{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2020:LWI6LGAU3OINEOWAGHASHWQTAW","short_pith_number":"pith:LWI6LGAU","canonical_record":{"source":{"id":"2010.13887","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MS","submitted_at":"2020-10-23T13:45:26Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7fe3fa4bd3046efaab337625921aa139aac4e5836acae1138f24ebd407c9ddff","abstract_canon_sha256":"f807fdd23be011f66fb4fa083bf69429ee133461bc20c7a06f173db3f89e1a7b"},"schema_version":"1.0"},"canonical_sha256":"5d91e59814db90d23ac031c123da1305b314104b5cf4ed4d9316ae1585e74b65","source":{"kind":"arxiv","id":"2010.13887","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2010.13887","created_at":"2026-07-05T02:34:09Z"},{"alias_kind":"arxiv_version","alias_value":"2010.13887v4","created_at":"2026-07-05T02:34:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.13887","created_at":"2026-07-05T02:34:09Z"},{"alias_kind":"pith_short_12","alias_value":"LWI6LGAU3OIN","created_at":"2026-07-05T02:34:09Z"},{"alias_kind":"pith_short_16","alias_value":"LWI6LGAU3OINEOWA","created_at":"2026-07-05T02:34:09Z"},{"alias_kind":"pith_short_8","alias_value":"LWI6LGAU","created_at":"2026-07-05T02:34:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2020:LWI6LGAU3OINEOWAGHASHWQTAW","target":"record","payload":{"canonical_record":{"source":{"id":"2010.13887","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MS","submitted_at":"2020-10-23T13:45:26Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7fe3fa4bd3046efaab337625921aa139aac4e5836acae1138f24ebd407c9ddff","abstract_canon_sha256":"f807fdd23be011f66fb4fa083bf69429ee133461bc20c7a06f173db3f89e1a7b"},"schema_version":"1.0"},"canonical_sha256":"5d91e59814db90d23ac031c123da1305b314104b5cf4ed4d9316ae1585e74b65","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:34:09.967155Z","signature_b64":"hZNsHSLD09CUFPyBEXRJILX64cvkssv/2/w2dlsGofH5UM6lugQNS1l/Mvd2fIF/CFJgk/b8PIJs1OThWG8bDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5d91e59814db90d23ac031c123da1305b314104b5cf4ed4d9316ae1585e74b65","last_reissued_at":"2026-07-05T02:34:09.966655Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:34:09.966655Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2010.13887","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T02:34:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9CQUzf51WBRY5pLFv6EtOcmt6KWL9yWNDpFccpo8/QlHOyaAV1SF0Z88rKhhrK10N4LjmGuhEc4NOFLlacWsCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T20:52:47.676611Z"},"content_sha256":"1e29389b65819e09404fb836e11d31aad27f85f169277cdb6affd6553b5664e7","schema_version":"1.0","event_id":"sha256:1e29389b65819e09404fb836e11d31aad27f85f169277cdb6affd6553b5664e7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2020:LWI6LGAU3OINEOWAGHASHWQTAW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LightSeq: A High Performance Inference Library for Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.MS","authors_text":"Lei Li, Mingxuan Wang, Xiaohui Wang, Yang Wei, Ying Xiong","submitted_at":"2020-10-23T13:45:26Z","abstract_excerpt":"Transformer, BERT and their variants have achieved great success in natural language processing. Since Transformer models are huge in size, serving these models is a challenge for real industrial applications. In this paper, we propose LightSeq, a highly efficient inference library for models in the Transformer family. LightSeq includes a series of GPU optimization techniques to to streamline the computation of neural layers and to reduce memory footprint. LightSeq can easily import models trained using PyTorch and Tensorflow. Experimental results on machine translation benchmarks show that Li"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.13887","kind":"arxiv","version":4},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2010.13887/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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":"2026-07-05T02:34:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AgQo6X6xndGdVYi7Mz5CmFbZkMY1v3NnLbWPzoZECQNZCiATQBB7pO2MSUAc+8g/pTSHgqbhb+quJJ0CfvgXBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T20:52:47.677873Z"},"content_sha256":"f1cbd14f0bb2046a74f289d605ab1285799a10dbb8c4c643a2a63479f22ce217","schema_version":"1.0","event_id":"sha256:f1cbd14f0bb2046a74f289d605ab1285799a10dbb8c4c643a2a63479f22ce217"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LWI6LGAU3OINEOWAGHASHWQTAW/bundle.json","state_url":"https://pith.science/pith/LWI6LGAU3OINEOWAGHASHWQTAW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LWI6LGAU3OINEOWAGHASHWQTAW/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":"2026-08-05T20:52:47Z","links":{"resolver":"https://pith.science/pith/LWI6LGAU3OINEOWAGHASHWQTAW","bundle":"https://pith.science/pith/LWI6LGAU3OINEOWAGHASHWQTAW/bundle.json","state":"https://pith.science/pith/LWI6LGAU3OINEOWAGHASHWQTAW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LWI6LGAU3OINEOWAGHASHWQTAW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2020:LWI6LGAU3OINEOWAGHASHWQTAW","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":"f807fdd23be011f66fb4fa083bf69429ee133461bc20c7a06f173db3f89e1a7b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MS","submitted_at":"2020-10-23T13:45:26Z","title_canon_sha256":"7fe3fa4bd3046efaab337625921aa139aac4e5836acae1138f24ebd407c9ddff"},"schema_version":"1.0","source":{"id":"2010.13887","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2010.13887","created_at":"2026-07-05T02:34:09Z"},{"alias_kind":"arxiv_version","alias_value":"2010.13887v4","created_at":"2026-07-05T02:34:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2010.13887","created_at":"2026-07-05T02:34:09Z"},{"alias_kind":"pith_short_12","alias_value":"LWI6LGAU3OIN","created_at":"2026-07-05T02:34:09Z"},{"alias_kind":"pith_short_16","alias_value":"LWI6LGAU3OINEOWA","created_at":"2026-07-05T02:34:09Z"},{"alias_kind":"pith_short_8","alias_value":"LWI6LGAU","created_at":"2026-07-05T02:34:09Z"}],"graph_snapshots":[{"event_id":"sha256:f1cbd14f0bb2046a74f289d605ab1285799a10dbb8c4c643a2a63479f22ce217","target":"graph","created_at":"2026-07-05T02:34:09Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2010.13887/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Transformer, BERT and their variants have achieved great success in natural language processing. Since Transformer models are huge in size, serving these models is a challenge for real industrial applications. In this paper, we propose LightSeq, a highly efficient inference library for models in the Transformer family. LightSeq includes a series of GPU optimization techniques to to streamline the computation of neural layers and to reduce memory footprint. LightSeq can easily import models trained using PyTorch and Tensorflow. Experimental results on machine translation benchmarks show that Li","authors_text":"Lei Li, Mingxuan Wang, Xiaohui Wang, Yang Wei, Ying Xiong","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MS","submitted_at":"2020-10-23T13:45:26Z","title":"LightSeq: A High Performance Inference Library for Transformers"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2010.13887","kind":"arxiv","version":4},"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":"sha256:1e29389b65819e09404fb836e11d31aad27f85f169277cdb6affd6553b5664e7","target":"record","created_at":"2026-07-05T02:34:09Z","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":"f807fdd23be011f66fb4fa083bf69429ee133461bc20c7a06f173db3f89e1a7b","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.MS","submitted_at":"2020-10-23T13:45:26Z","title_canon_sha256":"7fe3fa4bd3046efaab337625921aa139aac4e5836acae1138f24ebd407c9ddff"},"schema_version":"1.0","source":{"id":"2010.13887","kind":"arxiv","version":4}},"canonical_sha256":"5d91e59814db90d23ac031c123da1305b314104b5cf4ed4d9316ae1585e74b65","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5d91e59814db90d23ac031c123da1305b314104b5cf4ed4d9316ae1585e74b65","first_computed_at":"2026-07-05T02:34:09.966655Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:34:09.966655Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"hZNsHSLD09CUFPyBEXRJILX64cvkssv/2/w2dlsGofH5UM6lugQNS1l/Mvd2fIF/CFJgk/b8PIJs1OThWG8bDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T02:34:09.967155Z","signed_message":"canonical_sha256_bytes"},"source_id":"2010.13887","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1e29389b65819e09404fb836e11d31aad27f85f169277cdb6affd6553b5664e7","sha256:f1cbd14f0bb2046a74f289d605ab1285799a10dbb8c4c643a2a63479f22ce217"],"state_sha256":"a21e687a3c5740dff7f9179c421b27030d07a7628a031d1b5c1cd0de1d8f3d65"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"n8eL3OSuk64sxGbkAWmTi70NEgC3i5FFYSqJnHSuE1W1ZTzljbifbXZb2X/GbYjR2UxWyUYAe2UCJULDH9GYAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T20:52:47.684460Z","bundle_sha256":"26b5cb965e078b1724fa8c1a20b6fde0ebdfd28314618ed1d61295778bf031c3"}}