{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:2N3BK6JSB2HYFBWUSTF72TQUV3","short_pith_number":"pith:2N3BK6JS","schema_version":"1.0","canonical_sha256":"d3761579320e8f8286d494cbfd4e14aef12319cb9e7e23c6987a7ac7545cd914","source":{"kind":"arxiv","id":"2410.05849","version":2},"attestation_state":"computed","paper":{"title":"ModalPrompt: Towards Efficient Multimodal Continual Instruction Tuning with Dual-Modality Guided Prompt","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cheng-Lin Liu, Fanhu Zeng, Fei Zhu, Haiyang Guo, Xu-Yao Zhang","submitted_at":"2024-10-08T09:35:37Z","abstract_excerpt":"Large Multimodal Models (LMMs) exhibit remarkable multi-tasking ability by learning mixed instruction datasets. However, novel tasks would be encountered sequentially in dynamic world, which urges for equipping LMMs with multimodal continual instruction learning (MCIT) ability especially for diverse and challenging generative tasks. Existing MCIT methods do not fully exploit the unique attribute of LMMs and often gain performance at the expense of efficiency. In this paper, we propose a novel prompt learning framework for MCIT to effectively alleviate forgetting of previous knowledge while man"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2410.05849","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-10-08T09:35:37Z","cross_cats_sorted":[],"title_canon_sha256":"7976a5a65fa46fefdf7c2130ef6bef9753b8f488df39a3b30061386a104d5b6f","abstract_canon_sha256":"60ae17ad21f696b6c94516a1cfe81f336b0fdb051667c3332f88dce9eb462ce5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:58:20.987887Z","signature_b64":"tb2Qh45RRhDp7KoB5TrKu0cFiyYb9DHK3aw/GGjICQ1/4YgOgNJ7Lt/qppyvVv4ItMA+1YgV+TEdgF8i78boDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d3761579320e8f8286d494cbfd4e14aef12319cb9e7e23c6987a7ac7545cd914","last_reissued_at":"2026-07-05T11:58:20.987433Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:58:20.987433Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ModalPrompt: Towards Efficient Multimodal Continual Instruction Tuning with Dual-Modality Guided Prompt","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Cheng-Lin Liu, Fanhu Zeng, Fei Zhu, Haiyang Guo, Xu-Yao Zhang","submitted_at":"2024-10-08T09:35:37Z","abstract_excerpt":"Large Multimodal Models (LMMs) exhibit remarkable multi-tasking ability by learning mixed instruction datasets. However, novel tasks would be encountered sequentially in dynamic world, which urges for equipping LMMs with multimodal continual instruction learning (MCIT) ability especially for diverse and challenging generative tasks. Existing MCIT methods do not fully exploit the unique attribute of LMMs and often gain performance at the expense of efficiency. In this paper, we propose a novel prompt learning framework for MCIT to effectively alleviate forgetting of previous knowledge while man"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.05849","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/2410.05849/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2410.05849","created_at":"2026-07-05T11:58:20.987497+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.05849v2","created_at":"2026-07-05T11:58:20.987497+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.05849","created_at":"2026-07-05T11:58:20.987497+00:00"},{"alias_kind":"pith_short_12","alias_value":"2N3BK6JSB2HY","created_at":"2026-07-05T11:58:20.987497+00:00"},{"alias_kind":"pith_short_16","alias_value":"2N3BK6JSB2HYFBWU","created_at":"2026-07-05T11:58:20.987497+00:00"},{"alias_kind":"pith_short_8","alias_value":"2N3BK6JS","created_at":"2026-07-05T11:58:20.987497+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2412.13050","citing_title":"Modality-Inconsistent Continual Learning of Multimodal Large Language Models","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2510.20685","citing_title":"C-NAV: Towards Self-Evolving Continual Object Navigation in Open World","ref_index":50,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04857","citing_title":"The Blind Spot of Adaptation: Quantifying and Mitigating Forgetting in Fine-tuned Driving Models","ref_index":52,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2N3BK6JSB2HYFBWUSTF72TQUV3","json":"https://pith.science/pith/2N3BK6JSB2HYFBWUSTF72TQUV3.json","graph_json":"https://pith.science/api/pith-number/2N3BK6JSB2HYFBWUSTF72TQUV3/graph.json","events_json":"https://pith.science/api/pith-number/2N3BK6JSB2HYFBWUSTF72TQUV3/events.json","paper":"https://pith.science/paper/2N3BK6JS"},"agent_actions":{"view_html":"https://pith.science/pith/2N3BK6JSB2HYFBWUSTF72TQUV3","download_json":"https://pith.science/pith/2N3BK6JSB2HYFBWUSTF72TQUV3.json","view_paper":"https://pith.science/paper/2N3BK6JS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.05849&json=true","fetch_graph":"https://pith.science/api/pith-number/2N3BK6JSB2HYFBWUSTF72TQUV3/graph.json","fetch_events":"https://pith.science/api/pith-number/2N3BK6JSB2HYFBWUSTF72TQUV3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2N3BK6JSB2HYFBWUSTF72TQUV3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2N3BK6JSB2HYFBWUSTF72TQUV3/action/storage_attestation","attest_author":"https://pith.science/pith/2N3BK6JSB2HYFBWUSTF72TQUV3/action/author_attestation","sign_citation":"https://pith.science/pith/2N3BK6JSB2HYFBWUSTF72TQUV3/action/citation_signature","submit_replication":"https://pith.science/pith/2N3BK6JSB2HYFBWUSTF72TQUV3/action/replication_record"}},"created_at":"2026-07-05T11:58:20.987497+00:00","updated_at":"2026-07-05T11:58:20.987497+00:00"}