{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ML2B24V65HI43RRGOJRCRXZ6W3","short_pith_number":"pith:ML2B24V6","schema_version":"1.0","canonical_sha256":"62f41d72bee9d1cdc626726228df3eb6ce6bb9710ac9d811923909934a6483e3","source":{"kind":"arxiv","id":"2502.08696","version":3},"attestation_state":"computed","paper":{"title":"Scalable Discrete Diffusion Samplers: Combinatorial Optimization and Statistical Physics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cond-mat.stat-mech","cs.AI","physics.comp-ph","stat.ML"],"primary_cat":"cs.LG","authors_text":"Haoyu Peter Wang, Martin Ennemoser, Sebastian Lehner, Sebastian Sanokowski, Sepp Hochreiter, Wilhelm Berghammer","submitted_at":"2025-02-12T18:59:55Z","abstract_excerpt":"Learning to sample from complex unnormalized distributions over discrete domains emerged as a promising research direction with applications in statistical physics, variational inference, and combinatorial optimization. Recent work has demonstrated the potential of diffusion models in this domain. However, existing methods face limitations in memory scaling and thus the number of attainable diffusion steps since they require backpropagation through the entire generative process. To overcome these limitations we introduce two novel training methods for discrete diffusion samplers, one grounded "},"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":"2502.08696","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-02-12T18:59:55Z","cross_cats_sorted":["cond-mat.stat-mech","cs.AI","physics.comp-ph","stat.ML"],"title_canon_sha256":"c0600139eaa9a3de613a6d18d64dec705bcbc2e21bfd35187a36c6b18a1cf2b5","abstract_canon_sha256":"905dda9b5fe7d8122e0b8bc83a485ae8721adffa59dd1c793f88c89126898d66"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:33:19.948980Z","signature_b64":"ZKCM5nMxiLEG8Ejg3QHCW++WFYUckAhTRlMvqqj2DVZoo1FTZwZ857QfLzVZO71YSD8O6PtXdHGryKFqqs/pCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"62f41d72bee9d1cdc626726228df3eb6ce6bb9710ac9d811923909934a6483e3","last_reissued_at":"2026-07-05T11:33:19.948518Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:33:19.948518Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scalable Discrete Diffusion Samplers: Combinatorial Optimization and Statistical Physics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cond-mat.stat-mech","cs.AI","physics.comp-ph","stat.ML"],"primary_cat":"cs.LG","authors_text":"Haoyu Peter Wang, Martin Ennemoser, Sebastian Lehner, Sebastian Sanokowski, Sepp Hochreiter, Wilhelm Berghammer","submitted_at":"2025-02-12T18:59:55Z","abstract_excerpt":"Learning to sample from complex unnormalized distributions over discrete domains emerged as a promising research direction with applications in statistical physics, variational inference, and combinatorial optimization. Recent work has demonstrated the potential of diffusion models in this domain. However, existing methods face limitations in memory scaling and thus the number of attainable diffusion steps since they require backpropagation through the entire generative process. To overcome these limitations we introduce two novel training methods for discrete diffusion samplers, one grounded "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.08696","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/2502.08696/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":"2502.08696","created_at":"2026-07-05T11:33:19.948581+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.08696v3","created_at":"2026-07-05T11:33:19.948581+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.08696","created_at":"2026-07-05T11:33:19.948581+00:00"},{"alias_kind":"pith_short_12","alias_value":"ML2B24V65HI4","created_at":"2026-07-05T11:33:19.948581+00:00"},{"alias_kind":"pith_short_16","alias_value":"ML2B24V65HI43RRG","created_at":"2026-07-05T11:33:19.948581+00:00"},{"alias_kind":"pith_short_8","alias_value":"ML2B24V6","created_at":"2026-07-05T11:33:19.948581+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.00295","citing_title":"Adaptive Order Policies for Masked Diffusion","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2510.19544","citing_title":"Demonstrating Real Advantage of Machine-Learning-Enhanced Monte Carlo for Combinatorial Optimization","ref_index":47,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ML2B24V65HI43RRGOJRCRXZ6W3","json":"https://pith.science/pith/ML2B24V65HI43RRGOJRCRXZ6W3.json","graph_json":"https://pith.science/api/pith-number/ML2B24V65HI43RRGOJRCRXZ6W3/graph.json","events_json":"https://pith.science/api/pith-number/ML2B24V65HI43RRGOJRCRXZ6W3/events.json","paper":"https://pith.science/paper/ML2B24V6"},"agent_actions":{"view_html":"https://pith.science/pith/ML2B24V65HI43RRGOJRCRXZ6W3","download_json":"https://pith.science/pith/ML2B24V65HI43RRGOJRCRXZ6W3.json","view_paper":"https://pith.science/paper/ML2B24V6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.08696&json=true","fetch_graph":"https://pith.science/api/pith-number/ML2B24V65HI43RRGOJRCRXZ6W3/graph.json","fetch_events":"https://pith.science/api/pith-number/ML2B24V65HI43RRGOJRCRXZ6W3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ML2B24V65HI43RRGOJRCRXZ6W3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ML2B24V65HI43RRGOJRCRXZ6W3/action/storage_attestation","attest_author":"https://pith.science/pith/ML2B24V65HI43RRGOJRCRXZ6W3/action/author_attestation","sign_citation":"https://pith.science/pith/ML2B24V65HI43RRGOJRCRXZ6W3/action/citation_signature","submit_replication":"https://pith.science/pith/ML2B24V65HI43RRGOJRCRXZ6W3/action/replication_record"}},"created_at":"2026-07-05T11:33:19.948581+00:00","updated_at":"2026-07-05T11:33:19.948581+00:00"}