{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:TFBRISTNB7XN2P672PRXVXK5XY","short_pith_number":"pith:TFBRISTN","canonical_record":{"source":{"id":"2603.13952","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2026-03-14T14:01:45Z","cross_cats_sorted":["cs.AI","eess.AS"],"title_canon_sha256":"76a702a3269c2ef2055b0d5fb3d43ceeb44ea589bef353d0a7f6a4f0aeaedc6e","abstract_canon_sha256":"5a85345cf8037da4c1328bf58f7ee0cdcd7a73fc5739c5ced93df23939c299ef"},"schema_version":"1.0"},"canonical_sha256":"9943144a6d0feedd3fdfd3e37add5dbe27a8cc6f88b18c6ec4c48acbd37a3f43","source":{"kind":"arxiv","id":"2603.13952","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.13952","created_at":"2026-07-16T01:22:37Z"},{"alias_kind":"arxiv_version","alias_value":"2603.13952v3","created_at":"2026-07-16T01:22:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.13952","created_at":"2026-07-16T01:22:37Z"},{"alias_kind":"pith_short_12","alias_value":"TFBRISTNB7XN","created_at":"2026-07-16T01:22:37Z"},{"alias_kind":"pith_short_16","alias_value":"TFBRISTNB7XN2P67","created_at":"2026-07-16T01:22:37Z"},{"alias_kind":"pith_short_8","alias_value":"TFBRISTN","created_at":"2026-07-16T01:22:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:TFBRISTNB7XN2P672PRXVXK5XY","target":"record","payload":{"canonical_record":{"source":{"id":"2603.13952","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2026-03-14T14:01:45Z","cross_cats_sorted":["cs.AI","eess.AS"],"title_canon_sha256":"76a702a3269c2ef2055b0d5fb3d43ceeb44ea589bef353d0a7f6a4f0aeaedc6e","abstract_canon_sha256":"5a85345cf8037da4c1328bf58f7ee0cdcd7a73fc5739c5ced93df23939c299ef"},"schema_version":"1.0"},"canonical_sha256":"9943144a6d0feedd3fdfd3e37add5dbe27a8cc6f88b18c6ec4c48acbd37a3f43","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-16T01:22:37.384614Z","signature_b64":"Utu8jsH+kCvXsrsfcyl7Bf+yX8xXQocZvRCkkx+tsQ9PYgWQC8npKYbq0TcCy+PADJPis1rEv/ttAtAbULENBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9943144a6d0feedd3fdfd3e37add5dbe27a8cc6f88b18c6ec4c48acbd37a3f43","last_reissued_at":"2026-07-16T01:22:37.383715Z","signature_status":"signed_v1","first_computed_at":"2026-07-16T01:22:37.383715Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2603.13952","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":"2026-07-16T01:22:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"S98kZhd/jIjER0QGG0n5Tk77d3TdkHjdATuvEnis+XCu/yyTkge15wFm0SYw9K3uxG+YTE9B8uDMQpPEnK6aBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T07:43:42.934895Z"},"content_sha256":"821e67322883707ac8efcfbec717d7d158d6f1beb13270103c6d460a5cecb6e3","schema_version":"1.0","event_id":"sha256:821e67322883707ac8efcfbec717d7d158d6f1beb13270103c6d460a5cecb6e3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:TFBRISTNB7XN2P672PRXVXK5XY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LLM-Guided Reinforcement Learning for Audio-Visual Speech Enhancement","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.AS"],"primary_cat":"cs.SD","authors_text":"Chih-Ning Chen, Fan-Gang Zeng, Hsin-Min Wang, Jen-Cheng Hou, Shao-Yi Chien, Yu Tsao","submitted_at":"2026-03-14T14:01:45Z","abstract_excerpt":"In existing Audio-Visual Speech Enhancement (AVSE) methods, objectives such as Scale-Invariant Signal-to-Noise Ratio (SI-SNR) and Mean Squared Error (MSE) are widely used; however, their correlation with perceived speech quality is often suboptimal and provides limited interpretability for optimization. This work proposes a reinforcement learning-based AVSE framework with a Large Language Model (LLM)-based interpretable reward model. An audio LLM generates natural language descriptions of enhanced speech, which are converted by a sentiment analysis model into a 1-5 rating score serving as the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.13952","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/2603.13952/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":"2026-07-16T01:22:37Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ITNcpmefrGdaVJgIRMAtGBNxnLGt0Q2h7Fl1l0d2Wzs1iwAz7qOQqSYmpcJzrv8uNpfGrxFXW/wqmWoChOulBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T07:43:42.935518Z"},"content_sha256":"5860be90801e07aec9affeb6a4f5fdc9eb7a108bf7cee81cfe0d3909fa9c06e5","schema_version":"1.0","event_id":"sha256:5860be90801e07aec9affeb6a4f5fdc9eb7a108bf7cee81cfe0d3909fa9c06e5"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TFBRISTNB7XN2P672PRXVXK5XY/bundle.json","state_url":"https://pith.science/pith/TFBRISTNB7XN2P672PRXVXK5XY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TFBRISTNB7XN2P672PRXVXK5XY/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":"2026-08-04T07:43:42Z","links":{"resolver":"https://pith.science/pith/TFBRISTNB7XN2P672PRXVXK5XY","bundle":"https://pith.science/pith/TFBRISTNB7XN2P672PRXVXK5XY/bundle.json","state":"https://pith.science/pith/TFBRISTNB7XN2P672PRXVXK5XY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TFBRISTNB7XN2P672PRXVXK5XY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:TFBRISTNB7XN2P672PRXVXK5XY","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":"5a85345cf8037da4c1328bf58f7ee0cdcd7a73fc5739c5ced93df23939c299ef","cross_cats_sorted":["cs.AI","eess.AS"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2026-03-14T14:01:45Z","title_canon_sha256":"76a702a3269c2ef2055b0d5fb3d43ceeb44ea589bef353d0a7f6a4f0aeaedc6e"},"schema_version":"1.0","source":{"id":"2603.13952","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2603.13952","created_at":"2026-07-16T01:22:37Z"},{"alias_kind":"arxiv_version","alias_value":"2603.13952v3","created_at":"2026-07-16T01:22:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.13952","created_at":"2026-07-16T01:22:37Z"},{"alias_kind":"pith_short_12","alias_value":"TFBRISTNB7XN","created_at":"2026-07-16T01:22:37Z"},{"alias_kind":"pith_short_16","alias_value":"TFBRISTNB7XN2P67","created_at":"2026-07-16T01:22:37Z"},{"alias_kind":"pith_short_8","alias_value":"TFBRISTN","created_at":"2026-07-16T01:22:37Z"}],"graph_snapshots":[{"event_id":"sha256:5860be90801e07aec9affeb6a4f5fdc9eb7a108bf7cee81cfe0d3909fa9c06e5","target":"graph","created_at":"2026-07-16T01:22:37Z","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/2603.13952/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In existing Audio-Visual Speech Enhancement (AVSE) methods, objectives such as Scale-Invariant Signal-to-Noise Ratio (SI-SNR) and Mean Squared Error (MSE) are widely used; however, their correlation with perceived speech quality is often suboptimal and provides limited interpretability for optimization. This work proposes a reinforcement learning-based AVSE framework with a Large Language Model (LLM)-based interpretable reward model. An audio LLM generates natural language descriptions of enhanced speech, which are converted by a sentiment analysis model into a 1-5 rating score serving as the ","authors_text":"Chih-Ning Chen, Fan-Gang Zeng, Hsin-Min Wang, Jen-Cheng Hou, Shao-Yi Chien, Yu Tsao","cross_cats":["cs.AI","eess.AS"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2026-03-14T14:01:45Z","title":"LLM-Guided Reinforcement Learning for Audio-Visual Speech Enhancement"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.13952","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":"sha256:821e67322883707ac8efcfbec717d7d158d6f1beb13270103c6d460a5cecb6e3","target":"record","created_at":"2026-07-16T01:22:37Z","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":"5a85345cf8037da4c1328bf58f7ee0cdcd7a73fc5739c5ced93df23939c299ef","cross_cats_sorted":["cs.AI","eess.AS"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SD","submitted_at":"2026-03-14T14:01:45Z","title_canon_sha256":"76a702a3269c2ef2055b0d5fb3d43ceeb44ea589bef353d0a7f6a4f0aeaedc6e"},"schema_version":"1.0","source":{"id":"2603.13952","kind":"arxiv","version":3}},"canonical_sha256":"9943144a6d0feedd3fdfd3e37add5dbe27a8cc6f88b18c6ec4c48acbd37a3f43","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9943144a6d0feedd3fdfd3e37add5dbe27a8cc6f88b18c6ec4c48acbd37a3f43","first_computed_at":"2026-07-16T01:22:37.383715Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-16T01:22:37.383715Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Utu8jsH+kCvXsrsfcyl7Bf+yX8xXQocZvRCkkx+tsQ9PYgWQC8npKYbq0TcCy+PADJPis1rEv/ttAtAbULENBw==","signature_status":"signed_v1","signed_at":"2026-07-16T01:22:37.384614Z","signed_message":"canonical_sha256_bytes"},"source_id":"2603.13952","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:821e67322883707ac8efcfbec717d7d158d6f1beb13270103c6d460a5cecb6e3","sha256:5860be90801e07aec9affeb6a4f5fdc9eb7a108bf7cee81cfe0d3909fa9c06e5"],"state_sha256":"8def6884e50e2b98f4a147125ed136f80ed9ca275ac4efa7cb15a89036f612e3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"W7SC/s67PqB6Ljzj4oGAu8vQsDDJet8hgCnPy2KLHfsaDQU+wbtg52pieF972360Az3oBGNYHYSizXKSSxa2CA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T07:43:42.939856Z","bundle_sha256":"2a69d8665ef59d4a3aa6d6afef52cb3cfc697cc45f7658696d57bdce9d147c31"}}