{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:WBAGA3UJKA5INZPGYX7MRYYYMA","short_pith_number":"pith:WBAGA3UJ","schema_version":"1.0","canonical_sha256":"b040606e89503a86e5e6c5fec8e3186020bfc6a76bd7fe2e69960b916f103417","source":{"kind":"arxiv","id":"2301.02654","version":1},"attestation_state":"computed","paper":{"title":"Does compressing activations help model parallel training?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dacheng Li, Eric P. Xing, Hongyi Wang, Shivaram Venkataraman, Song Bian","submitted_at":"2023-01-06T18:58:09Z","abstract_excerpt":"Large-scale Transformer models are known for their exceptional performance in a range of tasks, but training them can be difficult due to the requirement for communication-intensive model parallelism. One way to improve training speed is to compress the message size in communication. Previous approaches have primarily focused on compressing gradients in a data parallelism setting, but compression in a model-parallel setting is an understudied area. We have discovered that model parallelism has fundamentally different characteristics than data parallelism. In this work, we present the first emp"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2301.02654","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-01-06T18:58:09Z","cross_cats_sorted":[],"title_canon_sha256":"64d77cf43963b958c7d42496c9b7b5646f46e79dbf69f9bb14ed952cc7179b85","abstract_canon_sha256":"140dea10ea468d5f767b21f796d91a8895ce0ce92f233c8b2d14b94c854f35b9"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:31:08.995290Z","signature_b64":"AeS6y8CM4GvYcUN/xK91hjws1F+SrWtoy8alkPxdfOgZ973+sX2nSy5Xbsh4x7Xx+6kauht/9jcJOeHeu1gZAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b040606e89503a86e5e6c5fec8e3186020bfc6a76bd7fe2e69960b916f103417","last_reissued_at":"2026-07-05T05:31:08.994841Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:31:08.994841Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Does compressing activations help model parallel training?","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dacheng Li, Eric P. Xing, Hongyi Wang, Shivaram Venkataraman, Song Bian","submitted_at":"2023-01-06T18:58:09Z","abstract_excerpt":"Large-scale Transformer models are known for their exceptional performance in a range of tasks, but training them can be difficult due to the requirement for communication-intensive model parallelism. One way to improve training speed is to compress the message size in communication. Previous approaches have primarily focused on compressing gradients in a data parallelism setting, but compression in a model-parallel setting is an understudied area. We have discovered that model parallelism has fundamentally different characteristics than data parallelism. In this work, we present the first emp"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.02654","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/2301.02654/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2301.02654","created_at":"2026-07-05T05:31:08.994903+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.02654v1","created_at":"2026-07-05T05:31:08.994903+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.02654","created_at":"2026-07-05T05:31:08.994903+00:00"},{"alias_kind":"pith_short_12","alias_value":"WBAGA3UJKA5I","created_at":"2026-07-05T05:31:08.994903+00:00"},{"alias_kind":"pith_short_16","alias_value":"WBAGA3UJKA5INZPG","created_at":"2026-07-05T05:31:08.994903+00:00"},{"alias_kind":"pith_short_8","alias_value":"WBAGA3UJ","created_at":"2026-07-05T05:31:08.994903+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WBAGA3UJKA5INZPGYX7MRYYYMA","json":"https://pith.science/pith/WBAGA3UJKA5INZPGYX7MRYYYMA.json","graph_json":"https://pith.science/api/pith-number/WBAGA3UJKA5INZPGYX7MRYYYMA/graph.json","events_json":"https://pith.science/api/pith-number/WBAGA3UJKA5INZPGYX7MRYYYMA/events.json","paper":"https://pith.science/paper/WBAGA3UJ"},"agent_actions":{"view_html":"https://pith.science/pith/WBAGA3UJKA5INZPGYX7MRYYYMA","download_json":"https://pith.science/pith/WBAGA3UJKA5INZPGYX7MRYYYMA.json","view_paper":"https://pith.science/paper/WBAGA3UJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.02654&json=true","fetch_graph":"https://pith.science/api/pith-number/WBAGA3UJKA5INZPGYX7MRYYYMA/graph.json","fetch_events":"https://pith.science/api/pith-number/WBAGA3UJKA5INZPGYX7MRYYYMA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WBAGA3UJKA5INZPGYX7MRYYYMA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WBAGA3UJKA5INZPGYX7MRYYYMA/action/storage_attestation","attest_author":"https://pith.science/pith/WBAGA3UJKA5INZPGYX7MRYYYMA/action/author_attestation","sign_citation":"https://pith.science/pith/WBAGA3UJKA5INZPGYX7MRYYYMA/action/citation_signature","submit_replication":"https://pith.science/pith/WBAGA3UJKA5INZPGYX7MRYYYMA/action/replication_record"}},"created_at":"2026-07-05T05:31:08.994903+00:00","updated_at":"2026-07-05T05:31:08.994903+00:00"}