{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:ZRG2YGSCE5KP7DX2SXVLYSOT6I","short_pith_number":"pith:ZRG2YGSC","canonical_record":{"source":{"id":"2107.07651","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-07-16T00:19:22Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"73abf5702cfe6296da282ec7214ae3e30c5c3776033e810241f25cd6df7904d4","abstract_canon_sha256":"c4075835e291128ae9a4cb2cc6b6f37f097d3fdeb3109ec4df54f990a43f2770"},"schema_version":"1.0"},"canonical_sha256":"cc4dac1a422754ff8efa95eabc49d3f2197d9c0ef1e4eabd4c4d5a3f0b9a7064","source":{"kind":"arxiv","id":"2107.07651","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2107.07651","created_at":"2026-07-05T03:20:36Z"},{"alias_kind":"arxiv_version","alias_value":"2107.07651v2","created_at":"2026-07-05T03:20:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.07651","created_at":"2026-07-05T03:20:36Z"},{"alias_kind":"pith_short_12","alias_value":"ZRG2YGSCE5KP","created_at":"2026-07-05T03:20:36Z"},{"alias_kind":"pith_short_16","alias_value":"ZRG2YGSCE5KP7DX2","created_at":"2026-07-05T03:20:36Z"},{"alias_kind":"pith_short_8","alias_value":"ZRG2YGSC","created_at":"2026-07-05T03:20:36Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:ZRG2YGSCE5KP7DX2SXVLYSOT6I","target":"record","payload":{"canonical_record":{"source":{"id":"2107.07651","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-07-16T00:19:22Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"73abf5702cfe6296da282ec7214ae3e30c5c3776033e810241f25cd6df7904d4","abstract_canon_sha256":"c4075835e291128ae9a4cb2cc6b6f37f097d3fdeb3109ec4df54f990a43f2770"},"schema_version":"1.0"},"canonical_sha256":"cc4dac1a422754ff8efa95eabc49d3f2197d9c0ef1e4eabd4c4d5a3f0b9a7064","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:20:36.971253Z","signature_b64":"WdSvHHKHQ1X63SVAlPfUlZtSQUPPOji+HNQQg2mBBMK8Ll3VN+ceku88j6EjEwvdRQcMIoxBvrD3gLZZBBwyAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cc4dac1a422754ff8efa95eabc49d3f2197d9c0ef1e4eabd4c4d5a3f0b9a7064","last_reissued_at":"2026-07-05T03:20:36.970700Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:20:36.970700Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2107.07651","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T03:20:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2SKN6MZeAqQGBFn1s4tYrofrksjZdEiUMJrOl8k1brQShDSaTPODsV/mNRpGysCpNxZNWbuKHTIpnqTUv/O8Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T12:24:18.381147Z"},"content_sha256":"e3fd10d2c1c34cb9ccf253efcd5f1df426e796bcea5fabe52106045f28307598","schema_version":"1.0","event_id":"sha256:e3fd10d2c1c34cb9ccf253efcd5f1df426e796bcea5fabe52106045f28307598"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:ZRG2YGSCE5KP7DX2SXVLYSOT6I","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Align before Fuse: Vision and Language Representation Learning with Momentum Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Akhilesh Deepak Gotmare, Caiming Xiong, Junnan Li, Ramprasaath R. Selvaraju, Shafiq Joty, Steven Hoi","submitted_at":"2021-07-16T00:19:22Z","abstract_excerpt":"Large-scale vision and language representation learning has shown promising improvements on various vision-language tasks. Most existing methods employ a transformer-based multimodal encoder to jointly model visual tokens (region-based image features) and word tokens. Because the visual tokens and word tokens are unaligned, it is challenging for the multimodal encoder to learn image-text interactions. In this paper, we introduce a contrastive loss to ALign the image and text representations BEfore Fusing (ALBEF) them through cross-modal attention, which enables more grounded vision and languag"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.07651","kind":"arxiv","version":2},"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/2107.07651/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":"2026-07-05T03:20:36Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JvbghfXDzmnTO64JBvR2vMkS5gMu/vdY7jzLQ/hxH8lR27OGsY5nc289nSkduXup0NLAWKUj5iIOZ/xZyRTrBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T12:24:18.382751Z"},"content_sha256":"38957dad0d5c77bdaccc2fbd1f43a20360b681ab3192382e21e29da60a2b55c0","schema_version":"1.0","event_id":"sha256:38957dad0d5c77bdaccc2fbd1f43a20360b681ab3192382e21e29da60a2b55c0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ZRG2YGSCE5KP7DX2SXVLYSOT6I/bundle.json","state_url":"https://pith.science/pith/ZRG2YGSCE5KP7DX2SXVLYSOT6I/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ZRG2YGSCE5KP7DX2SXVLYSOT6I/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":"2026-08-17T12:24:18Z","links":{"resolver":"https://pith.science/pith/ZRG2YGSCE5KP7DX2SXVLYSOT6I","bundle":"https://pith.science/pith/ZRG2YGSCE5KP7DX2SXVLYSOT6I/bundle.json","state":"https://pith.science/pith/ZRG2YGSCE5KP7DX2SXVLYSOT6I/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ZRG2YGSCE5KP7DX2SXVLYSOT6I/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:ZRG2YGSCE5KP7DX2SXVLYSOT6I","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":"c4075835e291128ae9a4cb2cc6b6f37f097d3fdeb3109ec4df54f990a43f2770","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-07-16T00:19:22Z","title_canon_sha256":"73abf5702cfe6296da282ec7214ae3e30c5c3776033e810241f25cd6df7904d4"},"schema_version":"1.0","source":{"id":"2107.07651","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2107.07651","created_at":"2026-07-05T03:20:36Z"},{"alias_kind":"arxiv_version","alias_value":"2107.07651v2","created_at":"2026-07-05T03:20:36Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.07651","created_at":"2026-07-05T03:20:36Z"},{"alias_kind":"pith_short_12","alias_value":"ZRG2YGSCE5KP","created_at":"2026-07-05T03:20:36Z"},{"alias_kind":"pith_short_16","alias_value":"ZRG2YGSCE5KP7DX2","created_at":"2026-07-05T03:20:36Z"},{"alias_kind":"pith_short_8","alias_value":"ZRG2YGSC","created_at":"2026-07-05T03:20:36Z"}],"graph_snapshots":[{"event_id":"sha256:38957dad0d5c77bdaccc2fbd1f43a20360b681ab3192382e21e29da60a2b55c0","target":"graph","created_at":"2026-07-05T03:20:36Z","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/2107.07651/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large-scale vision and language representation learning has shown promising improvements on various vision-language tasks. Most existing methods employ a transformer-based multimodal encoder to jointly model visual tokens (region-based image features) and word tokens. Because the visual tokens and word tokens are unaligned, it is challenging for the multimodal encoder to learn image-text interactions. In this paper, we introduce a contrastive loss to ALign the image and text representations BEfore Fusing (ALBEF) them through cross-modal attention, which enables more grounded vision and languag","authors_text":"Akhilesh Deepak Gotmare, Caiming Xiong, Junnan Li, Ramprasaath R. Selvaraju, Shafiq Joty, Steven Hoi","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-07-16T00:19:22Z","title":"Align before Fuse: Vision and Language Representation Learning with Momentum Distillation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.07651","kind":"arxiv","version":2},"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:e3fd10d2c1c34cb9ccf253efcd5f1df426e796bcea5fabe52106045f28307598","target":"record","created_at":"2026-07-05T03:20:36Z","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":"c4075835e291128ae9a4cb2cc6b6f37f097d3fdeb3109ec4df54f990a43f2770","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-07-16T00:19:22Z","title_canon_sha256":"73abf5702cfe6296da282ec7214ae3e30c5c3776033e810241f25cd6df7904d4"},"schema_version":"1.0","source":{"id":"2107.07651","kind":"arxiv","version":2}},"canonical_sha256":"cc4dac1a422754ff8efa95eabc49d3f2197d9c0ef1e4eabd4c4d5a3f0b9a7064","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"cc4dac1a422754ff8efa95eabc49d3f2197d9c0ef1e4eabd4c4d5a3f0b9a7064","first_computed_at":"2026-07-05T03:20:36.970700Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:20:36.970700Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"WdSvHHKHQ1X63SVAlPfUlZtSQUPPOji+HNQQg2mBBMK8Ll3VN+ceku88j6EjEwvdRQcMIoxBvrD3gLZZBBwyAg==","signature_status":"signed_v1","signed_at":"2026-07-05T03:20:36.971253Z","signed_message":"canonical_sha256_bytes"},"source_id":"2107.07651","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e3fd10d2c1c34cb9ccf253efcd5f1df426e796bcea5fabe52106045f28307598","sha256:38957dad0d5c77bdaccc2fbd1f43a20360b681ab3192382e21e29da60a2b55c0"],"state_sha256":"d1cb738386f4c3be82d8a43614e5ea303fe3a70d060130da0321daa720286672"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5CQ+0n6aVBCEve2rBtKqtO4HeUdcwwS/yaQpcqoX9jrDKFm/Qa/Zeq6NshwnSAPeRbMyM/pQKZ7QHppTwb8ADw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T12:24:18.400125Z","bundle_sha256":"e8b2bea1440a91f380eacda06bf47c932f84fab0b95b410ee035e2a0401633a1"}}