{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:GYSL5Z5UMKBYUJNCJPSNFNWOEL","short_pith_number":"pith:GYSL5Z5U","canonical_record":{"source":{"id":"2406.18060","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-26T04:33:13Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"3b4dbe7b316c9f411d4b472fe7c1c422498433127d8cbdeb0222a72438002e9f","abstract_canon_sha256":"8e32d2a34885541ca61c62d0fae888136843357750ae4af451bd9f68b24963a2"},"schema_version":"1.0"},"canonical_sha256":"3624bee7b462838a25a24be4d2b6ce22d485952609b1f8b392e4f64df5708d79","source":{"kind":"arxiv","id":"2406.18060","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.18060","created_at":"2026-07-05T09:43:30Z"},{"alias_kind":"arxiv_version","alias_value":"2406.18060v3","created_at":"2026-07-05T09:43:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.18060","created_at":"2026-07-05T09:43:30Z"},{"alias_kind":"pith_short_12","alias_value":"GYSL5Z5UMKBY","created_at":"2026-07-05T09:43:30Z"},{"alias_kind":"pith_short_16","alias_value":"GYSL5Z5UMKBYUJNC","created_at":"2026-07-05T09:43:30Z"},{"alias_kind":"pith_short_8","alias_value":"GYSL5Z5U","created_at":"2026-07-05T09:43:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:GYSL5Z5UMKBYUJNCJPSNFNWOEL","target":"record","payload":{"canonical_record":{"source":{"id":"2406.18060","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-26T04:33:13Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"3b4dbe7b316c9f411d4b472fe7c1c422498433127d8cbdeb0222a72438002e9f","abstract_canon_sha256":"8e32d2a34885541ca61c62d0fae888136843357750ae4af451bd9f68b24963a2"},"schema_version":"1.0"},"canonical_sha256":"3624bee7b462838a25a24be4d2b6ce22d485952609b1f8b392e4f64df5708d79","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:43:30.700439Z","signature_b64":"kQsDh35cVy6QqTLglSGLY2OdMUybQZn5oD2J57r6YYMNhCX6OCpsWBJbZF49pCH66U+PvxAFBVTNlf6u4r73Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3624bee7b462838a25a24be4d2b6ce22d485952609b1f8b392e4f64df5708d79","last_reissued_at":"2026-07-05T09:43:30.699878Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:43:30.699878Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.18060","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:43:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IhztjK/XigFKazBJgRxkjvHnmFziLen/1SGP8v0Y/5xNX7BoNQzRhAmoV1uU/GrekIxhDlWNFdfylBjR1nw6DA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T06:01:33.391298Z"},"content_sha256":"dbd403a68a9ca230a629a405e8862a14e86377896806ce4a08f55e974d4e9730","schema_version":"1.0","event_id":"sha256:dbd403a68a9ca230a629a405e8862a14e86377896806ce4a08f55e974d4e9730"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:GYSL5Z5UMKBYUJNCJPSNFNWOEL","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"AdaZeta: Adaptive Zeroth-Order Tensor-Train Adaption for Memory-Efficient Large Language Models Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Athanasios Mouchtaris, Ershad Banijamal, Kai Zhen, Yifan Yang, Zheng Zhang","submitted_at":"2024-06-26T04:33:13Z","abstract_excerpt":"Fine-tuning large language models (LLMs) has achieved remarkable performance across various natural language processing tasks, yet it demands more and more memory as model sizes keep growing. To address this issue, the recently proposed Memory-efficient Zeroth-order (MeZO) methods attempt to fine-tune LLMs using only forward passes, thereby avoiding the need for a backpropagation graph. However, significant performance drops and a high risk of divergence have limited their widespread adoption. In this paper, we propose the Adaptive Zeroth-order Tensor-Train Adaption (AdaZeta) framework, specif"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.18060","kind":"arxiv","version":3},"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/2406.18060/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":"2026-07-05T09:43:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Z8lzTEnxyz06GgehWr1C9BDmq8kx+J5cc2/R0fGBDMoinxDcL8i3kxPKylmLA60cldqEWGZQRRFK63xXcNBNAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T06:01:33.391684Z"},"content_sha256":"da455647035629ec6006435c125885e8aca2e1184f67260df330c7eec25ff27a","schema_version":"1.0","event_id":"sha256:da455647035629ec6006435c125885e8aca2e1184f67260df330c7eec25ff27a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GYSL5Z5UMKBYUJNCJPSNFNWOEL/bundle.json","state_url":"https://pith.science/pith/GYSL5Z5UMKBYUJNCJPSNFNWOEL/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GYSL5Z5UMKBYUJNCJPSNFNWOEL/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":"2026-07-07T06:01:33Z","links":{"resolver":"https://pith.science/pith/GYSL5Z5UMKBYUJNCJPSNFNWOEL","bundle":"https://pith.science/pith/GYSL5Z5UMKBYUJNCJPSNFNWOEL/bundle.json","state":"https://pith.science/pith/GYSL5Z5UMKBYUJNCJPSNFNWOEL/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GYSL5Z5UMKBYUJNCJPSNFNWOEL/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GYSL5Z5UMKBYUJNCJPSNFNWOEL","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":"8e32d2a34885541ca61c62d0fae888136843357750ae4af451bd9f68b24963a2","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-26T04:33:13Z","title_canon_sha256":"3b4dbe7b316c9f411d4b472fe7c1c422498433127d8cbdeb0222a72438002e9f"},"schema_version":"1.0","source":{"id":"2406.18060","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.18060","created_at":"2026-07-05T09:43:30Z"},{"alias_kind":"arxiv_version","alias_value":"2406.18060v3","created_at":"2026-07-05T09:43:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.18060","created_at":"2026-07-05T09:43:30Z"},{"alias_kind":"pith_short_12","alias_value":"GYSL5Z5UMKBY","created_at":"2026-07-05T09:43:30Z"},{"alias_kind":"pith_short_16","alias_value":"GYSL5Z5UMKBYUJNC","created_at":"2026-07-05T09:43:30Z"},{"alias_kind":"pith_short_8","alias_value":"GYSL5Z5U","created_at":"2026-07-05T09:43:30Z"}],"graph_snapshots":[{"event_id":"sha256:da455647035629ec6006435c125885e8aca2e1184f67260df330c7eec25ff27a","target":"graph","created_at":"2026-07-05T09:43:30Z","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/2406.18060/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Fine-tuning large language models (LLMs) has achieved remarkable performance across various natural language processing tasks, yet it demands more and more memory as model sizes keep growing. To address this issue, the recently proposed Memory-efficient Zeroth-order (MeZO) methods attempt to fine-tune LLMs using only forward passes, thereby avoiding the need for a backpropagation graph. However, significant performance drops and a high risk of divergence have limited their widespread adoption. In this paper, we propose the Adaptive Zeroth-order Tensor-Train Adaption (AdaZeta) framework, specif","authors_text":"Athanasios Mouchtaris, Ershad Banijamal, Kai Zhen, Yifan Yang, Zheng Zhang","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-26T04:33:13Z","title":"AdaZeta: Adaptive Zeroth-Order Tensor-Train Adaption for Memory-Efficient Large Language Models Fine-Tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.18060","kind":"arxiv","version":3},"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:dbd403a68a9ca230a629a405e8862a14e86377896806ce4a08f55e974d4e9730","target":"record","created_at":"2026-07-05T09:43:30Z","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":"8e32d2a34885541ca61c62d0fae888136843357750ae4af451bd9f68b24963a2","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-26T04:33:13Z","title_canon_sha256":"3b4dbe7b316c9f411d4b472fe7c1c422498433127d8cbdeb0222a72438002e9f"},"schema_version":"1.0","source":{"id":"2406.18060","kind":"arxiv","version":3}},"canonical_sha256":"3624bee7b462838a25a24be4d2b6ce22d485952609b1f8b392e4f64df5708d79","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3624bee7b462838a25a24be4d2b6ce22d485952609b1f8b392e4f64df5708d79","first_computed_at":"2026-07-05T09:43:30.699878Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:43:30.699878Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"kQsDh35cVy6QqTLglSGLY2OdMUybQZn5oD2J57r6YYMNhCX6OCpsWBJbZF49pCH66U+PvxAFBVTNlf6u4r73Dg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:43:30.700439Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.18060","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dbd403a68a9ca230a629a405e8862a14e86377896806ce4a08f55e974d4e9730","sha256:da455647035629ec6006435c125885e8aca2e1184f67260df330c7eec25ff27a"],"state_sha256":"9a8b30648c6076a4975b266d4b73f2e33d5a89cc6564d9e789f0798514db0744"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"75YA+8xZW03rJ46PLcfXdIa2OvWQKaM3PZHwOVqVddJ2eDSLz+Tlpx74rUdd2Mql82u44ztltba+ypb59b9xDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T06:01:33.393627Z","bundle_sha256":"22b6f240d2f79b05ddc6cd397e3cb944a8d1cf33de8762bfa749c5d8380607e0"}}