{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:AM5CTMEE2V7MJWXVZ2SV2PUNSC","short_pith_number":"pith:AM5CTMEE","canonical_record":{"source":{"id":"2605.18740","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-18T17:57:04Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"da30378a87b9eaa6421123f1cd2520aeb4bbcee9e465da22493134b6c2050675","abstract_canon_sha256":"07aace2f706900f01f124afa67312008c99f09ec4633c35fd72b868be220b86a"},"schema_version":"1.0"},"canonical_sha256":"033a29b084d57ec4daf5cea55d3e8d90bc88bbfcbb1e94a8bbb82b16a7fd624e","source":{"kind":"arxiv","id":"2605.18740","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.18740","created_at":"2026-05-20T00:06:18Z"},{"alias_kind":"arxiv_version","alias_value":"2605.18740v1","created_at":"2026-05-20T00:06:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18740","created_at":"2026-05-20T00:06:18Z"},{"alias_kind":"pith_short_12","alias_value":"AM5CTMEE2V7M","created_at":"2026-05-20T00:06:18Z"},{"alias_kind":"pith_short_16","alias_value":"AM5CTMEE2V7MJWXV","created_at":"2026-05-20T00:06:18Z"},{"alias_kind":"pith_short_8","alias_value":"AM5CTMEE","created_at":"2026-05-20T00:06:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:AM5CTMEE2V7MJWXVZ2SV2PUNSC","target":"record","payload":{"canonical_record":{"source":{"id":"2605.18740","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-18T17:57:04Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"da30378a87b9eaa6421123f1cd2520aeb4bbcee9e465da22493134b6c2050675","abstract_canon_sha256":"07aace2f706900f01f124afa67312008c99f09ec4633c35fd72b868be220b86a"},"schema_version":"1.0"},"canonical_sha256":"033a29b084d57ec4daf5cea55d3e8d90bc88bbfcbb1e94a8bbb82b16a7fd624e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:06:18.504169Z","signature_b64":"alVsX29CGkBDG4bqLfxdNBjAj9rk7ytgkhv2GR/yToEMTM4IZWkCAxJ/65+MSqpWcFfuhsXJn2r5ozExO80aCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"033a29b084d57ec4daf5cea55d3e8d90bc88bbfcbb1e94a8bbb82b16a7fd624e","last_reissued_at":"2026-05-20T00:06:18.503363Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:06:18.503363Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.18740","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:06:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GYp71M8SDuUOJelMiLe89MoQzZp/wRGpqiOBu5Ej/mg3KCJhXf5yQX4/utHPAyOI9AprAZVn9fP5DZxhwAshDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-20T20:09:56.327770Z"},"content_sha256":"261bfe18b8e0b42f906b330d04ada4407c711f0b2854e2d787513a78f95308d2","schema_version":"1.0","event_id":"sha256:261bfe18b8e0b42f906b330d04ada4407c711f0b2854e2d787513a78f95308d2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:AM5CTMEE2V7MJWXVZ2SV2PUNSC","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Hongyu Lin, Jie Lou, Le Sun, Qianhao Yuan, Xianpei Han, Xing Yu, Yaojie Lu","submitted_at":"2026-05-18T17:57:04Z","abstract_excerpt":"Multimodal Large Language Models (MLLMs) still struggle with fine-grained visual understanding, where answers often depend on small but decisive evidence in the full image. We observe a regional-to-global perception gap: the same MLLM answers fine-grained questions more accurately when conditioned on evidence-centered crops than on the corresponding full images, suggesting that many failures stem from difficulty to focus on relevant evidence rather than insufficient local recognition ability. Motivated by this observation, we propose Vision-OPD (Vision On-Policy Distillation), a regional-to-gl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18740","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/2605.18740/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T00:01:59.000399Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"399a1904d677a4c7b6c67a58d132b5fa5a0254287c5310fc523ee66ec505a85d"},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-20T00:06:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eTw41Luu179IA+/PXekf4pr5SQUJUKhkFy0AiDlZdqgpfiF9fUJ6KlS8T9x4mEuT8/lMAFMkhoKz9xNXkk5nAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-20T20:09:56.328240Z"},"content_sha256":"09836680dc44f3402f50632faf69d69b2ddde760454985784644f6fac56d6bcd","schema_version":"1.0","event_id":"sha256:09836680dc44f3402f50632faf69d69b2ddde760454985784644f6fac56d6bcd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/AM5CTMEE2V7MJWXVZ2SV2PUNSC/bundle.json","state_url":"https://pith.science/pith/AM5CTMEE2V7MJWXVZ2SV2PUNSC/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/AM5CTMEE2V7MJWXVZ2SV2PUNSC/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-20T20:09:56Z","links":{"resolver":"https://pith.science/pith/AM5CTMEE2V7MJWXVZ2SV2PUNSC","bundle":"https://pith.science/pith/AM5CTMEE2V7MJWXVZ2SV2PUNSC/bundle.json","state":"https://pith.science/pith/AM5CTMEE2V7MJWXVZ2SV2PUNSC/state.json","well_known_bundle":"https://pith.science/.well-known/pith/AM5CTMEE2V7MJWXVZ2SV2PUNSC/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:AM5CTMEE2V7MJWXVZ2SV2PUNSC","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"07aace2f706900f01f124afa67312008c99f09ec4633c35fd72b868be220b86a","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-18T17:57:04Z","title_canon_sha256":"da30378a87b9eaa6421123f1cd2520aeb4bbcee9e465da22493134b6c2050675"},"schema_version":"1.0","source":{"id":"2605.18740","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.18740","created_at":"2026-05-20T00:06:18Z"},{"alias_kind":"arxiv_version","alias_value":"2605.18740v1","created_at":"2026-05-20T00:06:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.18740","created_at":"2026-05-20T00:06:18Z"},{"alias_kind":"pith_short_12","alias_value":"AM5CTMEE2V7M","created_at":"2026-05-20T00:06:18Z"},{"alias_kind":"pith_short_16","alias_value":"AM5CTMEE2V7MJWXV","created_at":"2026-05-20T00:06:18Z"},{"alias_kind":"pith_short_8","alias_value":"AM5CTMEE","created_at":"2026-05-20T00:06:18Z"}],"graph_snapshots":[{"event_id":"sha256:09836680dc44f3402f50632faf69d69b2ddde760454985784644f6fac56d6bcd","target":"graph","created_at":"2026-05-20T00:06:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-20T00:01:59.000399Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.18740/integrity.json","findings":[],"snapshot_sha256":"399a1904d677a4c7b6c67a58d132b5fa5a0254287c5310fc523ee66ec505a85d","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multimodal Large Language Models (MLLMs) still struggle with fine-grained visual understanding, where answers often depend on small but decisive evidence in the full image. We observe a regional-to-global perception gap: the same MLLM answers fine-grained questions more accurately when conditioned on evidence-centered crops than on the corresponding full images, suggesting that many failures stem from difficulty to focus on relevant evidence rather than insufficient local recognition ability. Motivated by this observation, we propose Vision-OPD (Vision On-Policy Distillation), a regional-to-gl","authors_text":"Hongyu Lin, Jie Lou, Le Sun, Qianhao Yuan, Xianpei Han, Xing Yu, Yaojie Lu","cross_cats":["cs.AI","cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-18T17:57:04Z","title":"Vision-OPD: Learning to See Fine Details for Multimodal LLMs via On-Policy Self-Distillation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.18740","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:261bfe18b8e0b42f906b330d04ada4407c711f0b2854e2d787513a78f95308d2","target":"record","created_at":"2026-05-20T00:06:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"07aace2f706900f01f124afa67312008c99f09ec4633c35fd72b868be220b86a","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-18T17:57:04Z","title_canon_sha256":"da30378a87b9eaa6421123f1cd2520aeb4bbcee9e465da22493134b6c2050675"},"schema_version":"1.0","source":{"id":"2605.18740","kind":"arxiv","version":1}},"canonical_sha256":"033a29b084d57ec4daf5cea55d3e8d90bc88bbfcbb1e94a8bbb82b16a7fd624e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"033a29b084d57ec4daf5cea55d3e8d90bc88bbfcbb1e94a8bbb82b16a7fd624e","first_computed_at":"2026-05-20T00:06:18.503363Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-20T00:06:18.503363Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"alVsX29CGkBDG4bqLfxdNBjAj9rk7ytgkhv2GR/yToEMTM4IZWkCAxJ/65+MSqpWcFfuhsXJn2r5ozExO80aCw==","signature_status":"signed_v1","signed_at":"2026-05-20T00:06:18.504169Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.18740","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:261bfe18b8e0b42f906b330d04ada4407c711f0b2854e2d787513a78f95308d2","sha256:09836680dc44f3402f50632faf69d69b2ddde760454985784644f6fac56d6bcd"],"state_sha256":"f8076beaa029c696c0cd860c9bf50f5696c637b5d3074bb93abd86712a0d6139"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nchdFWBIhFMWxTD+/OlMyUfWPfLp3Au326ODjXLlDnNBs4r4kZS9Fl6V6VtS6xr/BkQFtgk86T20GgIioOssCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-20T20:09:56.332294Z","bundle_sha256":"29c0bb2c4967b8ccd21d0487a4d95fc99150fae43fc64ec77d1e1d7bc633b96a"}}