{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:ATFTBZLXXY5EQJO4QD5DIIW2KQ","short_pith_number":"pith:ATFTBZLX","canonical_record":{"source":{"id":"2512.02551","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-12-02T09:20:15Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ad00fa6fe9fe6fcf025096a2c66c8ec8381423ee7d918eb8331cc0bfe86635d0","abstract_canon_sha256":"094b84eb51a6aec703ac377fcb82a8722c52654abba1eecaa92e8e83d95454e0"},"schema_version":"1.0"},"canonical_sha256":"04cb30e577be3a4825dc80fa3422da5405e42d1c72f0dd911608c521718e6fe2","source":{"kind":"arxiv","id":"2512.02551","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.02551","created_at":"2026-07-14T01:20:49Z"},{"alias_kind":"arxiv_version","alias_value":"2512.02551v3","created_at":"2026-07-14T01:20:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.02551","created_at":"2026-07-14T01:20:49Z"},{"alias_kind":"pith_short_12","alias_value":"ATFTBZLXXY5E","created_at":"2026-07-14T01:20:49Z"},{"alias_kind":"pith_short_16","alias_value":"ATFTBZLXXY5EQJO4","created_at":"2026-07-14T01:20:49Z"},{"alias_kind":"pith_short_8","alias_value":"ATFTBZLX","created_at":"2026-07-14T01:20:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:ATFTBZLXXY5EQJO4QD5DIIW2KQ","target":"record","payload":{"canonical_record":{"source":{"id":"2512.02551","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-12-02T09:20:15Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ad00fa6fe9fe6fcf025096a2c66c8ec8381423ee7d918eb8331cc0bfe86635d0","abstract_canon_sha256":"094b84eb51a6aec703ac377fcb82a8722c52654abba1eecaa92e8e83d95454e0"},"schema_version":"1.0"},"canonical_sha256":"04cb30e577be3a4825dc80fa3422da5405e42d1c72f0dd911608c521718e6fe2","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-14T01:20:49.283837Z","signature_b64":"FcK7Bcy6yFNfYWJJB32dLRrqJ6OW8fcG9IZuIyD575aC9EuOawHxkUCLavzslmsYD95zhkQpqcmAIN0N4h61Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"04cb30e577be3a4825dc80fa3422da5405e42d1c72f0dd911608c521718e6fe2","last_reissued_at":"2026-07-14T01:20:49.282920Z","signature_status":"signed_v1","first_computed_at":"2026-07-14T01:20:49.282920Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2512.02551","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-14T01:20:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tpUAGjqt4E4B1lMoLAz0GJJGulcMJ5WSYTdheaELW5d8seMP/XnYofoua9mYJqAfnmROg1r+YyGj1vAWiTzeBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:12:11.006580Z"},"content_sha256":"6e4ce32d59698d0a2627cc42d2bde324fb156f08a8572a41b7aa89bb3c4ad832","schema_version":"1.0","event_id":"sha256:6e4ce32d59698d0a2627cc42d2bde324fb156f08a8572a41b7aa89bb3c4ad832"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:ATFTBZLXXY5EQJO4QD5DIIW2KQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CUDA-L2: Surpassing cuBLAS Performance for Matrix Multiplication through Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Albert Wang, Chris Shum, Guoyin Wang, Jiwei Li, Songqiao Su, Xiaoya Li","submitted_at":"2025-12-02T09:20:15Z","abstract_excerpt":"In this paper, we propose CUDA-L2, a system that combines large language models (LLMs) and reinforcement learning (RL) to automatically optimize Half-precision General Matrix Multiply (HGEMM) CUDA kernels. Using CUDA execution speed as the RL reward, CUDA-L2 automatically optimizes HGEMM kernels across 1,000 configurations. CUDA-L2 systematically outperforms major matmul baselines to date, from the widely-used torch.matmul to state-of-the-art Nvidia's closed-source libraries, i.e., cuBLAS, cuBLASLt. In offline mode, where kernels are executed consecutively without time intervals, CUDA-L2 yield"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.02551","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/2512.02551/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-14T01:20:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zCTJe2MbJh0K9aVfP2zEdkbDqpRR/nOEyroWyXIW6yTJ8eBqhYxWXBQXT4fG3ExXC1ErK3NNsViOrnpgz9prCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:12:11.007166Z"},"content_sha256":"2dc8563abef078ac4203a9483d32ff46a00b4cd326164857f9581dbd489f17c9","schema_version":"1.0","event_id":"sha256:2dc8563abef078ac4203a9483d32ff46a00b4cd326164857f9581dbd489f17c9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/ATFTBZLXXY5EQJO4QD5DIIW2KQ/bundle.json","state_url":"https://pith.science/pith/ATFTBZLXXY5EQJO4QD5DIIW2KQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/ATFTBZLXXY5EQJO4QD5DIIW2KQ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-03T16:12:11Z","links":{"resolver":"https://pith.science/pith/ATFTBZLXXY5EQJO4QD5DIIW2KQ","bundle":"https://pith.science/pith/ATFTBZLXXY5EQJO4QD5DIIW2KQ/bundle.json","state":"https://pith.science/pith/ATFTBZLXXY5EQJO4QD5DIIW2KQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/ATFTBZLXXY5EQJO4QD5DIIW2KQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:ATFTBZLXXY5EQJO4QD5DIIW2KQ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"094b84eb51a6aec703ac377fcb82a8722c52654abba1eecaa92e8e83d95454e0","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-12-02T09:20:15Z","title_canon_sha256":"ad00fa6fe9fe6fcf025096a2c66c8ec8381423ee7d918eb8331cc0bfe86635d0"},"schema_version":"1.0","source":{"id":"2512.02551","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2512.02551","created_at":"2026-07-14T01:20:49Z"},{"alias_kind":"arxiv_version","alias_value":"2512.02551v3","created_at":"2026-07-14T01:20:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.02551","created_at":"2026-07-14T01:20:49Z"},{"alias_kind":"pith_short_12","alias_value":"ATFTBZLXXY5E","created_at":"2026-07-14T01:20:49Z"},{"alias_kind":"pith_short_16","alias_value":"ATFTBZLXXY5EQJO4","created_at":"2026-07-14T01:20:49Z"},{"alias_kind":"pith_short_8","alias_value":"ATFTBZLX","created_at":"2026-07-14T01:20:49Z"}],"graph_snapshots":[{"event_id":"sha256:2dc8563abef078ac4203a9483d32ff46a00b4cd326164857f9581dbd489f17c9","target":"graph","created_at":"2026-07-14T01:20:49Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2512.02551/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In this paper, we propose CUDA-L2, a system that combines large language models (LLMs) and reinforcement learning (RL) to automatically optimize Half-precision General Matrix Multiply (HGEMM) CUDA kernels. Using CUDA execution speed as the RL reward, CUDA-L2 automatically optimizes HGEMM kernels across 1,000 configurations. CUDA-L2 systematically outperforms major matmul baselines to date, from the widely-used torch.matmul to state-of-the-art Nvidia's closed-source libraries, i.e., cuBLAS, cuBLASLt. In offline mode, where kernels are executed consecutively without time intervals, CUDA-L2 yield","authors_text":"Albert Wang, Chris Shum, Guoyin Wang, Jiwei Li, Songqiao Su, Xiaoya Li","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-12-02T09:20:15Z","title":"CUDA-L2: Surpassing cuBLAS Performance for Matrix Multiplication through Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.02551","kind":"arxiv","version":3},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:6e4ce32d59698d0a2627cc42d2bde324fb156f08a8572a41b7aa89bb3c4ad832","target":"record","created_at":"2026-07-14T01:20:49Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"094b84eb51a6aec703ac377fcb82a8722c52654abba1eecaa92e8e83d95454e0","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-12-02T09:20:15Z","title_canon_sha256":"ad00fa6fe9fe6fcf025096a2c66c8ec8381423ee7d918eb8331cc0bfe86635d0"},"schema_version":"1.0","source":{"id":"2512.02551","kind":"arxiv","version":3}},"canonical_sha256":"04cb30e577be3a4825dc80fa3422da5405e42d1c72f0dd911608c521718e6fe2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"04cb30e577be3a4825dc80fa3422da5405e42d1c72f0dd911608c521718e6fe2","first_computed_at":"2026-07-14T01:20:49.282920Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-14T01:20:49.282920Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"FcK7Bcy6yFNfYWJJB32dLRrqJ6OW8fcG9IZuIyD575aC9EuOawHxkUCLavzslmsYD95zhkQpqcmAIN0N4h61Cw==","signature_status":"signed_v1","signed_at":"2026-07-14T01:20:49.283837Z","signed_message":"canonical_sha256_bytes"},"source_id":"2512.02551","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6e4ce32d59698d0a2627cc42d2bde324fb156f08a8572a41b7aa89bb3c4ad832","sha256:2dc8563abef078ac4203a9483d32ff46a00b4cd326164857f9581dbd489f17c9"],"state_sha256":"c885ca9c3bd0cece894baefd0fd8c6d931fbb93b662f8ae45c43405153ab5027"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Zw5kEIx2jvK0ERc/aaoaXi08d0cy7Uik6iG/2NLRKWU1PmL5CXGfk+o/42gt0SlE0RPoR2S7CtDG/f5sv6F7Aw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T16:12:11.011986Z","bundle_sha256":"6ca1cbdfc1cd16260f75925e50b48b1d139b7b7fe97c283856b194e0753746a2"}}