{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:XZ5C3FQ4H5LF5SEQVZXOM26DYT","short_pith_number":"pith:XZ5C3FQ4","schema_version":"1.0","canonical_sha256":"be7a2d961c3f565ec890ae6ee66bc3c4df4ad4b997e345d530eb7946ad58b634","source":{"kind":"arxiv","id":"2406.07588","version":2},"attestation_state":"computed","paper":{"title":"AIM: Let Any Multi-modal Large Language Models Embrace Efficient In-Context Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.MM","authors_text":"Jun Gao, Qian Qiao, Wenjie Li, Zili Wang, Ziqiang Cao","submitted_at":"2024-06-11T08:12:43Z","abstract_excerpt":"In-context learning (ICL) facilitates Large Language Models (LLMs) exhibiting emergent ability on downstream tasks without updating billions of parameters. However, in the area of multi-modal Large Language Models (MLLMs), two problems hinder the application of multi-modal ICL: (1) Most primary MLLMs are only trained on single-image datasets, making them unable to read multi-modal demonstrations. (2) With the demonstrations increasing, thousands of visual tokens highly challenge hardware and degrade ICL performance. During preliminary explorations, we discovered that the inner LLM tends to foc"},"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":"2406.07588","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.MM","submitted_at":"2024-06-11T08:12:43Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"a760bd6b4b331d5f4c3d313e2e8f0fd73098411053cb84d5d59e279a37698377","abstract_canon_sha256":"e22e3e97973429ca723b469158967b8d959cd3462f0fe90801193ae7583690a7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:38:10.540065Z","signature_b64":"+BzsOrhn/gAd73KZCVw+vQYxcCsGnZ7d7BqpWxejJkmJ0mmVANXSW4/rd9iH4kLLdRQJ9NhihhUSzITrT11KAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"be7a2d961c3f565ec890ae6ee66bc3c4df4ad4b997e345d530eb7946ad58b634","last_reissued_at":"2026-07-05T08:38:10.539592Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:38:10.539592Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AIM: Let Any Multi-modal Large Language Models Embrace Efficient In-Context Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.MM","authors_text":"Jun Gao, Qian Qiao, Wenjie Li, Zili Wang, Ziqiang Cao","submitted_at":"2024-06-11T08:12:43Z","abstract_excerpt":"In-context learning (ICL) facilitates Large Language Models (LLMs) exhibiting emergent ability on downstream tasks without updating billions of parameters. However, in the area of multi-modal Large Language Models (MLLMs), two problems hinder the application of multi-modal ICL: (1) Most primary MLLMs are only trained on single-image datasets, making them unable to read multi-modal demonstrations. (2) With the demonstrations increasing, thousands of visual tokens highly challenge hardware and degrade ICL performance. During preliminary explorations, we discovered that the inner LLM tends to foc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.07588","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/2406.07588/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":"2406.07588","created_at":"2026-07-05T08:38:10.539657+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.07588v2","created_at":"2026-07-05T08:38:10.539657+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.07588","created_at":"2026-07-05T08:38:10.539657+00:00"},{"alias_kind":"pith_short_12","alias_value":"XZ5C3FQ4H5LF","created_at":"2026-07-05T08:38:10.539657+00:00"},{"alias_kind":"pith_short_16","alias_value":"XZ5C3FQ4H5LF5SEQ","created_at":"2026-07-05T08:38:10.539657+00:00"},{"alias_kind":"pith_short_8","alias_value":"XZ5C3FQ4","created_at":"2026-07-05T08:38:10.539657+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.13403","citing_title":"Why Multimodal In-Context Learning Lags Behind? Unveiling the Inner Mechanisms and Bottlenecks","ref_index":11,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XZ5C3FQ4H5LF5SEQVZXOM26DYT","json":"https://pith.science/pith/XZ5C3FQ4H5LF5SEQVZXOM26DYT.json","graph_json":"https://pith.science/api/pith-number/XZ5C3FQ4H5LF5SEQVZXOM26DYT/graph.json","events_json":"https://pith.science/api/pith-number/XZ5C3FQ4H5LF5SEQVZXOM26DYT/events.json","paper":"https://pith.science/paper/XZ5C3FQ4"},"agent_actions":{"view_html":"https://pith.science/pith/XZ5C3FQ4H5LF5SEQVZXOM26DYT","download_json":"https://pith.science/pith/XZ5C3FQ4H5LF5SEQVZXOM26DYT.json","view_paper":"https://pith.science/paper/XZ5C3FQ4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.07588&json=true","fetch_graph":"https://pith.science/api/pith-number/XZ5C3FQ4H5LF5SEQVZXOM26DYT/graph.json","fetch_events":"https://pith.science/api/pith-number/XZ5C3FQ4H5LF5SEQVZXOM26DYT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XZ5C3FQ4H5LF5SEQVZXOM26DYT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XZ5C3FQ4H5LF5SEQVZXOM26DYT/action/storage_attestation","attest_author":"https://pith.science/pith/XZ5C3FQ4H5LF5SEQVZXOM26DYT/action/author_attestation","sign_citation":"https://pith.science/pith/XZ5C3FQ4H5LF5SEQVZXOM26DYT/action/citation_signature","submit_replication":"https://pith.science/pith/XZ5C3FQ4H5LF5SEQVZXOM26DYT/action/replication_record"}},"created_at":"2026-07-05T08:38:10.539657+00:00","updated_at":"2026-07-05T08:38:10.539657+00:00"}