{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:PLOEPUQ2JY46TN3BIBDBXOGMXC","short_pith_number":"pith:PLOEPUQ2","canonical_record":{"source":{"id":"2401.04679","kind":"arxiv","version":7},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-01-09T17:09:01Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"698704d2d45ec75b642185e831bb219e4b41e740c2be92509f3288994ca77baa","abstract_canon_sha256":"ec2baf9b2fface44b69690f0d2c92b7952e90fedec370005c94a4a900852ad85"},"schema_version":"1.0"},"canonical_sha256":"7adc47d21a4e39e9b76140461bb8ccb88700e90a8a2d9558127921dee23c0655","source":{"kind":"arxiv","id":"2401.04679","version":7},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.04679","created_at":"2026-07-05T08:26:17Z"},{"alias_kind":"arxiv_version","alias_value":"2401.04679v7","created_at":"2026-07-05T08:26:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.04679","created_at":"2026-07-05T08:26:17Z"},{"alias_kind":"pith_short_12","alias_value":"PLOEPUQ2JY46","created_at":"2026-07-05T08:26:17Z"},{"alias_kind":"pith_short_16","alias_value":"PLOEPUQ2JY46TN3B","created_at":"2026-07-05T08:26:17Z"},{"alias_kind":"pith_short_8","alias_value":"PLOEPUQ2","created_at":"2026-07-05T08:26:17Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:PLOEPUQ2JY46TN3BIBDBXOGMXC","target":"record","payload":{"canonical_record":{"source":{"id":"2401.04679","kind":"arxiv","version":7},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-01-09T17:09:01Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"698704d2d45ec75b642185e831bb219e4b41e740c2be92509f3288994ca77baa","abstract_canon_sha256":"ec2baf9b2fface44b69690f0d2c92b7952e90fedec370005c94a4a900852ad85"},"schema_version":"1.0"},"canonical_sha256":"7adc47d21a4e39e9b76140461bb8ccb88700e90a8a2d9558127921dee23c0655","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:26:17.558392Z","signature_b64":"o33xAex5qI4h8SQIFe3+TMujhFTqXbCr1kAhXVrrDQtsw0BjJQYzbA8NXWbH7EiX7eQE7BlXckyIU7npeRuIAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7adc47d21a4e39e9b76140461bb8ccb88700e90a8a2d9558127921dee23c0655","last_reissued_at":"2026-07-05T08:26:17.557893Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:26:17.557893Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2401.04679","source_version":7,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:26:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ng2/q5ngWOC9m2CFAiTN0Q4awdVatXdRwZDo4wr/Chc7y68mM7x0SJQQWCY1xtDSgo88XPn+TPiwfCYCQ5zaDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T08:16:51.805425Z"},"content_sha256":"1d79a1a3933dde6400e1f19cf57a0b87172cab59320941c7c2ac8d653b31e0d4","schema_version":"1.0","event_id":"sha256:1d79a1a3933dde6400e1f19cf57a0b87172cab59320941c7c2ac8d653b31e0d4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:PLOEPUQ2JY46TN3BIBDBXOGMXC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RoSA: Accurate Parameter-Efficient Fine-Tuning via Robust Adaptation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Dan Alistarh, Elvir Crn\\v{c}evi\\'c, Mahdi Nikdan, Soroush Tabesh","submitted_at":"2024-01-09T17:09:01Z","abstract_excerpt":"We investigate parameter-efficient fine-tuning (PEFT) methods that can provide good accuracy under limited computational and memory budgets in the context of large language models (LLMs). We present a new PEFT method called Robust Adaptation (RoSA) inspired by robust principal component analysis that jointly trains $\\textit{low-rank}$ and $\\textit{highly-sparse}$ components on top of a set of fixed pretrained weights to efficiently approximate the performance of a full-fine-tuning (FFT) solution. Across a series of challenging generative tasks such as grade-school math and SQL query generation"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.04679","kind":"arxiv","version":7},"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/2401.04679/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":"2026-07-05T08:26:17Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Cl+Zu98Yb24Yx4jr1aovr1pHPRgRKnAlUGxjK8hVKx3PAOMK0cECxkayT0n+WIxra6bSa10w2feh1moxNSKWBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T08:16:51.805928Z"},"content_sha256":"1585ddedd0ac835cb34e0f20eab667c578f5997314dec90de68544878fff75ce","schema_version":"1.0","event_id":"sha256:1585ddedd0ac835cb34e0f20eab667c578f5997314dec90de68544878fff75ce"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PLOEPUQ2JY46TN3BIBDBXOGMXC/bundle.json","state_url":"https://pith.science/pith/PLOEPUQ2JY46TN3BIBDBXOGMXC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PLOEPUQ2JY46TN3BIBDBXOGMXC/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":"2026-08-05T08:16:51Z","links":{"resolver":"https://pith.science/pith/PLOEPUQ2JY46TN3BIBDBXOGMXC","bundle":"https://pith.science/pith/PLOEPUQ2JY46TN3BIBDBXOGMXC/bundle.json","state":"https://pith.science/pith/PLOEPUQ2JY46TN3BIBDBXOGMXC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PLOEPUQ2JY46TN3BIBDBXOGMXC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:PLOEPUQ2JY46TN3BIBDBXOGMXC","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":"ec2baf9b2fface44b69690f0d2c92b7952e90fedec370005c94a4a900852ad85","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-01-09T17:09:01Z","title_canon_sha256":"698704d2d45ec75b642185e831bb219e4b41e740c2be92509f3288994ca77baa"},"schema_version":"1.0","source":{"id":"2401.04679","kind":"arxiv","version":7}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2401.04679","created_at":"2026-07-05T08:26:17Z"},{"alias_kind":"arxiv_version","alias_value":"2401.04679v7","created_at":"2026-07-05T08:26:17Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2401.04679","created_at":"2026-07-05T08:26:17Z"},{"alias_kind":"pith_short_12","alias_value":"PLOEPUQ2JY46","created_at":"2026-07-05T08:26:17Z"},{"alias_kind":"pith_short_16","alias_value":"PLOEPUQ2JY46TN3B","created_at":"2026-07-05T08:26:17Z"},{"alias_kind":"pith_short_8","alias_value":"PLOEPUQ2","created_at":"2026-07-05T08:26:17Z"}],"graph_snapshots":[{"event_id":"sha256:1585ddedd0ac835cb34e0f20eab667c578f5997314dec90de68544878fff75ce","target":"graph","created_at":"2026-07-05T08:26:17Z","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/2401.04679/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We investigate parameter-efficient fine-tuning (PEFT) methods that can provide good accuracy under limited computational and memory budgets in the context of large language models (LLMs). We present a new PEFT method called Robust Adaptation (RoSA) inspired by robust principal component analysis that jointly trains $\\textit{low-rank}$ and $\\textit{highly-sparse}$ components on top of a set of fixed pretrained weights to efficiently approximate the performance of a full-fine-tuning (FFT) solution. Across a series of challenging generative tasks such as grade-school math and SQL query generation","authors_text":"Dan Alistarh, Elvir Crn\\v{c}evi\\'c, Mahdi Nikdan, Soroush Tabesh","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-01-09T17:09:01Z","title":"RoSA: Accurate Parameter-Efficient Fine-Tuning via Robust Adaptation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2401.04679","kind":"arxiv","version":7},"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":"sha256:1d79a1a3933dde6400e1f19cf57a0b87172cab59320941c7c2ac8d653b31e0d4","target":"record","created_at":"2026-07-05T08:26:17Z","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":"ec2baf9b2fface44b69690f0d2c92b7952e90fedec370005c94a4a900852ad85","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-01-09T17:09:01Z","title_canon_sha256":"698704d2d45ec75b642185e831bb219e4b41e740c2be92509f3288994ca77baa"},"schema_version":"1.0","source":{"id":"2401.04679","kind":"arxiv","version":7}},"canonical_sha256":"7adc47d21a4e39e9b76140461bb8ccb88700e90a8a2d9558127921dee23c0655","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7adc47d21a4e39e9b76140461bb8ccb88700e90a8a2d9558127921dee23c0655","first_computed_at":"2026-07-05T08:26:17.557893Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:26:17.557893Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"o33xAex5qI4h8SQIFe3+TMujhFTqXbCr1kAhXVrrDQtsw0BjJQYzbA8NXWbH7EiX7eQE7BlXckyIU7npeRuIAw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:26:17.558392Z","signed_message":"canonical_sha256_bytes"},"source_id":"2401.04679","source_kind":"arxiv","source_version":7}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1d79a1a3933dde6400e1f19cf57a0b87172cab59320941c7c2ac8d653b31e0d4","sha256:1585ddedd0ac835cb34e0f20eab667c578f5997314dec90de68544878fff75ce"],"state_sha256":"49edf820a41de8ea415b78031523d010dd0aa9682fcb1c3d6ead26c13b1b0ef4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0OUJ5S30zf3erEZiQCsNj4PJIS+uMrGgeoGX64EjY80gFEm1Tw6EeP9rFcirS5FqcVDoJ5p/wIN/u+U1B6UGBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T08:16:51.812079Z","bundle_sha256":"030d6076df2872a208445c82f3f025141aaa52cf221e725dae661c4c99018479"}}