{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:IA43GNDK6FS6IGFSHLKZ6RUGDU","short_pith_number":"pith:IA43GNDK","canonical_record":{"source":{"id":"2506.17267","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T20:20:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b8da4dc3cf7d15336381c5a1ed4e81bc790b81bd066aafbb4d11e6a11370285c","abstract_canon_sha256":"4eb1d56e314e3486624b1d993f9f3bc8b4f91f7ee09e2615247f3d93bf8f6180"},"schema_version":"1.0"},"canonical_sha256":"4039b3346af165e418b23ad59f46861d0b88e15f27a3eff2579954b58236c920","source":{"kind":"arxiv","id":"2506.17267","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.17267","created_at":"2026-07-05T11:25:14Z"},{"alias_kind":"arxiv_version","alias_value":"2506.17267v1","created_at":"2026-07-05T11:25:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.17267","created_at":"2026-07-05T11:25:14Z"},{"alias_kind":"pith_short_12","alias_value":"IA43GNDK6FS6","created_at":"2026-07-05T11:25:14Z"},{"alias_kind":"pith_short_16","alias_value":"IA43GNDK6FS6IGFS","created_at":"2026-07-05T11:25:14Z"},{"alias_kind":"pith_short_8","alias_value":"IA43GNDK","created_at":"2026-07-05T11:25:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:IA43GNDK6FS6IGFSHLKZ6RUGDU","target":"record","payload":{"canonical_record":{"source":{"id":"2506.17267","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T20:20:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"b8da4dc3cf7d15336381c5a1ed4e81bc790b81bd066aafbb4d11e6a11370285c","abstract_canon_sha256":"4eb1d56e314e3486624b1d993f9f3bc8b4f91f7ee09e2615247f3d93bf8f6180"},"schema_version":"1.0"},"canonical_sha256":"4039b3346af165e418b23ad59f46861d0b88e15f27a3eff2579954b58236c920","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:25:14.535098Z","signature_b64":"n2B0WZ2SPtXyblNaW3WS2hzLAxZuKH4geC8ozK2MY4706lbD4YWY3pvjyEFkSQLxLdQ9hk/PQzGpLgIChZt1Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4039b3346af165e418b23ad59f46861d0b88e15f27a3eff2579954b58236c920","last_reissued_at":"2026-07-05T11:25:14.534638Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:25:14.534638Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.17267","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":"2026-07-05T11:25:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gTJm3Ao+ulQ1rVHQZZVwCyHBFUCS7E7yFDvZ2e3qGqs9g/Mx6ciDnCehkxbF5Lh00LwpOUFigetCknwDuy5IAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T14:48:36.553596Z"},"content_sha256":"f5efcc1873c592f0899bed9c8f1e1e90682eac83dc7226f443364e9119d36f68","schema_version":"1.0","event_id":"sha256:f5efcc1873c592f0899bed9c8f1e1e90682eac83dc7226f443364e9119d36f68"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:IA43GNDK6FS6IGFSHLKZ6RUGDU","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CF-VLM:CounterFactual Vision-Language Fine-tuning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jian Wang, Jusheng Zhang, Kaitong Cai, Keze Wang, Yijia Fan","submitted_at":"2025-06-10T20:20:05Z","abstract_excerpt":"Recent advances in vision-language models (VLMs) have greatly improved cross-modal semantic understanding, yet significant limitations remain in fine-grained discrimination and deep causal reasoning tasks. Existing VLMs often rely on superficial statistical correlations, lacking the ability to capture the underlying causal logic between visual and textual content. To address this, we propose CounterFactual Vision-Language Fine-tuning (CF-VLM), a novel framework that enhances the causal reasoning capabilities of VLMs through the targeted use of counterfactual samples. CF-VLM introduces three co"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.17267","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/2506.17267/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":"2026-07-05T11:25:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kAQtzfxZV6ISqw9glitfVVmJaQFg6PCrI9eoWKa4DkLuZ+7i1u/ZeRk3Ks9ppS702PyL+zh0LIeO83TkwbtOBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T14:48:36.554503Z"},"content_sha256":"af310869ca49e0cae625c9ed230deb62381b5b3714d2050c128ce3fe9e0d0e4b","schema_version":"1.0","event_id":"sha256:af310869ca49e0cae625c9ed230deb62381b5b3714d2050c128ce3fe9e0d0e4b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IA43GNDK6FS6IGFSHLKZ6RUGDU/bundle.json","state_url":"https://pith.science/pith/IA43GNDK6FS6IGFSHLKZ6RUGDU/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IA43GNDK6FS6IGFSHLKZ6RUGDU/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":"2026-08-05T14:48:36Z","links":{"resolver":"https://pith.science/pith/IA43GNDK6FS6IGFSHLKZ6RUGDU","bundle":"https://pith.science/pith/IA43GNDK6FS6IGFSHLKZ6RUGDU/bundle.json","state":"https://pith.science/pith/IA43GNDK6FS6IGFSHLKZ6RUGDU/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IA43GNDK6FS6IGFSHLKZ6RUGDU/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:IA43GNDK6FS6IGFSHLKZ6RUGDU","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":"4eb1d56e314e3486624b1d993f9f3bc8b4f91f7ee09e2615247f3d93bf8f6180","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T20:20:05Z","title_canon_sha256":"b8da4dc3cf7d15336381c5a1ed4e81bc790b81bd066aafbb4d11e6a11370285c"},"schema_version":"1.0","source":{"id":"2506.17267","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.17267","created_at":"2026-07-05T11:25:14Z"},{"alias_kind":"arxiv_version","alias_value":"2506.17267v1","created_at":"2026-07-05T11:25:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.17267","created_at":"2026-07-05T11:25:14Z"},{"alias_kind":"pith_short_12","alias_value":"IA43GNDK6FS6","created_at":"2026-07-05T11:25:14Z"},{"alias_kind":"pith_short_16","alias_value":"IA43GNDK6FS6IGFS","created_at":"2026-07-05T11:25:14Z"},{"alias_kind":"pith_short_8","alias_value":"IA43GNDK","created_at":"2026-07-05T11:25:14Z"}],"graph_snapshots":[{"event_id":"sha256:af310869ca49e0cae625c9ed230deb62381b5b3714d2050c128ce3fe9e0d0e4b","target":"graph","created_at":"2026-07-05T11:25:14Z","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/2506.17267/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advances in vision-language models (VLMs) have greatly improved cross-modal semantic understanding, yet significant limitations remain in fine-grained discrimination and deep causal reasoning tasks. Existing VLMs often rely on superficial statistical correlations, lacking the ability to capture the underlying causal logic between visual and textual content. To address this, we propose CounterFactual Vision-Language Fine-tuning (CF-VLM), a novel framework that enhances the causal reasoning capabilities of VLMs through the targeted use of counterfactual samples. CF-VLM introduces three co","authors_text":"Jian Wang, Jusheng Zhang, Kaitong Cai, Keze Wang, Yijia Fan","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T20:20:05Z","title":"CF-VLM:CounterFactual Vision-Language Fine-tuning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.17267","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:f5efcc1873c592f0899bed9c8f1e1e90682eac83dc7226f443364e9119d36f68","target":"record","created_at":"2026-07-05T11:25:14Z","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":"4eb1d56e314e3486624b1d993f9f3bc8b4f91f7ee09e2615247f3d93bf8f6180","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T20:20:05Z","title_canon_sha256":"b8da4dc3cf7d15336381c5a1ed4e81bc790b81bd066aafbb4d11e6a11370285c"},"schema_version":"1.0","source":{"id":"2506.17267","kind":"arxiv","version":1}},"canonical_sha256":"4039b3346af165e418b23ad59f46861d0b88e15f27a3eff2579954b58236c920","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4039b3346af165e418b23ad59f46861d0b88e15f27a3eff2579954b58236c920","first_computed_at":"2026-07-05T11:25:14.534638Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:25:14.534638Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"n2B0WZ2SPtXyblNaW3WS2hzLAxZuKH4geC8ozK2MY4706lbD4YWY3pvjyEFkSQLxLdQ9hk/PQzGpLgIChZt1Dw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:25:14.535098Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.17267","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f5efcc1873c592f0899bed9c8f1e1e90682eac83dc7226f443364e9119d36f68","sha256:af310869ca49e0cae625c9ed230deb62381b5b3714d2050c128ce3fe9e0d0e4b"],"state_sha256":"d9def3bf339ccc247330f1e1f98ee2fbc0cde49849037ea07d4c8f9bc488959e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ir/yPO8IwriQ1B+mBPZc0X9UaETkqBxKLYixqOkHdTbMA52yaxIW0NTi7wwgZvyFwzoSbbtIkm/2Jbl41DnCBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T14:48:36.561847Z","bundle_sha256":"d13e4e61935892f0b57ab5c8dead9d25d24fbd7e2bde1da3731f61dd31c4c42f"}}