{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YYQ4HBU5SYM6CUFD24AHT6YOU4","short_pith_number":"pith:YYQ4HBU5","schema_version":"1.0","canonical_sha256":"c621c3869d9619e150a3d70079fb0ea72ff8618ba4950da63d2a34374e918ac0","source":{"kind":"arxiv","id":"2605.24799","version":1},"attestation_state":"computed","paper":{"title":"Divide-and-Conquer Inference for Large-Scale Visual Recognition with Multimodal Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Dawei Wang, Feng Jiang, Jiaqi Huang, Qian Qiao, Qiufeng Wang, Xihang Zhou, Yikang Duan, Zhipeng Ye","submitted_at":"2026-05-24T01:07:05Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) have demonstrated strong capabilities across a wide range of vision language tasks. However, when applied to large scale image classification, their performance degrades significantly as the label space expands a phenomenon we define as Performance Collapse in Long Sequence Recognition. Through an information theoretic analysis, we reveal that this collapse stems from a fundamental conflict between the escalating information entropy and the prominent attention dilution and decay within attention mechanisms, which impairs the model's ability to maintain "},"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":"2605.24799","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-24T01:07:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"77b62b15ccdc27ea364b9c98ea0739e9933f82186f5f26f9d3d516501f2f2f6a","abstract_canon_sha256":"bc07264b1ea0de1c64a19a3a272cff6dbc44372960065c0d8495d56ccbc15bfa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:03:58.956819Z","signature_b64":"QXiOBXpjB4/BcqCf5F1u339dC6qRXgbCfv1Zw2ZJz5w5eh9OvTDs2mm4lJY1D3XxGoaB98xs867Jl1zKAvZECQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c621c3869d9619e150a3d70079fb0ea72ff8618ba4950da63d2a34374e918ac0","last_reissued_at":"2026-05-26T01:03:58.956103Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:03:58.956103Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Divide-and-Conquer Inference for Large-Scale Visual Recognition with Multimodal Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Dawei Wang, Feng Jiang, Jiaqi Huang, Qian Qiao, Qiufeng Wang, Xihang Zhou, Yikang Duan, Zhipeng Ye","submitted_at":"2026-05-24T01:07:05Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) have demonstrated strong capabilities across a wide range of vision language tasks. However, when applied to large scale image classification, their performance degrades significantly as the label space expands a phenomenon we define as Performance Collapse in Long Sequence Recognition. Through an information theoretic analysis, we reveal that this collapse stems from a fundamental conflict between the escalating information entropy and the prominent attention dilution and decay within attention mechanisms, which impairs the model's ability to maintain "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24799","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/2605.24799/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":"2605.24799","created_at":"2026-05-26T01:03:58.956212+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.24799v1","created_at":"2026-05-26T01:03:58.956212+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24799","created_at":"2026-05-26T01:03:58.956212+00:00"},{"alias_kind":"pith_short_12","alias_value":"YYQ4HBU5SYM6","created_at":"2026-05-26T01:03:58.956212+00:00"},{"alias_kind":"pith_short_16","alias_value":"YYQ4HBU5SYM6CUFD","created_at":"2026-05-26T01:03:58.956212+00:00"},{"alias_kind":"pith_short_8","alias_value":"YYQ4HBU5","created_at":"2026-05-26T01:03:58.956212+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YYQ4HBU5SYM6CUFD24AHT6YOU4","json":"https://pith.science/pith/YYQ4HBU5SYM6CUFD24AHT6YOU4.json","graph_json":"https://pith.science/api/pith-number/YYQ4HBU5SYM6CUFD24AHT6YOU4/graph.json","events_json":"https://pith.science/api/pith-number/YYQ4HBU5SYM6CUFD24AHT6YOU4/events.json","paper":"https://pith.science/paper/YYQ4HBU5"},"agent_actions":{"view_html":"https://pith.science/pith/YYQ4HBU5SYM6CUFD24AHT6YOU4","download_json":"https://pith.science/pith/YYQ4HBU5SYM6CUFD24AHT6YOU4.json","view_paper":"https://pith.science/paper/YYQ4HBU5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.24799&json=true","fetch_graph":"https://pith.science/api/pith-number/YYQ4HBU5SYM6CUFD24AHT6YOU4/graph.json","fetch_events":"https://pith.science/api/pith-number/YYQ4HBU5SYM6CUFD24AHT6YOU4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YYQ4HBU5SYM6CUFD24AHT6YOU4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YYQ4HBU5SYM6CUFD24AHT6YOU4/action/storage_attestation","attest_author":"https://pith.science/pith/YYQ4HBU5SYM6CUFD24AHT6YOU4/action/author_attestation","sign_citation":"https://pith.science/pith/YYQ4HBU5SYM6CUFD24AHT6YOU4/action/citation_signature","submit_replication":"https://pith.science/pith/YYQ4HBU5SYM6CUFD24AHT6YOU4/action/replication_record"}},"created_at":"2026-05-26T01:03:58.956212+00:00","updated_at":"2026-05-26T01:03:58.956212+00:00"}