{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:TSMGA4GKVTXGEXP5T5ZPH623IQ","short_pith_number":"pith:TSMGA4GK","canonical_record":{"source":{"id":"2211.00262","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-01T04:07:11Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"7a8ec6f07e153bb005ed2c5d194aebd6851d6d4d982b3954fe31cad799747170","abstract_canon_sha256":"4809786c16946619b505d0420bd2ba91e079041eadb760bcf96746930093abc7"},"schema_version":"1.0"},"canonical_sha256":"9c986070caacee625dfd9f72f3fb5b440148e7300e11d0cd67aa4790d9fe0719","source":{"kind":"arxiv","id":"2211.00262","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2211.00262","created_at":"2026-07-05T05:12:20Z"},{"alias_kind":"arxiv_version","alias_value":"2211.00262v1","created_at":"2026-07-05T05:12:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.00262","created_at":"2026-07-05T05:12:20Z"},{"alias_kind":"pith_short_12","alias_value":"TSMGA4GKVTXG","created_at":"2026-07-05T05:12:20Z"},{"alias_kind":"pith_short_16","alias_value":"TSMGA4GKVTXGEXP5","created_at":"2026-07-05T05:12:20Z"},{"alias_kind":"pith_short_8","alias_value":"TSMGA4GK","created_at":"2026-07-05T05:12:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:TSMGA4GKVTXGEXP5T5ZPH623IQ","target":"record","payload":{"canonical_record":{"source":{"id":"2211.00262","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-01T04:07:11Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"7a8ec6f07e153bb005ed2c5d194aebd6851d6d4d982b3954fe31cad799747170","abstract_canon_sha256":"4809786c16946619b505d0420bd2ba91e079041eadb760bcf96746930093abc7"},"schema_version":"1.0"},"canonical_sha256":"9c986070caacee625dfd9f72f3fb5b440148e7300e11d0cd67aa4790d9fe0719","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:12:20.617122Z","signature_b64":"WDeFikUPytjkXTZzJLPCPBNr+YvqtM8BXQQG8zmgnvEIpBmAoVe4djUBE8HnxquP8AbiDxFfAIt3OjY1IAQNAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9c986070caacee625dfd9f72f3fb5b440148e7300e11d0cd67aa4790d9fe0719","last_reissued_at":"2026-07-05T05:12:20.616695Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:12:20.616695Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2211.00262","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":"2026-07-05T05:12:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"PoicsU0uYe/MxkOmWkDCzVEdd6eW2cn3p66tdRXVn6j7xBuw4adEZD+4S+xeqf4EHI8Xa9+neGskY+6TEgGNBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T23:24:20.378932Z"},"content_sha256":"ff99992aa4df03f3c7d33960942725170d36467531e8d54c3ab54f0804e983ce","schema_version":"1.0","event_id":"sha256:ff99992aa4df03f3c7d33960942725170d36467531e8d54c3ab54f0804e983ce"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:TSMGA4GKVTXGEXP5T5ZPH623IQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Training Vision-Language Models with Less Bimodal Supervision","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Ben Bogin, Elad Segal, Jonathan Berant","submitted_at":"2022-11-01T04:07:11Z","abstract_excerpt":"Standard practice in pretraining multimodal models, such as vision-language models, is to rely on pairs of aligned inputs from both modalities, for example, aligned image-text pairs. However, such pairs can be difficult to obtain in low-resource settings and for some modality pairs (e.g., structured tables and images). In this work, we investigate the extent to which we can reduce the reliance on such parallel data, which we term \\emph{bimodal supervision}, and use models that are pretrained on each modality independently. We experiment with a high-performing vision-language model, and analyze"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.00262","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/2211.00262/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":"2026-07-05T05:12:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"t9B8FN+1QFEJ5q3IZlnOt0ma/EBA2/FPDNMO0iwd2FBgfdQgFpTc2YwNeSSouorCB7qKgYrQPkPe9P7uiMbzAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T23:24:20.379309Z"},"content_sha256":"97218d3634728f26a29ce735166e11541a974c73a5a9fec9dd1b30a759206328","schema_version":"1.0","event_id":"sha256:97218d3634728f26a29ce735166e11541a974c73a5a9fec9dd1b30a759206328"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TSMGA4GKVTXGEXP5T5ZPH623IQ/bundle.json","state_url":"https://pith.science/pith/TSMGA4GKVTXGEXP5T5ZPH623IQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TSMGA4GKVTXGEXP5T5ZPH623IQ/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":"2026-07-06T23:24:20Z","links":{"resolver":"https://pith.science/pith/TSMGA4GKVTXGEXP5T5ZPH623IQ","bundle":"https://pith.science/pith/TSMGA4GKVTXGEXP5T5ZPH623IQ/bundle.json","state":"https://pith.science/pith/TSMGA4GKVTXGEXP5T5ZPH623IQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TSMGA4GKVTXGEXP5T5ZPH623IQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:TSMGA4GKVTXGEXP5T5ZPH623IQ","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":"4809786c16946619b505d0420bd2ba91e079041eadb760bcf96746930093abc7","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-01T04:07:11Z","title_canon_sha256":"7a8ec6f07e153bb005ed2c5d194aebd6851d6d4d982b3954fe31cad799747170"},"schema_version":"1.0","source":{"id":"2211.00262","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2211.00262","created_at":"2026-07-05T05:12:20Z"},{"alias_kind":"arxiv_version","alias_value":"2211.00262v1","created_at":"2026-07-05T05:12:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2211.00262","created_at":"2026-07-05T05:12:20Z"},{"alias_kind":"pith_short_12","alias_value":"TSMGA4GKVTXG","created_at":"2026-07-05T05:12:20Z"},{"alias_kind":"pith_short_16","alias_value":"TSMGA4GKVTXGEXP5","created_at":"2026-07-05T05:12:20Z"},{"alias_kind":"pith_short_8","alias_value":"TSMGA4GK","created_at":"2026-07-05T05:12:20Z"}],"graph_snapshots":[{"event_id":"sha256:97218d3634728f26a29ce735166e11541a974c73a5a9fec9dd1b30a759206328","target":"graph","created_at":"2026-07-05T05:12:20Z","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/2211.00262/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Standard practice in pretraining multimodal models, such as vision-language models, is to rely on pairs of aligned inputs from both modalities, for example, aligned image-text pairs. However, such pairs can be difficult to obtain in low-resource settings and for some modality pairs (e.g., structured tables and images). In this work, we investigate the extent to which we can reduce the reliance on such parallel data, which we term \\emph{bimodal supervision}, and use models that are pretrained on each modality independently. We experiment with a high-performing vision-language model, and analyze","authors_text":"Ben Bogin, Elad Segal, Jonathan Berant","cross_cats":["cs.CV"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-01T04:07:11Z","title":"Training Vision-Language Models with Less Bimodal Supervision"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2211.00262","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:ff99992aa4df03f3c7d33960942725170d36467531e8d54c3ab54f0804e983ce","target":"record","created_at":"2026-07-05T05:12:20Z","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":"4809786c16946619b505d0420bd2ba91e079041eadb760bcf96746930093abc7","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-11-01T04:07:11Z","title_canon_sha256":"7a8ec6f07e153bb005ed2c5d194aebd6851d6d4d982b3954fe31cad799747170"},"schema_version":"1.0","source":{"id":"2211.00262","kind":"arxiv","version":1}},"canonical_sha256":"9c986070caacee625dfd9f72f3fb5b440148e7300e11d0cd67aa4790d9fe0719","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9c986070caacee625dfd9f72f3fb5b440148e7300e11d0cd67aa4790d9fe0719","first_computed_at":"2026-07-05T05:12:20.616695Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:12:20.616695Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"WDeFikUPytjkXTZzJLPCPBNr+YvqtM8BXQQG8zmgnvEIpBmAoVe4djUBE8HnxquP8AbiDxFfAIt3OjY1IAQNAA==","signature_status":"signed_v1","signed_at":"2026-07-05T05:12:20.617122Z","signed_message":"canonical_sha256_bytes"},"source_id":"2211.00262","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ff99992aa4df03f3c7d33960942725170d36467531e8d54c3ab54f0804e983ce","sha256:97218d3634728f26a29ce735166e11541a974c73a5a9fec9dd1b30a759206328"],"state_sha256":"ac45c9234f6ed1f16d744512e8ecf5817c1fc827a4bbd10f2b49a3f27c8294b5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hIoDANi1Tp7SDsxVQvtSgjbQwsT3Nby+E77x54R6FSqf+u/ZuhmXzOeFUqLKKPZPrkUWJQp0HuVDEyw4lGTyDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T23:24:20.381424Z","bundle_sha256":"739fc0442442093d442241fff2da0a3644a262c7ea012b5254c5ab7d0e558090"}}