{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:NSD4MLLGKN5OAF2YFC743W6YX4","short_pith_number":"pith:NSD4MLLG","canonical_record":{"source":{"id":"1710.01329","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-10-03T18:15:10Z","cross_cats_sorted":[],"title_canon_sha256":"8469b62548def7629a1510c35ab565c182928e1d7880b63c920739e1536b87f3","abstract_canon_sha256":"6c4c510fc814ee234f9da6288000b86ee4b77673b8463a963c5c248442c77301"},"schema_version":"1.0"},"canonical_sha256":"6c87c62d66537ae0175828bfcddbd8bf04a3a57af52ce788bb4a803137d174c4","source":{"kind":"arxiv","id":"1710.01329","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1710.01329","created_at":"2026-05-18T00:18:09Z"},{"alias_kind":"arxiv_version","alias_value":"1710.01329v3","created_at":"2026-05-18T00:18:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1710.01329","created_at":"2026-05-18T00:18:09Z"},{"alias_kind":"pith_short_12","alias_value":"NSD4MLLGKN5O","created_at":"2026-05-18T12:31:34Z"},{"alias_kind":"pith_short_16","alias_value":"NSD4MLLGKN5OAF2Y","created_at":"2026-05-18T12:31:34Z"},{"alias_kind":"pith_short_8","alias_value":"NSD4MLLG","created_at":"2026-05-18T12:31:34Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:NSD4MLLGKN5OAF2YFC743W6YX4","target":"record","payload":{"canonical_record":{"source":{"id":"1710.01329","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-10-03T18:15:10Z","cross_cats_sorted":[],"title_canon_sha256":"8469b62548def7629a1510c35ab565c182928e1d7880b63c920739e1536b87f3","abstract_canon_sha256":"6c4c510fc814ee234f9da6288000b86ee4b77673b8463a963c5c248442c77301"},"schema_version":"1.0"},"canonical_sha256":"6c87c62d66537ae0175828bfcddbd8bf04a3a57af52ce788bb4a803137d174c4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:18:09.460789Z","signature_b64":"VExggtzNAkYl6YN5QJYPsPzZAXhRADfsvLHUY0j+PhOx/8piYPmVUCBkzbHTqPmIcYIIL/0Tt67SXcdgucYdDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6c87c62d66537ae0175828bfcddbd8bf04a3a57af52ce788bb4a803137d174c4","last_reissued_at":"2026-05-18T00:18:09.460282Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:18:09.460282Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1710.01329","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:18:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3qlVuP5Wf2HmNRBaNb+mUzX+t3j77bejzR+GwSsYAEsQWq0DLycqMOWQyr65Ucw/HQmdkqqaIfO8W9z+PgaQAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T02:58:45.125683Z"},"content_sha256":"51e57036bd3f6785ba8fc5cb92f58192e7e3941020224496b98689b7607c877e","schema_version":"1.0","event_id":"sha256:51e57036bd3f6785ba8fc5cb92f58192e7e3941020224496b98689b7607c877e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:NSD4MLLGKN5OAF2YFC743W6YX4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Improving Lexical Choice in Neural Machine Translation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"David Chiang, Toan Q. Nguyen","submitted_at":"2017-10-03T18:15:10Z","abstract_excerpt":"We explore two solutions to the problem of mistranslating rare words in neural machine translation. First, we argue that the standard output layer, which computes the inner product of a vector representing the context with all possible output word embeddings, rewards frequent words disproportionately, and we propose to fix the norms of both vectors to a constant value. Second, we integrate a simple lexical module which is jointly trained with the rest of the model. We evaluate our approaches on eight language pairs with data sizes ranging from 100k to 8M words, and achieve improvements of up t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1710.01329","kind":"arxiv","version":3},"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":"2026-05-18T00:18:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Q/QEKj3eD61oYbETI1fY0ehXL70nkeDdCfJ/sQnsV8IbtgMlLz4lAGd+pQZ02g5DAzbS4jmsjrAlbQI9jF1HCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T02:58:45.126294Z"},"content_sha256":"2b2dbb260b06b967145081cbb35701d70f2c6de507fc0c49aeed7a9819eb57fe","schema_version":"1.0","event_id":"sha256:2b2dbb260b06b967145081cbb35701d70f2c6de507fc0c49aeed7a9819eb57fe"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NSD4MLLGKN5OAF2YFC743W6YX4/bundle.json","state_url":"https://pith.science/pith/NSD4MLLGKN5OAF2YFC743W6YX4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NSD4MLLGKN5OAF2YFC743W6YX4/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":"2026-05-27T02:58:45Z","links":{"resolver":"https://pith.science/pith/NSD4MLLGKN5OAF2YFC743W6YX4","bundle":"https://pith.science/pith/NSD4MLLGKN5OAF2YFC743W6YX4/bundle.json","state":"https://pith.science/pith/NSD4MLLGKN5OAF2YFC743W6YX4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NSD4MLLGKN5OAF2YFC743W6YX4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:NSD4MLLGKN5OAF2YFC743W6YX4","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":"6c4c510fc814ee234f9da6288000b86ee4b77673b8463a963c5c248442c77301","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-10-03T18:15:10Z","title_canon_sha256":"8469b62548def7629a1510c35ab565c182928e1d7880b63c920739e1536b87f3"},"schema_version":"1.0","source":{"id":"1710.01329","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1710.01329","created_at":"2026-05-18T00:18:09Z"},{"alias_kind":"arxiv_version","alias_value":"1710.01329v3","created_at":"2026-05-18T00:18:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1710.01329","created_at":"2026-05-18T00:18:09Z"},{"alias_kind":"pith_short_12","alias_value":"NSD4MLLGKN5O","created_at":"2026-05-18T12:31:34Z"},{"alias_kind":"pith_short_16","alias_value":"NSD4MLLGKN5OAF2Y","created_at":"2026-05-18T12:31:34Z"},{"alias_kind":"pith_short_8","alias_value":"NSD4MLLG","created_at":"2026-05-18T12:31:34Z"}],"graph_snapshots":[{"event_id":"sha256:2b2dbb260b06b967145081cbb35701d70f2c6de507fc0c49aeed7a9819eb57fe","target":"graph","created_at":"2026-05-18T00:18:09Z","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 explore two solutions to the problem of mistranslating rare words in neural machine translation. First, we argue that the standard output layer, which computes the inner product of a vector representing the context with all possible output word embeddings, rewards frequent words disproportionately, and we propose to fix the norms of both vectors to a constant value. Second, we integrate a simple lexical module which is jointly trained with the rest of the model. We evaluate our approaches on eight language pairs with data sizes ranging from 100k to 8M words, and achieve improvements of up t","authors_text":"David Chiang, Toan Q. Nguyen","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-10-03T18:15:10Z","title":"Improving Lexical Choice in Neural Machine Translation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1710.01329","kind":"arxiv","version":3},"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":"sha256:51e57036bd3f6785ba8fc5cb92f58192e7e3941020224496b98689b7607c877e","target":"record","created_at":"2026-05-18T00:18:09Z","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":"6c4c510fc814ee234f9da6288000b86ee4b77673b8463a963c5c248442c77301","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-10-03T18:15:10Z","title_canon_sha256":"8469b62548def7629a1510c35ab565c182928e1d7880b63c920739e1536b87f3"},"schema_version":"1.0","source":{"id":"1710.01329","kind":"arxiv","version":3}},"canonical_sha256":"6c87c62d66537ae0175828bfcddbd8bf04a3a57af52ce788bb4a803137d174c4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6c87c62d66537ae0175828bfcddbd8bf04a3a57af52ce788bb4a803137d174c4","first_computed_at":"2026-05-18T00:18:09.460282Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:18:09.460282Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"VExggtzNAkYl6YN5QJYPsPzZAXhRADfsvLHUY0j+PhOx/8piYPmVUCBkzbHTqPmIcYIIL/0Tt67SXcdgucYdDQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:18:09.460789Z","signed_message":"canonical_sha256_bytes"},"source_id":"1710.01329","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:51e57036bd3f6785ba8fc5cb92f58192e7e3941020224496b98689b7607c877e","sha256:2b2dbb260b06b967145081cbb35701d70f2c6de507fc0c49aeed7a9819eb57fe"],"state_sha256":"c7d35249db1f02703e65a2de2ec7f44ca22693c7be3a6dc726c26527b7ff60bd"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"E/GFr3GqOAhTcFyKTVTlynwwf5u4uEjaCufY9wofKnEtvuV6ZkFBhAWIqSomFxZwvR3NobGxw/3ZbFBEhIwoAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T02:58:45.129561Z","bundle_sha256":"8115133e6f0543f10c43683cc5439ab4d1c2dc35b7e37749d1a14bd440112fcc"}}