{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:4L6ZCE7BZHSVCI7FGPBPBXSUNL","short_pith_number":"pith:4L6ZCE7B","schema_version":"1.0","canonical_sha256":"e2fd9113e1c9e55123e533c2f0de546af677c28e88c416953e5630c6c68ef3b6","source":{"kind":"arxiv","id":"1701.05517","version":1},"attestation_state":"computed","paper":{"title":"PixelCNN++: Improving the PixelCNN with Discretized Logistic Mixture Likelihood and Other Modifications","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrej Karpathy, Diederik P. Kingma, Tim Salimans, Xi Chen","submitted_at":"2017-01-19T17:29:06Z","abstract_excerpt":"PixelCNNs are a recently proposed class of powerful generative models with tractable likelihood. Here we discuss our implementation of PixelCNNs which we make available at https://github.com/openai/pixel-cnn. Our implementation contains a number of modifications to the original model that both simplify its structure and improve its performance. 1) We use a discretized logistic mixture likelihood on the pixels, rather than a 256-way softmax, which we find to speed up training. 2) We condition on whole pixels, rather than R/G/B sub-pixels, simplifying the model structure. 3) We use downsampling "},"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":"1701.05517","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-01-19T17:29:06Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"23938590f30b952f11344b7064e165eb6a6426dbc73d83f90bb04a9e47a0253b","abstract_canon_sha256":"3bde8f301fd60968a44727823a06c26bcd397a4aceb284c5dfd611e3402b47f4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:52:22.020234Z","signature_b64":"NFDBMp1d5IS3hMgysP9+m5fKip9wxkQZfGcb4rsyYH55o10IIdBXh00aVhJkl0KmDzcjf+RkzrP3Egnyv6D0Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e2fd9113e1c9e55123e533c2f0de546af677c28e88c416953e5630c6c68ef3b6","last_reissued_at":"2026-05-18T00:52:22.019643Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:52:22.019643Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PixelCNN++: Improving the PixelCNN with Discretized Logistic Mixture Likelihood and Other Modifications","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrej Karpathy, Diederik P. Kingma, Tim Salimans, Xi Chen","submitted_at":"2017-01-19T17:29:06Z","abstract_excerpt":"PixelCNNs are a recently proposed class of powerful generative models with tractable likelihood. Here we discuss our implementation of PixelCNNs which we make available at https://github.com/openai/pixel-cnn. Our implementation contains a number of modifications to the original model that both simplify its structure and improve its performance. 1) We use a discretized logistic mixture likelihood on the pixels, rather than a 256-way softmax, which we find to speed up training. 2) We condition on whole pixels, rather than R/G/B sub-pixels, simplifying the model structure. 3) We use downsampling "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1701.05517","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":""},"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":"1701.05517","created_at":"2026-05-18T00:52:22.019744+00:00"},{"alias_kind":"arxiv_version","alias_value":"1701.05517v1","created_at":"2026-05-18T00:52:22.019744+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1701.05517","created_at":"2026-05-18T00:52:22.019744+00:00"},{"alias_kind":"pith_short_12","alias_value":"4L6ZCE7BZHSV","created_at":"2026-05-18T12:31:00.734936+00:00"},{"alias_kind":"pith_short_16","alias_value":"4L6ZCE7BZHSVCI7F","created_at":"2026-05-18T12:31:00.734936+00:00"},{"alias_kind":"pith_short_8","alias_value":"4L6ZCE7B","created_at":"2026-05-18T12:31:00.734936+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"1906.09237","citing_title":"Shaping Belief States with Generative Environment Models for RL","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2510.17381","citing_title":"Beyond Binary Out-of-Distribution Detection: Characterizing Distributional Shifts with Multi-Statistic Diffusion Trajectories","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2104.10157","citing_title":"VideoGPT: Video Generation using VQ-VAE and Transformers","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2212.09748","citing_title":"Scalable Diffusion Models with Transformers","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2112.10752","citing_title":"High-Resolution Image Synthesis with Latent Diffusion Models","ref_index":73,"is_internal_anchor":false},{"citing_arxiv_id":"2111.00396","citing_title":"Efficiently Modeling Long Sequences with Structured State Spaces","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"1904.10509","citing_title":"Generating Long Sequences with Sparse Transformers","ref_index":20,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4L6ZCE7BZHSVCI7FGPBPBXSUNL","json":"https://pith.science/pith/4L6ZCE7BZHSVCI7FGPBPBXSUNL.json","graph_json":"https://pith.science/api/pith-number/4L6ZCE7BZHSVCI7FGPBPBXSUNL/graph.json","events_json":"https://pith.science/api/pith-number/4L6ZCE7BZHSVCI7FGPBPBXSUNL/events.json","paper":"https://pith.science/paper/4L6ZCE7B"},"agent_actions":{"view_html":"https://pith.science/pith/4L6ZCE7BZHSVCI7FGPBPBXSUNL","download_json":"https://pith.science/pith/4L6ZCE7BZHSVCI7FGPBPBXSUNL.json","view_paper":"https://pith.science/paper/4L6ZCE7B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1701.05517&json=true","fetch_graph":"https://pith.science/api/pith-number/4L6ZCE7BZHSVCI7FGPBPBXSUNL/graph.json","fetch_events":"https://pith.science/api/pith-number/4L6ZCE7BZHSVCI7FGPBPBXSUNL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4L6ZCE7BZHSVCI7FGPBPBXSUNL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4L6ZCE7BZHSVCI7FGPBPBXSUNL/action/storage_attestation","attest_author":"https://pith.science/pith/4L6ZCE7BZHSVCI7FGPBPBXSUNL/action/author_attestation","sign_citation":"https://pith.science/pith/4L6ZCE7BZHSVCI7FGPBPBXSUNL/action/citation_signature","submit_replication":"https://pith.science/pith/4L6ZCE7BZHSVCI7FGPBPBXSUNL/action/replication_record"}},"created_at":"2026-05-18T00:52:22.019744+00:00","updated_at":"2026-05-18T00:52:22.019744+00:00"}