{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:X4UDHKOTRJV2T2EYWV2QNIPZ2V","short_pith_number":"pith:X4UDHKOT","schema_version":"1.0","canonical_sha256":"bf2833a9d38a6ba9e898b57506a1f9d547cff763f95b03ab24f0073fe7cc934c","source":{"kind":"arxiv","id":"2605.29357","version":1},"attestation_state":"computed","paper":{"title":"PassNet: Scaling Large Language Models for Graph Compiler Pass Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.PL"],"primary_cat":"cs.AI","authors_text":"Dongyan Chen, Enrong Zheng, Honglei Qiu, Jingjing Wu, Ruqi Yang, Sijun He, Siqi Bao, Tai Liang, Weihan Yi, XinQi Li, Yingsheng Wu, Yiqun Liu, Yiwei Zhang, Yuhan Zhou","submitted_at":"2026-05-28T04:55:14Z","abstract_excerpt":"Modern tensor compilers such as TorchInductor deliver substantial speedups on mainstream models, yet face a systematic performance ceiling on long-tail workloads -- our profiling shows that 43% of real-world subgraphs experience end-to-end slowdowns under default compilation. While LLMs offer a path toward automated optimization, existing efforts focus on standalone kernel generation. We argue that pass generation -- where LLMs author structured graph transformations that integrate directly into compiler pipelines -- is the more appropriate abstraction. We propose PassNet, the first large-scal"},"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":"2605.29357","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-28T04:55:14Z","cross_cats_sorted":["cs.LG","cs.PL"],"title_canon_sha256":"ccc8bca3df89934d6261724b7db6efbbbba660607e457d67d4d3c936f40ec52e","abstract_canon_sha256":"74edb85c2fcd121c542895e7786d57df64ae252b7a30604359a3fcd9264a7b02"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:34.906508Z","signature_b64":"dexJUGUwjkjJ9SJjJ1bD1rbWGM8t2x1hDa4Z+2Thh4QQL4jxz6LbUr9gRASa+vxTbDNKzDA2hTIZoGwIYGPjBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bf2833a9d38a6ba9e898b57506a1f9d547cff763f95b03ab24f0073fe7cc934c","last_reissued_at":"2026-05-29T01:05:34.905934Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:34.905934Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PassNet: Scaling Large Language Models for Graph Compiler Pass Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.PL"],"primary_cat":"cs.AI","authors_text":"Dongyan Chen, Enrong Zheng, Honglei Qiu, Jingjing Wu, Ruqi Yang, Sijun He, Siqi Bao, Tai Liang, Weihan Yi, XinQi Li, Yingsheng Wu, Yiqun Liu, Yiwei Zhang, Yuhan Zhou","submitted_at":"2026-05-28T04:55:14Z","abstract_excerpt":"Modern tensor compilers such as TorchInductor deliver substantial speedups on mainstream models, yet face a systematic performance ceiling on long-tail workloads -- our profiling shows that 43% of real-world subgraphs experience end-to-end slowdowns under default compilation. While LLMs offer a path toward automated optimization, existing efforts focus on standalone kernel generation. We argue that pass generation -- where LLMs author structured graph transformations that integrate directly into compiler pipelines -- is the more appropriate abstraction. We propose PassNet, the first large-scal"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29357","kind":"arxiv","version":1},"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/2605.29357/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":"2605.29357","created_at":"2026-05-29T01:05:34.905999+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.29357v1","created_at":"2026-05-29T01:05:34.905999+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29357","created_at":"2026-05-29T01:05:34.905999+00:00"},{"alias_kind":"pith_short_12","alias_value":"X4UDHKOTRJV2","created_at":"2026-05-29T01:05:34.905999+00:00"},{"alias_kind":"pith_short_16","alias_value":"X4UDHKOTRJV2T2EY","created_at":"2026-05-29T01:05:34.905999+00:00"},{"alias_kind":"pith_short_8","alias_value":"X4UDHKOT","created_at":"2026-05-29T01:05:34.905999+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/X4UDHKOTRJV2T2EYWV2QNIPZ2V","json":"https://pith.science/pith/X4UDHKOTRJV2T2EYWV2QNIPZ2V.json","graph_json":"https://pith.science/api/pith-number/X4UDHKOTRJV2T2EYWV2QNIPZ2V/graph.json","events_json":"https://pith.science/api/pith-number/X4UDHKOTRJV2T2EYWV2QNIPZ2V/events.json","paper":"https://pith.science/paper/X4UDHKOT"},"agent_actions":{"view_html":"https://pith.science/pith/X4UDHKOTRJV2T2EYWV2QNIPZ2V","download_json":"https://pith.science/pith/X4UDHKOTRJV2T2EYWV2QNIPZ2V.json","view_paper":"https://pith.science/paper/X4UDHKOT","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.29357&json=true","fetch_graph":"https://pith.science/api/pith-number/X4UDHKOTRJV2T2EYWV2QNIPZ2V/graph.json","fetch_events":"https://pith.science/api/pith-number/X4UDHKOTRJV2T2EYWV2QNIPZ2V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/X4UDHKOTRJV2T2EYWV2QNIPZ2V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/X4UDHKOTRJV2T2EYWV2QNIPZ2V/action/storage_attestation","attest_author":"https://pith.science/pith/X4UDHKOTRJV2T2EYWV2QNIPZ2V/action/author_attestation","sign_citation":"https://pith.science/pith/X4UDHKOTRJV2T2EYWV2QNIPZ2V/action/citation_signature","submit_replication":"https://pith.science/pith/X4UDHKOTRJV2T2EYWV2QNIPZ2V/action/replication_record"}},"created_at":"2026-05-29T01:05:34.905999+00:00","updated_at":"2026-05-29T01:05:34.905999+00:00"}