{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:TJZESEOQHT2LMJQFCS44AUAXSV","short_pith_number":"pith:TJZESEOQ","canonical_record":{"source":{"id":"1903.04167","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-03-11T08:20:13Z","cross_cats_sorted":[],"title_canon_sha256":"531eda685cda3ebab003d36fb44bfa6bbf740cd66557ad7d5e313d6356dec572","abstract_canon_sha256":"d3b56728cecbcf531866b2d578c5f790b45ffe02844ddff1806555fac58d1f07"},"schema_version":"1.0"},"canonical_sha256":"9a724911d03cf4b6260514b9c050179571c4881a802798956343d32ff54dbcf9","source":{"kind":"arxiv","id":"1903.04167","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1903.04167","created_at":"2026-05-17T23:51:29Z"},{"alias_kind":"arxiv_version","alias_value":"1903.04167v2","created_at":"2026-05-17T23:51:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1903.04167","created_at":"2026-05-17T23:51:29Z"},{"alias_kind":"pith_short_12","alias_value":"TJZESEOQHT2L","created_at":"2026-05-18T12:33:27Z"},{"alias_kind":"pith_short_16","alias_value":"TJZESEOQHT2LMJQF","created_at":"2026-05-18T12:33:27Z"},{"alias_kind":"pith_short_8","alias_value":"TJZESEOQ","created_at":"2026-05-18T12:33:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:TJZESEOQHT2LMJQFCS44AUAXSV","target":"record","payload":{"canonical_record":{"source":{"id":"1903.04167","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-03-11T08:20:13Z","cross_cats_sorted":[],"title_canon_sha256":"531eda685cda3ebab003d36fb44bfa6bbf740cd66557ad7d5e313d6356dec572","abstract_canon_sha256":"d3b56728cecbcf531866b2d578c5f790b45ffe02844ddff1806555fac58d1f07"},"schema_version":"1.0"},"canonical_sha256":"9a724911d03cf4b6260514b9c050179571c4881a802798956343d32ff54dbcf9","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:51:29.481776Z","signature_b64":"3sb+PqonrAH0NjWXFzGemqDrIYxYWLYIULFkaPZZ9QDL78eOai4ZVQikHiIJMpJ4+ACzbcSwrWu9J9Yq6S0UBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9a724911d03cf4b6260514b9c050179571c4881a802798956343d32ff54dbcf9","last_reissued_at":"2026-05-17T23:51:29.481397Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:51:29.481397Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1903.04167","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":"2026-05-17T23:51:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"n0oa57lOdqVDh0GRkd1d2EiSOP6grrsGehesWUjlJ7YfwfsTCO6BJ1IMSKhpxj+cYx+kTI07Bpjehp7nemNlBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T05:47:43.582816Z"},"content_sha256":"f89f90bfaea406a4a1a14b9f1f0f7c2b2e6e543c5696887c273fbc7651de2b4b","schema_version":"1.0","event_id":"sha256:f89f90bfaea406a4a1a14b9f1f0f7c2b2e6e543c5696887c273fbc7651de2b4b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:TJZESEOQHT2LMJQFCS44AUAXSV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Partially Shuffling the Training Data to Improve Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ofir Press","submitted_at":"2019-03-11T08:20:13Z","abstract_excerpt":"Although SGD requires shuffling the training data between epochs, currently none of the word-level language modeling systems do this. Naively shuffling all sentences in the training data would not permit the model to learn inter-sentence dependencies. Here we present a method that partially shuffles the training data between epochs. This method makes each batch random, while keeping most sentence ordering intact. It achieves new state of the art results on word-level language modeling on both the Penn Treebank and WikiText-2 datasets."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1903.04167","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":"2026-05-17T23:51:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yH868VN4XX2nJc0DV2No+B+CDoJ0Hi836dGC+4sVEtmPtaYP4Z6CAAX9qpsH3PoR/4xliF/WZSFOv1gilK7XDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-16T05:47:43.583262Z"},"content_sha256":"403f4a6c6b7632de48be4cf9f6bfdcdbec52dcbc24dc071d20162414aebaba1a","schema_version":"1.0","event_id":"sha256:403f4a6c6b7632de48be4cf9f6bfdcdbec52dcbc24dc071d20162414aebaba1a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TJZESEOQHT2LMJQFCS44AUAXSV/bundle.json","state_url":"https://pith.science/pith/TJZESEOQHT2LMJQFCS44AUAXSV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TJZESEOQHT2LMJQFCS44AUAXSV/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":"2026-08-16T05:47:43Z","links":{"resolver":"https://pith.science/pith/TJZESEOQHT2LMJQFCS44AUAXSV","bundle":"https://pith.science/pith/TJZESEOQHT2LMJQFCS44AUAXSV/bundle.json","state":"https://pith.science/pith/TJZESEOQHT2LMJQFCS44AUAXSV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TJZESEOQHT2LMJQFCS44AUAXSV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:TJZESEOQHT2LMJQFCS44AUAXSV","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":"d3b56728cecbcf531866b2d578c5f790b45ffe02844ddff1806555fac58d1f07","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-03-11T08:20:13Z","title_canon_sha256":"531eda685cda3ebab003d36fb44bfa6bbf740cd66557ad7d5e313d6356dec572"},"schema_version":"1.0","source":{"id":"1903.04167","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1903.04167","created_at":"2026-05-17T23:51:29Z"},{"alias_kind":"arxiv_version","alias_value":"1903.04167v2","created_at":"2026-05-17T23:51:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1903.04167","created_at":"2026-05-17T23:51:29Z"},{"alias_kind":"pith_short_12","alias_value":"TJZESEOQHT2L","created_at":"2026-05-18T12:33:27Z"},{"alias_kind":"pith_short_16","alias_value":"TJZESEOQHT2LMJQF","created_at":"2026-05-18T12:33:27Z"},{"alias_kind":"pith_short_8","alias_value":"TJZESEOQ","created_at":"2026-05-18T12:33:27Z"}],"graph_snapshots":[{"event_id":"sha256:403f4a6c6b7632de48be4cf9f6bfdcdbec52dcbc24dc071d20162414aebaba1a","target":"graph","created_at":"2026-05-17T23:51:29Z","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":"Although SGD requires shuffling the training data between epochs, currently none of the word-level language modeling systems do this. Naively shuffling all sentences in the training data would not permit the model to learn inter-sentence dependencies. Here we present a method that partially shuffles the training data between epochs. This method makes each batch random, while keeping most sentence ordering intact. It achieves new state of the art results on word-level language modeling on both the Penn Treebank and WikiText-2 datasets.","authors_text":"Ofir Press","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-03-11T08:20:13Z","title":"Partially Shuffling the Training Data to Improve Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1903.04167","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:f89f90bfaea406a4a1a14b9f1f0f7c2b2e6e543c5696887c273fbc7651de2b4b","target":"record","created_at":"2026-05-17T23:51:29Z","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":"d3b56728cecbcf531866b2d578c5f790b45ffe02844ddff1806555fac58d1f07","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-03-11T08:20:13Z","title_canon_sha256":"531eda685cda3ebab003d36fb44bfa6bbf740cd66557ad7d5e313d6356dec572"},"schema_version":"1.0","source":{"id":"1903.04167","kind":"arxiv","version":2}},"canonical_sha256":"9a724911d03cf4b6260514b9c050179571c4881a802798956343d32ff54dbcf9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9a724911d03cf4b6260514b9c050179571c4881a802798956343d32ff54dbcf9","first_computed_at":"2026-05-17T23:51:29.481397Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:51:29.481397Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"3sb+PqonrAH0NjWXFzGemqDrIYxYWLYIULFkaPZZ9QDL78eOai4ZVQikHiIJMpJ4+ACzbcSwrWu9J9Yq6S0UBQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:51:29.481776Z","signed_message":"canonical_sha256_bytes"},"source_id":"1903.04167","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f89f90bfaea406a4a1a14b9f1f0f7c2b2e6e543c5696887c273fbc7651de2b4b","sha256:403f4a6c6b7632de48be4cf9f6bfdcdbec52dcbc24dc071d20162414aebaba1a"],"state_sha256":"a2a53cf67dc999489501ba072a77337982b3497ace26bc758d2a079520bb20c6"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HcjAAYDX2AFnpOnSgStjK4LNVE9BJpj16QtzDpKSDniYKCOOJg+yUGzXb6gkZ14S+F11kRz8M9VRuOkC/qrmBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-16T05:47:43.589031Z","bundle_sha256":"bfe4df2e180a4c0fae34ca3b85ef63cdec2f58b8da8fb4b72b924655aab848dc"}}