{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:RWP6ZAGIGZC6XQUHGSPBSN5AXW","short_pith_number":"pith:RWP6ZAGI","schema_version":"1.0","canonical_sha256":"8d9fec80c83645ebc287349e1937a0bd996b9d56ba75882dc6fd957a1b5299d4","source":{"kind":"arxiv","id":"2509.26476","version":2},"attestation_state":"computed","paper":{"title":"Regression Language Models for Code","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.PF","cs.SE"],"primary_cat":"cs.CL","authors_text":"Arissa Wongpanich, Bryan Lewandowski, Mohamed S. Abdelfattah, Xingyou Song, Yash Akhauri","submitted_at":"2025-09-30T16:25:23Z","abstract_excerpt":"We study code-to-metric regression: predicting numeric outcomes of code executions, a challenging task due to the open-ended nature of programming languages. While prior methods have resorted to heavy and domain-specific feature engineering, we show that a single unified Regression Language Model (RLM) using a frozen LLM encoder can simultaneously predict directly from text, (i) the memory footprint of code across multiple high-level languages such as Python and C++, (ii) the latency of Triton GPU kernels, and (iii) the accuracy and speed of trained neural networks represented in ONNX. In part"},"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":"2509.26476","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-09-30T16:25:23Z","cross_cats_sorted":["cs.AI","cs.LG","cs.PF","cs.SE"],"title_canon_sha256":"cb022bc415e997db5943ff47e2f2e6f5e2825fac6f129642b6a249672aa3d0c0","abstract_canon_sha256":"4499ef3d2325b96170010c8fc7d9956b2c90f3cd886696c9d3f4831960cbb371"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T02:04:43.174600Z","signature_b64":"KAZAJ6ZThybuEWTJXk++w6mnB9HtETYOchgM39L9nsnXBHOiawh/a5k2AgRJ1gL7PU3wgROot33UnvPMdF/LAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8d9fec80c83645ebc287349e1937a0bd996b9d56ba75882dc6fd957a1b5299d4","last_reissued_at":"2026-05-28T02:04:43.173808Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T02:04:43.173808Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Regression Language Models for Code","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","cs.PF","cs.SE"],"primary_cat":"cs.CL","authors_text":"Arissa Wongpanich, Bryan Lewandowski, Mohamed S. Abdelfattah, Xingyou Song, Yash Akhauri","submitted_at":"2025-09-30T16:25:23Z","abstract_excerpt":"We study code-to-metric regression: predicting numeric outcomes of code executions, a challenging task due to the open-ended nature of programming languages. While prior methods have resorted to heavy and domain-specific feature engineering, we show that a single unified Regression Language Model (RLM) using a frozen LLM encoder can simultaneously predict directly from text, (i) the memory footprint of code across multiple high-level languages such as Python and C++, (ii) the latency of Triton GPU kernels, and (iii) the accuracy and speed of trained neural networks represented in ONNX. In part"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.26476","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/2509.26476/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":"2509.26476","created_at":"2026-05-28T02:04:43.173912+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.26476v2","created_at":"2026-05-28T02:04:43.173912+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.26476","created_at":"2026-05-28T02:04:43.173912+00:00"},{"alias_kind":"pith_short_12","alias_value":"RWP6ZAGIGZC6","created_at":"2026-05-28T02:04:43.173912+00:00"},{"alias_kind":"pith_short_16","alias_value":"RWP6ZAGIGZC6XQUH","created_at":"2026-05-28T02:04:43.173912+00:00"},{"alias_kind":"pith_short_8","alias_value":"RWP6ZAGI","created_at":"2026-05-28T02:04:43.173912+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2512.19701","citing_title":"LASER: Language Model Regression for Semi-Structured Workflow Resource and Runtime Estimation","ref_index":1,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20740","citing_title":"Distribution-Aware Reward: Reinforcement Learning over Predictive Distributions for LLM Regression","ref_index":20,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RWP6ZAGIGZC6XQUHGSPBSN5AXW","json":"https://pith.science/pith/RWP6ZAGIGZC6XQUHGSPBSN5AXW.json","graph_json":"https://pith.science/api/pith-number/RWP6ZAGIGZC6XQUHGSPBSN5AXW/graph.json","events_json":"https://pith.science/api/pith-number/RWP6ZAGIGZC6XQUHGSPBSN5AXW/events.json","paper":"https://pith.science/paper/RWP6ZAGI"},"agent_actions":{"view_html":"https://pith.science/pith/RWP6ZAGIGZC6XQUHGSPBSN5AXW","download_json":"https://pith.science/pith/RWP6ZAGIGZC6XQUHGSPBSN5AXW.json","view_paper":"https://pith.science/paper/RWP6ZAGI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.26476&json=true","fetch_graph":"https://pith.science/api/pith-number/RWP6ZAGIGZC6XQUHGSPBSN5AXW/graph.json","fetch_events":"https://pith.science/api/pith-number/RWP6ZAGIGZC6XQUHGSPBSN5AXW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RWP6ZAGIGZC6XQUHGSPBSN5AXW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RWP6ZAGIGZC6XQUHGSPBSN5AXW/action/storage_attestation","attest_author":"https://pith.science/pith/RWP6ZAGIGZC6XQUHGSPBSN5AXW/action/author_attestation","sign_citation":"https://pith.science/pith/RWP6ZAGIGZC6XQUHGSPBSN5AXW/action/citation_signature","submit_replication":"https://pith.science/pith/RWP6ZAGIGZC6XQUHGSPBSN5AXW/action/replication_record"}},"created_at":"2026-05-28T02:04:43.173912+00:00","updated_at":"2026-05-28T02:04:43.173912+00:00"}