{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SYF4DYC3NPKTYTBEQUIIUOYSX6","short_pith_number":"pith:SYF4DYC3","schema_version":"1.0","canonical_sha256":"960bc1e05b6bd53c4c2485108a3b12bf93a34730ddd9118e791f98dccb839593","source":{"kind":"arxiv","id":"2504.08851","version":2},"attestation_state":"computed","paper":{"title":"Mimic In-Context Learning for Multimodal Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chenduo Hao, Jiale Fu, Xin Geng, Xinting Hu, Xu Yang, Yingzhe Peng, Yuchu Jiang","submitted_at":"2025-04-11T03:37:59Z","abstract_excerpt":"Recently, In-context Learning (ICL) has become a significant inference paradigm in Large Multimodal Models (LMMs), utilizing a few in-context demonstrations (ICDs) to prompt LMMs for new tasks. However, the synergistic effects in multimodal data increase the sensitivity of ICL performance to the configurations of ICDs, stimulating the need for a more stable and general mapping function. Mathematically, in Transformer-based models, ICDs act as \"shift vectors\" added to the hidden states of query tokens. Inspired by this, we introduce Mimic In-Context Learning (MimIC) to learn stable and generali"},"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":"2504.08851","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-11T03:37:59Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"82609c65f3cafcad2c19954d6a9b60cf20c8636aabff4d212584dc24d1858fbf","abstract_canon_sha256":"1285e71adf69a932bac91d7109b3388071fcf2e2ea4561e96d4f55fc3fd8da27"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:04:48.898962Z","signature_b64":"t1JVfiDtufgtm1hW6YPtE0s+kGO18Jeyg2bqXqfPym+TU6IRoG1/yYjN5RkGXpVekHfvkTwA9Z4E43Y6gZjRCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"960bc1e05b6bd53c4c2485108a3b12bf93a34730ddd9118e791f98dccb839593","last_reissued_at":"2026-07-05T11:04:48.898423Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:04:48.898423Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Mimic In-Context Learning for Multimodal Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Chenduo Hao, Jiale Fu, Xin Geng, Xinting Hu, Xu Yang, Yingzhe Peng, Yuchu Jiang","submitted_at":"2025-04-11T03:37:59Z","abstract_excerpt":"Recently, In-context Learning (ICL) has become a significant inference paradigm in Large Multimodal Models (LMMs), utilizing a few in-context demonstrations (ICDs) to prompt LMMs for new tasks. However, the synergistic effects in multimodal data increase the sensitivity of ICL performance to the configurations of ICDs, stimulating the need for a more stable and general mapping function. Mathematically, in Transformer-based models, ICDs act as \"shift vectors\" added to the hidden states of query tokens. Inspired by this, we introduce Mimic In-Context Learning (MimIC) to learn stable and generali"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.08851","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/2504.08851/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":"2504.08851","created_at":"2026-07-05T11:04:48.898487+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.08851v2","created_at":"2026-07-05T11:04:48.898487+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.08851","created_at":"2026-07-05T11:04:48.898487+00:00"},{"alias_kind":"pith_short_12","alias_value":"SYF4DYC3NPKT","created_at":"2026-07-05T11:04:48.898487+00:00"},{"alias_kind":"pith_short_16","alias_value":"SYF4DYC3NPKTYTBE","created_at":"2026-07-05T11:04:48.898487+00:00"},{"alias_kind":"pith_short_8","alias_value":"SYF4DYC3","created_at":"2026-07-05T11:04:48.898487+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2501.03012","citing_title":"Analyzing Finetuning Representation Shift for Multimodal LLMs Steering","ref_index":29,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SYF4DYC3NPKTYTBEQUIIUOYSX6","json":"https://pith.science/pith/SYF4DYC3NPKTYTBEQUIIUOYSX6.json","graph_json":"https://pith.science/api/pith-number/SYF4DYC3NPKTYTBEQUIIUOYSX6/graph.json","events_json":"https://pith.science/api/pith-number/SYF4DYC3NPKTYTBEQUIIUOYSX6/events.json","paper":"https://pith.science/paper/SYF4DYC3"},"agent_actions":{"view_html":"https://pith.science/pith/SYF4DYC3NPKTYTBEQUIIUOYSX6","download_json":"https://pith.science/pith/SYF4DYC3NPKTYTBEQUIIUOYSX6.json","view_paper":"https://pith.science/paper/SYF4DYC3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.08851&json=true","fetch_graph":"https://pith.science/api/pith-number/SYF4DYC3NPKTYTBEQUIIUOYSX6/graph.json","fetch_events":"https://pith.science/api/pith-number/SYF4DYC3NPKTYTBEQUIIUOYSX6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SYF4DYC3NPKTYTBEQUIIUOYSX6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SYF4DYC3NPKTYTBEQUIIUOYSX6/action/storage_attestation","attest_author":"https://pith.science/pith/SYF4DYC3NPKTYTBEQUIIUOYSX6/action/author_attestation","sign_citation":"https://pith.science/pith/SYF4DYC3NPKTYTBEQUIIUOYSX6/action/citation_signature","submit_replication":"https://pith.science/pith/SYF4DYC3NPKTYTBEQUIIUOYSX6/action/replication_record"}},"created_at":"2026-07-05T11:04:48.898487+00:00","updated_at":"2026-07-05T11:04:48.898487+00:00"}