{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:GY5ARHJLUBQ6SCG253AACRTQTG","short_pith_number":"pith:GY5ARHJL","canonical_record":{"source":{"id":"2407.08044","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-10T20:52:18Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"95f647d25de437a5d7d09d447bf32fa40e6492b24bf68a2f1d10c3759bcfc1c1","abstract_canon_sha256":"efe1bb408121f35b215fcc86197f8c3cf0ea1b2a361ac5ed3590dbd50f6a6056"},"schema_version":"1.0"},"canonical_sha256":"363a089d2ba061e908daeec001467099aae0b610272b8741f68f0b1f6a4370e3","source":{"kind":"arxiv","id":"2407.08044","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.08044","created_at":"2026-07-05T09:12:29Z"},{"alias_kind":"arxiv_version","alias_value":"2407.08044v2","created_at":"2026-07-05T09:12:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.08044","created_at":"2026-07-05T09:12:29Z"},{"alias_kind":"pith_short_12","alias_value":"GY5ARHJLUBQ6","created_at":"2026-07-05T09:12:29Z"},{"alias_kind":"pith_short_16","alias_value":"GY5ARHJLUBQ6SCG2","created_at":"2026-07-05T09:12:29Z"},{"alias_kind":"pith_short_8","alias_value":"GY5ARHJL","created_at":"2026-07-05T09:12:29Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:GY5ARHJLUBQ6SCG253AACRTQTG","target":"record","payload":{"canonical_record":{"source":{"id":"2407.08044","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-10T20:52:18Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"95f647d25de437a5d7d09d447bf32fa40e6492b24bf68a2f1d10c3759bcfc1c1","abstract_canon_sha256":"efe1bb408121f35b215fcc86197f8c3cf0ea1b2a361ac5ed3590dbd50f6a6056"},"schema_version":"1.0"},"canonical_sha256":"363a089d2ba061e908daeec001467099aae0b610272b8741f68f0b1f6a4370e3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:12:29.879878Z","signature_b64":"2b8YNmMrgrknpqe+PIbFZd1Yt9Zy0u2s7nX91iTmJbj9e8WFH01LKrrYlYxNmEMfOJHemQ/jUPDhcIrisVeNAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"363a089d2ba061e908daeec001467099aae0b610272b8741f68f0b1f6a4370e3","last_reissued_at":"2026-07-05T09:12:29.879377Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:12:29.879377Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2407.08044","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:12:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kCbhON89Tvt6nakbW7eGx3zi88Y1ejGBjadND94KaGGOoVRxRWT30O5aIQHwQrXbRLR6ohn59NAEmHzKF0ocBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T19:51:23.577332Z"},"content_sha256":"8f6dde91af9d167cb04edb421af3ea64ce498c73113651c161df99ce8bc22775","schema_version":"1.0","event_id":"sha256:8f6dde91af9d167cb04edb421af3ea64ce498c73113651c161df99ce8bc22775"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:GY5ARHJLUBQ6SCG253AACRTQTG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"RoLoRA: Fine-tuning Rotated Outlier-free LLMs for Effective Weight-Activation Quantization","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Kwang-Ting Cheng, Shih-Yang Liu, Xijie Huang, Zechun Liu","submitted_at":"2024-07-10T20:52:18Z","abstract_excerpt":"Low-Rank Adaptation (LoRA), as a representative Parameter-Efficient Fine-Tuning (PEFT)method, significantly enhances the training efficiency by updating only a small portion of the weights in Large Language Models (LLMs). Recently, weight-only quantization techniques have also been applied to LoRA methods to reduce the memory footprint of fine-tuning. However, applying weight-activation quantization to the LoRA pipeline is under-explored, and we observe substantial performance degradation primarily due to the presence of activation outliers. In this work, we propose RoLoRA, the first LoRA-base"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.08044","kind":"arxiv","version":2},"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/2407.08044/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":"2026-07-05T09:12:29Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CDwFEDDMDpVtkEW4Wy0NG53NfsD/oKJj7sr7EIuxXjsWDB+M6ZFWICIMHAhZIbJMqcsDy/sSBtWn/ULcHEVuCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T19:51:23.577850Z"},"content_sha256":"50ef9c9d056e58a3c40cfce341dc5c5852b764fa0207c6eba01573c962100738","schema_version":"1.0","event_id":"sha256:50ef9c9d056e58a3c40cfce341dc5c5852b764fa0207c6eba01573c962100738"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/GY5ARHJLUBQ6SCG253AACRTQTG/bundle.json","state_url":"https://pith.science/pith/GY5ARHJLUBQ6SCG253AACRTQTG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/GY5ARHJLUBQ6SCG253AACRTQTG/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":"2026-08-04T19:51:23Z","links":{"resolver":"https://pith.science/pith/GY5ARHJLUBQ6SCG253AACRTQTG","bundle":"https://pith.science/pith/GY5ARHJLUBQ6SCG253AACRTQTG/bundle.json","state":"https://pith.science/pith/GY5ARHJLUBQ6SCG253AACRTQTG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/GY5ARHJLUBQ6SCG253AACRTQTG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GY5ARHJLUBQ6SCG253AACRTQTG","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":"efe1bb408121f35b215fcc86197f8c3cf0ea1b2a361ac5ed3590dbd50f6a6056","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-10T20:52:18Z","title_canon_sha256":"95f647d25de437a5d7d09d447bf32fa40e6492b24bf68a2f1d10c3759bcfc1c1"},"schema_version":"1.0","source":{"id":"2407.08044","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.08044","created_at":"2026-07-05T09:12:29Z"},{"alias_kind":"arxiv_version","alias_value":"2407.08044v2","created_at":"2026-07-05T09:12:29Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.08044","created_at":"2026-07-05T09:12:29Z"},{"alias_kind":"pith_short_12","alias_value":"GY5ARHJLUBQ6","created_at":"2026-07-05T09:12:29Z"},{"alias_kind":"pith_short_16","alias_value":"GY5ARHJLUBQ6SCG2","created_at":"2026-07-05T09:12:29Z"},{"alias_kind":"pith_short_8","alias_value":"GY5ARHJL","created_at":"2026-07-05T09:12:29Z"}],"graph_snapshots":[{"event_id":"sha256:50ef9c9d056e58a3c40cfce341dc5c5852b764fa0207c6eba01573c962100738","target":"graph","created_at":"2026-07-05T09:12:29Z","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/2407.08044/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Low-Rank Adaptation (LoRA), as a representative Parameter-Efficient Fine-Tuning (PEFT)method, significantly enhances the training efficiency by updating only a small portion of the weights in Large Language Models (LLMs). Recently, weight-only quantization techniques have also been applied to LoRA methods to reduce the memory footprint of fine-tuning. However, applying weight-activation quantization to the LoRA pipeline is under-explored, and we observe substantial performance degradation primarily due to the presence of activation outliers. In this work, we propose RoLoRA, the first LoRA-base","authors_text":"Kwang-Ting Cheng, Shih-Yang Liu, Xijie Huang, Zechun Liu","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-10T20:52:18Z","title":"RoLoRA: Fine-tuning Rotated Outlier-free LLMs for Effective Weight-Activation Quantization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.08044","kind":"arxiv","version":2},"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":"sha256:8f6dde91af9d167cb04edb421af3ea64ce498c73113651c161df99ce8bc22775","target":"record","created_at":"2026-07-05T09:12:29Z","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":"efe1bb408121f35b215fcc86197f8c3cf0ea1b2a361ac5ed3590dbd50f6a6056","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-10T20:52:18Z","title_canon_sha256":"95f647d25de437a5d7d09d447bf32fa40e6492b24bf68a2f1d10c3759bcfc1c1"},"schema_version":"1.0","source":{"id":"2407.08044","kind":"arxiv","version":2}},"canonical_sha256":"363a089d2ba061e908daeec001467099aae0b610272b8741f68f0b1f6a4370e3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"363a089d2ba061e908daeec001467099aae0b610272b8741f68f0b1f6a4370e3","first_computed_at":"2026-07-05T09:12:29.879377Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:12:29.879377Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2b8YNmMrgrknpqe+PIbFZd1Yt9Zy0u2s7nX91iTmJbj9e8WFH01LKrrYlYxNmEMfOJHemQ/jUPDhcIrisVeNAg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:12:29.879878Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.08044","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8f6dde91af9d167cb04edb421af3ea64ce498c73113651c161df99ce8bc22775","sha256:50ef9c9d056e58a3c40cfce341dc5c5852b764fa0207c6eba01573c962100738"],"state_sha256":"5a304134a8c88c0ab11bae65350c6914b328d7d2ccf77a1492b1c17e7c880bee"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nxaQEkPk0MO0ax5DQwdOwrstbNy76hB4A1h4JvG4XVx0fF4TI9MTpw6qhdbe8DgThCfryUEiQ0E9M3ayzKHrDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T19:51:23.582111Z","bundle_sha256":"4ed52275094775aac0464291bd7894681c82ce73d1301326bd9915ae6c077b8e"}}