{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:TW7PL5QL4KQROQNYBLQMXVQFEN","short_pith_number":"pith:TW7PL5QL","canonical_record":{"source":{"id":"2109.08186","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2021-09-16T18:45:45Z","cross_cats_sorted":["cs.CL","cs.IR"],"title_canon_sha256":"2a6e3aec440201542dc7eb89a5968f468fd8eed148b5fd919387f64eebc8052e","abstract_canon_sha256":"1bcf3519289db239558982e5f28472664109bfc509fe72696a241e7193bd7a44"},"schema_version":"1.0"},"canonical_sha256":"9dbef5f60be2a11741b80ae0cbd60523461763fbf30a5044149fc3934abea5db","source":{"kind":"arxiv","id":"2109.08186","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2109.08186","created_at":"2026-07-05T04:01:31Z"},{"alias_kind":"arxiv_version","alias_value":"2109.08186v4","created_at":"2026-07-05T04:01:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.08186","created_at":"2026-07-05T04:01:31Z"},{"alias_kind":"pith_short_12","alias_value":"TW7PL5QL4KQR","created_at":"2026-07-05T04:01:31Z"},{"alias_kind":"pith_short_16","alias_value":"TW7PL5QL4KQROQNY","created_at":"2026-07-05T04:01:31Z"},{"alias_kind":"pith_short_8","alias_value":"TW7PL5QL","created_at":"2026-07-05T04:01:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:TW7PL5QL4KQROQNYBLQMXVQFEN","target":"record","payload":{"canonical_record":{"source":{"id":"2109.08186","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2021-09-16T18:45:45Z","cross_cats_sorted":["cs.CL","cs.IR"],"title_canon_sha256":"2a6e3aec440201542dc7eb89a5968f468fd8eed148b5fd919387f64eebc8052e","abstract_canon_sha256":"1bcf3519289db239558982e5f28472664109bfc509fe72696a241e7193bd7a44"},"schema_version":"1.0"},"canonical_sha256":"9dbef5f60be2a11741b80ae0cbd60523461763fbf30a5044149fc3934abea5db","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:01:31.970441Z","signature_b64":"YXlcyzhmG2nu6QfxIDYeeu3+LXMOgRQn5KMTBxRwFIk6/s2emWt0ncHxQwwa/elXypjHkPCkQIEPNOjAR/5cAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9dbef5f60be2a11741b80ae0cbd60523461763fbf30a5044149fc3934abea5db","last_reissued_at":"2026-07-05T04:01:31.970049Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:01:31.970049Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2109.08186","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T04:01:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NmY+T7r59b5lPTwhnPCjOetm8gdpKFmdfFDjbndSNj9YWvqI3QD81TBtepw5p1b1ImhFVpQweQQ+rlSjWHkdDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T18:41:18.104448Z"},"content_sha256":"9efedc10667de7a0b5d7dc700c30c8294576bc1df0dfee5e345295cf27ced5cb","schema_version":"1.0","event_id":"sha256:9efedc10667de7a0b5d7dc700c30c8294576bc1df0dfee5e345295cf27ced5cb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:TW7PL5QL4KQROQNYBLQMXVQFEN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Fast-Slow Transformer for Visually Grounding Speech","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.IR"],"primary_cat":"eess.AS","authors_text":"David Harwath, Puyuan Peng","submitted_at":"2021-09-16T18:45:45Z","abstract_excerpt":"We present Fast-Slow Transformer for Visually Grounding Speech, or FaST-VGS. FaST-VGS is a Transformer-based model for learning the associations between raw speech waveforms and visual images. The model unifies dual-encoder and cross-attention architectures into a single model, reaping the superior retrieval speed of the former along with the accuracy of the latter. FaST-VGS achieves state-of-the-art speech-image retrieval accuracy on benchmark datasets, and its learned representations exhibit strong performance on the ZeroSpeech 2021 phonetic and semantic tasks."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.08186","kind":"arxiv","version":4},"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/2109.08186/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":"2026-07-05T04:01:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AaRv5F42hsgAnWcVFloevElxCT2ATcE4kJUR/WWuuTaKy4jcdOc+TSz1FleSO1/fTX27bd8f9X9wzAx6RYP+AA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T18:41:18.104824Z"},"content_sha256":"4f2430d55033fa1eae8ec306f77fae9c43d88e209435dd61b1f0e5f461cffe53","schema_version":"1.0","event_id":"sha256:4f2430d55033fa1eae8ec306f77fae9c43d88e209435dd61b1f0e5f461cffe53"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TW7PL5QL4KQROQNYBLQMXVQFEN/bundle.json","state_url":"https://pith.science/pith/TW7PL5QL4KQROQNYBLQMXVQFEN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TW7PL5QL4KQROQNYBLQMXVQFEN/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":"2026-07-06T18:41:18Z","links":{"resolver":"https://pith.science/pith/TW7PL5QL4KQROQNYBLQMXVQFEN","bundle":"https://pith.science/pith/TW7PL5QL4KQROQNYBLQMXVQFEN/bundle.json","state":"https://pith.science/pith/TW7PL5QL4KQROQNYBLQMXVQFEN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TW7PL5QL4KQROQNYBLQMXVQFEN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:TW7PL5QL4KQROQNYBLQMXVQFEN","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":"1bcf3519289db239558982e5f28472664109bfc509fe72696a241e7193bd7a44","cross_cats_sorted":["cs.CL","cs.IR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2021-09-16T18:45:45Z","title_canon_sha256":"2a6e3aec440201542dc7eb89a5968f468fd8eed148b5fd919387f64eebc8052e"},"schema_version":"1.0","source":{"id":"2109.08186","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2109.08186","created_at":"2026-07-05T04:01:31Z"},{"alias_kind":"arxiv_version","alias_value":"2109.08186v4","created_at":"2026-07-05T04:01:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.08186","created_at":"2026-07-05T04:01:31Z"},{"alias_kind":"pith_short_12","alias_value":"TW7PL5QL4KQR","created_at":"2026-07-05T04:01:31Z"},{"alias_kind":"pith_short_16","alias_value":"TW7PL5QL4KQROQNY","created_at":"2026-07-05T04:01:31Z"},{"alias_kind":"pith_short_8","alias_value":"TW7PL5QL","created_at":"2026-07-05T04:01:31Z"}],"graph_snapshots":[{"event_id":"sha256:4f2430d55033fa1eae8ec306f77fae9c43d88e209435dd61b1f0e5f461cffe53","target":"graph","created_at":"2026-07-05T04:01:31Z","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/2109.08186/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We present Fast-Slow Transformer for Visually Grounding Speech, or FaST-VGS. FaST-VGS is a Transformer-based model for learning the associations between raw speech waveforms and visual images. The model unifies dual-encoder and cross-attention architectures into a single model, reaping the superior retrieval speed of the former along with the accuracy of the latter. FaST-VGS achieves state-of-the-art speech-image retrieval accuracy on benchmark datasets, and its learned representations exhibit strong performance on the ZeroSpeech 2021 phonetic and semantic tasks.","authors_text":"David Harwath, Puyuan Peng","cross_cats":["cs.CL","cs.IR"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2021-09-16T18:45:45Z","title":"Fast-Slow Transformer for Visually Grounding Speech"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.08186","kind":"arxiv","version":4},"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":"sha256:9efedc10667de7a0b5d7dc700c30c8294576bc1df0dfee5e345295cf27ced5cb","target":"record","created_at":"2026-07-05T04:01:31Z","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":"1bcf3519289db239558982e5f28472664109bfc509fe72696a241e7193bd7a44","cross_cats_sorted":["cs.CL","cs.IR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.AS","submitted_at":"2021-09-16T18:45:45Z","title_canon_sha256":"2a6e3aec440201542dc7eb89a5968f468fd8eed148b5fd919387f64eebc8052e"},"schema_version":"1.0","source":{"id":"2109.08186","kind":"arxiv","version":4}},"canonical_sha256":"9dbef5f60be2a11741b80ae0cbd60523461763fbf30a5044149fc3934abea5db","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9dbef5f60be2a11741b80ae0cbd60523461763fbf30a5044149fc3934abea5db","first_computed_at":"2026-07-05T04:01:31.970049Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T04:01:31.970049Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"YXlcyzhmG2nu6QfxIDYeeu3+LXMOgRQn5KMTBxRwFIk6/s2emWt0ncHxQwwa/elXypjHkPCkQIEPNOjAR/5cAg==","signature_status":"signed_v1","signed_at":"2026-07-05T04:01:31.970441Z","signed_message":"canonical_sha256_bytes"},"source_id":"2109.08186","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9efedc10667de7a0b5d7dc700c30c8294576bc1df0dfee5e345295cf27ced5cb","sha256:4f2430d55033fa1eae8ec306f77fae9c43d88e209435dd61b1f0e5f461cffe53"],"state_sha256":"6e4dd4d511d60ee4fb16cb173e50febb79118d719ded406bf39dc4647ea1b757"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"qeOAH9T2hcd1HJTy1vTTogpukolYNJEtFx/vA6yEFrXNZyn7AGtilENLMgWAC8SjloHyFofMRsshGfNiIvAlDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T18:41:18.107034Z","bundle_sha256":"2e3412ee5f4ef7e1bcd375d200ad79218a3e23a81c90ec370feddca12a72c9c1"}}