{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:OACF7TTEL4J6XRHQ75CWLCFMCW","short_pith_number":"pith:OACF7TTE","schema_version":"1.0","canonical_sha256":"70045fce645f13ebc4f0ff456588ac15b4329994dd73592df272c85b798fcddf","source":{"kind":"arxiv","id":"2310.02304","version":3},"attestation_state":"computed","paper":{"title":"Self-Taught Optimizer (STOP): Recursively Self-Improving Code Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Adam Tauman Kalai, Eliana Lorch, Eric Zelikman, Lester Mackey","submitted_at":"2023-10-03T17:59:32Z","abstract_excerpt":"Several recent advances in AI systems solve problems by providing a \"scaffolding\" program that structures multiple calls to language models (LMs) to generate better outputs. A scaffolding program is written in a programming language such as Python. In this work, we use a language-model-infused scaffolding program to improve itself. We start with a seed \"improver\" that improves an input program according to a given utility function by querying an LM several times and returning the best solution. We then run this seed improver to improve itself. Across a small set of downstream tasks, the result"},"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":"2310.02304","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-03T17:59:32Z","cross_cats_sorted":["cs.AI","cs.LG","stat.ML"],"title_canon_sha256":"3ee11214715971a47617d0b4767d09541b8287f18ad5645deda1d4671cb75e39","abstract_canon_sha256":"e5cbb3ecc55c1c27c3386ba319c298baa7d1d5e4ab179f6562a7763f856650a2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:56:02.523693Z","signature_b64":"gFbeKsog94hVSCWeQTvprl1fpS2+q4IWi3HGq2So95b7fH8HjQm2h0a2fMF7nGeJl2ZGD3stCb/0NxVs2Vw6BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"70045fce645f13ebc4f0ff456588ac15b4329994dd73592df272c85b798fcddf","last_reissued_at":"2026-07-05T08:56:02.523206Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:56:02.523206Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Self-Taught Optimizer (STOP): Recursively Self-Improving Code Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG","stat.ML"],"primary_cat":"cs.CL","authors_text":"Adam Tauman Kalai, Eliana Lorch, Eric Zelikman, Lester Mackey","submitted_at":"2023-10-03T17:59:32Z","abstract_excerpt":"Several recent advances in AI systems solve problems by providing a \"scaffolding\" program that structures multiple calls to language models (LMs) to generate better outputs. A scaffolding program is written in a programming language such as Python. In this work, we use a language-model-infused scaffolding program to improve itself. We start with a seed \"improver\" that improves an input program according to a given utility function by querying an LM several times and returning the best solution. We then run this seed improver to improve itself. Across a small set of downstream tasks, the result"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.02304","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/2310.02304/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":"2310.02304","created_at":"2026-07-05T08:56:02.523265+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.02304v3","created_at":"2026-07-05T08:56:02.523265+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.02304","created_at":"2026-07-05T08:56:02.523265+00:00"},{"alias_kind":"pith_short_12","alias_value":"OACF7TTEL4J6","created_at":"2026-07-05T08:56:02.523265+00:00"},{"alias_kind":"pith_short_16","alias_value":"OACF7TTEL4J6XRHQ","created_at":"2026-07-05T08:56:02.523265+00:00"},{"alias_kind":"pith_short_8","alias_value":"OACF7TTE","created_at":"2026-07-05T08:56:02.523265+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.08124","citing_title":"TTHE: Test-Time Harness Evolution","ref_index":76,"is_internal_anchor":true},{"citing_arxiv_id":"2606.26294","citing_title":"The Red Queen G\\\"odel Machine: Co-Evolving Agents and Their Evaluators","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2606.20475","citing_title":"Marginal Advantage Accumulation for Memory-Driven Agent Self-Evolution","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09498","citing_title":"Self-Harness: Harnesses That Improve Themselves","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2606.26294","citing_title":"The Red Queen G\\\"odel Machine: Co-Evolving Agents and Their Evaluators","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OACF7TTEL4J6XRHQ75CWLCFMCW","json":"https://pith.science/pith/OACF7TTEL4J6XRHQ75CWLCFMCW.json","graph_json":"https://pith.science/api/pith-number/OACF7TTEL4J6XRHQ75CWLCFMCW/graph.json","events_json":"https://pith.science/api/pith-number/OACF7TTEL4J6XRHQ75CWLCFMCW/events.json","paper":"https://pith.science/paper/OACF7TTE"},"agent_actions":{"view_html":"https://pith.science/pith/OACF7TTEL4J6XRHQ75CWLCFMCW","download_json":"https://pith.science/pith/OACF7TTEL4J6XRHQ75CWLCFMCW.json","view_paper":"https://pith.science/paper/OACF7TTE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.02304&json=true","fetch_graph":"https://pith.science/api/pith-number/OACF7TTEL4J6XRHQ75CWLCFMCW/graph.json","fetch_events":"https://pith.science/api/pith-number/OACF7TTEL4J6XRHQ75CWLCFMCW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OACF7TTEL4J6XRHQ75CWLCFMCW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OACF7TTEL4J6XRHQ75CWLCFMCW/action/storage_attestation","attest_author":"https://pith.science/pith/OACF7TTEL4J6XRHQ75CWLCFMCW/action/author_attestation","sign_citation":"https://pith.science/pith/OACF7TTEL4J6XRHQ75CWLCFMCW/action/citation_signature","submit_replication":"https://pith.science/pith/OACF7TTEL4J6XRHQ75CWLCFMCW/action/replication_record"}},"created_at":"2026-07-05T08:56:02.523265+00:00","updated_at":"2026-07-05T08:56:02.523265+00:00"}