{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:GHPEJSEIJ3ZATZLEWQ67UTQZAC","short_pith_number":"pith:GHPEJSEI","canonical_record":{"source":{"id":"1805.04623","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-05-12T00:26:29Z","cross_cats_sorted":[],"title_canon_sha256":"53addaa55793fada59e6102a54437c58e0c351785223505be836d18e458ed6f3","abstract_canon_sha256":"519286d8e7d10ff0926737cd195f8239b11cc41e2fd560d7c9f681843c2e0037"},"schema_version":"1.0"},"canonical_sha256":"31de44c8884ef209e564b43dfa4e19009da2cbb95bf6e57a0226aefb38fcf22a","source":{"kind":"arxiv","id":"1805.04623","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1805.04623","created_at":"2026-05-18T00:16:06Z"},{"alias_kind":"arxiv_version","alias_value":"1805.04623v1","created_at":"2026-05-18T00:16:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1805.04623","created_at":"2026-05-18T00:16:06Z"},{"alias_kind":"pith_short_12","alias_value":"GHPEJSEIJ3ZA","created_at":"2026-05-18T12:32:25Z"},{"alias_kind":"pith_short_16","alias_value":"GHPEJSEIJ3ZATZLE","created_at":"2026-05-18T12:32:25Z"},{"alias_kind":"pith_short_8","alias_value":"GHPEJSEI","created_at":"2026-05-18T12:32:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:GHPEJSEIJ3ZATZLEWQ67UTQZAC","target":"record","payload":{"canonical_record":{"source":{"id":"1805.04623","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-05-12T00:26:29Z","cross_cats_sorted":[],"title_canon_sha256":"53addaa55793fada59e6102a54437c58e0c351785223505be836d18e458ed6f3","abstract_canon_sha256":"519286d8e7d10ff0926737cd195f8239b11cc41e2fd560d7c9f681843c2e0037"},"schema_version":"1.0"},"canonical_sha256":"31de44c8884ef209e564b43dfa4e19009da2cbb95bf6e57a0226aefb38fcf22a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:16:06.856999Z","signature_b64":"tLi8qiEtaDAERl03DUTMiK157rcE6SMo6mEcEkEvcBXTd8LZhEO/cJciUO6jZwK+LPu+SqgJbE9YWorQtpnRAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"31de44c8884ef209e564b43dfa4e19009da2cbb95bf6e57a0226aefb38fcf22a","last_reissued_at":"2026-05-18T00:16:06.856297Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:16:06.856297Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1805.04623","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:16:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yZ7l1UxahpyjUhp8QrUWIhY3uThjODRsFTX0d4CZjWun5Jga3Cg9h/WdMc+n4+5rQK8rkX9Ux4/ByzEgXiC7CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T20:27:34.346354Z"},"content_sha256":"a8565193d7edf9e0802b94306bc134a8eafb17c4812a4a0a370d6b08ad2f3187","schema_version":"1.0","event_id":"sha256:a8565193d7edf9e0802b94306bc134a8eafb17c4812a4a0a370d6b08ad2f3187"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:GHPEJSEIJ3ZATZLEWQ67UTQZAC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Sharp Nearby, Fuzzy Far Away: How Neural Language Models Use Context","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dan Jurafsky, He He, Peng Qi, Urvashi Khandelwal","submitted_at":"2018-05-12T00:26:29Z","abstract_excerpt":"We know very little about how neural language models (LM) use prior linguistic context. In this paper, we investigate the role of context in an LSTM LM, through ablation studies. Specifically, we analyze the increase in perplexity when prior context words are shuffled, replaced, or dropped. On two standard datasets, Penn Treebank and WikiText-2, we find that the model is capable of using about 200 tokens of context on average, but sharply distinguishes nearby context (recent 50 tokens) from the distant history. The model is highly sensitive to the order of words within the most recent sentence"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1805.04623","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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:16:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9tGs6eFjE/5HSuQKDrHKDe84tU+KTe5m3JhVCWJd16WaFg66I0gPOzqNG/45FFtRjZnsdS1YpCy2kwTEoAtKAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T20:27:34.346730Z"},"content_sha256":"5e398c8a7f674a8fc1221791e5fd4359d572ab08cffb7c2b8562b2e52fde9236","schema_version":"1.0","event_id":"sha256:5e398c8a7f674a8fc1221791e5fd4359d572ab08cffb7c2b8562b2e52fde9236"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GHPEJSEIJ3ZATZLEWQ67UTQZAC/bundle.json","state_url":"https://pith.science/pith/GHPEJSEIJ3ZATZLEWQ67UTQZAC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GHPEJSEIJ3ZATZLEWQ67UTQZAC/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":"2026-06-04T20:27:34Z","links":{"resolver":"https://pith.science/pith/GHPEJSEIJ3ZATZLEWQ67UTQZAC","bundle":"https://pith.science/pith/GHPEJSEIJ3ZATZLEWQ67UTQZAC/bundle.json","state":"https://pith.science/pith/GHPEJSEIJ3ZATZLEWQ67UTQZAC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GHPEJSEIJ3ZATZLEWQ67UTQZAC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:GHPEJSEIJ3ZATZLEWQ67UTQZAC","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":"519286d8e7d10ff0926737cd195f8239b11cc41e2fd560d7c9f681843c2e0037","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-05-12T00:26:29Z","title_canon_sha256":"53addaa55793fada59e6102a54437c58e0c351785223505be836d18e458ed6f3"},"schema_version":"1.0","source":{"id":"1805.04623","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1805.04623","created_at":"2026-05-18T00:16:06Z"},{"alias_kind":"arxiv_version","alias_value":"1805.04623v1","created_at":"2026-05-18T00:16:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1805.04623","created_at":"2026-05-18T00:16:06Z"},{"alias_kind":"pith_short_12","alias_value":"GHPEJSEIJ3ZA","created_at":"2026-05-18T12:32:25Z"},{"alias_kind":"pith_short_16","alias_value":"GHPEJSEIJ3ZATZLE","created_at":"2026-05-18T12:32:25Z"},{"alias_kind":"pith_short_8","alias_value":"GHPEJSEI","created_at":"2026-05-18T12:32:25Z"}],"graph_snapshots":[{"event_id":"sha256:5e398c8a7f674a8fc1221791e5fd4359d572ab08cffb7c2b8562b2e52fde9236","target":"graph","created_at":"2026-05-18T00:16:06Z","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 know very little about how neural language models (LM) use prior linguistic context. In this paper, we investigate the role of context in an LSTM LM, through ablation studies. Specifically, we analyze the increase in perplexity when prior context words are shuffled, replaced, or dropped. On two standard datasets, Penn Treebank and WikiText-2, we find that the model is capable of using about 200 tokens of context on average, but sharply distinguishes nearby context (recent 50 tokens) from the distant history. The model is highly sensitive to the order of words within the most recent sentence","authors_text":"Dan Jurafsky, He He, Peng Qi, Urvashi Khandelwal","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-05-12T00:26:29Z","title":"Sharp Nearby, Fuzzy Far Away: How Neural Language Models Use Context"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1805.04623","kind":"arxiv","version":1},"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:a8565193d7edf9e0802b94306bc134a8eafb17c4812a4a0a370d6b08ad2f3187","target":"record","created_at":"2026-05-18T00:16:06Z","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":"519286d8e7d10ff0926737cd195f8239b11cc41e2fd560d7c9f681843c2e0037","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-05-12T00:26:29Z","title_canon_sha256":"53addaa55793fada59e6102a54437c58e0c351785223505be836d18e458ed6f3"},"schema_version":"1.0","source":{"id":"1805.04623","kind":"arxiv","version":1}},"canonical_sha256":"31de44c8884ef209e564b43dfa4e19009da2cbb95bf6e57a0226aefb38fcf22a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"31de44c8884ef209e564b43dfa4e19009da2cbb95bf6e57a0226aefb38fcf22a","first_computed_at":"2026-05-18T00:16:06.856297Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:16:06.856297Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"tLi8qiEtaDAERl03DUTMiK157rcE6SMo6mEcEkEvcBXTd8LZhEO/cJciUO6jZwK+LPu+SqgJbE9YWorQtpnRAw==","signature_status":"signed_v1","signed_at":"2026-05-18T00:16:06.856999Z","signed_message":"canonical_sha256_bytes"},"source_id":"1805.04623","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a8565193d7edf9e0802b94306bc134a8eafb17c4812a4a0a370d6b08ad2f3187","sha256:5e398c8a7f674a8fc1221791e5fd4359d572ab08cffb7c2b8562b2e52fde9236"],"state_sha256":"10b944d3c5774d2a10eb76b0dafac5dd991dd2347ac1422bfc5e267b2682e21e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VgMZPp9LBuZ8dvOSv35WjLOG9t+Lvv4++NwQqCaER9NPDB+rsifVn8KgiWTjZlMWz+X7gvZgtR0QHO1xufEzCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-04T20:27:34.349072Z","bundle_sha256":"3797d53a7d0303ae9ac7c2e636e535d9efffdc7a3d1bb4af78e5aca92cc5a283"}}