{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:GD2CKQXVWKBZDDLSHJQKM55NHV","short_pith_number":"pith:GD2CKQXV","schema_version":"1.0","canonical_sha256":"30f42542f5b283918d723a60a677ad3d6dd936050ac6aca8d6051f4ad8f5fe1d","source":{"kind":"arxiv","id":"2301.08243","version":3},"attestation_state":"computed","paper":{"title":"Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.IV"],"primary_cat":"cs.CV","authors_text":"Ishan Misra, Mahmoud Assran, Michael Rabbat, Nicolas Ballas, Pascal Vincent, Piotr Bojanowski, Quentin Duval, Yann LeCun","submitted_at":"2023-01-19T18:59:01Z","abstract_excerpt":"This paper demonstrates an approach for learning highly semantic image representations without relying on hand-crafted data-augmentations. We introduce the Image-based Joint-Embedding Predictive Architecture (I-JEPA), a non-generative approach for self-supervised learning from images. The idea behind I-JEPA is simple: from a single context block, predict the representations of various target blocks in the same image. A core design choice to guide I-JEPA towards producing semantic representations is the masking strategy; specifically, it is crucial to (a) sample target blocks with sufficiently "},"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":"2301.08243","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-01-19T18:59:01Z","cross_cats_sorted":["cs.AI","cs.LG","eess.IV"],"title_canon_sha256":"7a8fc9d7a76a458defafd37db0ad29bec84e7df59d21635463a9b7979a975cca","abstract_canon_sha256":"67cf0652724d16c857db46345d6781de0825f96641e01790f967fe704a46a44a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:00:51.264606Z","signature_b64":"rm/seR4HDZZ3pZAyLd3anF+2kgni5XazxKSmrCh9pidvO0hur2kzkiaT7xZDtqFBVNZtY2GdQhe9MleZQG8CBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"30f42542f5b283918d723a60a677ad3d6dd936050ac6aca8d6051f4ad8f5fe1d","last_reissued_at":"2026-07-05T06:00:51.264115Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:00:51.264115Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-Supervised Learning from Images with a Joint-Embedding Predictive Architecture","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","eess.IV"],"primary_cat":"cs.CV","authors_text":"Ishan Misra, Mahmoud Assran, Michael Rabbat, Nicolas Ballas, Pascal Vincent, Piotr Bojanowski, Quentin Duval, Yann LeCun","submitted_at":"2023-01-19T18:59:01Z","abstract_excerpt":"This paper demonstrates an approach for learning highly semantic image representations without relying on hand-crafted data-augmentations. We introduce the Image-based Joint-Embedding Predictive Architecture (I-JEPA), a non-generative approach for self-supervised learning from images. The idea behind I-JEPA is simple: from a single context block, predict the representations of various target blocks in the same image. A core design choice to guide I-JEPA towards producing semantic representations is the masking strategy; specifically, it is crucial to (a) sample target blocks with sufficiently "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.08243","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/2301.08243/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":"2301.08243","created_at":"2026-07-05T06:00:51.264173+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.08243v3","created_at":"2026-07-05T06:00:51.264173+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.08243","created_at":"2026-07-05T06:00:51.264173+00:00"},{"alias_kind":"pith_short_12","alias_value":"GD2CKQXVWKBZ","created_at":"2026-07-05T06:00:51.264173+00:00"},{"alias_kind":"pith_short_16","alias_value":"GD2CKQXVWKBZDDLS","created_at":"2026-07-05T06:00:51.264173+00:00"},{"alias_kind":"pith_short_8","alias_value":"GD2CKQXV","created_at":"2026-07-05T06:00:51.264173+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":29,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06629","citing_title":"STST-JEPA: Shallow-Target Spatio-Temporal Joint Embedding Prediction Architecture For EEG Self-Supervised Learning","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2606.26410","citing_title":"Neural Voxel Dynamics: Learning Implicit 3D Physics via Volumetric Feature Advection","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20781","citing_title":"World Action Models: A Survey","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00412","citing_title":"Physically Native World Models: A Hamiltonian Perspective on Generative World Modeling","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09646","citing_title":"Do Video Foundation Models Understand Intuitive Physics? A Layerwise Probing Analysis","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2607.00958","citing_title":"LeNEPA: No-Augmentation Next-Latent Prediction for Time-Series Representation Learning","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01443","citing_title":"UR-JEPA: Uniform Rectifiability as a Regularizer for Joint-Embedding Predictive Architectures","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00412","citing_title":"Physically Native World Models: A Hamiltonian Perspective on Generative World Modeling","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15466","citing_title":"Entity-Centric World Models: Interaction-Aware Masking for Causal Video Prediction","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18714","citing_title":"Semantic Generative Tuning for Unified Multimodal Models","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25012","citing_title":"Learning from Semantic Dictionaries: Discriminative Codebook Contrastive Learning for Unified Visual Representation and Generation","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29723","citing_title":"ScaleAware-JEPA: Latent Representation for Discovery in Multiscale Physical Fields","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30724","citing_title":"Research progress on quantum neural networks and quantum machine learning","ref_index":267,"is_internal_anchor":false},{"citing_arxiv_id":"2606.16009","citing_title":"Bridging the Usability Gap: Lessons from Interpreting Studies for Machine Interpreting Design","ref_index":88,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23699","citing_title":"CRONOS: Benchmarking Counterfactual Physical Consistency in Video Models","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.21075","citing_title":"SpectralEarth-FM: Bringing Hyperspectral Imagery into Multimodal Earth Observation Pretraining","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16477","citing_title":"Seeking the Unfamiliar but Memorable: Conceptual Creativity as Meta-Learning","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18714","citing_title":"Semantic Generative Tuning for Unified Multimodal Models","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19462","citing_title":"Quantifying the Pre-training Dividend: Generative versus Latent Self-Supervised Learning for Time Series Foundation Models","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15394","citing_title":"Representation Without Reward: A JEPA Audit for LLM Fine-Tuning","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15466","citing_title":"Entity-Centric World Models: Interaction-Aware Masking for Causal Video Prediction","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2602.06912","citing_title":"PANC: Prior-Aware Normalized Cut via Anchor-Augmented Token Graphs","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08078","citing_title":"Normalizing Trajectory Models","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00412","citing_title":"Physically Native World Models: A Hamiltonian Perspective on Generative World Modeling","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01694","citing_title":"Latent State Design for World Models under Sufficiency Constraints","ref_index":2,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GD2CKQXVWKBZDDLSHJQKM55NHV","json":"https://pith.science/pith/GD2CKQXVWKBZDDLSHJQKM55NHV.json","graph_json":"https://pith.science/api/pith-number/GD2CKQXVWKBZDDLSHJQKM55NHV/graph.json","events_json":"https://pith.science/api/pith-number/GD2CKQXVWKBZDDLSHJQKM55NHV/events.json","paper":"https://pith.science/paper/GD2CKQXV"},"agent_actions":{"view_html":"https://pith.science/pith/GD2CKQXVWKBZDDLSHJQKM55NHV","download_json":"https://pith.science/pith/GD2CKQXVWKBZDDLSHJQKM55NHV.json","view_paper":"https://pith.science/paper/GD2CKQXV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.08243&json=true","fetch_graph":"https://pith.science/api/pith-number/GD2CKQXVWKBZDDLSHJQKM55NHV/graph.json","fetch_events":"https://pith.science/api/pith-number/GD2CKQXVWKBZDDLSHJQKM55NHV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GD2CKQXVWKBZDDLSHJQKM55NHV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GD2CKQXVWKBZDDLSHJQKM55NHV/action/storage_attestation","attest_author":"https://pith.science/pith/GD2CKQXVWKBZDDLSHJQKM55NHV/action/author_attestation","sign_citation":"https://pith.science/pith/GD2CKQXVWKBZDDLSHJQKM55NHV/action/citation_signature","submit_replication":"https://pith.science/pith/GD2CKQXVWKBZDDLSHJQKM55NHV/action/replication_record"}},"created_at":"2026-07-05T06:00:51.264173+00:00","updated_at":"2026-07-05T06:00:51.264173+00:00"}