{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6Y7JFZ3WDEBJUE2RTXGU5Q6VM2","short_pith_number":"pith:6Y7JFZ3W","schema_version":"1.0","canonical_sha256":"f63e92e77619029a13519dcd4ec3d566a484078111627f6ecf932c4897236a9b","source":{"kind":"arxiv","id":"2505.03906","version":3},"attestation_state":"computed","paper":{"title":"MARCO: Multi-Agent Code Optimization with Real-Time Knowledge Integration for High-Performance Computing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.SE"],"primary_cat":"cs.DC","authors_text":"Aidan Walters, Aneesh Tummeti, Asif Rahman, Bryan Torres, Denise Cooney, Dimitrios S. Nikolopoulos, Kathleen Reece, Margaret Ellis, Veljko Cvetkovic, Yasir Hassan","submitted_at":"2025-05-06T18:22:38Z","abstract_excerpt":"Large language models (LLMs) have transformed software development through code generation capabilities, yet their effectiveness for high-performance computing (HPC) remains limited. HPC code requires specialized optimizations for parallelism, memory efficiency, and architecture-specific considerations that general-purpose LLMs often overlook. We present MARCO (Multi-Agent Reactive Code Optimizer), a novel framework that enhances LLM-generated code for HPC through a specialized multi-agent architecture. MARCO employs separate agents for code generation and performance evaluation, connected by "},"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":"2505.03906","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.DC","submitted_at":"2025-05-06T18:22:38Z","cross_cats_sorted":["cs.LG","cs.SE"],"title_canon_sha256":"ae5d55aaf6e4941bfd52d304a44a355ac3ce9ecd54bdf7943996071506856af4","abstract_canon_sha256":"3de3db0edc514abc346b3498422af6fc5603af60dca95f74cfa94159efb98fd3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:26:51.003637Z","signature_b64":"p+6hVQd0pdzaENW9sH9eDOiR2Glt0cXHU//8lPY9I8kdS1PgGSspIQTMUdtv1HOesjTq9MTUB9o2+qnVTg4WDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f63e92e77619029a13519dcd4ec3d566a484078111627f6ecf932c4897236a9b","last_reissued_at":"2026-07-05T11:26:51.003141Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:26:51.003141Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MARCO: Multi-Agent Code Optimization with Real-Time Knowledge Integration for High-Performance Computing","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG","cs.SE"],"primary_cat":"cs.DC","authors_text":"Aidan Walters, Aneesh Tummeti, Asif Rahman, Bryan Torres, Denise Cooney, Dimitrios S. Nikolopoulos, Kathleen Reece, Margaret Ellis, Veljko Cvetkovic, Yasir Hassan","submitted_at":"2025-05-06T18:22:38Z","abstract_excerpt":"Large language models (LLMs) have transformed software development through code generation capabilities, yet their effectiveness for high-performance computing (HPC) remains limited. HPC code requires specialized optimizations for parallelism, memory efficiency, and architecture-specific considerations that general-purpose LLMs often overlook. We present MARCO (Multi-Agent Reactive Code Optimizer), a novel framework that enhances LLM-generated code for HPC through a specialized multi-agent architecture. MARCO employs separate agents for code generation and performance evaluation, connected by "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.03906","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/2505.03906/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":"2505.03906","created_at":"2026-07-05T11:26:51.003199+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.03906v3","created_at":"2026-07-05T11:26:51.003199+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.03906","created_at":"2026-07-05T11:26:51.003199+00:00"},{"alias_kind":"pith_short_12","alias_value":"6Y7JFZ3WDEBJ","created_at":"2026-07-05T11:26:51.003199+00:00"},{"alias_kind":"pith_short_16","alias_value":"6Y7JFZ3WDEBJUE2R","created_at":"2026-07-05T11:26:51.003199+00:00"},{"alias_kind":"pith_short_8","alias_value":"6Y7JFZ3W","created_at":"2026-07-05T11:26:51.003199+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.29372","citing_title":"On the Road to Personalized Code Intelligence: Portraiting and Assisting Developers Based on Their In-IDE Behaviors","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2506.01249","citing_title":"SysLLMatic: Large Language Models are Software System Optimizers","ref_index":76,"is_internal_anchor":false},{"citing_arxiv_id":"2601.12538","citing_title":"Agentic Reasoning for Large Language Models","ref_index":100,"is_internal_anchor":false},{"citing_arxiv_id":"2508.07407","citing_title":"A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systems","ref_index":76,"is_internal_anchor":false},{"citing_arxiv_id":"2511.20857","citing_title":"Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory","ref_index":235,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6Y7JFZ3WDEBJUE2RTXGU5Q6VM2","json":"https://pith.science/pith/6Y7JFZ3WDEBJUE2RTXGU5Q6VM2.json","graph_json":"https://pith.science/api/pith-number/6Y7JFZ3WDEBJUE2RTXGU5Q6VM2/graph.json","events_json":"https://pith.science/api/pith-number/6Y7JFZ3WDEBJUE2RTXGU5Q6VM2/events.json","paper":"https://pith.science/paper/6Y7JFZ3W"},"agent_actions":{"view_html":"https://pith.science/pith/6Y7JFZ3WDEBJUE2RTXGU5Q6VM2","download_json":"https://pith.science/pith/6Y7JFZ3WDEBJUE2RTXGU5Q6VM2.json","view_paper":"https://pith.science/paper/6Y7JFZ3W","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.03906&json=true","fetch_graph":"https://pith.science/api/pith-number/6Y7JFZ3WDEBJUE2RTXGU5Q6VM2/graph.json","fetch_events":"https://pith.science/api/pith-number/6Y7JFZ3WDEBJUE2RTXGU5Q6VM2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6Y7JFZ3WDEBJUE2RTXGU5Q6VM2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6Y7JFZ3WDEBJUE2RTXGU5Q6VM2/action/storage_attestation","attest_author":"https://pith.science/pith/6Y7JFZ3WDEBJUE2RTXGU5Q6VM2/action/author_attestation","sign_citation":"https://pith.science/pith/6Y7JFZ3WDEBJUE2RTXGU5Q6VM2/action/citation_signature","submit_replication":"https://pith.science/pith/6Y7JFZ3WDEBJUE2RTXGU5Q6VM2/action/replication_record"}},"created_at":"2026-07-05T11:26:51.003199+00:00","updated_at":"2026-07-05T11:26:51.003199+00:00"}