{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:A5K2HDWUFIXNKOWPZ2IW6ESI4H","short_pith_number":"pith:A5K2HDWU","schema_version":"1.0","canonical_sha256":"0755a38ed42a2ed53acfce916f1248e1f2ecaf3f7bca329d74256ac7a07a858f","source":{"kind":"arxiv","id":"2509.24728","version":2},"attestation_state":"computed","paper":{"title":"Beyond Softmax: A Natural Parameterization for Categorical Random Variables","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alessandro Manenti, Cesare Alippi","submitted_at":"2025-09-29T12:55:50Z","abstract_excerpt":"Latent categorical variables are frequently found in deep learning architectures. They can model actions in discrete reinforcement-learning environments, represent categories in latent-variable models, or express relations in graph neural networks. Despite their widespread use, their discrete nature poses significant challenges to gradient-descent learning algorithms. While a substantial body of work has offered improved gradient estimation techniques, we take a complementary approach. Specifically, we: 1) revisit the ubiquitous $\\textit{softmax}$ function and demonstrate its limitations from "},"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":"2509.24728","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-09-29T12:55:50Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"55871e89d4022a48c003c9b9b516aaf64a632bcfd915d3a4f0c2ab9d4c1916e0","abstract_canon_sha256":"360fe965ca91acdebba5d1caff92e97523bfd113a0d933e830603a1260847a66"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:10:12.178510Z","signature_b64":"0L33ooyzO/VST1OvrS77m00U34MgebTiBsX4rhABQAJH9q/S0u1jds8rbHHRA4qAeW56RM6BZgyVc1svGqSqCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0755a38ed42a2ed53acfce916f1248e1f2ecaf3f7bca329d74256ac7a07a858f","last_reissued_at":"2026-05-18T03:10:12.177850Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:10:12.177850Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Beyond Softmax: A Natural Parameterization for Categorical Random Variables","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Alessandro Manenti, Cesare Alippi","submitted_at":"2025-09-29T12:55:50Z","abstract_excerpt":"Latent categorical variables are frequently found in deep learning architectures. They can model actions in discrete reinforcement-learning environments, represent categories in latent-variable models, or express relations in graph neural networks. Despite their widespread use, their discrete nature poses significant challenges to gradient-descent learning algorithms. While a substantial body of work has offered improved gradient estimation techniques, we take a complementary approach. Specifically, we: 1) revisit the ubiquitous $\\textit{softmax}$ function and demonstrate its limitations from "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.24728","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":""},"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":"2509.24728","created_at":"2026-05-18T03:10:12.177968+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.24728v2","created_at":"2026-05-18T03:10:12.177968+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.24728","created_at":"2026-05-18T03:10:12.177968+00:00"},{"alias_kind":"pith_short_12","alias_value":"A5K2HDWUFIXN","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"A5K2HDWUFIXNKOWP","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"A5K2HDWU","created_at":"2026-05-18T12:33:37.589309+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/A5K2HDWUFIXNKOWPZ2IW6ESI4H","json":"https://pith.science/pith/A5K2HDWUFIXNKOWPZ2IW6ESI4H.json","graph_json":"https://pith.science/api/pith-number/A5K2HDWUFIXNKOWPZ2IW6ESI4H/graph.json","events_json":"https://pith.science/api/pith-number/A5K2HDWUFIXNKOWPZ2IW6ESI4H/events.json","paper":"https://pith.science/paper/A5K2HDWU"},"agent_actions":{"view_html":"https://pith.science/pith/A5K2HDWUFIXNKOWPZ2IW6ESI4H","download_json":"https://pith.science/pith/A5K2HDWUFIXNKOWPZ2IW6ESI4H.json","view_paper":"https://pith.science/paper/A5K2HDWU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.24728&json=true","fetch_graph":"https://pith.science/api/pith-number/A5K2HDWUFIXNKOWPZ2IW6ESI4H/graph.json","fetch_events":"https://pith.science/api/pith-number/A5K2HDWUFIXNKOWPZ2IW6ESI4H/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A5K2HDWUFIXNKOWPZ2IW6ESI4H/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A5K2HDWUFIXNKOWPZ2IW6ESI4H/action/storage_attestation","attest_author":"https://pith.science/pith/A5K2HDWUFIXNKOWPZ2IW6ESI4H/action/author_attestation","sign_citation":"https://pith.science/pith/A5K2HDWUFIXNKOWPZ2IW6ESI4H/action/citation_signature","submit_replication":"https://pith.science/pith/A5K2HDWUFIXNKOWPZ2IW6ESI4H/action/replication_record"}},"created_at":"2026-05-18T03:10:12.177968+00:00","updated_at":"2026-05-18T03:10:12.177968+00:00"}