{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:2ES3ABWKDJFV6IXKHNPVUHUFQN","short_pith_number":"pith:2ES3ABWK","schema_version":"1.0","canonical_sha256":"d125b006ca1a4b5f22ea3b5f5a1e85834794dcd21c84fcbef153b997769b3f10","source":{"kind":"arxiv","id":"2112.01527","version":3},"attestation_state":"computed","paper":{"title":"Masked-attention Mask Transformer for Universal Image Segmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Alexander G. Schwing, Alexander Kirillov, Bowen Cheng, Ishan Misra, Rohit Girdhar","submitted_at":"2021-12-02T18:59:58Z","abstract_excerpt":"Image segmentation is about grouping pixels with different semantics, e.g., category or instance membership, where each choice of semantics defines a task. While only the semantics of each task differ, current research focuses on designing specialized architectures for each task. We present Masked-attention Mask Transformer (Mask2Former), a new architecture capable of addressing any image segmentation task (panoptic, instance or semantic). Its key components include masked attention, which extracts localized features by constraining cross-attention within predicted mask regions. In addition to"},"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":"2112.01527","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2021-12-02T18:59:58Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"c34d5a0c1a2e312b11f745b2344ec5df49a47dd394cef21e1abd4bc0bb87ec8a","abstract_canon_sha256":"5f9f5bb722f790ad2b439b53fb86bf14407cda87db415bd1cc2292ad8cfe4a8e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:32:13.488278Z","signature_b64":"L/dNaXshpIYjtm7ckOiFfPK59cr6P05PW0cXfQLiLLYFIvqacyjTO8MGsVcO0iZ1Y5pGOriG8xSmQOqAzQlcCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d125b006ca1a4b5f22ea3b5f5a1e85834794dcd21c84fcbef153b997769b3f10","last_reissued_at":"2026-07-05T04:32:13.487771Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:32:13.487771Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Masked-attention Mask Transformer for Universal Image Segmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Alexander G. Schwing, Alexander Kirillov, Bowen Cheng, Ishan Misra, Rohit Girdhar","submitted_at":"2021-12-02T18:59:58Z","abstract_excerpt":"Image segmentation is about grouping pixels with different semantics, e.g., category or instance membership, where each choice of semantics defines a task. While only the semantics of each task differ, current research focuses on designing specialized architectures for each task. We present Masked-attention Mask Transformer (Mask2Former), a new architecture capable of addressing any image segmentation task (panoptic, instance or semantic). Its key components include masked attention, which extracts localized features by constraining cross-attention within predicted mask regions. In addition to"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.01527","kind":"arxiv","version":3},"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/2112.01527/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":"2112.01527","created_at":"2026-07-05T04:32:13.487832+00:00"},{"alias_kind":"arxiv_version","alias_value":"2112.01527v3","created_at":"2026-07-05T04:32:13.487832+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.01527","created_at":"2026-07-05T04:32:13.487832+00:00"},{"alias_kind":"pith_short_12","alias_value":"2ES3ABWKDJFV","created_at":"2026-07-05T04:32:13.487832+00:00"},{"alias_kind":"pith_short_16","alias_value":"2ES3ABWKDJFV6IXK","created_at":"2026-07-05T04:32:13.487832+00:00"},{"alias_kind":"pith_short_8","alias_value":"2ES3ABWK","created_at":"2026-07-05T04:32:13.487832+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":17,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2607.07001","citing_title":"Ego-Human Motion Prediction with 3D-Aware LLM","ref_index":11,"is_internal_anchor":true},{"citing_arxiv_id":"2607.06464","citing_title":"Hilti-Trimble-Oxford Dataset: 360 Visual-Inertial Benchmark with Floor Plan Priors for SLAM and Localization","ref_index":41,"is_internal_anchor":true},{"citing_arxiv_id":"2606.20272","citing_title":"Efficiently Linking Real Scenes with Synthetic Data Generation for AI-based Cognitive Robotics and Computer Vision Applications","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2606.17631","citing_title":"Better Queries, Cheaper Attention: Adapting Transformers for Efficient Sparse Reconstruction","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08517","citing_title":"A Joint Finite-Sample Certificate for Adaptive Selective Conformal Risk Control","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07415","citing_title":"ChartREG++: Towards Benchmarking and Improving Chart Referring Expression Grounding under Diverse referring clues and Multi-Target Referring","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29391","citing_title":"Phast: Simultaneous reconstruction of photoelectron count and time profiles from PMT waveforms via machine learning","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.26295","citing_title":"Beyond Aesthetics: Quantifying Information Loss in Turbid Scenes","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23747","citing_title":"Revitalizing Dense Material Segmentation: Stabilized Vision Transformers and the Generalization Paradox","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17630","citing_title":"SegRAG: Training-Free Retrieval-Augmented Semantic Segmentation","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16393","citing_title":"Vision Transformer-Conditioned UNet for Domain-Adaptive Semantic Segmentation","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17630","citing_title":"SegRAG: Training-Free Retrieval-Augmented Semantic Segmentation","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2505.05472","citing_title":"Mogao: An Omni Foundation Model for Interleaved Multi-Modal Generation","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2604.26869","citing_title":"KAYRA: A Microservice Architecture for AI-Assisted Karyotyping with Cloud and On-Premise Deployment","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25072","citing_title":"Beyond Accuracy: Benchmarking Cross-Task Consistency in Unified Multimodal Models","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.10333","citing_title":"Zero-shot World Models Are Developmentally Efficient Learners","ref_index":64,"is_internal_anchor":false},{"citing_arxiv_id":"2605.07415","citing_title":"ChartREG++: Towards Benchmarking and Improving Chart Referring Expression Grounding under Diverse referring clues and Multi-Target Referring","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2ES3ABWKDJFV6IXKHNPVUHUFQN","json":"https://pith.science/pith/2ES3ABWKDJFV6IXKHNPVUHUFQN.json","graph_json":"https://pith.science/api/pith-number/2ES3ABWKDJFV6IXKHNPVUHUFQN/graph.json","events_json":"https://pith.science/api/pith-number/2ES3ABWKDJFV6IXKHNPVUHUFQN/events.json","paper":"https://pith.science/paper/2ES3ABWK"},"agent_actions":{"view_html":"https://pith.science/pith/2ES3ABWKDJFV6IXKHNPVUHUFQN","download_json":"https://pith.science/pith/2ES3ABWKDJFV6IXKHNPVUHUFQN.json","view_paper":"https://pith.science/paper/2ES3ABWK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2112.01527&json=true","fetch_graph":"https://pith.science/api/pith-number/2ES3ABWKDJFV6IXKHNPVUHUFQN/graph.json","fetch_events":"https://pith.science/api/pith-number/2ES3ABWKDJFV6IXKHNPVUHUFQN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2ES3ABWKDJFV6IXKHNPVUHUFQN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2ES3ABWKDJFV6IXKHNPVUHUFQN/action/storage_attestation","attest_author":"https://pith.science/pith/2ES3ABWKDJFV6IXKHNPVUHUFQN/action/author_attestation","sign_citation":"https://pith.science/pith/2ES3ABWKDJFV6IXKHNPVUHUFQN/action/citation_signature","submit_replication":"https://pith.science/pith/2ES3ABWKDJFV6IXKHNPVUHUFQN/action/replication_record"}},"created_at":"2026-07-05T04:32:13.487832+00:00","updated_at":"2026-07-05T04:32:13.487832+00:00"}