{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2015:UXE7A2VVOOOOGYN7THKPFJT6P3","short_pith_number":"pith:UXE7A2VV","canonical_record":{"source":{"id":"1511.01574","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2015-11-05T01:45:29Z","cross_cats_sorted":[],"title_canon_sha256":"c0626740ed780b8bc41969c38187b7d04165597a58dc421503b943a3d7fa3549","abstract_canon_sha256":"c51b5caed5266057c739a9166052b4610d9efb70f034fc9cff110fffde56627d"},"schema_version":"1.0"},"canonical_sha256":"a5c9f06ab5739ce361bf99d4f2a67e7eca6412576618fb4a07c649918b8f34d8","source":{"kind":"arxiv","id":"1511.01574","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1511.01574","created_at":"2026-05-18T01:20:16Z"},{"alias_kind":"arxiv_version","alias_value":"1511.01574v2","created_at":"2026-05-18T01:20:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1511.01574","created_at":"2026-05-18T01:20:16Z"},{"alias_kind":"pith_short_12","alias_value":"UXE7A2VVOOOO","created_at":"2026-05-18T12:29:44Z"},{"alias_kind":"pith_short_16","alias_value":"UXE7A2VVOOOOGYN7","created_at":"2026-05-18T12:29:44Z"},{"alias_kind":"pith_short_8","alias_value":"UXE7A2VV","created_at":"2026-05-18T12:29:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2015:UXE7A2VVOOOOGYN7THKPFJT6P3","target":"record","payload":{"canonical_record":{"source":{"id":"1511.01574","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2015-11-05T01:45:29Z","cross_cats_sorted":[],"title_canon_sha256":"c0626740ed780b8bc41969c38187b7d04165597a58dc421503b943a3d7fa3549","abstract_canon_sha256":"c51b5caed5266057c739a9166052b4610d9efb70f034fc9cff110fffde56627d"},"schema_version":"1.0"},"canonical_sha256":"a5c9f06ab5739ce361bf99d4f2a67e7eca6412576618fb4a07c649918b8f34d8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:20:16.583218Z","signature_b64":"rzpiWeKkQZEddsUiu6xkLNgUcjRY2yKirGCmJtFxcoAtaqdx1S1KzcE/+fArQ65YqRTfIDoZOL1H6x33cGu2Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a5c9f06ab5739ce361bf99d4f2a67e7eca6412576618fb4a07c649918b8f34d8","last_reissued_at":"2026-05-18T01:20:16.582586Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:20:16.582586Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1511.01574","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":"2026-05-18T01:20:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"F6nXkvpAUkylujXp7Vn/7MivBef5XscJY4uRWXf8apLDNZGg9BOUJ17FBHmfk/DURaNuh6BSw4kuDGQXXbDzCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-08T21:02:40.859999Z"},"content_sha256":"d1ebfe8a234ad682a981b2cb7fcb62c66b55587f7c803abe7cf9c221d06ec7d6","schema_version":"1.0","event_id":"sha256:d1ebfe8a234ad682a981b2cb7fcb62c66b55587f7c803abe7cf9c221d06ec7d6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2015:UXE7A2VVOOOOGYN7THKPFJT6P3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Multinomial Loss on Held-out Data for the Sparse Non-negative Matrix Language Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ciprian Chelba, Fernando Pereira","submitted_at":"2015-11-05T01:45:29Z","abstract_excerpt":"We describe Sparse Non-negative Matrix (SNM) language model estimation using multinomial loss on held-out data.\n  Being able to train on held-out data is important in practical situations where the training data is usually mismatched from the held-out/test data. It is also less constrained than the previous training algorithm using leave-one-out on training data: it allows the use of richer meta-features in the adjustment model, e.g. the diversity counts used by Kneser-Ney smoothing which would be difficult to deal with correctly in leave-one-out training.\n  In experiments on the one billion w"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1511.01574","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":"2026-05-18T01:20:16Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qMQ7QSFR1RqDvd3w4IuAbOCipgwnP+WWO4fC0Af81EsTtXMdWzAlLBeVEyZtvtiugaEY6kxKX8sHz8U5quVKDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-08T21:02:40.860725Z"},"content_sha256":"6f732b78030d000aad1709b984a418e05ab081102a0d29a4daf45f3009848b47","schema_version":"1.0","event_id":"sha256:6f732b78030d000aad1709b984a418e05ab081102a0d29a4daf45f3009848b47"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UXE7A2VVOOOOGYN7THKPFJT6P3/bundle.json","state_url":"https://pith.science/pith/UXE7A2VVOOOOGYN7THKPFJT6P3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UXE7A2VVOOOOGYN7THKPFJT6P3/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":"2026-06-08T21:02:40Z","links":{"resolver":"https://pith.science/pith/UXE7A2VVOOOOGYN7THKPFJT6P3","bundle":"https://pith.science/pith/UXE7A2VVOOOOGYN7THKPFJT6P3/bundle.json","state":"https://pith.science/pith/UXE7A2VVOOOOGYN7THKPFJT6P3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UXE7A2VVOOOOGYN7THKPFJT6P3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2015:UXE7A2VVOOOOGYN7THKPFJT6P3","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":"c51b5caed5266057c739a9166052b4610d9efb70f034fc9cff110fffde56627d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2015-11-05T01:45:29Z","title_canon_sha256":"c0626740ed780b8bc41969c38187b7d04165597a58dc421503b943a3d7fa3549"},"schema_version":"1.0","source":{"id":"1511.01574","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1511.01574","created_at":"2026-05-18T01:20:16Z"},{"alias_kind":"arxiv_version","alias_value":"1511.01574v2","created_at":"2026-05-18T01:20:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1511.01574","created_at":"2026-05-18T01:20:16Z"},{"alias_kind":"pith_short_12","alias_value":"UXE7A2VVOOOO","created_at":"2026-05-18T12:29:44Z"},{"alias_kind":"pith_short_16","alias_value":"UXE7A2VVOOOOGYN7","created_at":"2026-05-18T12:29:44Z"},{"alias_kind":"pith_short_8","alias_value":"UXE7A2VV","created_at":"2026-05-18T12:29:44Z"}],"graph_snapshots":[{"event_id":"sha256:6f732b78030d000aad1709b984a418e05ab081102a0d29a4daf45f3009848b47","target":"graph","created_at":"2026-05-18T01:20:16Z","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":"We describe Sparse Non-negative Matrix (SNM) language model estimation using multinomial loss on held-out data.\n  Being able to train on held-out data is important in practical situations where the training data is usually mismatched from the held-out/test data. It is also less constrained than the previous training algorithm using leave-one-out on training data: it allows the use of richer meta-features in the adjustment model, e.g. the diversity counts used by Kneser-Ney smoothing which would be difficult to deal with correctly in leave-one-out training.\n  In experiments on the one billion w","authors_text":"Ciprian Chelba, Fernando Pereira","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2015-11-05T01:45:29Z","title":"Multinomial Loss on Held-out Data for the Sparse Non-negative Matrix Language Model"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1511.01574","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:d1ebfe8a234ad682a981b2cb7fcb62c66b55587f7c803abe7cf9c221d06ec7d6","target":"record","created_at":"2026-05-18T01:20:16Z","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":"c51b5caed5266057c739a9166052b4610d9efb70f034fc9cff110fffde56627d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2015-11-05T01:45:29Z","title_canon_sha256":"c0626740ed780b8bc41969c38187b7d04165597a58dc421503b943a3d7fa3549"},"schema_version":"1.0","source":{"id":"1511.01574","kind":"arxiv","version":2}},"canonical_sha256":"a5c9f06ab5739ce361bf99d4f2a67e7eca6412576618fb4a07c649918b8f34d8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a5c9f06ab5739ce361bf99d4f2a67e7eca6412576618fb4a07c649918b8f34d8","first_computed_at":"2026-05-18T01:20:16.582586Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T01:20:16.582586Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"rzpiWeKkQZEddsUiu6xkLNgUcjRY2yKirGCmJtFxcoAtaqdx1S1KzcE/+fArQ65YqRTfIDoZOL1H6x33cGu2Dw==","signature_status":"signed_v1","signed_at":"2026-05-18T01:20:16.583218Z","signed_message":"canonical_sha256_bytes"},"source_id":"1511.01574","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d1ebfe8a234ad682a981b2cb7fcb62c66b55587f7c803abe7cf9c221d06ec7d6","sha256:6f732b78030d000aad1709b984a418e05ab081102a0d29a4daf45f3009848b47"],"state_sha256":"cb67415a571f776b5de506fd4df47acffe19043f4983c9aafa8e40d8f2c78bab"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jklZyHu0EuRaKlUOBK5Xc91XYVJR25zzUA2dckepdUTlIFbw//mIUdUQxw6jXCHGrmOJCzvZ7gHCUom1chuhAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-08T21:02:40.864602Z","bundle_sha256":"e0b9b4b3c5cce8d783f362466ca33b88213ee69bff14a66af287e905193015cf"}}