{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2015:KU4RHFNTAPSV6IPFQ7A75J4SYJ","short_pith_number":"pith:KU4RHFNT","canonical_record":{"source":{"id":"1506.04147","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2015-06-12T20:00:29Z","cross_cats_sorted":["cs.CL","cs.LG","stat.ME"],"title_canon_sha256":"f16581f9bb42f6cc49aa41925b9f69902fc4dd6c52fb45ac80099925d71e6666","abstract_canon_sha256":"e1a41db4c8a798f85ecbf10d7458fa4e3301320b426ff7267704c9f32ca80ef4"},"schema_version":"1.0"},"canonical_sha256":"55391395b303e55f21e587c1fea792c2751ddad1b41c5c693f79caf67d6df223","source":{"kind":"arxiv","id":"1506.04147","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1506.04147","created_at":"2026-05-18T01:44:10Z"},{"alias_kind":"arxiv_version","alias_value":"1506.04147v2","created_at":"2026-05-18T01:44:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1506.04147","created_at":"2026-05-18T01:44:10Z"},{"alias_kind":"pith_short_12","alias_value":"KU4RHFNTAPSV","created_at":"2026-05-18T12:29:29Z"},{"alias_kind":"pith_short_16","alias_value":"KU4RHFNTAPSV6IPF","created_at":"2026-05-18T12:29:29Z"},{"alias_kind":"pith_short_8","alias_value":"KU4RHFNT","created_at":"2026-05-18T12:29:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2015:KU4RHFNTAPSV6IPFQ7A75J4SYJ","target":"record","payload":{"canonical_record":{"source":{"id":"1506.04147","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2015-06-12T20:00:29Z","cross_cats_sorted":["cs.CL","cs.LG","stat.ME"],"title_canon_sha256":"f16581f9bb42f6cc49aa41925b9f69902fc4dd6c52fb45ac80099925d71e6666","abstract_canon_sha256":"e1a41db4c8a798f85ecbf10d7458fa4e3301320b426ff7267704c9f32ca80ef4"},"schema_version":"1.0"},"canonical_sha256":"55391395b303e55f21e587c1fea792c2751ddad1b41c5c693f79caf67d6df223","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:44:10.798861Z","signature_b64":"SpqYHZQ6cFCDDlu19OaZ/WKE6js0lNubh8p3a1xM4nXMyjurt29RBdZVZlQ/wI39bhCQJbbHgF0g4pe4FTvlDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"55391395b303e55f21e587c1fea792c2751ddad1b41c5c693f79caf67d6df223","last_reissued_at":"2026-05-18T01:44:10.798141Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:44:10.798141Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1506.04147","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T01:44:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DUn0ArXW2LwUsGd3jQ7Odimkw8d8j9NFxsCM/GdFZN6iQM0SZsHbHzkFBolV9qVjGeSImMhO/RdrfawE0R8gCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T03:55:37.657537Z"},"content_sha256":"7873001ac4070e8aaa517e8171ca9a776494682f6c215ab8e62a5d405e2448eb","schema_version":"1.0","event_id":"sha256:7873001ac4070e8aaa517e8171ca9a776494682f6c215ab8e62a5d405e2448eb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2015:KU4RHFNTAPSV6IPFQ7A75J4SYJ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On the accuracy of self-normalized log-linear models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","stat.ME"],"primary_cat":"stat.ML","authors_text":"Dan Klein, Jacob Andreas, Maxim Rabinovich, Michael I. Jordan","submitted_at":"2015-06-12T20:00:29Z","abstract_excerpt":"Calculation of the log-normalizer is a major computational obstacle in applications of log-linear models with large output spaces. The problem of fast normalizer computation has therefore attracted significant attention in the theoretical and applied machine learning literature. In this paper, we analyze a recently proposed technique known as \"self-normalization\", which introduces a regularization term in training to penalize log normalizers for deviating from zero. This makes it possible to use unnormalized model scores as approximate probabilities. Empirical evidence suggests that self-norma"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1506.04147","kind":"arxiv","version":2},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T01:44:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CZQbb9uiEmlgCzoVZqVyh0t+fZMrMkBku2QeUjgPZ2OSompinnMNodADJiRVZaGQYI6eRzbA7vkt5ik/pi9kCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T03:55:37.657914Z"},"content_sha256":"27f367ae242063ff3976700fd37bd2526310eb8961c748ae66d5dc5ba7f7c97c","schema_version":"1.0","event_id":"sha256:27f367ae242063ff3976700fd37bd2526310eb8961c748ae66d5dc5ba7f7c97c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KU4RHFNTAPSV6IPFQ7A75J4SYJ/bundle.json","state_url":"https://pith.science/pith/KU4RHFNTAPSV6IPFQ7A75J4SYJ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KU4RHFNTAPSV6IPFQ7A75J4SYJ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-26T03:55:37Z","links":{"resolver":"https://pith.science/pith/KU4RHFNTAPSV6IPFQ7A75J4SYJ","bundle":"https://pith.science/pith/KU4RHFNTAPSV6IPFQ7A75J4SYJ/bundle.json","state":"https://pith.science/pith/KU4RHFNTAPSV6IPFQ7A75J4SYJ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KU4RHFNTAPSV6IPFQ7A75J4SYJ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2015:KU4RHFNTAPSV6IPFQ7A75J4SYJ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"e1a41db4c8a798f85ecbf10d7458fa4e3301320b426ff7267704c9f32ca80ef4","cross_cats_sorted":["cs.CL","cs.LG","stat.ME"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2015-06-12T20:00:29Z","title_canon_sha256":"f16581f9bb42f6cc49aa41925b9f69902fc4dd6c52fb45ac80099925d71e6666"},"schema_version":"1.0","source":{"id":"1506.04147","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1506.04147","created_at":"2026-05-18T01:44:10Z"},{"alias_kind":"arxiv_version","alias_value":"1506.04147v2","created_at":"2026-05-18T01:44:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1506.04147","created_at":"2026-05-18T01:44:10Z"},{"alias_kind":"pith_short_12","alias_value":"KU4RHFNTAPSV","created_at":"2026-05-18T12:29:29Z"},{"alias_kind":"pith_short_16","alias_value":"KU4RHFNTAPSV6IPF","created_at":"2026-05-18T12:29:29Z"},{"alias_kind":"pith_short_8","alias_value":"KU4RHFNT","created_at":"2026-05-18T12:29:29Z"}],"graph_snapshots":[{"event_id":"sha256:27f367ae242063ff3976700fd37bd2526310eb8961c748ae66d5dc5ba7f7c97c","target":"graph","created_at":"2026-05-18T01:44:10Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Calculation of the log-normalizer is a major computational obstacle in applications of log-linear models with large output spaces. The problem of fast normalizer computation has therefore attracted significant attention in the theoretical and applied machine learning literature. In this paper, we analyze a recently proposed technique known as \"self-normalization\", which introduces a regularization term in training to penalize log normalizers for deviating from zero. This makes it possible to use unnormalized model scores as approximate probabilities. Empirical evidence suggests that self-norma","authors_text":"Dan Klein, Jacob Andreas, Maxim Rabinovich, Michael I. Jordan","cross_cats":["cs.CL","cs.LG","stat.ME"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2015-06-12T20:00:29Z","title":"On the accuracy of self-normalized log-linear models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1506.04147","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:7873001ac4070e8aaa517e8171ca9a776494682f6c215ab8e62a5d405e2448eb","target":"record","created_at":"2026-05-18T01:44:10Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"e1a41db4c8a798f85ecbf10d7458fa4e3301320b426ff7267704c9f32ca80ef4","cross_cats_sorted":["cs.CL","cs.LG","stat.ME"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2015-06-12T20:00:29Z","title_canon_sha256":"f16581f9bb42f6cc49aa41925b9f69902fc4dd6c52fb45ac80099925d71e6666"},"schema_version":"1.0","source":{"id":"1506.04147","kind":"arxiv","version":2}},"canonical_sha256":"55391395b303e55f21e587c1fea792c2751ddad1b41c5c693f79caf67d6df223","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"55391395b303e55f21e587c1fea792c2751ddad1b41c5c693f79caf67d6df223","first_computed_at":"2026-05-18T01:44:10.798141Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T01:44:10.798141Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SpqYHZQ6cFCDDlu19OaZ/WKE6js0lNubh8p3a1xM4nXMyjurt29RBdZVZlQ/wI39bhCQJbbHgF0g4pe4FTvlDg==","signature_status":"signed_v1","signed_at":"2026-05-18T01:44:10.798861Z","signed_message":"canonical_sha256_bytes"},"source_id":"1506.04147","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7873001ac4070e8aaa517e8171ca9a776494682f6c215ab8e62a5d405e2448eb","sha256:27f367ae242063ff3976700fd37bd2526310eb8961c748ae66d5dc5ba7f7c97c"],"state_sha256":"64ff10fe6440fb9935661dc02d5304750c36f3d97fde1c95618a5ccae0e129d5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sDc4R8I436iYArQpWW1AXg9tuBfG8gjCWnEXN8Wq6vU4RRusMjK6aUoT1gpTA/7L6AMUmLWDvyT67Gj6uxIRDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T03:55:37.660700Z","bundle_sha256":"c008276484459d644ae1d7dedb448eb9336938b0c2a9252d73b2bc769e7d8cc9"}}