{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:HOCODXIWSARME27A6ZR7V2GYVQ","short_pith_number":"pith:HOCODXIW","canonical_record":{"source":{"id":"1708.08712","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-08-29T11:56:41Z","cross_cats_sorted":[],"title_canon_sha256":"aa14057bfc02595b04df847be5b95475c9ba964127d14b89f274004d896fc85b","abstract_canon_sha256":"49710031f5ca52f9f833448be81f2c52689184b5417b4604ffae33ba0a3751d9"},"schema_version":"1.0"},"canonical_sha256":"3b84e1dd169022c26be0f663fae8d8ac34becdae6493189d566a7be4726977d1","source":{"kind":"arxiv","id":"1708.08712","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1708.08712","created_at":"2026-05-18T00:00:22Z"},{"alias_kind":"arxiv_version","alias_value":"1708.08712v3","created_at":"2026-05-18T00:00:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1708.08712","created_at":"2026-05-18T00:00:22Z"},{"alias_kind":"pith_short_12","alias_value":"HOCODXIWSARM","created_at":"2026-05-18T12:31:18Z"},{"alias_kind":"pith_short_16","alias_value":"HOCODXIWSARME27A","created_at":"2026-05-18T12:31:18Z"},{"alias_kind":"pith_short_8","alias_value":"HOCODXIW","created_at":"2026-05-18T12:31:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:HOCODXIWSARME27A6ZR7V2GYVQ","target":"record","payload":{"canonical_record":{"source":{"id":"1708.08712","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-08-29T11:56:41Z","cross_cats_sorted":[],"title_canon_sha256":"aa14057bfc02595b04df847be5b95475c9ba964127d14b89f274004d896fc85b","abstract_canon_sha256":"49710031f5ca52f9f833448be81f2c52689184b5417b4604ffae33ba0a3751d9"},"schema_version":"1.0"},"canonical_sha256":"3b84e1dd169022c26be0f663fae8d8ac34becdae6493189d566a7be4726977d1","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:00:22.490179Z","signature_b64":"0bttI2SLy4ort2pzfzYjPN18mS9yCc9G+L1hdFfbIACLIjJrsMc8NGWzxFh+UFk7TbKWqE3vSBqZ1j6sOnVaAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3b84e1dd169022c26be0f663fae8d8ac34becdae6493189d566a7be4726977d1","last_reissued_at":"2026-05-18T00:00:22.489550Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:00:22.489550Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1708.08712","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":"2026-05-18T00:00:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SHE3kvGlLG2WtfJVCJP92UXnetSMMpefnbVo2IjT7pB9Q+Ai01XkJWE/hIgLzFstQBcTiiTIcDLp4T84Og+5Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T01:15:57.604654Z"},"content_sha256":"cde2b262164a26a3cdc40c409f5fb486097035a1822c850a98bcf20ecf569695","schema_version":"1.0","event_id":"sha256:cde2b262164a26a3cdc40c409f5fb486097035a1822c850a98bcf20ecf569695"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:HOCODXIWSARME27A6ZR7V2GYVQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Neural Machine Translation Training in a Multi-Domain Scenario","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Fahim Dalvi, Hassan Sajjad, Nadir Durrani, Stephan Vogel, Yonatan Belinkov","submitted_at":"2017-08-29T11:56:41Z","abstract_excerpt":"In this paper, we explore alternative ways to train a neural machine translation system in a multi-domain scenario. We investigate data concatenation (with fine tuning), model stacking (multi-level fine tuning), data selection and multi-model ensemble. Our findings show that the best translation quality can be achieved by building an initial system on a concatenation of available out-of-domain data and then fine-tuning it on in-domain data. Model stacking works best when training begins with the furthest out-of-domain data and the model is incrementally fine-tuned with the next furthest domain"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1708.08712","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":"2026-05-18T00:00:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+uZUAdD/Fduaj2SN7DeV1Y/EXF/4TkzcEOv1wdDeI60oElY6gnDtj6DfgVlu5ITyOYQXQfW+UNxANYtf6zfGBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T01:15:57.605346Z"},"content_sha256":"f53b435ce46ebaeb8efc1a5f67a492aa4a265d355a4fd92afcf63cb8d8adb3d6","schema_version":"1.0","event_id":"sha256:f53b435ce46ebaeb8efc1a5f67a492aa4a265d355a4fd92afcf63cb8d8adb3d6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HOCODXIWSARME27A6ZR7V2GYVQ/bundle.json","state_url":"https://pith.science/pith/HOCODXIWSARME27A6ZR7V2GYVQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HOCODXIWSARME27A6ZR7V2GYVQ/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":"2026-05-31T01:15:57Z","links":{"resolver":"https://pith.science/pith/HOCODXIWSARME27A6ZR7V2GYVQ","bundle":"https://pith.science/pith/HOCODXIWSARME27A6ZR7V2GYVQ/bundle.json","state":"https://pith.science/pith/HOCODXIWSARME27A6ZR7V2GYVQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HOCODXIWSARME27A6ZR7V2GYVQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:HOCODXIWSARME27A6ZR7V2GYVQ","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":"49710031f5ca52f9f833448be81f2c52689184b5417b4604ffae33ba0a3751d9","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-08-29T11:56:41Z","title_canon_sha256":"aa14057bfc02595b04df847be5b95475c9ba964127d14b89f274004d896fc85b"},"schema_version":"1.0","source":{"id":"1708.08712","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1708.08712","created_at":"2026-05-18T00:00:22Z"},{"alias_kind":"arxiv_version","alias_value":"1708.08712v3","created_at":"2026-05-18T00:00:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1708.08712","created_at":"2026-05-18T00:00:22Z"},{"alias_kind":"pith_short_12","alias_value":"HOCODXIWSARM","created_at":"2026-05-18T12:31:18Z"},{"alias_kind":"pith_short_16","alias_value":"HOCODXIWSARME27A","created_at":"2026-05-18T12:31:18Z"},{"alias_kind":"pith_short_8","alias_value":"HOCODXIW","created_at":"2026-05-18T12:31:18Z"}],"graph_snapshots":[{"event_id":"sha256:f53b435ce46ebaeb8efc1a5f67a492aa4a265d355a4fd92afcf63cb8d8adb3d6","target":"graph","created_at":"2026-05-18T00:00:22Z","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":"In this paper, we explore alternative ways to train a neural machine translation system in a multi-domain scenario. We investigate data concatenation (with fine tuning), model stacking (multi-level fine tuning), data selection and multi-model ensemble. Our findings show that the best translation quality can be achieved by building an initial system on a concatenation of available out-of-domain data and then fine-tuning it on in-domain data. Model stacking works best when training begins with the furthest out-of-domain data and the model is incrementally fine-tuned with the next furthest domain","authors_text":"Fahim Dalvi, Hassan Sajjad, Nadir Durrani, Stephan Vogel, Yonatan Belinkov","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-08-29T11:56:41Z","title":"Neural Machine Translation Training in a Multi-Domain Scenario"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1708.08712","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:cde2b262164a26a3cdc40c409f5fb486097035a1822c850a98bcf20ecf569695","target":"record","created_at":"2026-05-18T00:00:22Z","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":"49710031f5ca52f9f833448be81f2c52689184b5417b4604ffae33ba0a3751d9","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2017-08-29T11:56:41Z","title_canon_sha256":"aa14057bfc02595b04df847be5b95475c9ba964127d14b89f274004d896fc85b"},"schema_version":"1.0","source":{"id":"1708.08712","kind":"arxiv","version":3}},"canonical_sha256":"3b84e1dd169022c26be0f663fae8d8ac34becdae6493189d566a7be4726977d1","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3b84e1dd169022c26be0f663fae8d8ac34becdae6493189d566a7be4726977d1","first_computed_at":"2026-05-18T00:00:22.489550Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:00:22.489550Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"0bttI2SLy4ort2pzfzYjPN18mS9yCc9G+L1hdFfbIACLIjJrsMc8NGWzxFh+UFk7TbKWqE3vSBqZ1j6sOnVaAw==","signature_status":"signed_v1","signed_at":"2026-05-18T00:00:22.490179Z","signed_message":"canonical_sha256_bytes"},"source_id":"1708.08712","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:cde2b262164a26a3cdc40c409f5fb486097035a1822c850a98bcf20ecf569695","sha256:f53b435ce46ebaeb8efc1a5f67a492aa4a265d355a4fd92afcf63cb8d8adb3d6"],"state_sha256":"7a8eea8df80020e497d0de465c8a6e6cd4eb3269d97ddb46253bf2dc02179a55"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iL0nArYvfBRWppAqrOb1aDvTwcx0Z8dQ33uZ+wHy7UpgX/yUa1r1sbzk9fOE3Ti+un/vookawkJzg9/XBCHIDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T01:15:57.608649Z","bundle_sha256":"221ff013faa20b41a52f4a94ea65a8552dbc6ec1cc85c3b211c9d68dcd0595a9"}}