{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:4PVEPCN5QIS3TVBXZWVZLEAFNV","short_pith_number":"pith:4PVEPCN5","canonical_record":{"source":{"id":"2109.08668","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-09-17T17:50:39Z","cross_cats_sorted":["cs.AI","cs.CL","cs.NE"],"title_canon_sha256":"adf5d6af2c24c4d193cc82391b0fa21bd1ed423bc104ec2b61f7f3194b057e85","abstract_canon_sha256":"bde3083a991521284c24f46648bdaed516a5065792d036691ffd78af20ae5773"},"schema_version":"1.0"},"canonical_sha256":"e3ea4789bd8225b9d437cdab9590056d6ac937fe94fd42859fe1c0f9dbf153c6","source":{"kind":"arxiv","id":"2109.08668","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2109.08668","created_at":"2026-07-05T03:51:09Z"},{"alias_kind":"arxiv_version","alias_value":"2109.08668v2","created_at":"2026-07-05T03:51:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.08668","created_at":"2026-07-05T03:51:09Z"},{"alias_kind":"pith_short_12","alias_value":"4PVEPCN5QIS3","created_at":"2026-07-05T03:51:09Z"},{"alias_kind":"pith_short_16","alias_value":"4PVEPCN5QIS3TVBX","created_at":"2026-07-05T03:51:09Z"},{"alias_kind":"pith_short_8","alias_value":"4PVEPCN5","created_at":"2026-07-05T03:51:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:4PVEPCN5QIS3TVBXZWVZLEAFNV","target":"record","payload":{"canonical_record":{"source":{"id":"2109.08668","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-09-17T17:50:39Z","cross_cats_sorted":["cs.AI","cs.CL","cs.NE"],"title_canon_sha256":"adf5d6af2c24c4d193cc82391b0fa21bd1ed423bc104ec2b61f7f3194b057e85","abstract_canon_sha256":"bde3083a991521284c24f46648bdaed516a5065792d036691ffd78af20ae5773"},"schema_version":"1.0"},"canonical_sha256":"e3ea4789bd8225b9d437cdab9590056d6ac937fe94fd42859fe1c0f9dbf153c6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:51:09.771192Z","signature_b64":"98Mk5rPyhynUsok5Z1x3UYWEUZw2LQhsquVxtZ5snLj2f05wDy5/HZo6RbcYzqZnaBhS9eN79Pn78WTkTXU2Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e3ea4789bd8225b9d437cdab9590056d6ac937fe94fd42859fe1c0f9dbf153c6","last_reissued_at":"2026-07-05T03:51:09.770729Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:51:09.770729Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2109.08668","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T03:51:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"M+J7r1IsA2XfMyUT2IvyjeB2uqPZrt8oOKOxHrD+NZbDoYFB0LpBf21cM01BrfTgilepj5qDsN72DXlU8Nx0Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T16:51:33.300563Z"},"content_sha256":"1ce4be92991d22dda539ef21f33e63cf2f36c1af0c6ea086906ec654cb37d0f4","schema_version":"1.0","event_id":"sha256:1ce4be92991d22dda539ef21f33e63cf2f36c1af0c6ea086906ec654cb37d0f4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:4PVEPCN5QIS3TVBXZWVZLEAFNV","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Primer: Searching for Efficient Transformers for Language Modeling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.NE"],"primary_cat":"cs.LG","authors_text":"David R. So, Hanxiao Liu, Noam Shazeer, Quoc V. Le, Wojciech Ma\\'nke, Zihang Dai","submitted_at":"2021-09-17T17:50:39Z","abstract_excerpt":"Large Transformer models have been central to recent advances in natural language processing. The training and inference costs of these models, however, have grown rapidly and become prohibitively expensive. Here we aim to reduce the costs of Transformers by searching for a more efficient variant. Compared to previous approaches, our search is performed at a lower level, over the primitives that define a Transformer TensorFlow program. We identify an architecture, named Primer, that has a smaller training cost than the original Transformer and other variants for auto-regressive language modeli"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.08668","kind":"arxiv","version":2},"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/2109.08668/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":"2026-07-05T03:51:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LgpcuOz8TAlqHWcrBDCsCf0nulx8N2o1ISak0cB7OaOHyzEqlF7VeqGaexkOdFm2bZ205ZCLp7S+5O9A8ePeDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T16:51:33.301119Z"},"content_sha256":"5eca2158629eed9c0f0089a97fdb494d0252035d708e0835d447cc3db1f4578b","schema_version":"1.0","event_id":"sha256:5eca2158629eed9c0f0089a97fdb494d0252035d708e0835d447cc3db1f4578b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4PVEPCN5QIS3TVBXZWVZLEAFNV/bundle.json","state_url":"https://pith.science/pith/4PVEPCN5QIS3TVBXZWVZLEAFNV/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4PVEPCN5QIS3TVBXZWVZLEAFNV/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":"2026-08-04T16:51:33Z","links":{"resolver":"https://pith.science/pith/4PVEPCN5QIS3TVBXZWVZLEAFNV","bundle":"https://pith.science/pith/4PVEPCN5QIS3TVBXZWVZLEAFNV/bundle.json","state":"https://pith.science/pith/4PVEPCN5QIS3TVBXZWVZLEAFNV/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4PVEPCN5QIS3TVBXZWVZLEAFNV/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:4PVEPCN5QIS3TVBXZWVZLEAFNV","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":"bde3083a991521284c24f46648bdaed516a5065792d036691ffd78af20ae5773","cross_cats_sorted":["cs.AI","cs.CL","cs.NE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-09-17T17:50:39Z","title_canon_sha256":"adf5d6af2c24c4d193cc82391b0fa21bd1ed423bc104ec2b61f7f3194b057e85"},"schema_version":"1.0","source":{"id":"2109.08668","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2109.08668","created_at":"2026-07-05T03:51:09Z"},{"alias_kind":"arxiv_version","alias_value":"2109.08668v2","created_at":"2026-07-05T03:51:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.08668","created_at":"2026-07-05T03:51:09Z"},{"alias_kind":"pith_short_12","alias_value":"4PVEPCN5QIS3","created_at":"2026-07-05T03:51:09Z"},{"alias_kind":"pith_short_16","alias_value":"4PVEPCN5QIS3TVBX","created_at":"2026-07-05T03:51:09Z"},{"alias_kind":"pith_short_8","alias_value":"4PVEPCN5","created_at":"2026-07-05T03:51:09Z"}],"graph_snapshots":[{"event_id":"sha256:5eca2158629eed9c0f0089a97fdb494d0252035d708e0835d447cc3db1f4578b","target":"graph","created_at":"2026-07-05T03:51:09Z","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/2109.08668/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large Transformer models have been central to recent advances in natural language processing. The training and inference costs of these models, however, have grown rapidly and become prohibitively expensive. Here we aim to reduce the costs of Transformers by searching for a more efficient variant. Compared to previous approaches, our search is performed at a lower level, over the primitives that define a Transformer TensorFlow program. We identify an architecture, named Primer, that has a smaller training cost than the original Transformer and other variants for auto-regressive language modeli","authors_text":"David R. So, Hanxiao Liu, Noam Shazeer, Quoc V. Le, Wojciech Ma\\'nke, Zihang Dai","cross_cats":["cs.AI","cs.CL","cs.NE"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-09-17T17:50:39Z","title":"Primer: Searching for Efficient Transformers for Language Modeling"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.08668","kind":"arxiv","version":2},"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":"sha256:1ce4be92991d22dda539ef21f33e63cf2f36c1af0c6ea086906ec654cb37d0f4","target":"record","created_at":"2026-07-05T03:51:09Z","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":"bde3083a991521284c24f46648bdaed516a5065792d036691ffd78af20ae5773","cross_cats_sorted":["cs.AI","cs.CL","cs.NE"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-09-17T17:50:39Z","title_canon_sha256":"adf5d6af2c24c4d193cc82391b0fa21bd1ed423bc104ec2b61f7f3194b057e85"},"schema_version":"1.0","source":{"id":"2109.08668","kind":"arxiv","version":2}},"canonical_sha256":"e3ea4789bd8225b9d437cdab9590056d6ac937fe94fd42859fe1c0f9dbf153c6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e3ea4789bd8225b9d437cdab9590056d6ac937fe94fd42859fe1c0f9dbf153c6","first_computed_at":"2026-07-05T03:51:09.770729Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:51:09.770729Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"98Mk5rPyhynUsok5Z1x3UYWEUZw2LQhsquVxtZ5snLj2f05wDy5/HZo6RbcYzqZnaBhS9eN79Pn78WTkTXU2Cw==","signature_status":"signed_v1","signed_at":"2026-07-05T03:51:09.771192Z","signed_message":"canonical_sha256_bytes"},"source_id":"2109.08668","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1ce4be92991d22dda539ef21f33e63cf2f36c1af0c6ea086906ec654cb37d0f4","sha256:5eca2158629eed9c0f0089a97fdb494d0252035d708e0835d447cc3db1f4578b"],"state_sha256":"46db977e47148eadf617b5f56592d362eff2f7c87a3d95a7edf18d1adc998e0d"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kPP+ywc5GF4eYy781ZVaf7H8hvdy71JD0/Ub9bYHRECi9/38ods9GgW6QdlzD2MXJFR2qHhJm81wYEJG7qqUDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T16:51:33.304813Z","bundle_sha256":"4a6116e39f36372ef3b175d56ec7957985396e56f831360b12e866f9f7fe877b"}}