{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2022:27PNU4SO5MCXZKUYDXCANDSZZB","short_pith_number":"pith:27PNU4SO","canonical_record":{"source":{"id":"2210.16428","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2022-10-28T22:45:41Z","cross_cats_sorted":["cs.AI","cs.MM","cs.SD"],"title_canon_sha256":"4d51abe7040d197f6df8998ac1f6ddb43f0a4397ba84489156e735e25d4cf6b0","abstract_canon_sha256":"dfe614f3f5a4073cf4fcf524e490fa701e369bee55312b0eaf209675a2e8065b"},"schema_version":"1.0"},"canonical_sha256":"d7deda724eeb057caa981dc4068e59c87f1e907e6e3921d37f3e9cb19271b32c","source":{"kind":"arxiv","id":"2210.16428","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2210.16428","created_at":"2026-07-05T06:14:37Z"},{"alias_kind":"arxiv_version","alias_value":"2210.16428v3","created_at":"2026-07-05T06:14:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.16428","created_at":"2026-07-05T06:14:37Z"},{"alias_kind":"pith_short_12","alias_value":"27PNU4SO5MCX","created_at":"2026-07-05T06:14:37Z"},{"alias_kind":"pith_short_16","alias_value":"27PNU4SO5MCXZKUY","created_at":"2026-07-05T06:14:37Z"},{"alias_kind":"pith_short_8","alias_value":"27PNU4SO","created_at":"2026-07-05T06:14:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2022:27PNU4SO5MCXZKUYDXCANDSZZB","target":"record","payload":{"canonical_record":{"source":{"id":"2210.16428","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2022-10-28T22:45:41Z","cross_cats_sorted":["cs.AI","cs.MM","cs.SD"],"title_canon_sha256":"4d51abe7040d197f6df8998ac1f6ddb43f0a4397ba84489156e735e25d4cf6b0","abstract_canon_sha256":"dfe614f3f5a4073cf4fcf524e490fa701e369bee55312b0eaf209675a2e8065b"},"schema_version":"1.0"},"canonical_sha256":"d7deda724eeb057caa981dc4068e59c87f1e907e6e3921d37f3e9cb19271b32c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:14:37.497345Z","signature_b64":"+cEb+caKAEcAMRKGB/QVf+lIGSo7maXWWKIo+UYrgFrjoDy9QILisc5yyYV/VtV4muZejSOt2tsfdtbvGCmzCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d7deda724eeb057caa981dc4068e59c87f1e907e6e3921d37f3e9cb19271b32c","last_reissued_at":"2026-07-05T06:14:37.496890Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:14:37.496890Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2210.16428","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:14:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bpdz/+3eocjCzGgVNy5c4sLrO208g2hIlDB0dZXS+ihNOtvOJJv8il42RO1O2ilTcPetHKeANnHVJIsFK+BwDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T21:58:45.468741Z"},"content_sha256":"4157f0817e469b99c16e915acd2be49cf410e3fcf83d21827e8cbd79f546e66a","schema_version":"1.0","event_id":"sha256:4157f0817e469b99c16e915acd2be49cf410e3fcf83d21827e8cbd79f546e66a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2022:27PNU4SO5MCXZKUYDXCANDSZZB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Visually-Aware Audio Captioning With Adaptive Audio-Visual Attention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.MM","cs.SD"],"primary_cat":"eess.AS","authors_text":"Haohe Liu, Jianyuan Sun, Lilian H. Tang, Mark D. Plumbley, Qiuqiang Kong, Qiushi Huang, Shengchen Li, Tom Ko, Volkan K{\\i}l{\\i}\\c{c}, Wenwu Wang, Xinhao Mei, Xubo Liu, Yu Zhang","submitted_at":"2022-10-28T22:45:41Z","abstract_excerpt":"Audio captioning aims to generate text descriptions of audio clips. In the real world, many objects produce similar sounds. How to accurately recognize ambiguous sounds is a major challenge for audio captioning. In this work, inspired by inherent human multimodal perception, we propose visually-aware audio captioning, which makes use of visual information to help the description of ambiguous sounding objects. Specifically, we introduce an off-the-shelf visual encoder to extract video features and incorporate the visual features into an audio captioning system. Furthermore, to better exploit co"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.16428","kind":"arxiv","version":3},"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/2210.16428/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":"2026-07-05T06:14:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JYZFDDsNsEmIrHIPxAMxQey0HKjTn624rHZ5UrGNRTPamVhQGtjlqIcqObz6Gd2Cw9ADJQduNGvPwsnNpFDDAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-07T21:58:45.469140Z"},"content_sha256":"8a79529fe6005c7d726b00eaa1934231c83ab8071d98f9134b2914ae5bf03c8c","schema_version":"1.0","event_id":"sha256:8a79529fe6005c7d726b00eaa1934231c83ab8071d98f9134b2914ae5bf03c8c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/27PNU4SO5MCXZKUYDXCANDSZZB/bundle.json","state_url":"https://pith.science/pith/27PNU4SO5MCXZKUYDXCANDSZZB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/27PNU4SO5MCXZKUYDXCANDSZZB/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":"2026-07-07T21:58:45Z","links":{"resolver":"https://pith.science/pith/27PNU4SO5MCXZKUYDXCANDSZZB","bundle":"https://pith.science/pith/27PNU4SO5MCXZKUYDXCANDSZZB/bundle.json","state":"https://pith.science/pith/27PNU4SO5MCXZKUYDXCANDSZZB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/27PNU4SO5MCXZKUYDXCANDSZZB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2022:27PNU4SO5MCXZKUYDXCANDSZZB","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":"dfe614f3f5a4073cf4fcf524e490fa701e369bee55312b0eaf209675a2e8065b","cross_cats_sorted":["cs.AI","cs.MM","cs.SD"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2022-10-28T22:45:41Z","title_canon_sha256":"4d51abe7040d197f6df8998ac1f6ddb43f0a4397ba84489156e735e25d4cf6b0"},"schema_version":"1.0","source":{"id":"2210.16428","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2210.16428","created_at":"2026-07-05T06:14:37Z"},{"alias_kind":"arxiv_version","alias_value":"2210.16428v3","created_at":"2026-07-05T06:14:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2210.16428","created_at":"2026-07-05T06:14:37Z"},{"alias_kind":"pith_short_12","alias_value":"27PNU4SO5MCX","created_at":"2026-07-05T06:14:37Z"},{"alias_kind":"pith_short_16","alias_value":"27PNU4SO5MCXZKUY","created_at":"2026-07-05T06:14:37Z"},{"alias_kind":"pith_short_8","alias_value":"27PNU4SO","created_at":"2026-07-05T06:14:37Z"}],"graph_snapshots":[{"event_id":"sha256:8a79529fe6005c7d726b00eaa1934231c83ab8071d98f9134b2914ae5bf03c8c","target":"graph","created_at":"2026-07-05T06:14:37Z","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/2210.16428/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Audio captioning aims to generate text descriptions of audio clips. In the real world, many objects produce similar sounds. How to accurately recognize ambiguous sounds is a major challenge for audio captioning. In this work, inspired by inherent human multimodal perception, we propose visually-aware audio captioning, which makes use of visual information to help the description of ambiguous sounding objects. Specifically, we introduce an off-the-shelf visual encoder to extract video features and incorporate the visual features into an audio captioning system. Furthermore, to better exploit co","authors_text":"Haohe Liu, Jianyuan Sun, Lilian H. Tang, Mark D. Plumbley, Qiuqiang Kong, Qiushi Huang, Shengchen Li, Tom Ko, Volkan K{\\i}l{\\i}\\c{c}, Wenwu Wang, Xinhao Mei, Xubo Liu, Yu Zhang","cross_cats":["cs.AI","cs.MM","cs.SD"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2022-10-28T22:45:41Z","title":"Visually-Aware Audio Captioning With Adaptive Audio-Visual Attention"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2210.16428","kind":"arxiv","version":3},"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":"sha256:4157f0817e469b99c16e915acd2be49cf410e3fcf83d21827e8cbd79f546e66a","target":"record","created_at":"2026-07-05T06:14:37Z","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":"dfe614f3f5a4073cf4fcf524e490fa701e369bee55312b0eaf209675a2e8065b","cross_cats_sorted":["cs.AI","cs.MM","cs.SD"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.AS","submitted_at":"2022-10-28T22:45:41Z","title_canon_sha256":"4d51abe7040d197f6df8998ac1f6ddb43f0a4397ba84489156e735e25d4cf6b0"},"schema_version":"1.0","source":{"id":"2210.16428","kind":"arxiv","version":3}},"canonical_sha256":"d7deda724eeb057caa981dc4068e59c87f1e907e6e3921d37f3e9cb19271b32c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d7deda724eeb057caa981dc4068e59c87f1e907e6e3921d37f3e9cb19271b32c","first_computed_at":"2026-07-05T06:14:37.496890Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:14:37.496890Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+cEb+caKAEcAMRKGB/QVf+lIGSo7maXWWKIo+UYrgFrjoDy9QILisc5yyYV/VtV4muZejSOt2tsfdtbvGCmzCg==","signature_status":"signed_v1","signed_at":"2026-07-05T06:14:37.497345Z","signed_message":"canonical_sha256_bytes"},"source_id":"2210.16428","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4157f0817e469b99c16e915acd2be49cf410e3fcf83d21827e8cbd79f546e66a","sha256:8a79529fe6005c7d726b00eaa1934231c83ab8071d98f9134b2914ae5bf03c8c"],"state_sha256":"0a1ba6ddbae808acf95a54e608b780d19049d1da5dfe4974e5065aedfa698467"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"S5tuVbMSdtwR8oE0/o+fPx6sdgCHz7+SFlnY607MmpYdhR728bVKck8+z0nMKxXBF2aP6MYGmb78P7vBCabQBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-07T21:58:45.471524Z","bundle_sha256":"3557cb71729d16988c5021057649c8406ccc3110deaaaa9f5e05c91795bfee24"}}