{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:O2I6WL3SGV5BNV4EKNTZFUMDPF","short_pith_number":"pith:O2I6WL3S","canonical_record":{"source":{"id":"2411.10928","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-17T01:16:37Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c109691348d1900af52d527da878e9f71f5af2d44b11cda74795d6669ac7931d","abstract_canon_sha256":"5466c0c261cf15ebc56b7aa86dd5cfc10cd97608b91a107e38d361d14a3245f0"},"schema_version":"1.0"},"canonical_sha256":"7691eb2f72357a16d784536792d1837948d45b37637bf704e3c2c8635da64a6d","source":{"kind":"arxiv","id":"2411.10928","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.10928","created_at":"2026-07-05T09:36:25Z"},{"alias_kind":"arxiv_version","alias_value":"2411.10928v1","created_at":"2026-07-05T09:36:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.10928","created_at":"2026-07-05T09:36:25Z"},{"alias_kind":"pith_short_12","alias_value":"O2I6WL3SGV5B","created_at":"2026-07-05T09:36:25Z"},{"alias_kind":"pith_short_16","alias_value":"O2I6WL3SGV5BNV4E","created_at":"2026-07-05T09:36:25Z"},{"alias_kind":"pith_short_8","alias_value":"O2I6WL3S","created_at":"2026-07-05T09:36:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:O2I6WL3SGV5BNV4EKNTZFUMDPF","target":"record","payload":{"canonical_record":{"source":{"id":"2411.10928","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-17T01:16:37Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c109691348d1900af52d527da878e9f71f5af2d44b11cda74795d6669ac7931d","abstract_canon_sha256":"5466c0c261cf15ebc56b7aa86dd5cfc10cd97608b91a107e38d361d14a3245f0"},"schema_version":"1.0"},"canonical_sha256":"7691eb2f72357a16d784536792d1837948d45b37637bf704e3c2c8635da64a6d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:36:25.183393Z","signature_b64":"TteRKHNtqkwuifYDfv4w1OOgssVO3eKi8U7FTs2cyi4tzJLsXTWJuHEOJVn6qrpN26/x6dqZ177c85oMH4W6Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7691eb2f72357a16d784536792d1837948d45b37637bf704e3c2c8635da64a6d","last_reissued_at":"2026-07-05T09:36:25.182967Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:36:25.182967Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2411.10928","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":"2026-07-05T09:36:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GSVr2AgCX0aML7vMlEBnl8jNO/ANO5QPpqmkXj9yzLf5nRoqzXEfU9/SjXGdUc29ZIJPdmOOnYuEeI7RWp0nBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-09T13:11:47.174240Z"},"content_sha256":"44064abb300fbb1ab5da74b59e858c7e3a584c7fa87b90d13c5f91db23d4989f","schema_version":"1.0","event_id":"sha256:44064abb300fbb1ab5da74b59e858c7e3a584c7fa87b90d13c5f91db23d4989f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:O2I6WL3SGV5BNV4EKNTZFUMDPF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Learn from Downstream and Be Yourself in Multimodal Large Language Model Fine-Tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bo Du, Dacheng Tao, Didi Zhu, Guancheng Wan, He Li, Jian Liang, Mang Ye, Wenke Huang, Zekun Shi","submitted_at":"2024-11-17T01:16:37Z","abstract_excerpt":"Multimodal Large Language Model (MLLM) have demonstrated strong generalization capabilities across diverse distributions and tasks, largely due to extensive pre-training datasets. Fine-tuning MLLM has become a common practice to improve performance on specific downstream tasks. However, during fine-tuning, MLLM often faces the risk of forgetting knowledge acquired during pre-training, which can result in a decline in generalization abilities. To balance the trade-off between generalization and specialization, we propose measuring the parameter importance for both pre-trained and fine-tuning di"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.10928","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/2411.10928/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":"2026-07-05T09:36:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3xZHhOirC/NpX+rzmmtyJQunY7Y1f3nyrxoWbQ9+t3Sghu3gOlhnTnaLRjzUr7BY0WSLRz6NF5OhEc2mQeWOAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-09T13:11:47.174628Z"},"content_sha256":"11b83a23aae6ccf35683a4d56c387b578fd522c1e8e7c714989e108ab95ab771","schema_version":"1.0","event_id":"sha256:11b83a23aae6ccf35683a4d56c387b578fd522c1e8e7c714989e108ab95ab771"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/O2I6WL3SGV5BNV4EKNTZFUMDPF/bundle.json","state_url":"https://pith.science/pith/O2I6WL3SGV5BNV4EKNTZFUMDPF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/O2I6WL3SGV5BNV4EKNTZFUMDPF/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":"2026-07-09T13:11:47Z","links":{"resolver":"https://pith.science/pith/O2I6WL3SGV5BNV4EKNTZFUMDPF","bundle":"https://pith.science/pith/O2I6WL3SGV5BNV4EKNTZFUMDPF/bundle.json","state":"https://pith.science/pith/O2I6WL3SGV5BNV4EKNTZFUMDPF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/O2I6WL3SGV5BNV4EKNTZFUMDPF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:O2I6WL3SGV5BNV4EKNTZFUMDPF","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":"5466c0c261cf15ebc56b7aa86dd5cfc10cd97608b91a107e38d361d14a3245f0","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-17T01:16:37Z","title_canon_sha256":"c109691348d1900af52d527da878e9f71f5af2d44b11cda74795d6669ac7931d"},"schema_version":"1.0","source":{"id":"2411.10928","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.10928","created_at":"2026-07-05T09:36:25Z"},{"alias_kind":"arxiv_version","alias_value":"2411.10928v1","created_at":"2026-07-05T09:36:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.10928","created_at":"2026-07-05T09:36:25Z"},{"alias_kind":"pith_short_12","alias_value":"O2I6WL3SGV5B","created_at":"2026-07-05T09:36:25Z"},{"alias_kind":"pith_short_16","alias_value":"O2I6WL3SGV5BNV4E","created_at":"2026-07-05T09:36:25Z"},{"alias_kind":"pith_short_8","alias_value":"O2I6WL3S","created_at":"2026-07-05T09:36:25Z"}],"graph_snapshots":[{"event_id":"sha256:11b83a23aae6ccf35683a4d56c387b578fd522c1e8e7c714989e108ab95ab771","target":"graph","created_at":"2026-07-05T09:36:25Z","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/2411.10928/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multimodal Large Language Model (MLLM) have demonstrated strong generalization capabilities across diverse distributions and tasks, largely due to extensive pre-training datasets. Fine-tuning MLLM has become a common practice to improve performance on specific downstream tasks. However, during fine-tuning, MLLM often faces the risk of forgetting knowledge acquired during pre-training, which can result in a decline in generalization abilities. To balance the trade-off between generalization and specialization, we propose measuring the parameter importance for both pre-trained and fine-tuning di","authors_text":"Bo Du, Dacheng Tao, Didi Zhu, Guancheng Wan, He Li, Jian Liang, Mang Ye, Wenke Huang, Zekun Shi","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-17T01:16:37Z","title":"Learn from Downstream and Be Yourself in Multimodal Large Language Model Fine-Tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.10928","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":"sha256:44064abb300fbb1ab5da74b59e858c7e3a584c7fa87b90d13c5f91db23d4989f","target":"record","created_at":"2026-07-05T09:36:25Z","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":"5466c0c261cf15ebc56b7aa86dd5cfc10cd97608b91a107e38d361d14a3245f0","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-11-17T01:16:37Z","title_canon_sha256":"c109691348d1900af52d527da878e9f71f5af2d44b11cda74795d6669ac7931d"},"schema_version":"1.0","source":{"id":"2411.10928","kind":"arxiv","version":1}},"canonical_sha256":"7691eb2f72357a16d784536792d1837948d45b37637bf704e3c2c8635da64a6d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7691eb2f72357a16d784536792d1837948d45b37637bf704e3c2c8635da64a6d","first_computed_at":"2026-07-05T09:36:25.182967Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:36:25.182967Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"TteRKHNtqkwuifYDfv4w1OOgssVO3eKi8U7FTs2cyi4tzJLsXTWJuHEOJVn6qrpN26/x6dqZ177c85oMH4W6Aw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:36:25.183393Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.10928","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:44064abb300fbb1ab5da74b59e858c7e3a584c7fa87b90d13c5f91db23d4989f","sha256:11b83a23aae6ccf35683a4d56c387b578fd522c1e8e7c714989e108ab95ab771"],"state_sha256":"81a56184ea992897d3c6199f4a1af9ce319e92974f3f9b2e2fad0b23fa208a65"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MC/SjG2k5aXCf+22dc38G5GYb7uDIOxx9skbxeKjC+Jm7GSebnLmmHL5FN6bOSigkPF0/dIw3twebffpsRAxAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-09T13:11:47.177618Z","bundle_sha256":"d7585357f3974903a2a2436838c50179c0c33d2c13f9dd56abeb6610e2806597"}}