{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:AWGIJE6LR3SWVYMZ4BZVXFMBM6","short_pith_number":"pith:AWGIJE6L","schema_version":"1.0","canonical_sha256":"058c8493cb8ee56ae199e0735b958167865edb2c76b18e8cd97c4659a87172d2","source":{"kind":"arxiv","id":"2410.06264","version":2},"attestation_state":"computed","paper":{"title":"Think While You Generate: Discrete Diffusion with Planned Denoising","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrew Campbell, Hannes St\\\"ark, Juno Nam, Rafael G\\'omez-Bombarelli, Sulin Liu, Tommi Jaakkola, Yilun Xu","submitted_at":"2024-10-08T18:03:34Z","abstract_excerpt":"Discrete diffusion has achieved state-of-the-art performance, outperforming or approaching autoregressive models on standard benchmarks. In this work, we introduce Discrete Diffusion with Planned Denoising (DDPD), a novel framework that separates the generation process into two models: a planner and a denoiser. At inference time, the planner selects which positions to denoise next by identifying the most corrupted positions in need of denoising, including both initially corrupted and those requiring additional refinement. This plan-and-denoise approach enables more efficient reconstruction dur"},"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":"2410.06264","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-10-08T18:03:34Z","cross_cats_sorted":["cs.AI","cs.CL","cs.CV","stat.ML"],"title_canon_sha256":"47669ea4a4440f69358232d98df10b35fc17e04572c6269985ce082ad75e96f4","abstract_canon_sha256":"6e977f2149195c13c23501bee16099f3e3cf8aa7c0e9be6389b0344f7bc6d4dd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:46:57.137779Z","signature_b64":"Vt53b7Mt28IVxjV3aSGLsV2LTf0TF92NoayGqKU/AsqjKMsQ7cajteOAW21fve7uoAJHiDT0vmnyBweYNvA2BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"058c8493cb8ee56ae199e0735b958167865edb2c76b18e8cd97c4659a87172d2","last_reissued_at":"2026-07-05T10:46:57.137272Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:46:57.137272Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Think While You Generate: Discrete Diffusion with Planned Denoising","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV","stat.ML"],"primary_cat":"cs.LG","authors_text":"Andrew Campbell, Hannes St\\\"ark, Juno Nam, Rafael G\\'omez-Bombarelli, Sulin Liu, Tommi Jaakkola, Yilun Xu","submitted_at":"2024-10-08T18:03:34Z","abstract_excerpt":"Discrete diffusion has achieved state-of-the-art performance, outperforming or approaching autoregressive models on standard benchmarks. In this work, we introduce Discrete Diffusion with Planned Denoising (DDPD), a novel framework that separates the generation process into two models: a planner and a denoiser. At inference time, the planner selects which positions to denoise next by identifying the most corrupted positions in need of denoising, including both initially corrupted and those requiring additional refinement. This plan-and-denoise approach enables more efficient reconstruction dur"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.06264","kind":"arxiv","version":2},"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/2410.06264/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":"2410.06264","created_at":"2026-07-05T10:46:57.137343+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.06264v2","created_at":"2026-07-05T10:46:57.137343+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.06264","created_at":"2026-07-05T10:46:57.137343+00:00"},{"alias_kind":"pith_short_12","alias_value":"AWGIJE6LR3SW","created_at":"2026-07-05T10:46:57.137343+00:00"},{"alias_kind":"pith_short_16","alias_value":"AWGIJE6LR3SWVYMZ","created_at":"2026-07-05T10:46:57.137343+00:00"},{"alias_kind":"pith_short_8","alias_value":"AWGIJE6L","created_at":"2026-07-05T10:46:57.137343+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.00773","citing_title":"Accelerating Discrete Diffusion Models with Parallel-In-Time Sampling","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05645","citing_title":"Discrete-WAM: Unified Discrete Vision-Action Token Editing for World-Policy Learning","ref_index":54,"is_internal_anchor":false},{"citing_arxiv_id":"2605.31215","citing_title":"Fixed-Point Masked Generative Modeling","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22765","citing_title":"Uniform Diffusion Models Revisited: Leave-One-Out Denoiser and Absorbing State Reformulation","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2502.09992","citing_title":"Large Language Diffusion Models","ref_index":81,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AWGIJE6LR3SWVYMZ4BZVXFMBM6","json":"https://pith.science/pith/AWGIJE6LR3SWVYMZ4BZVXFMBM6.json","graph_json":"https://pith.science/api/pith-number/AWGIJE6LR3SWVYMZ4BZVXFMBM6/graph.json","events_json":"https://pith.science/api/pith-number/AWGIJE6LR3SWVYMZ4BZVXFMBM6/events.json","paper":"https://pith.science/paper/AWGIJE6L"},"agent_actions":{"view_html":"https://pith.science/pith/AWGIJE6LR3SWVYMZ4BZVXFMBM6","download_json":"https://pith.science/pith/AWGIJE6LR3SWVYMZ4BZVXFMBM6.json","view_paper":"https://pith.science/paper/AWGIJE6L","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.06264&json=true","fetch_graph":"https://pith.science/api/pith-number/AWGIJE6LR3SWVYMZ4BZVXFMBM6/graph.json","fetch_events":"https://pith.science/api/pith-number/AWGIJE6LR3SWVYMZ4BZVXFMBM6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AWGIJE6LR3SWVYMZ4BZVXFMBM6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AWGIJE6LR3SWVYMZ4BZVXFMBM6/action/storage_attestation","attest_author":"https://pith.science/pith/AWGIJE6LR3SWVYMZ4BZVXFMBM6/action/author_attestation","sign_citation":"https://pith.science/pith/AWGIJE6LR3SWVYMZ4BZVXFMBM6/action/citation_signature","submit_replication":"https://pith.science/pith/AWGIJE6LR3SWVYMZ4BZVXFMBM6/action/replication_record"}},"created_at":"2026-07-05T10:46:57.137343+00:00","updated_at":"2026-07-05T10:46:57.137343+00:00"}