{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:DNM54WKWKYRRRS23SJTE2OS6MU","short_pith_number":"pith:DNM54WKW","canonical_record":{"source":{"id":"2402.12048","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T11:02:05Z","cross_cats_sorted":[],"title_canon_sha256":"d38261723303467e77b5d8b57da8f73386a89b771442489918644594cf4e022e","abstract_canon_sha256":"dfb24461ec263bc71e869aec54dccd26b7f252a5434193c2d9bfdc3a33852578"},"schema_version":"1.0"},"canonical_sha256":"1b59de5956562318cb5b92664d3a5e650e052ec18149f4290627fe06204580c2","source":{"kind":"arxiv","id":"2402.12048","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.12048","created_at":"2026-07-05T07:46:48Z"},{"alias_kind":"arxiv_version","alias_value":"2402.12048v1","created_at":"2026-07-05T07:46:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.12048","created_at":"2026-07-05T07:46:48Z"},{"alias_kind":"pith_short_12","alias_value":"DNM54WKWKYRR","created_at":"2026-07-05T07:46:48Z"},{"alias_kind":"pith_short_16","alias_value":"DNM54WKWKYRRRS23","created_at":"2026-07-05T07:46:48Z"},{"alias_kind":"pith_short_8","alias_value":"DNM54WKW","created_at":"2026-07-05T07:46:48Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:DNM54WKWKYRRRS23SJTE2OS6MU","target":"record","payload":{"canonical_record":{"source":{"id":"2402.12048","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T11:02:05Z","cross_cats_sorted":[],"title_canon_sha256":"d38261723303467e77b5d8b57da8f73386a89b771442489918644594cf4e022e","abstract_canon_sha256":"dfb24461ec263bc71e869aec54dccd26b7f252a5434193c2d9bfdc3a33852578"},"schema_version":"1.0"},"canonical_sha256":"1b59de5956562318cb5b92664d3a5e650e052ec18149f4290627fe06204580c2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:46:48.033306Z","signature_b64":"vtD80Vk8ULJq+WePB2P+2hL0ilZMyjRg3mYW1ZOApM23Wijp+udS9uXSOYWxITaflz7SlVL5lgW8N8WRphXJBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1b59de5956562318cb5b92664d3a5e650e052ec18149f4290627fe06204580c2","last_reissued_at":"2026-07-05T07:46:48.032806Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:46:48.032806Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.12048","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:46:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"UAr7FF/63P7/ki3jcrl4vLkla1VABZXVkiHkJVl8/38OypI/2IC0xR2hOY35REOZ6ghYMDGXAyY/cc9iU5dvDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T08:44:32.120124Z"},"content_sha256":"98fe52672685636410e13ecde23e3712a393a82faf9f2431118935b233daa63f","schema_version":"1.0","event_id":"sha256:98fe52672685636410e13ecde23e3712a393a82faf9f2431118935b233daa63f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:DNM54WKWKYRRRS23SJTE2OS6MU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Model Tailor: Mitigating Catastrophic Forgetting in Multi-modal Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chao Wu, Didi Zhu, Ke Yan, Kun Kuang, Shouhong Ding, Tao Shen, Zexi Li, Zhongyi Sun","submitted_at":"2024-02-19T11:02:05Z","abstract_excerpt":"Catastrophic forgetting emerges as a critical challenge when fine-tuning multi-modal large language models (MLLMs), where improving performance on unseen tasks often leads to a significant performance drop on the original tasks. This paper presents a comprehensive analysis of catastrophic forgetting in MLLMs and introduces a post-training adjustment method called Model Tailor. Our method primarily preserves the pre-trained parameters while replacing a small number ($\\leq$ 10\\%) of fine-tuned parameters, maintaining $\\sim$ 99\\% effectiveness on original tasks versus pre-training, and achieving "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.12048","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/2402.12048/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":"2026-07-05T07:46:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AA2z4K2BJrKvimqj1fEpZG5zx7IW/CtYYWmSw22QM0/93lpJFZXDRCNzkDoMK7XmZZW4MIWgCv24F7PigscUCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T08:44:32.120627Z"},"content_sha256":"384061c03e34cbd526f257b18671eca0e07db663ec92179e88b912c9ed062777","schema_version":"1.0","event_id":"sha256:384061c03e34cbd526f257b18671eca0e07db663ec92179e88b912c9ed062777"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DNM54WKWKYRRRS23SJTE2OS6MU/bundle.json","state_url":"https://pith.science/pith/DNM54WKWKYRRRS23SJTE2OS6MU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DNM54WKWKYRRRS23SJTE2OS6MU/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":"2026-08-06T08:44:32Z","links":{"resolver":"https://pith.science/pith/DNM54WKWKYRRRS23SJTE2OS6MU","bundle":"https://pith.science/pith/DNM54WKWKYRRRS23SJTE2OS6MU/bundle.json","state":"https://pith.science/pith/DNM54WKWKYRRRS23SJTE2OS6MU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DNM54WKWKYRRRS23SJTE2OS6MU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:DNM54WKWKYRRRS23SJTE2OS6MU","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":"dfb24461ec263bc71e869aec54dccd26b7f252a5434193c2d9bfdc3a33852578","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T11:02:05Z","title_canon_sha256":"d38261723303467e77b5d8b57da8f73386a89b771442489918644594cf4e022e"},"schema_version":"1.0","source":{"id":"2402.12048","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.12048","created_at":"2026-07-05T07:46:48Z"},{"alias_kind":"arxiv_version","alias_value":"2402.12048v1","created_at":"2026-07-05T07:46:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.12048","created_at":"2026-07-05T07:46:48Z"},{"alias_kind":"pith_short_12","alias_value":"DNM54WKWKYRR","created_at":"2026-07-05T07:46:48Z"},{"alias_kind":"pith_short_16","alias_value":"DNM54WKWKYRRRS23","created_at":"2026-07-05T07:46:48Z"},{"alias_kind":"pith_short_8","alias_value":"DNM54WKW","created_at":"2026-07-05T07:46:48Z"}],"graph_snapshots":[{"event_id":"sha256:384061c03e34cbd526f257b18671eca0e07db663ec92179e88b912c9ed062777","target":"graph","created_at":"2026-07-05T07:46:48Z","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/2402.12048/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Catastrophic forgetting emerges as a critical challenge when fine-tuning multi-modal large language models (MLLMs), where improving performance on unseen tasks often leads to a significant performance drop on the original tasks. This paper presents a comprehensive analysis of catastrophic forgetting in MLLMs and introduces a post-training adjustment method called Model Tailor. Our method primarily preserves the pre-trained parameters while replacing a small number ($\\leq$ 10\\%) of fine-tuned parameters, maintaining $\\sim$ 99\\% effectiveness on original tasks versus pre-training, and achieving ","authors_text":"Chao Wu, Didi Zhu, Ke Yan, Kun Kuang, Shouhong Ding, Tao Shen, Zexi Li, Zhongyi Sun","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T11:02:05Z","title":"Model Tailor: Mitigating Catastrophic Forgetting in Multi-modal Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.12048","kind":"arxiv","version":1},"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":"sha256:98fe52672685636410e13ecde23e3712a393a82faf9f2431118935b233daa63f","target":"record","created_at":"2026-07-05T07:46:48Z","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":"dfb24461ec263bc71e869aec54dccd26b7f252a5434193c2d9bfdc3a33852578","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-19T11:02:05Z","title_canon_sha256":"d38261723303467e77b5d8b57da8f73386a89b771442489918644594cf4e022e"},"schema_version":"1.0","source":{"id":"2402.12048","kind":"arxiv","version":1}},"canonical_sha256":"1b59de5956562318cb5b92664d3a5e650e052ec18149f4290627fe06204580c2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1b59de5956562318cb5b92664d3a5e650e052ec18149f4290627fe06204580c2","first_computed_at":"2026-07-05T07:46:48.032806Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:46:48.032806Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vtD80Vk8ULJq+WePB2P+2hL0ilZMyjRg3mYW1ZOApM23Wijp+udS9uXSOYWxITaflz7SlVL5lgW8N8WRphXJBg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:46:48.033306Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.12048","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:98fe52672685636410e13ecde23e3712a393a82faf9f2431118935b233daa63f","sha256:384061c03e34cbd526f257b18671eca0e07db663ec92179e88b912c9ed062777"],"state_sha256":"5541148ac91c6fcd2ff7cc5ff7a4b2d1963ae0b841fc94a4b64c3595a8973ba5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ne6Zr6oJcE2EV29+8D/nziuYdLFhUlW4xBaWHYp2D4DH71ciolOxr8/ePLxorht3ngdP63oC6Aew0m4N3LJMDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T08:44:32.127637Z","bundle_sha256":"18e4d28c7b34b22ff75f855c507ea885b19be15137c51b4fd7497e22cc36cf20"}}