{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:Q4KOS5N5KBHMDNP3YXLKVSC4PX","short_pith_number":"pith:Q4KOS5N5","schema_version":"1.0","canonical_sha256":"8714e975bd504ec1b5fbc5d6aac85c7ddd5c5a2c4306a83e5dde9a6c55d2e9ad","source":{"kind":"arxiv","id":"2606.03820","version":1},"attestation_state":"computed","paper":{"title":"A Quantitative Approximation Framework for Flow Distillation in Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Hanfei Zhou, Lei Shi, Ming Li, Weiguo Gao","submitted_at":"2026-06-02T16:00:43Z","abstract_excerpt":"We develop a quantitative approximation framework for diffusion distillation, viewing few-step sampling as error propagation under compositions of learned flow maps. Focusing on trajectory distillation for the probability-flow ODE, we show that local approximation errors can be strongly amplified in low-noise multimodal regimes, where the underlying dynamics become stiff. In an analytically tractable Gaussian-mixture Ornstein--Uhlenbeck setting, we separate two core difficulties: approximating the time-dependent score field and controlling the dynamical amplification governed by the time-integ"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.03820","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2026-06-02T16:00:43Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"d2b510134b5b4187a10ee19bfd52b019bae301b7dc485d125c67f800254f68f7","abstract_canon_sha256":"01ad40faea247f2ee5f53c2674d35496fe54ac7ad17da478d3c2fcfe4dbfcf99"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T02:06:03.322448Z","signature_b64":"uRCbBSOaY3LH5SoTZPzd+VAqsUFIBug5Q7pmgBzbkbD4zJxkNUk/jNbgFuuaAs1XZWRbU/FbtVU0MkF31XEGCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8714e975bd504ec1b5fbc5d6aac85c7ddd5c5a2c4306a83e5dde9a6c55d2e9ad","last_reissued_at":"2026-06-03T02:06:03.322085Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T02:06:03.322085Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"A Quantitative Approximation Framework for Flow Distillation in Diffusion Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Hanfei Zhou, Lei Shi, Ming Li, Weiguo Gao","submitted_at":"2026-06-02T16:00:43Z","abstract_excerpt":"We develop a quantitative approximation framework for diffusion distillation, viewing few-step sampling as error propagation under compositions of learned flow maps. Focusing on trajectory distillation for the probability-flow ODE, we show that local approximation errors can be strongly amplified in low-noise multimodal regimes, where the underlying dynamics become stiff. In an analytically tractable Gaussian-mixture Ornstein--Uhlenbeck setting, we separate two core difficulties: approximating the time-dependent score field and controlling the dynamical amplification governed by the time-integ"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.03820","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/2606.03820/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.03820","created_at":"2026-06-03T02:06:03.322140+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.03820v1","created_at":"2026-06-03T02:06:03.322140+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.03820","created_at":"2026-06-03T02:06:03.322140+00:00"},{"alias_kind":"pith_short_12","alias_value":"Q4KOS5N5KBHM","created_at":"2026-06-03T02:06:03.322140+00:00"},{"alias_kind":"pith_short_16","alias_value":"Q4KOS5N5KBHMDNP3","created_at":"2026-06-03T02:06:03.322140+00:00"},{"alias_kind":"pith_short_8","alias_value":"Q4KOS5N5","created_at":"2026-06-03T02:06:03.322140+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/Q4KOS5N5KBHMDNP3YXLKVSC4PX","json":"https://pith.science/pith/Q4KOS5N5KBHMDNP3YXLKVSC4PX.json","graph_json":"https://pith.science/api/pith-number/Q4KOS5N5KBHMDNP3YXLKVSC4PX/graph.json","events_json":"https://pith.science/api/pith-number/Q4KOS5N5KBHMDNP3YXLKVSC4PX/events.json","paper":"https://pith.science/paper/Q4KOS5N5"},"agent_actions":{"view_html":"https://pith.science/pith/Q4KOS5N5KBHMDNP3YXLKVSC4PX","download_json":"https://pith.science/pith/Q4KOS5N5KBHMDNP3YXLKVSC4PX.json","view_paper":"https://pith.science/paper/Q4KOS5N5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.03820&json=true","fetch_graph":"https://pith.science/api/pith-number/Q4KOS5N5KBHMDNP3YXLKVSC4PX/graph.json","fetch_events":"https://pith.science/api/pith-number/Q4KOS5N5KBHMDNP3YXLKVSC4PX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Q4KOS5N5KBHMDNP3YXLKVSC4PX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Q4KOS5N5KBHMDNP3YXLKVSC4PX/action/storage_attestation","attest_author":"https://pith.science/pith/Q4KOS5N5KBHMDNP3YXLKVSC4PX/action/author_attestation","sign_citation":"https://pith.science/pith/Q4KOS5N5KBHMDNP3YXLKVSC4PX/action/citation_signature","submit_replication":"https://pith.science/pith/Q4KOS5N5KBHMDNP3YXLKVSC4PX/action/replication_record"}},"created_at":"2026-06-03T02:06:03.322140+00:00","updated_at":"2026-06-03T02:06:03.322140+00:00"}