# PAT

Source: [PAT](<https://www.sota2.com/research/dataset/pat>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Benchmarks

Page 1 of 26. This is an index page, not the full collection.

- [Next index page](<https://www.sota2.com/research/dataset/pat.md?page=2>): Follow this link to discover more pages.

- [Image Captioning on MS COCO Karpathy (test)](<https://www.sota2.com/research/sota/image-captioning-on-ms-coco-karpathy-test.md>): Image Captioning; MS COCO Karpathy (test).

- [Spatial Reasoning on EmbSpatial](<https://www.sota2.com/research/sota/spatial-reasoning-on-embspatial.md>): Spatial Reasoning; EmbSpatial.

- [Spatial Reasoning on Viewspatial](<https://www.sota2.com/research/sota/spatial-reasoning-on-viewspatial.md>): Spatial Reasoning; Viewspatial.

- [Automatic Speech Recognition on NKI-RUG-UMCG (patient speakers)](<https://www.sota2.com/research/sota/automatic-speech-recognition-on-nki-rug-umcg-patient-speakers.md>): Automatic Speech Recognition; NKI-RUG-UMCG (patient speakers).

- [Medical Visual Question Answering on PathVQA (Open and Closed Metrics)](<https://www.sota2.com/research/sota/medical-visual-question-answering-on-pathvqa-open-and-closed-metrics.md>): Medical Visual Question Answering; PathVQA.

- [Node Classification on PATTERN (test)](<https://www.sota2.com/research/sota/node-classification-on-pattern-test.md>): Node Classification; PATTERN (test).

- [Shortest Path Optimization on 5x5 Grid](<https://www.sota2.com/research/sota/shortest-path-optimization-on-5x5-grid.md>): Shortest path optimization; 5x5 grid shortest path problem.

- [Medical Visual Question Answering on PathVQA](<https://www.sota2.com/research/sota/medical-visual-question-answering-on-pathvqa.md>): Medical Visual Question Answering; PathVQA.

- [Robotic Manipulation on LIBERO Spatial Object Goal Long](<https://www.sota2.com/research/sota/robotic-manipulation-on-libero-spatial-object-goal-long.md>): Robotic Manipulation; LIBERO Spatial Object Goal Long.

- [Image Classification on PatchCamelyon](<https://www.sota2.com/research/sota/image-classification-on-patchcamelyon.md>): Image Classification; PatchCamelyon.

- [Homography Estimation on HPatches (Pixel Accuracy Thresholds)](<https://www.sota2.com/research/sota/homography-estimation-on-hpatches-pixel-accuracy-thresholds.md>): Homography Estimation; HPatches.

- [Image Captioning on Flickr30K (Karpathy split)](<https://www.sota2.com/research/sota/image-captioning-on-flickr30k-karpathy-split.md>): Image Captioning; Flickr30K (Karpathy split).

- [Image Captioning on COCO (Karpathy split)](<https://www.sota2.com/research/sota/image-captioning-on-coco-karpathy-split.md>): Image Captioning; COCO (Karpathy split).

- [Targeted Adversarial Attack on ImageNet-Compatible](<https://www.sota2.com/research/sota/targeted-adversarial-attack-on-imagenet-compatible.md>): Targeted Adversarial Attack; ImageNet-Compatible.

- [Medical Image Segmentation on ACDC (100 patients total test)](<https://www.sota2.com/research/sota/medical-image-segmentation-on-acdc-100-patients-total-test.md>): Medical Image Segmentation; ACDC 100 patients total (test).

- [Image Classification on PatternNet](<https://www.sota2.com/research/sota/image-classification-on-patternnet.md>): Image Classification; PatternNet.

- [Caption Retrieval on MS COCO Karpathy 1k (test)](<https://www.sota2.com/research/sota/caption-retrieval-on-ms-coco-karpathy-1k-test.md>): Caption Retrieval; MS COCO Karpathy 1k (test).

- [Medical Image Classification on PathMNIST](<https://www.sota2.com/research/sota/medical-image-classification-on-pathmnist.md>): Medical Image Classification; PathMNIST.

- [Summarization on bigPatent](<https://www.sota2.com/research/sota/summarization-on-bigpatent.md>): Summarization; bigPatent.

- [Image Classification on PathMNIST (Adversarial Robustness)](<https://www.sota2.com/research/sota/image-classification-on-pathmnist-adversarial-robustness.md>): Image Classification; PathMNIST.

- [Image-to-text retrieval on Flickr30K 1K Karpathy (test)](<https://www.sota2.com/research/sota/image-to-text-retrieval-on-flickr30k-1k-karpathy-test.md>): Image-to-text retrieval; Flickr30K 1K Karpathy (test).

- [Fairness-aware Conformal Prediction on Fitzpatrick](<https://www.sota2.com/research/sota/fairness-aware-conformal-prediction-on-fitzpatrick.md>): Fairness-aware Conformal Prediction; Fitzpatrick.

- [Image Captioning on MS COCO (Karpathy)](<https://www.sota2.com/research/sota/image-captioning-on-ms-coco-karpathy.md>): Image Captioning; MS COCO (Karpathy).

- [Medical Visual Question Answering on PathVQA (test)](<https://www.sota2.com/research/sota/medical-visual-question-answering-on-pathvqa-test.md>): Medical Visual Question Answering; PathVQA (test).

- [Homography estimation on HPatches](<https://www.sota2.com/research/sota/homography-estimation-on-hpatches.md>): Homography estimation; HPatches.

- [Spatial Reasoning on OmniSpatial (test)](<https://www.sota2.com/research/sota/spatial-reasoning-on-omnispatial-test.md>): Spatial Reasoning; OmniSpatial (test).

- [Degradation Severity Estimation on CT Patients: L506, L192, L310 (test)](<https://www.sota2.com/research/sota/degradation-severity-estimation-on-ct-patients-l506-l192-l310-test.md>): Degradation Severity Estimation; CT Patients: L506, L192, L310 (test).

- [Occupational AI Exposure Assessment on Occupations Claude usage most observed AI exposure Top 10 1.0](<https://www.sota2.com/research/sota/occupational-ai-exposure-assessment-on-occupations-claude-usage-most-observed-ai-exposure-top-10-1-0.md>): Occupational AI Exposure Assessment; Occupations Claude usage most observed AI exposure Top 10 1.0.

- [Emotion Classification on EMPATHETICDIALOGUES (test)](<https://www.sota2.com/research/sota/emotion-classification-on-empatheticdialogues-test.md>): Emotion Classification; EMPATHETICDIALOGUES (test).

- [Medical Image Classification on PATHMNIST (Adversarial Robustness)](<https://www.sota2.com/research/sota/medical-image-classification-on-pathmnist-adversarial-robustness.md>): Medical Image Classification; PATHMNIST.

- [Accuracy on RoboSpatial Spatial Reasoning](<https://www.sota2.com/research/sota/accuracy-on-robospatial-spatial-reasoning.md>): Spatial Reasoning; RoboSpatial.

- [Spatial Reasoning on Visual Spatial Reasoning (VSR)](<https://www.sota2.com/research/sota/spatial-reasoning-on-visual-spatial-reasoning-vsr.md>): Spatial Reasoning; Visual Spatial Reasoning (VSR).

- [Multiple Choice Question Answering on PathMMU (val)](<https://www.sota2.com/research/sota/multiple-choice-question-answering-on-pathmmu-val.md>): Multiple Choice Question Answering; PathMMU (val).

- [Image Classification on PathMNIST](<https://www.sota2.com/research/sota/image-classification-on-pathmnist.md>): Image Classification; PathMNIST.

- [Landmark Identification on RESECT (patient-wise)](<https://www.sota2.com/research/sota/landmark-identification-on-resect-patient-wise.md>): Landmark identification; RESECT (patient-wise).

- [3D Geometry Estimation and Reconstruction on SpatialBench Average across settings](<https://www.sota2.com/research/sota/3d-geometry-estimation-and-reconstruction-on-spatialbench-average-across-settings.md>): 3D Geometry Estimation and Reconstruction; SpatialBench Average across settings.

- [3D Geometry Estimation and Reconstruction on SpatialBench Medium](<https://www.sota2.com/research/sota/3d-geometry-estimation-and-reconstruction-on-spatialbench-medium.md>): 3D Geometry Estimation and Reconstruction; SpatialBench Medium.

- [3D Geometry Estimation and Reconstruction on SpatialBench Sparse](<https://www.sota2.com/research/sota/3d-geometry-estimation-and-reconstruction-on-spatialbench-sparse.md>): 3D Geometry Estimation and Reconstruction; SpatialBench Sparse.

- [3D Geometry Estimation and Reconstruction on SpatialBench Single Frame (AbsRel)](<https://www.sota2.com/research/sota/3d-geometry-estimation-and-reconstruction-on-spatialbench-single-frame-absrel.md>): 3D Geometry Estimation and Reconstruction; SpatialBench Single Frame.

- [Lesion Grading on AIDPATH-G](<https://www.sota2.com/research/sota/lesion-grading-on-aidpath-g.md>): Lesion Grading; AIDPATH-G.

- [Spatial Logical Reasoning on SpatiaLQA](<https://www.sota2.com/research/sota/spatial-logical-reasoning-on-spatialqa.md>): Spatial Logical Reasoning; SpatiaLQA.

- [Multimodal Robotic Control on Fetch-PickAndPlace Patch corruptions (test)](<https://www.sota2.com/research/sota/multimodal-robotic-control-on-fetch-pickandplace-patch-corruptions-test.md>): Multimodal Robotic Control; Fetch-PickAndPlace Patch corruptions (test).

- [Visual Question Answering on PathMMU Tiny 1.0 (test)](<https://www.sota2.com/research/sota/visual-question-answering-on-pathmmu-tiny-1-0-test.md>): Visual Question Answering; PathMMU Tiny 1.0 (test).

- [Node Classification on snap-patents (test)](<https://www.sota2.com/research/sota/node-classification-on-snap-patents-test.md>): Node Classification; snap-patents (test).

- [Medical Image Segmentation on CHD 68 patients total (test)](<https://www.sota2.com/research/sota/medical-image-segmentation-on-chd-68-patients-total-test.md>): Medical Image Segmentation; CHD 68 patients total (test).

- [Robot Manipulation on LIBERO Spatial](<https://www.sota2.com/research/sota/robot-manipulation-on-libero-spatial.md>): Robot Manipulation; LIBERO Spatial.

- [Steroid-free clinical remission (SFCR) prediction on IFX arm and ADA patients cohort (SFCR outcome)](<https://www.sota2.com/research/sota/steroid-free-clinical-remission-sfcr-prediction-on-ifx-arm-and-ada-patients-cohort-sfcr-outcome.md>): Steroid-free clinical remission (SFCR) prediction; IFX arm and ADA patients cohort (SFCR outcome).

- [Sample-level Membership Inference on LIBERO Spatial](<https://www.sota2.com/research/sota/sample-level-membership-inference-on-libero-spatial.md>): Sample-level Membership Inference; LIBERO Spatial.

- [Spatial Understanding on EmbSpatial](<https://www.sota2.com/research/sota/spatial-understanding-on-embspatial.md>): Spatial Understanding; EmbSpatial.

- [Spatial Reasoning on RefSpatial](<https://www.sota2.com/research/sota/spatial-reasoning-on-refspatial.md>): Spatial Reasoning; RefSpatial.

- [Adversarial Attack on ImageNet-compatible Stable Diffusion context v1.4 (test)](<https://www.sota2.com/research/sota/adversarial-attack-on-imagenet-compatible-stable-diffusion-context-v1-4-test.md>): Adversarial Attack; ImageNet-compatible Stable Diffusion context v1.4 (test).

- [Localization on Blackbird clov, patr, thri, wint, amp, dice, half, sid trajectories (test)](<https://www.sota2.com/research/sota/localization-on-blackbird-clov-patr-thri-wint-amp-dice-half-sid-trajectories-test.md>): Localization; Blackbird clov, patr, thri, wint, amp, dice, half, sid trajectories (test).

- [Feature Matching on HPatches](<https://www.sota2.com/research/sota/feature-matching-on-hpatches.md>): Feature Matching; HPatches.

- [Mirage Detection on VQA Mirage Detection Chest, Pathology, Natural, Document, Infographic (test)](<https://www.sota2.com/research/sota/mirage-detection-on-vqa-mirage-detection-chest-pathology-natural-document-infographic-test.md>): Mirage Detection; VQA Mirage Detection Chest, Pathology, Natural, Document, Infographic (test).

- [Image Classification on CINIC-10 Dir(0.01), 50 clients, 20% participation](<https://www.sota2.com/research/sota/image-classification-on-cinic-10-dir-0-01-50-clients-20-percent-participation.md>): Image Classification; CINIC-10 Dir(0.01), 50 clients, 20% participation.

- [Spatial Reasoning on RoboSpatial](<https://www.sota2.com/research/sota/spatial-reasoning-on-robospatial.md>): Spatial Reasoning; RoboSpatial.

- [Few-shot classification on PatternNet (19-19)](<https://www.sota2.com/research/sota/few-shot-classification-on-patternnet-19-19.md>): Few-shot classification; PatternNet (19-19).

- [Image Classification on PathMNIST v1 (test)](<https://www.sota2.com/research/sota/image-classification-on-pathmnist-v1-test.md>): Image Classification; PathMNIST v1 (test).

- [Shortest Path Optimization on 5x5 Grid (test)](<https://www.sota2.com/research/sota/shortest-path-optimization-on-5x5-grid-test.md>): Shortest path optimization; Shortest path problem 5x5 grid (test).

- [Image Classification on CIFAR-100 Pathological](<https://www.sota2.com/research/sota/image-classification-on-cifar-100-pathological.md>): Image Classification; CIFAR-100 Pathological.

- [Spatial Reasoning on ViewSpatial-Bench](<https://www.sota2.com/research/sota/spatial-reasoning-on-viewspatial-bench.md>): Spatial Reasoning; ViewSpatial-Bench.

- [Satellite Image Generation on Git-Spatial 15k](<https://www.sota2.com/research/sota/satellite-image-generation-on-git-spatial-15k.md>): Satellite image generation; Git-Spatial 15k.

- [Node Classification on PATTERN](<https://www.sota2.com/research/sota/node-classification-on-pattern.md>): Node classification; PATTERN.

- [Image Classification on PathMNIST v2 (test)](<https://www.sota2.com/research/sota/image-classification-on-pathmnist-v2-test.md>): Image Classification; PathMNIST v2 (test).

- [Medical Visual Question Answering on PathVQA closed-end](<https://www.sota2.com/research/sota/medical-visual-question-answering-on-pathvqa-closed-end.md>): Medical Visual Question Answering; PathVQA closed-end.

- [Node Classification on snap-patents](<https://www.sota2.com/research/sota/node-classification-on-snap-patents.md>): Node Classification; snap-patents.

- [Planning on LIBERO Spatial Suite](<https://www.sota2.com/research/sota/planning-on-libero-spatial-suite.md>): Planning; LIBERO Spatial Suite.

- [Geometric Matching on HPatches 240 x 240](<https://www.sota2.com/research/sota/geometric-matching-on-hpatches-240-x-240.md>): Geometric Matching; HPatches 240 x 240.

- [Shortest Path on Shortest Path Degree 2, 4, 6, 8 (test)](<https://www.sota2.com/research/sota/shortest-path-on-shortest-path-degree-2-4-6-8-test.md>): Shortest Path; Shortest Path Degree 2, 4, 6, 8 (test).

- [Age on Patchscopes few-shot 1.0 (test)](<https://www.sota2.com/research/sota/age-on-patchscopes-few-shot-1-0-test.md>): Age; Patchscopes few-shot 1.0 (test).

- [Culture on Patchscopes few-shot 1.0 (test)](<https://www.sota2.com/research/sota/culture-on-patchscopes-few-shot-1-0-test.md>): Culture; Patchscopes few-shot 1.0 (test).

- [Gender on Patchscopes few-shot 1.0 (test)](<https://www.sota2.com/research/sota/gender-on-patchscopes-few-shot-1-0-test.md>): Gender; Patchscopes few-shot 1.0 (test).

- [Color on Patchscopes few-shot 1.0 (test)](<https://www.sota2.com/research/sota/color-on-patchscopes-few-shot-1-0-test.md>): Color; Patchscopes few-shot 1.0 (test).

- [Age on Patchscopes (test)](<https://www.sota2.com/research/sota/age-on-patchscopes-test.md>): Age; Patchscopes (test).

- [Culture on Patchscopes (test)](<https://www.sota2.com/research/sota/culture-on-patchscopes-test.md>): Culture; Patchscopes (test).

- [Gender on Patchscopes (test)](<https://www.sota2.com/research/sota/gender-on-patchscopes-test.md>): Gender; Patchscopes (test).

- [Color on Patchscopes (test)](<https://www.sota2.com/research/sota/color-on-patchscopes-test.md>): Color; Patchscopes (test).

- [9-class Classification on PathMNIST](<https://www.sota2.com/research/sota/9-class-classification-on-pathmnist.md>): 9-class classification; PathMNIST.

- [Image Denoising on Kodak24 Spatially-correlated noise](<https://www.sota2.com/research/sota/image-denoising-on-kodak24-spatially-correlated-noise.md>): Image Denoising; Kodak24 Spatially-correlated noise.

- [Text Retrieval on Flickr30k Karpathy (test)](<https://www.sota2.com/research/sota/text-retrieval-on-flickr30k-karpathy-test.md>): Text Retrieval; Flickr30k Karpathy (test).