{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LC6CWMAYVROQYZB5FUALP55MGU","short_pith_number":"pith:LC6CWMAY","schema_version":"1.0","canonical_sha256":"58bc2b3018ac5d0c643d2d00b7f7ac351f892cdf341a8d5b9567282301f78766","source":{"kind":"arxiv","id":"2603.17863","version":2},"attestation_state":"computed","paper":{"title":"DiscoGen: Procedural Generation of Algorithm Discovery Tasks in Machine Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Adrian Hayler, Aleksandra Kalisz, Alexander D. Goldie, Alistair Letcher, Charles O'Neill, Clarisse Wibault, Deepak Nathani, Edan Toledo, Hannah Erlebach, Jakob N. Foerster, Ken Thampiratwong, Michael Beukman, Nicholas Roberts, Roberta Raileanu, Saeed Rahmani, Shashank Reddy, Shimon Whiteson, Theo Wolf, Uljad Berdica, Zilin Wang","submitted_at":"2026-03-18T15:49:32Z","abstract_excerpt":"Automating the development of machine learning algorithms has the potential to unlock new breakthroughs. However, our ability to improve and evaluate algorithm discovery systems has thus far been limited by existing task suites. They suffer from many issues, such as: poor evaluation methodologies; data contamination; and containing saturated or very similar problems. Here, we introduce DiscoGen, a procedural generator of algorithm discovery tasks for machine learning, such as developing optimisers for reinforcement learning or loss functions for image classification. Motivated by the success o"},"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":"2603.17863","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-03-18T15:49:32Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"39cbd8f5ef93dd932003e4c44acbcbaf3244ed1ddb7b9a9ced1f0f703b958ba1","abstract_canon_sha256":"96d37639d73c2fa8b40fd7c5d6d18795e796a9e747651d54319f8fef1491c3f2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T01:17:37.103853Z","signature_b64":"kQCGr37f6DZoGrFYo6CVCQA1QsTNx0xDNJV++7y2Bwa0jqY57U3HSU+Hbi8zeU4VgRezK0O40E6mAU+xRqIaCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"58bc2b3018ac5d0c643d2d00b7f7ac351f892cdf341a8d5b9567282301f78766","last_reissued_at":"2026-06-30T01:17:37.103118Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T01:17:37.103118Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DiscoGen: Procedural Generation of Algorithm Discovery Tasks in Machine Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Adrian Hayler, Aleksandra Kalisz, Alexander D. Goldie, Alistair Letcher, Charles O'Neill, Clarisse Wibault, Deepak Nathani, Edan Toledo, Hannah Erlebach, Jakob N. Foerster, Ken Thampiratwong, Michael Beukman, Nicholas Roberts, Roberta Raileanu, Saeed Rahmani, Shashank Reddy, Shimon Whiteson, Theo Wolf, Uljad Berdica, Zilin Wang","submitted_at":"2026-03-18T15:49:32Z","abstract_excerpt":"Automating the development of machine learning algorithms has the potential to unlock new breakthroughs. However, our ability to improve and evaluate algorithm discovery systems has thus far been limited by existing task suites. They suffer from many issues, such as: poor evaluation methodologies; data contamination; and containing saturated or very similar problems. Here, we introduce DiscoGen, a procedural generator of algorithm discovery tasks for machine learning, such as developing optimisers for reinforcement learning or loss functions for image classification. Motivated by the success o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2603.17863","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/2603.17863/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":"2603.17863","created_at":"2026-06-30T01:17:37.103216+00:00"},{"alias_kind":"arxiv_version","alias_value":"2603.17863v2","created_at":"2026-06-30T01:17:37.103216+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2603.17863","created_at":"2026-06-30T01:17:37.103216+00:00"},{"alias_kind":"pith_short_12","alias_value":"LC6CWMAYVROQ","created_at":"2026-06-30T01:17:37.103216+00:00"},{"alias_kind":"pith_short_16","alias_value":"LC6CWMAYVROQYZB5","created_at":"2026-06-30T01:17:37.103216+00:00"},{"alias_kind":"pith_short_8","alias_value":"LC6CWMAY","created_at":"2026-06-30T01:17:37.103216+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2606.25198","citing_title":"Heuresis: Search Strategies for Autonomous AI Research Agents Across Quality, Diversity and Novelty","ref_index":17,"is_internal_anchor":true},{"citing_arxiv_id":"2606.25198","citing_title":"Heuresis: Search Strategies for Autonomous AI Research Agents Across Quality, Diversity and Novelty","ref_index":17,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LC6CWMAYVROQYZB5FUALP55MGU","json":"https://pith.science/pith/LC6CWMAYVROQYZB5FUALP55MGU.json","graph_json":"https://pith.science/api/pith-number/LC6CWMAYVROQYZB5FUALP55MGU/graph.json","events_json":"https://pith.science/api/pith-number/LC6CWMAYVROQYZB5FUALP55MGU/events.json","paper":"https://pith.science/paper/LC6CWMAY"},"agent_actions":{"view_html":"https://pith.science/pith/LC6CWMAYVROQYZB5FUALP55MGU","download_json":"https://pith.science/pith/LC6CWMAYVROQYZB5FUALP55MGU.json","view_paper":"https://pith.science/paper/LC6CWMAY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2603.17863&json=true","fetch_graph":"https://pith.science/api/pith-number/LC6CWMAYVROQYZB5FUALP55MGU/graph.json","fetch_events":"https://pith.science/api/pith-number/LC6CWMAYVROQYZB5FUALP55MGU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LC6CWMAYVROQYZB5FUALP55MGU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LC6CWMAYVROQYZB5FUALP55MGU/action/storage_attestation","attest_author":"https://pith.science/pith/LC6CWMAYVROQYZB5FUALP55MGU/action/author_attestation","sign_citation":"https://pith.science/pith/LC6CWMAYVROQYZB5FUALP55MGU/action/citation_signature","submit_replication":"https://pith.science/pith/LC6CWMAYVROQYZB5FUALP55MGU/action/replication_record"}},"created_at":"2026-06-30T01:17:37.103216+00:00","updated_at":"2026-06-30T01:17:37.103216+00:00"}