</> Code with Papers by Syntology

Papers › Open-World Semi-Supervised Learning

Open-World Semi-Supervised Learning

6 Feb 2021ICLR 2022 4arXiv:2102.03526archive 2025-07-28

Kaidi Cao, Maria Brbic, Jure Leskovec

A fundamental limitation of applying semi-supervised learning in real-world settings is the assumption that unlabeled test data contains only classes previously encountered in the labeled training data. However, this assumption rarely holds for data in-the-wild, where instances belonging to novel classes may appear at testing time. Here, we introduce a novel open-world semi-supervised learning setting that formalizes the notion that novel classes may appear in the unlabeled test data. In this novel setting, the goal is to solve the class distribution mismatch between labeled and unlabeled data, where at the test time every input instance either needs to be classified into one of the existing classes or a new unseen class needs to be initialized. To tackle this challenging problem, we propose ORCA, an end-to-end deep learning approach that introduces uncertainty adaptive margin mechanism to circumvent the bias towards seen classes caused by learning discriminative features for seen classes faster than for the novel classes. In this way, ORCA reduces the gap between intra-class variance of seen with respect to novel classes. Experiments on image classification datasets and a single-cell annotation dataset demonstrate that ORCA consistently outperforms alternative baselines, achieving 25% improvement on seen and 96% improvement on novel classes of the ImageNet dataset.

PaperPDFConference PDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

snap-stanford/orca officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Image ClassificationNovel Object DetectionOpen-World Semi-Supervised Learningimage-classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Novel Object Detection LVIS v1.0 val ORCA Cao et al. (2022) All mAP 2.03 #4 of 5 Archive leaderboard report
Novel Object Detection LVIS v1.0 val ORCA Cao et al. (2022) Known mAP 20.57 #4 of 5 Archive leaderboard report
Novel Object Detection LVIS v1.0 val ORCA Cao et al. (2022) Novel mAP 0.49 #4 of 5 Archive leaderboard report
Open-World Semi-Supervised Learning CIFAR-10 ORCA (ResNet-18) All accuracy (10% Labeled) 84.1 #3 of 5 Archive leaderboard report
Open-World Semi-Supervised Learning CIFAR-10 ORCA (ResNet-18) All accuracy (50% Labeled) 89.7 #3 of 5 Archive leaderboard report
Open-World Semi-Supervised Learning CIFAR-10 ORCA (ResNet-18) Novel accuracy (10% Labeled) 85.5 #3 of 5 Archive leaderboard report
Open-World Semi-Supervised Learning CIFAR-10 ORCA (ResNet-18) Novel accuracy (50% Labeled) 90.4 #3 of 5 Archive leaderboard report
Open-World Semi-Supervised Learning CIFAR-10 ORCA (ResNet-18) Seen accuracy (10% Labeled) 82.8 #3 of 5 Archive leaderboard report
Open-World Semi-Supervised Learning CIFAR-10 ORCA (ResNet-18) Seen accuracy (50% Labeled) 88.2 #3 of 5 Archive leaderboard report
Open-World Semi-Supervised Learning CIFAR-100 ORCA (ResNet-18) All accuracy (10% Labeled) 38.6 #3 of 3 Archive leaderboard report
Open-World Semi-Supervised Learning CIFAR-100 ORCA (ResNet-18) All accuracy (50% Labeled) 48.1 #3 of 3 Archive leaderboard report
Open-World Semi-Supervised Learning CIFAR-100 ORCA (ResNet-18) Novel accuracy (10% Labeled) 31.8 #3 of 3 Archive leaderboard report
Open-World Semi-Supervised Learning CIFAR-100 ORCA (ResNet-18) Novel accuracy (50% Labeled) 43.0 #3 of 3 Archive leaderboard report
Open-World Semi-Supervised Learning CIFAR-100 ORCA (ResNet-18) Seen accuracy (10% Labeled) 52.5 #3 of 3 Archive leaderboard report
Open-World Semi-Supervised Learning CIFAR-100 ORCA (ResNet-18) Seen accuracy (50% Labeled) 66.9 #3 of 3 Archive leaderboard report
Open-World Semi-Supervised Learning ImageNet-100 (TEMI Split) ORCA (ResNet-50) All accuracy (10% Labeled) 69.7 #2 of 5 Archive leaderboard report
Open-World Semi-Supervised Learning ImageNet-100 (TEMI Split) ORCA (ResNet-50) All accuracy (50% Labeled) 77.8 #2 of 5 Archive leaderboard report
Open-World Semi-Supervised Learning ImageNet-100 (TEMI Split) ORCA (ResNet-50) Novel accuracy (10% Labeled) 60.5 #2 of 5 Archive leaderboard report
Open-World Semi-Supervised Learning ImageNet-100 (TEMI Split) ORCA (ResNet-50) Novel accuracy (50% Labeled) 72.1 #2 of 5 Archive leaderboard report
Open-World Semi-Supervised Learning ImageNet-100 (TEMI Split) ORCA (ResNet-50) Seen accuracy (10% Labeled) 83.9 #2 of 5 Archive leaderboard report
Open-World Semi-Supervised Learning ImageNet-100 (TEMI Split) ORCA (ResNet-50) Seen accuracy (50% Labeled) 89.1 #2 of 5 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

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