{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:PVM4J7MLWXZLZ7PM5FOR3LAQSD","short_pith_number":"pith:PVM4J7ML","canonical_record":{"source":{"id":"2501.05122","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-09T10:26:14Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"71b18e38f195627809a2029ad0a1381875bec8c2525be01fac8304956de66aef","abstract_canon_sha256":"0ce010e385fe97f7aa95535bcad0b96fbed59ea6c9bfdf1dc33fa0fd8316c74e"},"schema_version":"1.0"},"canonical_sha256":"7d59c4fd8bb5f2bcfdece95d1dac1090e301aa3e5a06fb9eb0d188a9d9899e69","source":{"kind":"arxiv","id":"2501.05122","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.05122","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"arxiv_version","alias_value":"2501.05122v1","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.05122","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"pith_short_12","alias_value":"PVM4J7MLWXZL","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"pith_short_16","alias_value":"PVM4J7MLWXZLZ7PM","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"pith_short_8","alias_value":"PVM4J7ML","created_at":"2026-07-05T09:59:04Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:PVM4J7MLWXZLZ7PM5FOR3LAQSD","target":"record","payload":{"canonical_record":{"source":{"id":"2501.05122","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-09T10:26:14Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"71b18e38f195627809a2029ad0a1381875bec8c2525be01fac8304956de66aef","abstract_canon_sha256":"0ce010e385fe97f7aa95535bcad0b96fbed59ea6c9bfdf1dc33fa0fd8316c74e"},"schema_version":"1.0"},"canonical_sha256":"7d59c4fd8bb5f2bcfdece95d1dac1090e301aa3e5a06fb9eb0d188a9d9899e69","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:59:04.746806Z","signature_b64":"D4VssX/Q7FIBR9wFwRkodPS3bTwo/cxD3Fu71jb8lp9IBCQLK2HqINIJmD0NIWPRexqPuBzJIpN/P44PXDX4Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7d59c4fd8bb5f2bcfdece95d1dac1090e301aa3e5a06fb9eb0d188a9d9899e69","last_reissued_at":"2026-07-05T09:59:04.746392Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:59:04.746392Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2501.05122","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":"2026-07-05T09:59:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Py5en34uuZqCeYYuQYpBbfEndQx96dxANlfZT8b2GseNhNd8y0IzwCjkW1OtiSvb8p0Z79QWhBaDDuKXd+X7Ag==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T18:42:56.675955Z"},"content_sha256":"6d0799c01ed9d8acbce46984810d22e3a5aa56b32cd896d462c752c7d9e1b74a","schema_version":"1.0","event_id":"sha256:6d0799c01ed9d8acbce46984810d22e3a5aa56b32cd896d462c752c7d9e1b74a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:PVM4J7MLWXZLZ7PM5FOR3LAQSD","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Centurio: On Drivers of Multilingual Ability of Large Vision-Language Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Anne Lauscher, Carolin Holtermann, Chris Biemann, Florian Schneider, Goran Glava\\v{s}, Gregor Geigle, Radu Timofte","submitted_at":"2025-01-09T10:26:14Z","abstract_excerpt":"Most Large Vision-Language Models (LVLMs) to date are trained predominantly on English data, which makes them struggle to understand non-English input and fail to generate output in the desired target language. Existing efforts mitigate these issues by adding multilingual training data, but do so in a largely ad-hoc manner, lacking insight into how different training mixes tip the scale for different groups of languages. In this work, we present a comprehensive investigation into the training strategies for massively multilingual LVLMs. First, we conduct a series of multi-stage experiments spa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.05122","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/2501.05122/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:59:04Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GboEPfCL23keGhmK3wnKrTHQdpUHuIzQfEZzp/hwOURAwuVtGrfhJrYlnEvMiX3Nh8hNrxL70PmG7MHLqFvhCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T18:42:56.676826Z"},"content_sha256":"01d4db7e503d2e23ee93cd49a3b466923c1d782f6ffa7bb56734b13bf8997df8","schema_version":"1.0","event_id":"sha256:01d4db7e503d2e23ee93cd49a3b466923c1d782f6ffa7bb56734b13bf8997df8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/PVM4J7MLWXZLZ7PM5FOR3LAQSD/bundle.json","state_url":"https://pith.science/pith/PVM4J7MLWXZLZ7PM5FOR3LAQSD/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/PVM4J7MLWXZLZ7PM5FOR3LAQSD/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":"2026-08-12T18:42:56Z","links":{"resolver":"https://pith.science/pith/PVM4J7MLWXZLZ7PM5FOR3LAQSD","bundle":"https://pith.science/pith/PVM4J7MLWXZLZ7PM5FOR3LAQSD/bundle.json","state":"https://pith.science/pith/PVM4J7MLWXZLZ7PM5FOR3LAQSD/state.json","well_known_bundle":"https://pith.science/.well-known/pith/PVM4J7MLWXZLZ7PM5FOR3LAQSD/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:PVM4J7MLWXZLZ7PM5FOR3LAQSD","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":"0ce010e385fe97f7aa95535bcad0b96fbed59ea6c9bfdf1dc33fa0fd8316c74e","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-09T10:26:14Z","title_canon_sha256":"71b18e38f195627809a2029ad0a1381875bec8c2525be01fac8304956de66aef"},"schema_version":"1.0","source":{"id":"2501.05122","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.05122","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"arxiv_version","alias_value":"2501.05122v1","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.05122","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"pith_short_12","alias_value":"PVM4J7MLWXZL","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"pith_short_16","alias_value":"PVM4J7MLWXZLZ7PM","created_at":"2026-07-05T09:59:04Z"},{"alias_kind":"pith_short_8","alias_value":"PVM4J7ML","created_at":"2026-07-05T09:59:04Z"}],"graph_snapshots":[{"event_id":"sha256:01d4db7e503d2e23ee93cd49a3b466923c1d782f6ffa7bb56734b13bf8997df8","target":"graph","created_at":"2026-07-05T09:59:04Z","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":[],"endpoint":"/pith/2501.05122/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Most Large Vision-Language Models (LVLMs) to date are trained predominantly on English data, which makes them struggle to understand non-English input and fail to generate output in the desired target language. Existing efforts mitigate these issues by adding multilingual training data, but do so in a largely ad-hoc manner, lacking insight into how different training mixes tip the scale for different groups of languages. In this work, we present a comprehensive investigation into the training strategies for massively multilingual LVLMs. First, we conduct a series of multi-stage experiments spa","authors_text":"Anne Lauscher, Carolin Holtermann, Chris Biemann, Florian Schneider, Goran Glava\\v{s}, Gregor Geigle, Radu Timofte","cross_cats":["cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-09T10:26:14Z","title":"Centurio: On Drivers of Multilingual Ability of Large Vision-Language Model"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.05122","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":"sha256:6d0799c01ed9d8acbce46984810d22e3a5aa56b32cd896d462c752c7d9e1b74a","target":"record","created_at":"2026-07-05T09:59:04Z","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":"0ce010e385fe97f7aa95535bcad0b96fbed59ea6c9bfdf1dc33fa0fd8316c74e","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-01-09T10:26:14Z","title_canon_sha256":"71b18e38f195627809a2029ad0a1381875bec8c2525be01fac8304956de66aef"},"schema_version":"1.0","source":{"id":"2501.05122","kind":"arxiv","version":1}},"canonical_sha256":"7d59c4fd8bb5f2bcfdece95d1dac1090e301aa3e5a06fb9eb0d188a9d9899e69","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"7d59c4fd8bb5f2bcfdece95d1dac1090e301aa3e5a06fb9eb0d188a9d9899e69","first_computed_at":"2026-07-05T09:59:04.746392Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:59:04.746392Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"D4VssX/Q7FIBR9wFwRkodPS3bTwo/cxD3Fu71jb8lp9IBCQLK2HqINIJmD0NIWPRexqPuBzJIpN/P44PXDX4Aw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:59:04.746806Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.05122","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6d0799c01ed9d8acbce46984810d22e3a5aa56b32cd896d462c752c7d9e1b74a","sha256:01d4db7e503d2e23ee93cd49a3b466923c1d782f6ffa7bb56734b13bf8997df8"],"state_sha256":"ffabda3437212258dda40cbac52d0ab5151cf8b19fc9a17f24d4ca8f1d21eaa5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"a4lPMC0pE+cWaI6p4ZlfAnxD7DvY/lRiMuQCus3qZLzNegkCGR0Biw2m1zRoVxnTg1KHkuQ0VNwkS2KFrf/fDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-12T18:42:56.681309Z","bundle_sha256":"21126984960eccd2dffc428d2f86c2779aa7df4300245152362ce2edda17a4ea"}}