{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:JB2RKQ6ZL7ACFOIVDX2TWDPMMZ","short_pith_number":"pith:JB2RKQ6Z","canonical_record":{"source":{"id":"2205.05198","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-05-10T22:40:17Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"ca9e4db5fbb76544c2b8651a3dc7fdda471bf9f3338b303163efd65d58348b3d","abstract_canon_sha256":"e4ace58f724b13ae2426e6e7b63b73fef3ce98753d51a34ead639bed3dd4d40c"},"schema_version":"1.0"},"canonical_sha256":"48751543d95fc022b9151df53b0dec66710b2ba070cae4de6ee01d55adc38298","source":{"kind":"arxiv","id":"2205.05198","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2205.05198","created_at":"2026-07-05T04:22:23Z"},{"alias_kind":"arxiv_version","alias_value":"2205.05198v1","created_at":"2026-07-05T04:22:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.05198","created_at":"2026-07-05T04:22:23Z"},{"alias_kind":"pith_short_12","alias_value":"JB2RKQ6ZL7AC","created_at":"2026-07-05T04:22:23Z"},{"alias_kind":"pith_short_16","alias_value":"JB2RKQ6ZL7ACFOIV","created_at":"2026-07-05T04:22:23Z"},{"alias_kind":"pith_short_8","alias_value":"JB2RKQ6Z","created_at":"2026-07-05T04:22:23Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:JB2RKQ6ZL7ACFOIVDX2TWDPMMZ","target":"record","payload":{"canonical_record":{"source":{"id":"2205.05198","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-05-10T22:40:17Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"ca9e4db5fbb76544c2b8651a3dc7fdda471bf9f3338b303163efd65d58348b3d","abstract_canon_sha256":"e4ace58f724b13ae2426e6e7b63b73fef3ce98753d51a34ead639bed3dd4d40c"},"schema_version":"1.0"},"canonical_sha256":"48751543d95fc022b9151df53b0dec66710b2ba070cae4de6ee01d55adc38298","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:22:23.156461Z","signature_b64":"q1bCacqNNGJlk8X4XihNOF+Dpq4Z+bJyCJvMushOJaRaR2EhdPsVz3S+L6DbGCYx0guj77PmP7Tykk2sQatIAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"48751543d95fc022b9151df53b0dec66710b2ba070cae4de6ee01d55adc38298","last_reissued_at":"2026-07-05T04:22:23.156018Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:22:23.156018Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2205.05198","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":"2026-07-05T04:22:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2QY8fCd9pvUEFmJ7SUbuxkQYcNYm9aWb3dn1H/2qyMj4pRW/XGH261UJVLqrMnHoeejnBfql8UBxaiQcSjbfCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T06:02:55.227885Z"},"content_sha256":"2175985c0a9c2c656c8c9f9ba10988df93f61b45f586b85d4c613781511c7083","schema_version":"1.0","event_id":"sha256:2175985c0a9c2c656c8c9f9ba10988df93f61b45f586b85d4c613781511c7083"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:JB2RKQ6ZL7ACFOIVDX2TWDPMMZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Reducing Activation Recomputation in Large Transformer Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Bryan Catanzaro, Jared Casper, Lawrence McAfee, Michael Andersch, Mohammad Shoeybi, Sangkug Lym, Vijay Korthikanti","submitted_at":"2022-05-10T22:40:17Z","abstract_excerpt":"Training large transformer models is one of the most important computational challenges of modern AI. In this paper, we show how to significantly accelerate training of large transformer models by reducing activation recomputation. Activation recomputation is commonly used to work around memory capacity constraints. Rather than storing activations for backpropagation, they are traditionally recomputed, which saves memory but adds redundant compute. In this work, we show most of this redundant compute is unnecessary because we can reduce memory consumption sufficiently without it. We present tw"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.05198","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2205.05198/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":"2026-07-05T04:22:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ak8oy/dNyWdlWWs3bl5n6uh5kWOFzX4TA5yZI1LBBsf0xXhUI5xiCkQiXzlykEMxGXPJoDaZ5EkAxYMPrYDmAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T06:02:55.228423Z"},"content_sha256":"361c64996b0b145b1d956cf33efeb17b0ee3fd8b36255be5183c993350d3effd","schema_version":"1.0","event_id":"sha256:361c64996b0b145b1d956cf33efeb17b0ee3fd8b36255be5183c993350d3effd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JB2RKQ6ZL7ACFOIVDX2TWDPMMZ/bundle.json","state_url":"https://pith.science/pith/JB2RKQ6ZL7ACFOIVDX2TWDPMMZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JB2RKQ6ZL7ACFOIVDX2TWDPMMZ/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":"2026-08-07T06:02:55Z","links":{"resolver":"https://pith.science/pith/JB2RKQ6ZL7ACFOIVDX2TWDPMMZ","bundle":"https://pith.science/pith/JB2RKQ6ZL7ACFOIVDX2TWDPMMZ/bundle.json","state":"https://pith.science/pith/JB2RKQ6ZL7ACFOIVDX2TWDPMMZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JB2RKQ6ZL7ACFOIVDX2TWDPMMZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:JB2RKQ6ZL7ACFOIVDX2TWDPMMZ","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":"e4ace58f724b13ae2426e6e7b63b73fef3ce98753d51a34ead639bed3dd4d40c","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-05-10T22:40:17Z","title_canon_sha256":"ca9e4db5fbb76544c2b8651a3dc7fdda471bf9f3338b303163efd65d58348b3d"},"schema_version":"1.0","source":{"id":"2205.05198","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2205.05198","created_at":"2026-07-05T04:22:23Z"},{"alias_kind":"arxiv_version","alias_value":"2205.05198v1","created_at":"2026-07-05T04:22:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2205.05198","created_at":"2026-07-05T04:22:23Z"},{"alias_kind":"pith_short_12","alias_value":"JB2RKQ6ZL7AC","created_at":"2026-07-05T04:22:23Z"},{"alias_kind":"pith_short_16","alias_value":"JB2RKQ6ZL7ACFOIV","created_at":"2026-07-05T04:22:23Z"},{"alias_kind":"pith_short_8","alias_value":"JB2RKQ6Z","created_at":"2026-07-05T04:22:23Z"}],"graph_snapshots":[{"event_id":"sha256:361c64996b0b145b1d956cf33efeb17b0ee3fd8b36255be5183c993350d3effd","target":"graph","created_at":"2026-07-05T04:22:23Z","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/2205.05198/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Training large transformer models is one of the most important computational challenges of modern AI. In this paper, we show how to significantly accelerate training of large transformer models by reducing activation recomputation. Activation recomputation is commonly used to work around memory capacity constraints. Rather than storing activations for backpropagation, they are traditionally recomputed, which saves memory but adds redundant compute. In this work, we show most of this redundant compute is unnecessary because we can reduce memory consumption sufficiently without it. We present tw","authors_text":"Bryan Catanzaro, Jared Casper, Lawrence McAfee, Michael Andersch, Mohammad Shoeybi, Sangkug Lym, Vijay Korthikanti","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-05-10T22:40:17Z","title":"Reducing Activation Recomputation in Large Transformer Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2205.05198","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":"sha256:2175985c0a9c2c656c8c9f9ba10988df93f61b45f586b85d4c613781511c7083","target":"record","created_at":"2026-07-05T04:22:23Z","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":"e4ace58f724b13ae2426e6e7b63b73fef3ce98753d51a34ead639bed3dd4d40c","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-05-10T22:40:17Z","title_canon_sha256":"ca9e4db5fbb76544c2b8651a3dc7fdda471bf9f3338b303163efd65d58348b3d"},"schema_version":"1.0","source":{"id":"2205.05198","kind":"arxiv","version":1}},"canonical_sha256":"48751543d95fc022b9151df53b0dec66710b2ba070cae4de6ee01d55adc38298","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"48751543d95fc022b9151df53b0dec66710b2ba070cae4de6ee01d55adc38298","first_computed_at":"2026-07-05T04:22:23.156018Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:22:23.156018Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"q1bCacqNNGJlk8X4XihNOF+Dpq4Z+bJyCJvMushOJaRaR2EhdPsVz3S+L6DbGCYx0guj77PmP7Tykk2sQatIAw==","signature_status":"signed_v1","signed_at":"2026-07-05T04:22:23.156461Z","signed_message":"canonical_sha256_bytes"},"source_id":"2205.05198","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2175985c0a9c2c656c8c9f9ba10988df93f61b45f586b85d4c613781511c7083","sha256:361c64996b0b145b1d956cf33efeb17b0ee3fd8b36255be5183c993350d3effd"],"state_sha256":"b02eaf2b49c8c39a527790e6a42924903d9d5e5af5531c06f95960933f79f73e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MQ9omKKV3v1J7xb9RrJu/WpJSJDMm60poRRgdyn5WlcCRN8ztI6CjoCrFklX1bqDeEvs/vO//gI+Llng9Tv7Bg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T06:02:55.233323Z","bundle_sha256":"238a947019863a839c90c40455d7e2cd98934e5daa569cd36fa709ac6730445d"}}