{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:L3V3E4ECXX6BV7XEEICIW47E6O","short_pith_number":"pith:L3V3E4EC","canonical_record":{"source":{"id":"2604.19635","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2026-04-21T16:25:22Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"06403645674df1044a0055482b07b8d8a22512ae778fbdcca77f8d19dd5ef91d","abstract_canon_sha256":"0947e4415e43a1097348e2c32ae3a479d21be1f89aa020d32dd800462577071c"},"schema_version":"1.0"},"canonical_sha256":"5eebb27082bdfc1afee422048b73e4f3977b67b0c29b60a373627a733b96bc0f","source":{"kind":"arxiv","id":"2604.19635","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.19635","created_at":"2026-07-07T02:18:41Z"},{"alias_kind":"arxiv_version","alias_value":"2604.19635v2","created_at":"2026-07-07T02:18:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.19635","created_at":"2026-07-07T02:18:41Z"},{"alias_kind":"pith_short_12","alias_value":"L3V3E4ECXX6B","created_at":"2026-07-07T02:18:41Z"},{"alias_kind":"pith_short_16","alias_value":"L3V3E4ECXX6BV7XE","created_at":"2026-07-07T02:18:41Z"},{"alias_kind":"pith_short_8","alias_value":"L3V3E4EC","created_at":"2026-07-07T02:18:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:L3V3E4ECXX6BV7XEEICIW47E6O","target":"record","payload":{"canonical_record":{"source":{"id":"2604.19635","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2026-04-21T16:25:22Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"06403645674df1044a0055482b07b8d8a22512ae778fbdcca77f8d19dd5ef91d","abstract_canon_sha256":"0947e4415e43a1097348e2c32ae3a479d21be1f89aa020d32dd800462577071c"},"schema_version":"1.0"},"canonical_sha256":"5eebb27082bdfc1afee422048b73e4f3977b67b0c29b60a373627a733b96bc0f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T02:18:41.022902Z","signature_b64":"FoNn6vSNod+rQeGX6CcyjqWkYVZR23pkckvRJDwp/uvF6GGkML5sA6E4r41rkxygrFGf2AY7cZBsE3CKfLOgBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5eebb27082bdfc1afee422048b73e4f3977b67b0c29b60a373627a733b96bc0f","last_reissued_at":"2026-07-07T02:18:41.021900Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T02:18:41.021900Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.19635","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":"2026-07-07T02:18:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HhPeg0IQoUK/mbZD8ohlLVusOP5bfWYlTjCLhxnOMoOjzFBYkO4kSTE8d+DemcinWMNR2wnwzpbPTzKvLG97Cw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T11:23:31.635641Z"},"content_sha256":"5a9bc459ec47e3c1824601ed394fde49928fef5bb64b82183f6a9d10b456b951","schema_version":"1.0","event_id":"sha256:5a9bc459ec47e3c1824601ed394fde49928fef5bb64b82183f6a9d10b456b951"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:L3V3E4ECXX6BV7XEEICIW47E6O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"StarTSE: Towards Streaming Target Speaker Extraction via Chunk-wise Interleaved Splicing of Autoregressive Language Model","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Chunk-wise interleaved splicing lets autoregressive models perform stable streaming target speaker extraction.","cross_cats":["cs.AI"],"primary_cat":"cs.SD","authors_text":"Guiping Zhong, Haiyun Li, Hui Lu, Huimeng Wang, Jiakui Li, Jinjiang Liu, Liang Cao, Liyang Chen, Shiyin Kang, Shuhai Peng, Zhiyong Wu","submitted_at":"2026-04-21T16:25:22Z","abstract_excerpt":"While generative models have set new benchmarks for Target Speaker Extraction (TSE), their inherent reliance on global context precludes deployment in real-time applications. Direct adaptation to streaming scenarios often leads to catastrophic inference performance degradation due to the severe mismatch between training and streaming inference. To bridge this gap, we present the first autoregressive (AR) models tailored for streaming TSE. Our approach introduces a Chunk-wise Interleaved Splicing Paradigm that ensures highly efficient and stable streaming inference. To ensure the coherence betw"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Our approach maintains 100% stability and superior intelligibility while achieving streaming results comparable to or even surpassing offline baselines, with RTF of 0.248.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the Chunk-wise Interleaved Splicing Paradigm plus historical context refinement fully resolves the training-inference mismatch for autoregressive models without introducing new artifacts at chunk boundaries.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Chunk-wise Interleaved Splicing with historical context refinement lets autoregressive models deliver stable streaming target speaker extraction on Libri2Mix, matching or exceeding offline baselines at RTF 0.248.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Chunk-wise interleaved splicing lets autoregressive models perform stable streaming target speaker extraction.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"d5cf18df10e4bfa0eaaa189731415afd4b60db04945162d7728da5b5b3a1ef42"},"source":{"id":"2604.19635","kind":"arxiv","version":2},"verdict":{"id":"13ce1bd6-242a-4cf3-89b0-d8e035211665","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T00:55:55.859069Z","strongest_claim":"Our approach maintains 100% stability and superior intelligibility while achieving streaming results comparable to or even surpassing offline baselines, with RTF of 0.248.","one_line_summary":"Chunk-wise Interleaved Splicing with historical context refinement lets autoregressive models deliver stable streaming target speaker extraction on Libri2Mix, matching or exceeding offline baselines at RTF 0.248.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the Chunk-wise Interleaved Splicing Paradigm plus historical context refinement fully resolves the training-inference mismatch for autoregressive models without introducing new artifacts at chunk boundaries.","pith_extraction_headline":"Chunk-wise interleaved splicing lets autoregressive models perform stable streaming target speaker extraction."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.19635/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-20T02:41:09.372225Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"4fd26f1844898ed2ad9cf7f926c64877c0e70aca25c39c992a18bec76aea4ace"},"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":"13ce1bd6-242a-4cf3-89b0-d8e035211665"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-07T02:18:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mVfVspLLDXi945mSMbSPRrRpf6j68SoySx8wuMAjRVi3dbPcEmWgi///u/Mq/TN313q03tjEGCd0WZF/s3ncAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T11:23:31.636287Z"},"content_sha256":"39eaa10d6c370f2355fa419b7d1552ace0d5c6d5435dbfa35c9ec23b62a2f182","schema_version":"1.0","event_id":"sha256:39eaa10d6c370f2355fa419b7d1552ace0d5c6d5435dbfa35c9ec23b62a2f182"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/L3V3E4ECXX6BV7XEEICIW47E6O/bundle.json","state_url":"https://pith.science/pith/L3V3E4ECXX6BV7XEEICIW47E6O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/L3V3E4ECXX6BV7XEEICIW47E6O/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":"2026-08-04T11:23:31Z","links":{"resolver":"https://pith.science/pith/L3V3E4ECXX6BV7XEEICIW47E6O","bundle":"https://pith.science/pith/L3V3E4ECXX6BV7XEEICIW47E6O/bundle.json","state":"https://pith.science/pith/L3V3E4ECXX6BV7XEEICIW47E6O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/L3V3E4ECXX6BV7XEEICIW47E6O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:L3V3E4ECXX6BV7XEEICIW47E6O","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":"0947e4415e43a1097348e2c32ae3a479d21be1f89aa020d32dd800462577071c","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2026-04-21T16:25:22Z","title_canon_sha256":"06403645674df1044a0055482b07b8d8a22512ae778fbdcca77f8d19dd5ef91d"},"schema_version":"1.0","source":{"id":"2604.19635","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.19635","created_at":"2026-07-07T02:18:41Z"},{"alias_kind":"arxiv_version","alias_value":"2604.19635v2","created_at":"2026-07-07T02:18:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.19635","created_at":"2026-07-07T02:18:41Z"},{"alias_kind":"pith_short_12","alias_value":"L3V3E4ECXX6B","created_at":"2026-07-07T02:18:41Z"},{"alias_kind":"pith_short_16","alias_value":"L3V3E4ECXX6BV7XE","created_at":"2026-07-07T02:18:41Z"},{"alias_kind":"pith_short_8","alias_value":"L3V3E4EC","created_at":"2026-07-07T02:18:41Z"}],"graph_snapshots":[{"event_id":"sha256:39eaa10d6c370f2355fa419b7d1552ace0d5c6d5435dbfa35c9ec23b62a2f182","target":"graph","created_at":"2026-07-07T02:18:41Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Our approach maintains 100% stability and superior intelligibility while achieving streaming results comparable to or even surpassing offline baselines, with RTF of 0.248."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the Chunk-wise Interleaved Splicing Paradigm plus historical context refinement fully resolves the training-inference mismatch for autoregressive models without introducing new artifacts at chunk boundaries."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Chunk-wise Interleaved Splicing with historical context refinement lets autoregressive models deliver stable streaming target speaker extraction on Libri2Mix, matching or exceeding offline baselines at RTF 0.248."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Chunk-wise interleaved splicing lets autoregressive models perform stable streaming target speaker extraction."}],"snapshot_sha256":"d5cf18df10e4bfa0eaaa189731415afd4b60db04945162d7728da5b5b3a1ef42"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-20T02:41:09.372225Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2604.19635/integrity.json","findings":[],"snapshot_sha256":"4fd26f1844898ed2ad9cf7f926c64877c0e70aca25c39c992a18bec76aea4ace","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While generative models have set new benchmarks for Target Speaker Extraction (TSE), their inherent reliance on global context precludes deployment in real-time applications. Direct adaptation to streaming scenarios often leads to catastrophic inference performance degradation due to the severe mismatch between training and streaming inference. To bridge this gap, we present the first autoregressive (AR) models tailored for streaming TSE. Our approach introduces a Chunk-wise Interleaved Splicing Paradigm that ensures highly efficient and stable streaming inference. To ensure the coherence betw","authors_text":"Guiping Zhong, Haiyun Li, Hui Lu, Huimeng Wang, Jiakui Li, Jinjiang Liu, Liang Cao, Liyang Chen, Shiyin Kang, Shuhai Peng, Zhiyong Wu","cross_cats":["cs.AI"],"headline":"Chunk-wise interleaved splicing lets autoregressive models perform stable streaming target speaker extraction.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2026-04-21T16:25:22Z","title":"StarTSE: Towards Streaming Target Speaker Extraction via Chunk-wise Interleaved Splicing of Autoregressive Language Model"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.19635","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-10T00:55:55.859069Z","id":"13ce1bd6-242a-4cf3-89b0-d8e035211665","model_set":{"reader":"grok-4.3"},"one_line_summary":"Chunk-wise Interleaved Splicing with historical context refinement lets autoregressive models deliver stable streaming target speaker extraction on Libri2Mix, matching or exceeding offline baselines at RTF 0.248.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Chunk-wise interleaved splicing lets autoregressive models perform stable streaming target speaker extraction.","strongest_claim":"Our approach maintains 100% stability and superior intelligibility while achieving streaming results comparable to or even surpassing offline baselines, with RTF of 0.248.","weakest_assumption":"That the Chunk-wise Interleaved Splicing Paradigm plus historical context refinement fully resolves the training-inference mismatch for autoregressive models without introducing new artifacts at chunk boundaries."}},"verdict_id":"13ce1bd6-242a-4cf3-89b0-d8e035211665"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:5a9bc459ec47e3c1824601ed394fde49928fef5bb64b82183f6a9d10b456b951","target":"record","created_at":"2026-07-07T02:18:41Z","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":"0947e4415e43a1097348e2c32ae3a479d21be1f89aa020d32dd800462577071c","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SD","submitted_at":"2026-04-21T16:25:22Z","title_canon_sha256":"06403645674df1044a0055482b07b8d8a22512ae778fbdcca77f8d19dd5ef91d"},"schema_version":"1.0","source":{"id":"2604.19635","kind":"arxiv","version":2}},"canonical_sha256":"5eebb27082bdfc1afee422048b73e4f3977b67b0c29b60a373627a733b96bc0f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5eebb27082bdfc1afee422048b73e4f3977b67b0c29b60a373627a733b96bc0f","first_computed_at":"2026-07-07T02:18:41.021900Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-07T02:18:41.021900Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"FoNn6vSNod+rQeGX6CcyjqWkYVZR23pkckvRJDwp/uvF6GGkML5sA6E4r41rkxygrFGf2AY7cZBsE3CKfLOgBQ==","signature_status":"signed_v1","signed_at":"2026-07-07T02:18:41.022902Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.19635","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5a9bc459ec47e3c1824601ed394fde49928fef5bb64b82183f6a9d10b456b951","sha256:39eaa10d6c370f2355fa419b7d1552ace0d5c6d5435dbfa35c9ec23b62a2f182"],"state_sha256":"2a944ec4e5e57d275eed7155bd6cf65e4944739cd8359c7d310bdd4d8ca792e9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uGRWk+QMxbXeZxO9yoi1xYncENv42ba3bza5GaouOzFAiTdgvYIWpRka1uu289fq2Q3LhS3LAby7rTTN4WN2DA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T11:23:31.640861Z","bundle_sha256":"55ec56752560f190a576b9b2327817cb848310c1aa4d67cb216c5a06205c0d33"}}