GTA5 to Cityscapes (28 rows) HALO Hyperbolic Active Learning for Semantic Segmentation under Domain Shift code Syntology official (archive's flag): 3 ran · 3 ran (of which 0 constructed an object rather than computing a result; 2 with no instrument failure: 2 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 5 unverified Compare
VisDA2017 (28 rows) FFTAT Feature Fusion Transferability Aware Transformer for Unsupervised... code — Compare
ImageCLEF-DA (17 rows) CMKD Unsupervised Domain Adaption Harnessing Vision-Language Pre-training code — Compare
Cityscapes to ACDC (16 rows) SoRA SoRA: Singular Value Decomposed Low-Rank Adaptation for Domain... code — Compare
MNIST-to-USPS (14 rows) FACT FACT: Federated Adversarial Cross Training code — Compare
SVHN-to-MNIST (14 rows) Mean teacher Self-ensembling for visual domain adaptation code — Compare
USPS-to-MNIST (14 rows) FAMCD Unsupervised domain adaptation using feature aligned maximum... — — Compare
SVNH-to-MNIST (9 rows) SRDA (RAN) Learning Smooth Representation for Unsupervised Domain Adaptation code — Compare
MoLane (8 rows) UFLD-SGPCS-ResNet18 CARLANE: A Lane Detection Benchmark for Unsupervised Domain... code — Compare
MuLane (8 rows) UFLD-SGADA-ResNet32 CARLANE: A Lane Detection Benchmark for Unsupervised Domain... code — Compare
Office-Caltech (8 rows) SPL Unsupervised Domain Adaptation via Structured Prediction Based... code — Compare
TuLane (8 rows) UFLD-SGPCS-ResNet32 CARLANE: A Lane Detection Benchmark for Unsupervised Domain... code — Compare
Cityscapes-to-FoggyZurich (6 rows) CoDA CoDA: Instructive Chain-of-Domain Adaptation with Severity-Aware... code — Compare
GTAV+Synscapes to Cityscapes (6 rows) DDB Deliberated Domain Bridging for Domain Adaptive Semantic Segmentation code — Compare
SYNSIG-to-GTSRB (6 rows) DFA-MCD Discriminative Feature Alignment: Improving Transferability of... code Syntology official: harvested, nothing ran · 0 ran · 1 unverified Compare
Cityscapes-to-FoggyDriving (5 rows) CoDA CoDA: Instructive Chain-of-Domain Adaptation with Severity-Aware... code — Compare
GTA5+Synscapes to Cityscapes (5 rows) MSCL Multi-Source Domain Adaptation with Collaborative Learning for... — — Compare
HMDBfull-to-UCF (5 rows) UNITE Unsupervised Video Domain Adaptation with Masked Pre-Training and... code — Compare
MNIST-to-MNIST-M (5 rows) DRANet DRANet: Disentangling Representation and Adaptation Networks for... code Syntology official (archive's flag): 1 ran · 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result Compare
Panoptic SYNTHIA-to-Cityscapes (5 rows) MC-PanDA MC-PanDA: Mask Confidence for Panoptic Domain Adaptation code — Compare
Panoptic SYNTHIA-to-Mapillary (5 rows) MC-PanDA MC-PanDA: Mask Confidence for Panoptic Domain Adaptation code — Compare
UCF --> HMDB (full) (5 rows) UNITE Unsupervised Video Domain Adaptation with Masked Pre-Training and... code — Compare
UCF-to-HMDBfull (5 rows) UNITE Unsupervised Video Domain Adaptation with Masked Pre-Training and... code — Compare
DomainNet (4 rows) SPM Shuffle PatchMix Augmentation with Confidence-Margin Weighted... code — Compare
GTAV to Cityscapes+Mapillary (4 rows) Rein Stronger, Fewer, & Superior: Harnessing Vision Foundation Models... code — Compare
HMDB --> UCF (full) (4 rows) TranSVAE — — — Compare
Synth Digits-to-SVHN (4 rows) DSN (DANN) Domain Separation Networks code Syntology community repositories only · 10 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 5 where Syntology's instrument failed) · 4 unverified Compare
Synth Signs-to-GTSRB (4 rows) Mean teacher Self-ensembling for visual domain adaptation code — Compare
HMDBsmall-to-UCF (3 rows) TA3N Temporal Attentive Alignment for Video Domain Adaptation code — Compare
Olympic-to-HMDBsmall (3 rows) TA3N Temporal Attentive Alignment for Video Domain Adaptation code — Compare
Synscapes-to-Cityscapes (3 rows) ProDA+CRA Cross-Region Domain Adaptation for Class-level Alignment — — Compare
UCF-to-HMDBsmall (3 rows) TA3N Temporal Attentive Alignment for Video Domain Adaptation code — Compare
UCF-to-Olympic (3 rows) TA3N Temporal Attentive Alignment for Video Domain Adaptation code — Compare
Comic2k (2 rows) DDT Diffusion Domain Teacher: Diffusion Guided Domain Adaptive Object Detector code — Compare
Rotating MNIST (2 rows) PCIDA Continuously Indexed Domain Adaptation code — Compare
Canon RAW Low Light (1 row) FSDA-LL Sony -> Canon Few-Shot Domain Adaptation for Low Light RAW Image Enhancement code Syntology official (archive's flag): 7 ran · 7 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified Compare
Foggy Cityscapes (1 row) MILA MILA: Memory-Based Instance-Level Adaptation for Cross-Domain... code — Compare
GTA-to-FoggyCityscapes (1 row) FogAdapt+ FogAdapt: Self-Supervised Domain Adaptation for Semantic... — — Compare
LeukemiaAttri (1 row) ConfMix [23] L_100x_C2 A Large-scale Multi Domain Leukemia Dataset for the White Blood... code — Compare
MNIST-M-to-MNIST (1 row) DRANet DRANet: Disentangling Representation and Adaptation Networks for... code Syntology official (archive's flag): 1 ran · 1 ran (of which 1 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified; the one sample that ran constructed an object rather than computing a result Compare
MSDA (1 row) MetaSelf-Learning Meta Self-Learning for Multi-Source Domain Adaptation: A Benchmark code — Compare
Nikon RAW Low Light (1 row) FSDA-LL Sony -> Nikon Few-Shot Domain Adaptation for Low Light RAW Image Enhancement code Syntology official (archive's flag): 7 ran · 7 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified Compare
Noisy-Amazon (20%) (1 row) Butterfly Butterfly: One-step Approach towards Wildly Unsupervised Domain Adaptation code Syntology 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified Compare
Noisy-Amazon (45%) (1 row) Butterfly Butterfly: One-step Approach towards Wildly Unsupervised Domain Adaptation code Syntology 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified Compare
Noisy-MNIST-to-SYND (1 row) Butterfly Butterfly: One-step Approach towards Wildly Unsupervised Domain Adaptation code Syntology 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified Compare
Noisy-SYND-to-MNIST (1 row) Butterfly Butterfly: One-step Approach towards Wildly Unsupervised Domain Adaptation code Syntology 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 0 honoured, 0 violated, 1 with no contract checked; 0 where Syntology's instrument failed) · 2 unverified Compare
Office-Caltech-10 (1 row) MEDA Visual Domain Adaptation with Manifold Embedded Distribution Alignment code — Compare
PACS (1 row) SSGEN Learning to Generalize across Domains on Single Test Samples code — Compare
S2RDA-49 (1 row) PGA Enhancing Domain Adaptation through Prompt Gradient Alignment code Syntology official (archive's flag): 7 ran · 7 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 2 honoured, 1 violated, 2 with no contract checked; 2 where Syntology's instrument failed) · 5 unverified Compare
S2RDA-MS-39 (1 row) PGA Enhancing Domain Adaptation through Prompt Gradient Alignment code Syntology official (archive's flag): 7 ran · 7 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 2 honoured, 1 violated, 2 with no contract checked; 2 where Syntology's instrument failed) · 5 unverified Compare
Sim10k (1 row) MILA MILA: Memory-Based Instance-Level Adaptation for Cross-Domain... code — Compare
Synth Objects-to-LINEMOD (1 row) DSN (DANN) Domain Separation Networks code Syntology community repositories only · 10 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 0 honoured, 0 violated, 5 with no contract checked; 5 where Syntology's instrument failed) · 4 unverified Compare
SYNTHIA-to-FoggyCityscapes (1 row) FogAdapt+ FogAdapt: Self-Supervised Domain Adaptation for Semantic... — — Compare
SYNTHIA-to-Cityscapes Labels (1 row) MRNet Unsupervised Scene Adaptation with Memory Regularization in vivo code Syntology official (archive's flag): 2 ran · 3 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 2 where Syntology's instrument failed) · 0 unverified Compare
VIPER-to-Cityscapes (1 row) STPL Spatio-Temporal Pixel-Level Contrastive Learning-based Source-Free... code — Compare

Syntology column: samples harvested from the paper's repositories and executed on synthesized fixtures; “ran” is not a correctness claim and does not order the table. Each line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the row's paper, the line starts with that repository's state (“official” is the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from. A dash means no Syntology record for that paper, not a recorded non-run. Each “ran” line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Libraries

Not in the archive: the export carries no per-task library table, so there is nothing to show at snapshot 2025-07-28.

Datasets archive 2025-07-28

96 datasets whose archive record lists this task, ordered by the archive's paper count. 30 shown of 96 until expanded.

Subtasks archive 2025-07-28

13 subtasks in the archive's task tree.

Papers archive 2025-07-28

archive papers tagged: 6,439 · with a code link: 2,400 · where Syntology ran a sample: 517 (429 with a run with no instrument failure, 88 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (517 of 6,439 tagged: 429 with a run with no instrument failure, 88 where every run was a failure of Syntology's instrument)

Page 1 of 65: papers 1 to 100 of 6,439, in archive order: by repositories listed in the archive (most first), then newest first, not by stars (the archive holds no stars, so PwC's “Social” and “Latest” sorts cannot be reproduced). Papers that list no repository come after every paper that lists one.

Papers without a page here are shown as plain text. A Syntology line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the figure “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code); hover it for the repositories the samples that ran came from. The first sentence of the abstract is shown for the first 30 papers; every abstract is on its paper's page.

Syntology lines on 63 of the papers shown; no Syntology record for the others (a paper without an arXiv id cannot be joined to the graph, and absence from the graph layer is not a recorded non-run). “Ran” means the sample executed on a synthesized fixture, not that the paper's result was reproduced; each line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections