{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:U5TMGCJRBMDJSPL34S2XD52LVS","short_pith_number":"pith:U5TMGCJR","canonical_record":{"source":{"id":"2606.22766","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-22T02:05:38Z","cross_cats_sorted":[],"title_canon_sha256":"7c51a19d99cf391d066a60d8319cc394ae258c773590db81ba0b77858996b907","abstract_canon_sha256":"8baacf5223dbecc01fc5e8c12f08e6d00323d9aacd49f3aabeae62bfbe996fe8"},"schema_version":"1.0"},"canonical_sha256":"a766c309310b06993d7be4b571f74bac989f04f254bc04dfe8b1ab6ddc86306e","source":{"kind":"arxiv","id":"2606.22766","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.22766","created_at":"2026-06-23T02:13:53Z"},{"alias_kind":"arxiv_version","alias_value":"2606.22766v1","created_at":"2026-06-23T02:13:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.22766","created_at":"2026-06-23T02:13:53Z"},{"alias_kind":"pith_short_12","alias_value":"U5TMGCJRBMDJ","created_at":"2026-06-23T02:13:53Z"},{"alias_kind":"pith_short_16","alias_value":"U5TMGCJRBMDJSPL3","created_at":"2026-06-23T02:13:53Z"},{"alias_kind":"pith_short_8","alias_value":"U5TMGCJR","created_at":"2026-06-23T02:13:53Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:U5TMGCJRBMDJSPL34S2XD52LVS","target":"record","payload":{"canonical_record":{"source":{"id":"2606.22766","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-22T02:05:38Z","cross_cats_sorted":[],"title_canon_sha256":"7c51a19d99cf391d066a60d8319cc394ae258c773590db81ba0b77858996b907","abstract_canon_sha256":"8baacf5223dbecc01fc5e8c12f08e6d00323d9aacd49f3aabeae62bfbe996fe8"},"schema_version":"1.0"},"canonical_sha256":"a766c309310b06993d7be4b571f74bac989f04f254bc04dfe8b1ab6ddc86306e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T02:13:53.937101Z","signature_b64":"l7wOGm21vUfxNJbmgu6rpDfq7XQSNCRn2YwSxi8ZNMA1iPZfwEOv5b5C1E+utvqfQVrVjDvNuSNgbbUI9Dt3Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a766c309310b06993d7be4b571f74bac989f04f254bc04dfe8b1ab6ddc86306e","last_reissued_at":"2026-06-23T02:13:53.936684Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T02:13:53.936684Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.22766","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":"2026-06-23T02:13:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4tmUN4SJN4gLJqyCM1iV2oDrPrucKK07aE4vaqwNlgh23QY7PF3r9zV3r/goO6maWS+8gc1W8OX03Tx2vAr9CQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T23:42:03.061228Z"},"content_sha256":"9971983bd79e345975c987cbf9ca88c9328f214f8ebd262fc5ef64ebb617b2c9","schema_version":"1.0","event_id":"sha256:9971983bd79e345975c987cbf9ca88c9328f214f8ebd262fc5ef64ebb617b2c9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:U5TMGCJRBMDJSPL34S2XD52LVS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"READ More than What You See: Reinforcement Learning for Accurate and Coherent Audio Description Generations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Antoni B. Chan, Bo Fang, Hang Zhou, Hui Zhang, Xinyao Zhang, Yuxin Song","submitted_at":"2026-06-22T02:05:38Z","abstract_excerpt":"Audio Description aims to generate concise narrations of essential visual content in audio-visual media for blind and low-vision audiences. Existing methods either rely on prompting off-the-shelf multimodal models, which often mismatch AD style, or partially optimize training-based systems with next-token prediction, which under-explores model capacity and biases generation toward generic expressions. We present READ, the first reinforcement-learning (RL) framework for training-based AD generation. READ formulates AD as sequence-level optimization with reference-matching, length, and format re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.22766","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/2606.22766/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":"2026-06-23T02:13:53Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1qbZtdma34k3ceVG0f5v0FUCqkhwjnxlw0GVMMvgY2QScBE9i7zWAnnA4ZrZwaWDoLwc8yJr/ZWgHZ/icQreAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-04T23:42:03.061614Z"},"content_sha256":"43f3eac981d72451d8cbde44b62155822353aa33f2e3e5ac426744a61aa32dbf","schema_version":"1.0","event_id":"sha256:43f3eac981d72451d8cbde44b62155822353aa33f2e3e5ac426744a61aa32dbf"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/U5TMGCJRBMDJSPL34S2XD52LVS/bundle.json","state_url":"https://pith.science/pith/U5TMGCJRBMDJSPL34S2XD52LVS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/U5TMGCJRBMDJSPL34S2XD52LVS/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":"2026-07-04T23:42:03Z","links":{"resolver":"https://pith.science/pith/U5TMGCJRBMDJSPL34S2XD52LVS","bundle":"https://pith.science/pith/U5TMGCJRBMDJSPL34S2XD52LVS/bundle.json","state":"https://pith.science/pith/U5TMGCJRBMDJSPL34S2XD52LVS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/U5TMGCJRBMDJSPL34S2XD52LVS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:U5TMGCJRBMDJSPL34S2XD52LVS","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":"8baacf5223dbecc01fc5e8c12f08e6d00323d9aacd49f3aabeae62bfbe996fe8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-22T02:05:38Z","title_canon_sha256":"7c51a19d99cf391d066a60d8319cc394ae258c773590db81ba0b77858996b907"},"schema_version":"1.0","source":{"id":"2606.22766","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.22766","created_at":"2026-06-23T02:13:53Z"},{"alias_kind":"arxiv_version","alias_value":"2606.22766v1","created_at":"2026-06-23T02:13:53Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.22766","created_at":"2026-06-23T02:13:53Z"},{"alias_kind":"pith_short_12","alias_value":"U5TMGCJRBMDJ","created_at":"2026-06-23T02:13:53Z"},{"alias_kind":"pith_short_16","alias_value":"U5TMGCJRBMDJSPL3","created_at":"2026-06-23T02:13:53Z"},{"alias_kind":"pith_short_8","alias_value":"U5TMGCJR","created_at":"2026-06-23T02:13:53Z"}],"graph_snapshots":[{"event_id":"sha256:43f3eac981d72451d8cbde44b62155822353aa33f2e3e5ac426744a61aa32dbf","target":"graph","created_at":"2026-06-23T02:13:53Z","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/2606.22766/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Audio Description aims to generate concise narrations of essential visual content in audio-visual media for blind and low-vision audiences. Existing methods either rely on prompting off-the-shelf multimodal models, which often mismatch AD style, or partially optimize training-based systems with next-token prediction, which under-explores model capacity and biases generation toward generic expressions. We present READ, the first reinforcement-learning (RL) framework for training-based AD generation. READ formulates AD as sequence-level optimization with reference-matching, length, and format re","authors_text":"Antoni B. Chan, Bo Fang, Hang Zhou, Hui Zhang, Xinyao Zhang, Yuxin Song","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-22T02:05:38Z","title":"READ More than What You See: Reinforcement Learning for Accurate and Coherent Audio Description Generations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.22766","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":"sha256:9971983bd79e345975c987cbf9ca88c9328f214f8ebd262fc5ef64ebb617b2c9","target":"record","created_at":"2026-06-23T02:13:53Z","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":"8baacf5223dbecc01fc5e8c12f08e6d00323d9aacd49f3aabeae62bfbe996fe8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-22T02:05:38Z","title_canon_sha256":"7c51a19d99cf391d066a60d8319cc394ae258c773590db81ba0b77858996b907"},"schema_version":"1.0","source":{"id":"2606.22766","kind":"arxiv","version":1}},"canonical_sha256":"a766c309310b06993d7be4b571f74bac989f04f254bc04dfe8b1ab6ddc86306e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a766c309310b06993d7be4b571f74bac989f04f254bc04dfe8b1ab6ddc86306e","first_computed_at":"2026-06-23T02:13:53.936684Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-23T02:13:53.936684Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"l7wOGm21vUfxNJbmgu6rpDfq7XQSNCRn2YwSxi8ZNMA1iPZfwEOv5b5C1E+utvqfQVrVjDvNuSNgbbUI9Dt3Aw==","signature_status":"signed_v1","signed_at":"2026-06-23T02:13:53.937101Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.22766","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9971983bd79e345975c987cbf9ca88c9328f214f8ebd262fc5ef64ebb617b2c9","sha256:43f3eac981d72451d8cbde44b62155822353aa33f2e3e5ac426744a61aa32dbf"],"state_sha256":"85cab49809ac130745535513741d4895ff6fc06bd5ddb76ebd169f7bb0e521fe"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Mq7mlnhR6dv46QcOsSBtRnPyyxvN3f77m1wuTWIPUzoRdWNhWaMEvKqXrOl3UfbUW4O+aDGhfuUI9Fnd3abDDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-04T23:42:03.063531Z","bundle_sha256":"64e6215288607a0c5067a14813f21f9110602b011697119fe37e586da93d8459"}}