{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:MVKERD7WBTA4UJR7KQLCZ7JZPM","short_pith_number":"pith:MVKERD7W","schema_version":"1.0","canonical_sha256":"6554488ff60cc1ca263f54162cfd397b261dae6212c10bf4531a706e927b792f","source":{"kind":"arxiv","id":"2104.02008","version":1},"attestation_state":"computed","paper":{"title":"Domain Generalization with MixStyle","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Kaiyang Zhou, Tao Xiang, Yongxin Yang, Yu Qiao","submitted_at":"2021-04-05T16:58:09Z","abstract_excerpt":"Though convolutional neural networks (CNNs) have demonstrated remarkable ability in learning discriminative features, they often generalize poorly to unseen domains. Domain generalization aims to address this problem by learning from a set of source domains a model that is generalizable to any unseen domain. In this paper, a novel approach is proposed based on probabilistically mixing instance-level feature statistics of training samples across source domains. Our method, termed MixStyle, is motivated by the observation that visual domain is closely related to image style (e.g., photo vs.~sket"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2104.02008","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-04-05T16:58:09Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"3c651f8993e27bfb15eba358f0e4f3b5ff4426109bbdbc2acee203019d767cd2","abstract_canon_sha256":"19a7fd92e228387508ab5a6b02be01f28fc76202c4934a77580af855f068c615"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:29:06.401004Z","signature_b64":"sbgHwTqYZS4uO/meNwLtu7A5mfC4IdMwGsVqCZsraw26Ob/hoszW8QSHUoPDaoKUvK28AVVY6xsmEx7KQoSQAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6554488ff60cc1ca263f54162cfd397b261dae6212c10bf4531a706e927b792f","last_reissued_at":"2026-07-05T02:29:06.400516Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:29:06.400516Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Domain Generalization with MixStyle","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Kaiyang Zhou, Tao Xiang, Yongxin Yang, Yu Qiao","submitted_at":"2021-04-05T16:58:09Z","abstract_excerpt":"Though convolutional neural networks (CNNs) have demonstrated remarkable ability in learning discriminative features, they often generalize poorly to unseen domains. Domain generalization aims to address this problem by learning from a set of source domains a model that is generalizable to any unseen domain. In this paper, a novel approach is proposed based on probabilistically mixing instance-level feature statistics of training samples across source domains. Our method, termed MixStyle, is motivated by the observation that visual domain is closely related to image style (e.g., photo vs.~sket"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2104.02008","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2104.02008/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2104.02008","created_at":"2026-07-05T02:29:06.400576+00:00"},{"alias_kind":"arxiv_version","alias_value":"2104.02008v1","created_at":"2026-07-05T02:29:06.400576+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2104.02008","created_at":"2026-07-05T02:29:06.400576+00:00"},{"alias_kind":"pith_short_12","alias_value":"MVKERD7WBTA4","created_at":"2026-07-05T02:29:06.400576+00:00"},{"alias_kind":"pith_short_16","alias_value":"MVKERD7WBTA4UJR7","created_at":"2026-07-05T02:29:06.400576+00:00"},{"alias_kind":"pith_short_8","alias_value":"MVKERD7W","created_at":"2026-07-05T02:29:06.400576+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":14,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.21736","citing_title":"Adversarial Domain Prompt Tuning and Generation for Single Domain Generalization","ref_index":77,"is_internal_anchor":false},{"citing_arxiv_id":"2606.07646","citing_title":"DOME: Learning Transferable Domain Variables from Sparse Supervision for Test-Time Adaptation","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24588","citing_title":"HeartBeatAI: An Interpretable and Robust Deep Learning Framework for Multi-Label ECG Arrhythmia Detection","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09881","citing_title":"Toward Calibrated, Fair, and accurate Deepfake Detection","ref_index":297,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14654","citing_title":"Beyond Instance-Level Self-Supervision in 3D Multi-Modal Medical Imaging","ref_index":181,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02564","citing_title":"Why Invariance is Not Enough for Biomedical Domain Generalization and How to Fix It","ref_index":75,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12435","citing_title":"Environment-Adaptive Preference Optimization for Wildfire Prediction","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2604.26820","citing_title":"Bridge: Basis-Driven Causal Inference Marries VFMs for Domain Generalization","ref_index":82,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09925","citing_title":"Frequency Adapter with SAM for Generalized Medical Image Segmentation","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04528","citing_title":"YOTOnet: Zero-Shot Cross-Domain Fault Diagnosis via Domain-Conditioned Mixture of Experts","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01967","citing_title":"MER-DG: Modality-Entropy Regularization for Multimodal Domain Generalization","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10524","citing_title":"FGML-DG: Feynman-Inspired Cognitive Science Paradigm for Cross-Domain Medical Image Segmentation","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2604.05335","citing_title":"Cross-Machine Anomaly Detection Leveraging Pre-trained Time-series Model","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2604.16892","citing_title":"CrossFlowDG: Bridging the Modality Gap with Cross-modal Flow Matching for Domain Generalization","ref_index":56,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MVKERD7WBTA4UJR7KQLCZ7JZPM","json":"https://pith.science/pith/MVKERD7WBTA4UJR7KQLCZ7JZPM.json","graph_json":"https://pith.science/api/pith-number/MVKERD7WBTA4UJR7KQLCZ7JZPM/graph.json","events_json":"https://pith.science/api/pith-number/MVKERD7WBTA4UJR7KQLCZ7JZPM/events.json","paper":"https://pith.science/paper/MVKERD7W"},"agent_actions":{"view_html":"https://pith.science/pith/MVKERD7WBTA4UJR7KQLCZ7JZPM","download_json":"https://pith.science/pith/MVKERD7WBTA4UJR7KQLCZ7JZPM.json","view_paper":"https://pith.science/paper/MVKERD7W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2104.02008&json=true","fetch_graph":"https://pith.science/api/pith-number/MVKERD7WBTA4UJR7KQLCZ7JZPM/graph.json","fetch_events":"https://pith.science/api/pith-number/MVKERD7WBTA4UJR7KQLCZ7JZPM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MVKERD7WBTA4UJR7KQLCZ7JZPM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MVKERD7WBTA4UJR7KQLCZ7JZPM/action/storage_attestation","attest_author":"https://pith.science/pith/MVKERD7WBTA4UJR7KQLCZ7JZPM/action/author_attestation","sign_citation":"https://pith.science/pith/MVKERD7WBTA4UJR7KQLCZ7JZPM/action/citation_signature","submit_replication":"https://pith.science/pith/MVKERD7WBTA4UJR7KQLCZ7JZPM/action/replication_record"}},"created_at":"2026-07-05T02:29:06.400576+00:00","updated_at":"2026-07-05T02:29:06.400576+00:00"}