{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:I4BF7DSGGRCUZDE2GJKTUA3IX4","short_pith_number":"pith:I4BF7DSG","schema_version":"1.0","canonical_sha256":"47025f8e4634454c8c9a32553a0368bf17d2ea13615ff7b104d9a8d0b712d0ae","source":{"kind":"arxiv","id":"2505.21465","version":1},"attestation_state":"computed","paper":{"title":"ID-Align: RoPE-Conscious Position Remapping for Dynamic High-Resolution Adaptation in Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Bozhou Li, Wentao Zhang","submitted_at":"2025-05-27T17:36:23Z","abstract_excerpt":"Currently, a prevalent approach for enhancing Vision-Language Models (VLMs) performance is to encode both the high-resolution version and the thumbnail of an image simultaneously. While effective, this method generates a large number of image tokens. When combined with the widely used Rotary Position Embedding (RoPE), its long-term decay property hinders the interaction between high-resolution tokens and thumbnail tokens, as well as between text and image. To address these issues, we propose ID-Align, which alleviates these problems by reordering position IDs. In this method, high-resolution t"},"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":"2505.21465","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-05-27T17:36:23Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"1c635c82c852bda35c29d0de8aeb8405766de849ce8df912609bdab5764a0734","abstract_canon_sha256":"7e2c0464e73e4849f905a874a00d8e25777503e69b74114b3ddb9b82440a967e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:10:42.425954Z","signature_b64":"rqu2j9WWbFIzaSV23HVlIQUceRsDOzB2I+IeUN1IW9NnlW+bQk/hzZG4tfBRLYMGQzddWIHHqsOPgY48AbZsCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"47025f8e4634454c8c9a32553a0368bf17d2ea13615ff7b104d9a8d0b712d0ae","last_reissued_at":"2026-07-05T11:10:42.425471Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:10:42.425471Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ID-Align: RoPE-Conscious Position Remapping for Dynamic High-Resolution Adaptation in Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Bozhou Li, Wentao Zhang","submitted_at":"2025-05-27T17:36:23Z","abstract_excerpt":"Currently, a prevalent approach for enhancing Vision-Language Models (VLMs) performance is to encode both the high-resolution version and the thumbnail of an image simultaneously. While effective, this method generates a large number of image tokens. When combined with the widely used Rotary Position Embedding (RoPE), its long-term decay property hinders the interaction between high-resolution tokens and thumbnail tokens, as well as between text and image. To address these issues, we propose ID-Align, which alleviates these problems by reordering position IDs. In this method, high-resolution t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21465","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/2505.21465/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":"2505.21465","created_at":"2026-07-05T11:10:42.425531+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.21465v1","created_at":"2026-07-05T11:10:42.425531+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21465","created_at":"2026-07-05T11:10:42.425531+00:00"},{"alias_kind":"pith_short_12","alias_value":"I4BF7DSGGRCU","created_at":"2026-07-05T11:10:42.425531+00:00"},{"alias_kind":"pith_short_16","alias_value":"I4BF7DSGGRCUZDE2","created_at":"2026-07-05T11:10:42.425531+00:00"},{"alias_kind":"pith_short_8","alias_value":"I4BF7DSG","created_at":"2026-07-05T11:10:42.425531+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/I4BF7DSGGRCUZDE2GJKTUA3IX4","json":"https://pith.science/pith/I4BF7DSGGRCUZDE2GJKTUA3IX4.json","graph_json":"https://pith.science/api/pith-number/I4BF7DSGGRCUZDE2GJKTUA3IX4/graph.json","events_json":"https://pith.science/api/pith-number/I4BF7DSGGRCUZDE2GJKTUA3IX4/events.json","paper":"https://pith.science/paper/I4BF7DSG"},"agent_actions":{"view_html":"https://pith.science/pith/I4BF7DSGGRCUZDE2GJKTUA3IX4","download_json":"https://pith.science/pith/I4BF7DSGGRCUZDE2GJKTUA3IX4.json","view_paper":"https://pith.science/paper/I4BF7DSG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.21465&json=true","fetch_graph":"https://pith.science/api/pith-number/I4BF7DSGGRCUZDE2GJKTUA3IX4/graph.json","fetch_events":"https://pith.science/api/pith-number/I4BF7DSGGRCUZDE2GJKTUA3IX4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I4BF7DSGGRCUZDE2GJKTUA3IX4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I4BF7DSGGRCUZDE2GJKTUA3IX4/action/storage_attestation","attest_author":"https://pith.science/pith/I4BF7DSGGRCUZDE2GJKTUA3IX4/action/author_attestation","sign_citation":"https://pith.science/pith/I4BF7DSGGRCUZDE2GJKTUA3IX4/action/citation_signature","submit_replication":"https://pith.science/pith/I4BF7DSGGRCUZDE2GJKTUA3IX4/action/replication_record"}},"created_at":"2026-07-05T11:10:42.425531+00:00","updated_at":"2026-07-05T11:10:42.425531+00:00"}