{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:IHYKQSPUC7GCJSO6PASQX3AHNU","short_pith_number":"pith:IHYKQSPU","schema_version":"1.0","canonical_sha256":"41f0a849f417cc24c9de78250bec076d295db158292575f75ffc5f6430299077","source":{"kind":"arxiv","id":"2605.29734","version":1},"attestation_state":"computed","paper":{"title":"HTAM: Hierarchical Transition-Attended Memory for Operator Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chengqing Zong, Chen Wang, Mingyang Yi, Tianhe Jia, Xuwen Xiang, Yining Zhang, Yue Wang, Zedong Dan","submitted_at":"2026-05-28T10:29:01Z","abstract_excerpt":"High-performance GPU kernels are essential for efficient LLM deployment, yet optimizing them remains expertise-intensive. Recent LLM-based code generation makes automatic GPU operator generation promising, but operator optimization remains a hardware-aware search problem. Existing LLM-based methods face a granularity mismatch: coarse hints are reusable but hard to execute, whereas detailed memories are actionable but enlarge the search space and obscure optimization bottlenecks. The key challenge is therefore to organize optimization experience at an appropriate granularity. To address this is"},"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":"2605.29734","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-28T10:29:01Z","cross_cats_sorted":[],"title_canon_sha256":"119f47fd213716912b09209324b53aab96fc02e3175311ac1725a1fbee047740","abstract_canon_sha256":"b5b282375784d1c596bf4174367b12fe9f606ec585d7ec7efd77f280b6f48162"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:57.325381Z","signature_b64":"ttA0g3dxgtHkSTUGhj6cQCrMAQu/esD5EGPxtkBTA8ue2ybVL5c09F9rr0Z646QyozVnrRYCW8MDfmzRLzSTAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"41f0a849f417cc24c9de78250bec076d295db158292575f75ffc5f6430299077","last_reissued_at":"2026-05-29T01:05:57.324799Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:57.324799Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"HTAM: Hierarchical Transition-Attended Memory for Operator Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chengqing Zong, Chen Wang, Mingyang Yi, Tianhe Jia, Xuwen Xiang, Yining Zhang, Yue Wang, Zedong Dan","submitted_at":"2026-05-28T10:29:01Z","abstract_excerpt":"High-performance GPU kernels are essential for efficient LLM deployment, yet optimizing them remains expertise-intensive. Recent LLM-based code generation makes automatic GPU operator generation promising, but operator optimization remains a hardware-aware search problem. Existing LLM-based methods face a granularity mismatch: coarse hints are reusable but hard to execute, whereas detailed memories are actionable but enlarge the search space and obscure optimization bottlenecks. The key challenge is therefore to organize optimization experience at an appropriate granularity. To address this is"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29734","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/2605.29734/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":"2605.29734","created_at":"2026-05-29T01:05:57.324888+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.29734v1","created_at":"2026-05-29T01:05:57.324888+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29734","created_at":"2026-05-29T01:05:57.324888+00:00"},{"alias_kind":"pith_short_12","alias_value":"IHYKQSPUC7GC","created_at":"2026-05-29T01:05:57.324888+00:00"},{"alias_kind":"pith_short_16","alias_value":"IHYKQSPUC7GCJSO6","created_at":"2026-05-29T01:05:57.324888+00:00"},{"alias_kind":"pith_short_8","alias_value":"IHYKQSPU","created_at":"2026-05-29T01:05:57.324888+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/IHYKQSPUC7GCJSO6PASQX3AHNU","json":"https://pith.science/pith/IHYKQSPUC7GCJSO6PASQX3AHNU.json","graph_json":"https://pith.science/api/pith-number/IHYKQSPUC7GCJSO6PASQX3AHNU/graph.json","events_json":"https://pith.science/api/pith-number/IHYKQSPUC7GCJSO6PASQX3AHNU/events.json","paper":"https://pith.science/paper/IHYKQSPU"},"agent_actions":{"view_html":"https://pith.science/pith/IHYKQSPUC7GCJSO6PASQX3AHNU","download_json":"https://pith.science/pith/IHYKQSPUC7GCJSO6PASQX3AHNU.json","view_paper":"https://pith.science/paper/IHYKQSPU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.29734&json=true","fetch_graph":"https://pith.science/api/pith-number/IHYKQSPUC7GCJSO6PASQX3AHNU/graph.json","fetch_events":"https://pith.science/api/pith-number/IHYKQSPUC7GCJSO6PASQX3AHNU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IHYKQSPUC7GCJSO6PASQX3AHNU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IHYKQSPUC7GCJSO6PASQX3AHNU/action/storage_attestation","attest_author":"https://pith.science/pith/IHYKQSPUC7GCJSO6PASQX3AHNU/action/author_attestation","sign_citation":"https://pith.science/pith/IHYKQSPUC7GCJSO6PASQX3AHNU/action/citation_signature","submit_replication":"https://pith.science/pith/IHYKQSPUC7GCJSO6PASQX3AHNU/action/replication_record"}},"created_at":"2026-05-29T01:05:57.324888+00:00","updated_at":"2026-05-29T01:05:57.324888+00:00"}