{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:JR3SU2CMWGHQSDPC6CB47N5F6O","short_pith_number":"pith:JR3SU2CM","canonical_record":{"source":{"id":"1808.08987","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-08-27T18:31:33Z","cross_cats_sorted":[],"title_canon_sha256":"0679e3f19834605d1fb0cccb2592c235b5245b7a4b62cb8d2a6f9b16e317bfa9","abstract_canon_sha256":"42020ccc8264029cf6725eed9369c481697691022ed6af07024374239201bed6"},"schema_version":"1.0"},"canonical_sha256":"4c772a684cb18f090de2f083cfb7a5f3ba79f8edf0f1551a6b749e6a1176efa4","source":{"kind":"arxiv","id":"1808.08987","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1808.08987","created_at":"2026-05-18T00:07:04Z"},{"alias_kind":"arxiv_version","alias_value":"1808.08987v1","created_at":"2026-05-18T00:07:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1808.08987","created_at":"2026-05-18T00:07:04Z"},{"alias_kind":"pith_short_12","alias_value":"JR3SU2CMWGHQ","created_at":"2026-05-18T12:32:31Z"},{"alias_kind":"pith_short_16","alias_value":"JR3SU2CMWGHQSDPC","created_at":"2026-05-18T12:32:31Z"},{"alias_kind":"pith_short_8","alias_value":"JR3SU2CM","created_at":"2026-05-18T12:32:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:JR3SU2CMWGHQSDPC6CB47N5F6O","target":"record","payload":{"canonical_record":{"source":{"id":"1808.08987","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-08-27T18:31:33Z","cross_cats_sorted":[],"title_canon_sha256":"0679e3f19834605d1fb0cccb2592c235b5245b7a4b62cb8d2a6f9b16e317bfa9","abstract_canon_sha256":"42020ccc8264029cf6725eed9369c481697691022ed6af07024374239201bed6"},"schema_version":"1.0"},"canonical_sha256":"4c772a684cb18f090de2f083cfb7a5f3ba79f8edf0f1551a6b749e6a1176efa4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:07:04.969048Z","signature_b64":"oEPudxSp3bDmpLVredafl93WFQLDsDxA3peMWy7HTklzhBFraYa0L1B7O/n0xW+4oxUohzEOdXg9MHK/z4T+DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4c772a684cb18f090de2f083cfb7a5f3ba79f8edf0f1551a6b749e6a1176efa4","last_reissued_at":"2026-05-18T00:07:04.968363Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:07:04.968363Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1808.08987","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":"2026-05-18T00:07:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SCMfohGUsB4xfqOAdGR7w9hc3MrW6xiHCUF02Q118MPJxsINPQnd7lNKvyVS9bQlK5m4NK18dHJZx5+RauJFDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T07:56:21.650467Z"},"content_sha256":"fb235fa658c830f1878169d7032ddfa60e189e1030edc3349e775bb528bc9ad9","schema_version":"1.0","event_id":"sha256:fb235fa658c830f1878169d7032ddfa60e189e1030edc3349e775bb528bc9ad9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:JR3SU2CMWGHQSDPC6CB47N5F6O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Large Margin Neural Language Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jiaji Huang, Liang Huang, Wei Ping, Yi Li","submitted_at":"2018-08-27T18:31:33Z","abstract_excerpt":"We propose a large margin criterion for training neural language models. Conventionally, neural language models are trained by minimizing perplexity (PPL) on grammatical sentences. However, we demonstrate that PPL may not be the best metric to optimize in some tasks, and further propose a large margin formulation. The proposed method aims to enlarge the margin between the \"good\" and \"bad\" sentences in a task-specific sense. It is trained end-to-end and can be widely applied to tasks that involve re-scoring of generated text. Compared with minimum-PPL training, our method gains up to 1.1 WER re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1808.08987","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":"2026-05-18T00:07:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"n7IgdNnW0M4bZ6UJ0hbZogAGdTTHF+wEAxmys8HdHfqDawS39iuLmx/eee2rtJmJBcdl7IipDG3f0lYA2urQCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T07:56:21.651084Z"},"content_sha256":"c7704c14ff8f6dd42f4a77879dc8e21ba916c96adddd256e27d70ecdab98ee94","schema_version":"1.0","event_id":"sha256:c7704c14ff8f6dd42f4a77879dc8e21ba916c96adddd256e27d70ecdab98ee94"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JR3SU2CMWGHQSDPC6CB47N5F6O/bundle.json","state_url":"https://pith.science/pith/JR3SU2CMWGHQSDPC6CB47N5F6O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JR3SU2CMWGHQSDPC6CB47N5F6O/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":"2026-05-31T07:56:21Z","links":{"resolver":"https://pith.science/pith/JR3SU2CMWGHQSDPC6CB47N5F6O","bundle":"https://pith.science/pith/JR3SU2CMWGHQSDPC6CB47N5F6O/bundle.json","state":"https://pith.science/pith/JR3SU2CMWGHQSDPC6CB47N5F6O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JR3SU2CMWGHQSDPC6CB47N5F6O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:JR3SU2CMWGHQSDPC6CB47N5F6O","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":"42020ccc8264029cf6725eed9369c481697691022ed6af07024374239201bed6","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-08-27T18:31:33Z","title_canon_sha256":"0679e3f19834605d1fb0cccb2592c235b5245b7a4b62cb8d2a6f9b16e317bfa9"},"schema_version":"1.0","source":{"id":"1808.08987","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1808.08987","created_at":"2026-05-18T00:07:04Z"},{"alias_kind":"arxiv_version","alias_value":"1808.08987v1","created_at":"2026-05-18T00:07:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1808.08987","created_at":"2026-05-18T00:07:04Z"},{"alias_kind":"pith_short_12","alias_value":"JR3SU2CMWGHQ","created_at":"2026-05-18T12:32:31Z"},{"alias_kind":"pith_short_16","alias_value":"JR3SU2CMWGHQSDPC","created_at":"2026-05-18T12:32:31Z"},{"alias_kind":"pith_short_8","alias_value":"JR3SU2CM","created_at":"2026-05-18T12:32:31Z"}],"graph_snapshots":[{"event_id":"sha256:c7704c14ff8f6dd42f4a77879dc8e21ba916c96adddd256e27d70ecdab98ee94","target":"graph","created_at":"2026-05-18T00:07:04Z","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 propose a large margin criterion for training neural language models. Conventionally, neural language models are trained by minimizing perplexity (PPL) on grammatical sentences. However, we demonstrate that PPL may not be the best metric to optimize in some tasks, and further propose a large margin formulation. The proposed method aims to enlarge the margin between the \"good\" and \"bad\" sentences in a task-specific sense. It is trained end-to-end and can be widely applied to tasks that involve re-scoring of generated text. Compared with minimum-PPL training, our method gains up to 1.1 WER re","authors_text":"Jiaji Huang, Liang Huang, Wei Ping, Yi Li","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-08-27T18:31:33Z","title":"Large Margin Neural Language Model"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1808.08987","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:fb235fa658c830f1878169d7032ddfa60e189e1030edc3349e775bb528bc9ad9","target":"record","created_at":"2026-05-18T00:07:04Z","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":"42020ccc8264029cf6725eed9369c481697691022ed6af07024374239201bed6","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2018-08-27T18:31:33Z","title_canon_sha256":"0679e3f19834605d1fb0cccb2592c235b5245b7a4b62cb8d2a6f9b16e317bfa9"},"schema_version":"1.0","source":{"id":"1808.08987","kind":"arxiv","version":1}},"canonical_sha256":"4c772a684cb18f090de2f083cfb7a5f3ba79f8edf0f1551a6b749e6a1176efa4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4c772a684cb18f090de2f083cfb7a5f3ba79f8edf0f1551a6b749e6a1176efa4","first_computed_at":"2026-05-18T00:07:04.968363Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:07:04.968363Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"oEPudxSp3bDmpLVredafl93WFQLDsDxA3peMWy7HTklzhBFraYa0L1B7O/n0xW+4oxUohzEOdXg9MHK/z4T+DA==","signature_status":"signed_v1","signed_at":"2026-05-18T00:07:04.969048Z","signed_message":"canonical_sha256_bytes"},"source_id":"1808.08987","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fb235fa658c830f1878169d7032ddfa60e189e1030edc3349e775bb528bc9ad9","sha256:c7704c14ff8f6dd42f4a77879dc8e21ba916c96adddd256e27d70ecdab98ee94"],"state_sha256":"635fb6ce1f3e2f492abd2731541eb9fe508090d1921a606fbf1f8b069677832a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"l6783qCSBYnFWF/DWAuYZP64OIYWxYcPnnfHVo9z0OtThSOV2pCfF95bIoKkvvlp9TzOV2pGYCLdXXhkVItrDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T07:56:21.654593Z","bundle_sha256":"8cf6937d38ef0534a8febe882f901636da05d864fb08be1503ad3290362b9ceb"}}