{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:4QKSDADI7R2MMKJNHRR75ZDYTX","short_pith_number":"pith:4QKSDADI","schema_version":"1.0","canonical_sha256":"e415218068fc74c6292d3c63fee4789dd8ede7493d42e8b828a4715ad5603acf","source":{"kind":"arxiv","id":"2503.01890","version":1},"attestation_state":"computed","paper":{"title":"AutoHete: An Automatic and Efficient Heterogeneous Training System for LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Chubo Liu, Fei Huang, Juan Hu, Kenli Li, Wei Yang Bryan Lim, Xin He, Yong Jiang, Zihao Zeng","submitted_at":"2025-02-27T14:46:22Z","abstract_excerpt":"Transformer-based large language models (LLMs) have demonstrated exceptional capabilities in sequence modeling and text generation, with improvements scaling proportionally with model size. However, the limitations of GPU memory have restricted LLM training accessibility for many researchers. Existing heterogeneous training methods significantly expand the scale of trainable models but introduce substantial communication overheads and CPU workloads. In this work, we propose AutoHete, an automatic and efficient heterogeneous training system compatible with both single-GPU and multi-GPU environm"},"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":"2503.01890","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-27T14:46:22Z","cross_cats_sorted":[],"title_canon_sha256":"979d0fbe68c2f6236ce36f647ee2ff31ddd36b1e5573c9f6b220eb792c0bacbb","abstract_canon_sha256":"e279bc8ba985278e0a2a93ea640d0b9f6215a3c9d15d25f5b43107634eac017d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:23:31.890216Z","signature_b64":"qGun18fSU/doqbhHxNelxYEPxjncrb1AsYBbSYQdrCDH9pjZ6HTmInsgz96V2vkPqBoBurWXBd4vjxJKU544Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e415218068fc74c6292d3c63fee4789dd8ede7493d42e8b828a4715ad5603acf","last_reissued_at":"2026-07-05T10:23:31.889348Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:23:31.889348Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AutoHete: An Automatic and Efficient Heterogeneous Training System for LLMs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Chubo Liu, Fei Huang, Juan Hu, Kenli Li, Wei Yang Bryan Lim, Xin He, Yong Jiang, Zihao Zeng","submitted_at":"2025-02-27T14:46:22Z","abstract_excerpt":"Transformer-based large language models (LLMs) have demonstrated exceptional capabilities in sequence modeling and text generation, with improvements scaling proportionally with model size. However, the limitations of GPU memory have restricted LLM training accessibility for many researchers. Existing heterogeneous training methods significantly expand the scale of trainable models but introduce substantial communication overheads and CPU workloads. In this work, we propose AutoHete, an automatic and efficient heterogeneous training system compatible with both single-GPU and multi-GPU environm"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.01890","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/2503.01890/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":"2503.01890","created_at":"2026-07-05T10:23:31.889459+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.01890v1","created_at":"2026-07-05T10:23:31.889459+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.01890","created_at":"2026-07-05T10:23:31.889459+00:00"},{"alias_kind":"pith_short_12","alias_value":"4QKSDADI7R2M","created_at":"2026-07-05T10:23:31.889459+00:00"},{"alias_kind":"pith_short_16","alias_value":"4QKSDADI7R2MMKJN","created_at":"2026-07-05T10:23:31.889459+00:00"},{"alias_kind":"pith_short_8","alias_value":"4QKSDADI","created_at":"2026-07-05T10:23:31.889459+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/4QKSDADI7R2MMKJNHRR75ZDYTX","json":"https://pith.science/pith/4QKSDADI7R2MMKJNHRR75ZDYTX.json","graph_json":"https://pith.science/api/pith-number/4QKSDADI7R2MMKJNHRR75ZDYTX/graph.json","events_json":"https://pith.science/api/pith-number/4QKSDADI7R2MMKJNHRR75ZDYTX/events.json","paper":"https://pith.science/paper/4QKSDADI"},"agent_actions":{"view_html":"https://pith.science/pith/4QKSDADI7R2MMKJNHRR75ZDYTX","download_json":"https://pith.science/pith/4QKSDADI7R2MMKJNHRR75ZDYTX.json","view_paper":"https://pith.science/paper/4QKSDADI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.01890&json=true","fetch_graph":"https://pith.science/api/pith-number/4QKSDADI7R2MMKJNHRR75ZDYTX/graph.json","fetch_events":"https://pith.science/api/pith-number/4QKSDADI7R2MMKJNHRR75ZDYTX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4QKSDADI7R2MMKJNHRR75ZDYTX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4QKSDADI7R2MMKJNHRR75ZDYTX/action/storage_attestation","attest_author":"https://pith.science/pith/4QKSDADI7R2MMKJNHRR75ZDYTX/action/author_attestation","sign_citation":"https://pith.science/pith/4QKSDADI7R2MMKJNHRR75ZDYTX/action/citation_signature","submit_replication":"https://pith.science/pith/4QKSDADI7R2MMKJNHRR75ZDYTX/action/replication_record"}},"created_at":"2026-07-05T10:23:31.889459+00:00","updated_at":"2026-07-05T10:23:31.889459+00:00"}