{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PJWOFGZQOL3UJJ2VLGM54AWC42","short_pith_number":"pith:PJWOFGZQ","schema_version":"1.0","canonical_sha256":"7a6ce29b3072f744a7555999de02c2e6ab86d99cd3a1af38d99b74759dd35c8c","source":{"kind":"arxiv","id":"2504.05108","version":4},"attestation_state":"computed","paper":{"title":"Algorithm Discovery With LLMs: Evolutionary Search Meets Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.NE"],"primary_cat":"cs.AI","authors_text":"Amal Seddas, Amin Mansouri, Anja Surina, Caglar Gulcehre, Emmanuel Abbe, Lars Quaedvlieg, Maryna Viazovska","submitted_at":"2025-04-07T14:14:15Z","abstract_excerpt":"Discovering efficient algorithms for solving complex problems has been an outstanding challenge in mathematics and computer science, requiring substantial human expertise over the years. Recent advancements in evolutionary search with large language models (LLMs) have shown promise in accelerating the discovery of algorithms across various domains, particularly in mathematics and optimization. However, existing approaches treat the LLM as a static generator, missing the opportunity to update the model with the signal obtained from evolutionary exploration. In this work, we propose to augment L"},"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":"2504.05108","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-04-07T14:14:15Z","cross_cats_sorted":["cs.LG","cs.NE"],"title_canon_sha256":"bbd190d2a31d4718e116ab2c8827331980a85f5c496f7ebfdfcecc827f6af163","abstract_canon_sha256":"0d0389c6ed277f71650a903014968127d6d066b28a867f198853b4016b00817a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T02:02:59.824798Z","signature_b64":"/3LQZCoQDflWbBj8Bm+D/MuAnV7Tjuze7PYRAIunkiUG0GU54KYqHDTK9k8YGsenDcT+PrmarPesHjjh/YiwBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7a6ce29b3072f744a7555999de02c2e6ab86d99cd3a1af38d99b74759dd35c8c","last_reissued_at":"2026-05-26T02:02:59.823785Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T02:02:59.823785Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Algorithm Discovery With LLMs: Evolutionary Search Meets Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.NE"],"primary_cat":"cs.AI","authors_text":"Amal Seddas, Amin Mansouri, Anja Surina, Caglar Gulcehre, Emmanuel Abbe, Lars Quaedvlieg, Maryna Viazovska","submitted_at":"2025-04-07T14:14:15Z","abstract_excerpt":"Discovering efficient algorithms for solving complex problems has been an outstanding challenge in mathematics and computer science, requiring substantial human expertise over the years. Recent advancements in evolutionary search with large language models (LLMs) have shown promise in accelerating the discovery of algorithms across various domains, particularly in mathematics and optimization. However, existing approaches treat the LLM as a static generator, missing the opportunity to update the model with the signal obtained from evolutionary exploration. In this work, we propose to augment L"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.05108","kind":"arxiv","version":4},"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/2504.05108/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":"2504.05108","created_at":"2026-05-26T02:02:59.823933+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.05108v4","created_at":"2026-05-26T02:02:59.823933+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.05108","created_at":"2026-05-26T02:02:59.823933+00:00"},{"alias_kind":"pith_short_12","alias_value":"PJWOFGZQOL3U","created_at":"2026-05-26T02:02:59.823933+00:00"},{"alias_kind":"pith_short_16","alias_value":"PJWOFGZQOL3UJJ2V","created_at":"2026-05-26T02:02:59.823933+00:00"},{"alias_kind":"pith_short_8","alias_value":"PJWOFGZQ","created_at":"2026-05-26T02:02:59.823933+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":10,"sample":[{"citing_arxiv_id":"2605.18004","citing_title":"RL4RLA: Teaching ML to Discover Randomized Linear Algebra Algorithms Through Curriculum Design and Graph-Based Search","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2510.16978","citing_title":"Lark: Biologically Inspired Neuroevolution for Multi-Stakeholder LLM Agents","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2601.16175","citing_title":"Learning to Discover at Test Time","ref_index":69,"is_internal_anchor":true},{"citing_arxiv_id":"2604.02721","citing_title":"GrandCode: Achieving Grandmaster Level in Competitive Programming via Agentic Reinforcement Learning","ref_index":28,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08678","citing_title":"MLS-Bench: A Holistic and Rigorous Assessment of AI Systems on Building Better AI","ref_index":90,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08756","citing_title":"AHD Agent: Agentic Reinforcement Learning for Automatic Heuristic Design","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10598","citing_title":"Budget-Efficient Automatic Algorithm Design via Code Graph","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06123","citing_title":"Back to the Beginning of Heuristic Design: Bridging Code and Knowledge with LLMs","ref_index":54,"is_internal_anchor":true},{"citing_arxiv_id":"2604.07240","citing_title":"$k$-server-bench: Automating Potential Discovery for the $k$-Server Conjecture","ref_index":39,"is_internal_anchor":true},{"citing_arxiv_id":"2506.13131","citing_title":"AlphaEvolve: A coding agent for scientific and algorithmic discovery","ref_index":98,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PJWOFGZQOL3UJJ2VLGM54AWC42","json":"https://pith.science/pith/PJWOFGZQOL3UJJ2VLGM54AWC42.json","graph_json":"https://pith.science/api/pith-number/PJWOFGZQOL3UJJ2VLGM54AWC42/graph.json","events_json":"https://pith.science/api/pith-number/PJWOFGZQOL3UJJ2VLGM54AWC42/events.json","paper":"https://pith.science/paper/PJWOFGZQ"},"agent_actions":{"view_html":"https://pith.science/pith/PJWOFGZQOL3UJJ2VLGM54AWC42","download_json":"https://pith.science/pith/PJWOFGZQOL3UJJ2VLGM54AWC42.json","view_paper":"https://pith.science/paper/PJWOFGZQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.05108&json=true","fetch_graph":"https://pith.science/api/pith-number/PJWOFGZQOL3UJJ2VLGM54AWC42/graph.json","fetch_events":"https://pith.science/api/pith-number/PJWOFGZQOL3UJJ2VLGM54AWC42/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PJWOFGZQOL3UJJ2VLGM54AWC42/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PJWOFGZQOL3UJJ2VLGM54AWC42/action/storage_attestation","attest_author":"https://pith.science/pith/PJWOFGZQOL3UJJ2VLGM54AWC42/action/author_attestation","sign_citation":"https://pith.science/pith/PJWOFGZQOL3UJJ2VLGM54AWC42/action/citation_signature","submit_replication":"https://pith.science/pith/PJWOFGZQOL3UJJ2VLGM54AWC42/action/replication_record"}},"created_at":"2026-05-26T02:02:59.823933+00:00","updated_at":"2026-05-26T02:02:59.823933+00:00"}