{"as_of":"2026-08-17T20:40:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:6863222358295ca0c711b6211e9ee39c6f13b35a973c96dea21dd20b56478840","coverage":[{"denominator":44,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":44,"source":"paper_references, paper_reference_links","source_observed_at":"2026-05-21T10:16:15.305706Z","state":"measured"},{"denominator":46,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":46,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-17T06:30:58.91139+00:00","state":"measured"},{"denominator":2,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":2,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-02T08:11:51.592817Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2603.23566","snapshot_observed_at":"2026-08-01T10:42:40.072916Z","title":"CoRR , volume =","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.20145","last_updated":"2026-07-30T09:30:44Z","snapshot_observed_at":"2026-08-15T12:14:51.147122Z","submitted_at":"2026-07-22T13:49:17Z","title":"SLAI T-Rex: Full-Parameter Post-training of the DeepSeek-V4 Family on Ascend SuperPOD","version":2},"reference_index":46,"source":"arxiv_source","source_observed_at":"2026-08-01T10:42:40.072916Z"},"links":{"cited_paper":"/paper/2603.23566","citing_paper":"/paper/2607.20145"},"observation_digest":"sha256:5505961792ecf297d262e4b701f84d2b9629245013f3eb4d04f063d8395c61ec","observation_id":"8f70b291-131c-4bcb-af34-b4b1ede46179","resolution":{"observed_at":"2026-08-01T10:42:40.072916Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2603.23566","snapshot_observed_at":"2026-08-02T08:11:51.592817Z","title":"AscendOp- timizer: Episodic agent for Ascend NPU operator optimization.arXiv preprint arXiv:2603.23566, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.20518","last_updated":"2026-07-08T02:22:22Z","snapshot_observed_at":"2026-08-14T17:25:19.834750Z","submitted_at":"2026-07-08T02:22:22Z","title":"CANN Bench: Benchmarking Agent Generated Kernels against Real NPU and Algorithmic Limits","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-02T08:11:51.592817Z"},"links":{"cited_paper":"/paper/2603.23566","citing_paper":"/paper/2607.20518"},"observation_digest":"sha256:79a20bc875db0df30d3373e1ccda2ca026d91c500232c080266d15dca5be69f7","observation_id":"1a3f8c0d-0bf6-4022-8c4f-df6d6299f2f4","resolution":{"observed_at":"2026-08-02T08:11:51.592817Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2603.23566/citation-record","integrity":"/paper/2603.23566/integrity","json":"/paper/2603.23566/citation-record.json","paper":"/paper/2603.23566"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"NeutronAscend: Op- timizing GNN training with Ascend AI processors.ACM Transactions on Architecture and Code Optimization, 22(4):1–26","venue":null,"work_id":"aa60e114-2717-4215-a0a8-45ee12bf2497","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:59fc3b6d781c80fcdcc5deaab8ee7173236f6cf1729a807712a9c9f5d23b7d1d","observation_id":"e9d11abd-f7da-4dd9-b7df-52ae006db54d","resolution":{"observed_at":"2026-05-21T10:20:00.703695Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2506.20807","last_updated":"2025-08-22T08:48:16Z","snapshot_observed_at":"2026-08-11T23:09:56.467501Z","submitted_at":"2025-06-25T19:59:34Z","title":"GPU Kernel Scientist: An LLM-Driven Framework for Iterative Kernel Optimization","version":2},"cited_work":{"arxiv_id":"2506.20807","doi":"10.48550/arxiv.2506.20807","metadata_source":"arxiv_reference","pith_arxiv_id":"2506.20807","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"GPU Kernel Scientist: An LLM-driven framework for iterative kernel optimization","venue":"ArXiv.org","work_id":"f94889ab-c418-4636-8b23-e86bffaa1609","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"cited_paper":"/paper/2506.20807","citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:8090724dce17b11605792c6cba0c28efb46332204002a007e31fb485f2269293","observation_id":"8c322bb4-a551-431a-91c3-0f9d0beb48d0","resolution":{"observed_at":"2026-05-21T10:20:00.050268Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Tiramisu: A polyhedral compiler for expressing fast and portable code","venue":null,"work_id":"7d1fec17-d2ba-4463-a3f5-32b3a14c5d22","year":2019},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:7f564d1d1370fc65f90339d515708a6e393d56c704c2c6182a9a3a8091f4264f","observation_id":"047fdc06-5d17-4948-9ed0-594b51feed50","resolution":{"observed_at":"2026-05-21T10:20:00.712101Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2507.11948","last_updated":"2025-07-16T06:33:07Z","snapshot_observed_at":"2026-08-09T05:05:38.345869Z","submitted_at":"2025-07-16T06:33:07Z","title":"Kevin: Multi-Turn RL for Generating CUDA Kernels","version":1},"cited_work":{"arxiv_id":"2507.11948","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2507.11948","snapshot_observed_at":"2026-07-03T22:29:00.387986Z","title":"Kevin: Multi-turn rl for generating cuda kernels","venue":null,"work_id":"5adf1cd8-3a29-431d-9a02-7eb2150b195d","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"cited_paper":"/paper/2507.11948","citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:fcfcbd919c10ec9d0f55f3cbb3cab5bcd27bcbd82f29fff504769422996853d9","observation_id":"9fba3d5f-733d-4ee9-99d1-7f48f8b58973","resolution":{"observed_at":"2026-05-21T10:20:00.045508Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"A practical automatic polyhedral parallelizer and locality optimizer","venue":null,"work_id":"a7065f8e-2663-4e91-837d-9d179cec2554","year":2008},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:9b179b80f6f16ffe0ad7a41239892cc203b27fc9f29852ad4f52a6fe4c7c94c5","observation_id":"db46928d-91c5-44dc-8651-bdeabf10f8f7","resolution":{"observed_at":"2026-05-21T10:20:00.720885Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2601.07160","last_updated":"2026-04-17T02:12:07Z","snapshot_observed_at":"2026-07-06T22:41:24.948774Z","submitted_at":"2026-01-12T03:12:58Z","title":"AscendKernelGen: A Systematic Study of LLM-Based Kernel Generation for Neural Processing Units","version":2},"cited_work":{"arxiv_id":"2601.07160","doi":null,"metadata_source":"pith","pith_arxiv_id":"2601.07160","snapshot_observed_at":"2026-07-03T15:08:32.457405Z","title":"AscendKernelGen: A Systematic Study of LLM-Based Kernel Generation for Neural Processing Units","venue":"cs.AI","work_id":"6935b3e5-79f9-489d-8b0a-ce0bec72cefb","year":2026},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"cited_paper":"/paper/2601.07160","citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:1f3408848ffc26dfc44412c5491666a0e5688852d35f747896d170f85bed4762","observation_id":"06f1982f-eb29-4f84-90d2-fde9dda89c7e","resolution":{"observed_at":"2026-05-21T10:20:00.092462Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"TVM: An automated end-to-end optimizing compiler for deep learning","venue":null,"work_id":"c2450e9c-de0c-4e01-96f6-0d67222ccbba","year":2018},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:967ddfed74184290adc878aff04c3c1088dd4a43ea0f50a987fa289c73a7603d","observation_id":"20854c5e-7ad7-4653-9553-9a03390cc6c8","resolution":{"observed_at":"2026-05-21T10:20:00.665632Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Flashattention: Fast and memory-efficient exact attention with IO-awareness.the 36th International Conference on Neural Information Processing Systems (NeurIPS)","venue":null,"work_id":"17abe0af-d6b2-48af-80d1-e8f16066d29a","year":2022},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:fb6571a742746045c0b1d20263092baf4a8120213b489c0ddd8de76c220b11b8","observation_id":"b043dedf-348e-4c04-8626-a1f0bc81341c","resolution":{"observed_at":"2026-05-21T10:20:00.660193Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2510.16996","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-04T04:59:36.336849Z","title":"STARK: Strategic team of agents for refining kernels","venue":null,"work_id":"65a0246a-b1aa-4c6f-878d-ae92c8aace6d","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:5496d193e87cd3dc59fcb8f7f4411c6e6bf492f18d27b3a39421f7d0c27a9093","observation_id":"5659c11b-3485-4cdd-a356-42a5690686a2","resolution":{"observed_at":"2026-05-21T10:20:00.082652Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2510.03760","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-01T13:55:46.665108Z","title":"EvoEngineer: Mastering automated CUDA kernel code evolution with large language models","venue":null,"work_id":"0d477d94-308e-4930-a1f6-d5bf9db1b4a6","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:1a297ede3204b15297248be3ddd2365ad75f49ca548096f119fdc320752e480b","observation_id":"d96cc372-0955-466d-bc40-6afb8d792f0e","resolution":{"observed_at":"2026-05-21T10:20:00.059737Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2511.06345","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"PRAGMA: A profiling-reasoned multi-agent framework for automatic kernel optimization","venue":null,"work_id":"4b6a5b8b-7612-49bd-a34b-0820dc8830ff","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:43dd66de0756a97e315ca2c7aa1c1a99670644dc1dd1ef13ba82f7b8b4234de5","observation_id":"6213f072-17ac-4a8c-845a-9973aa37b9a0","resolution":{"observed_at":"2026-05-21T10:20:00.078041Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2512.09196","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-03T22:29:00.452744Z","title":"Tritonforge: Profiling-guided framework for automated triton kernel optimization","venue":null,"work_id":"074b984f-e344-4648-a409-ed476d502a61","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:4668978f4fc6f1e49aeb60aa32a128533c6ece3f8993c1a9ad650439fcc298fd","observation_id":"f3545ae0-6a5e-41b9-bcad-1063adea01df","resolution":{"observed_at":"2026-05-21T10:20:00.032073Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"TritonBench: Benchmarking large language model capabilities for generating Triton operators","venue":null,"work_id":"53a15148-07e2-49d9-b924-a416cb7f86d1","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:1f004a500736abad826fc57d603b1735391230f9596c3e88602a2f45df94f921","observation_id":"b2eb99bf-3937-4005-bf1f-38b0638c17e4","resolution":{"observed_at":"2026-05-21T10:20:00.714972Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"The deep learning compiler: A comprehensive survey.IEEE Transactions on Parallel and Distributed Systems, 32(3):708–727","venue":null,"work_id":"3ca28123-710c-4310-8711-f1df5fdecd9f","year":2020},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:c5866dadb8e7f3efa426717d099c6d09df6b4b3bbe98a116642a55dee231ff29","observation_id":"0f3ded8b-6200-41b0-bd62-c41c4e3c7ea5","resolution":{"observed_at":"2026-05-21T10:20:00.718023Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2507.05687","last_updated":"2025-07-08T05:38:24Z","snapshot_observed_at":"2026-08-13T17:13:57.451550Z","submitted_at":"2025-07-08T05:38:24Z","title":"AutoTriton: Automatic Triton Programming with Reinforcement Learning in LLMs","version":1},"cited_work":{"arxiv_id":"2507.05687","doi":"10.48550/arxiv.2507.05687","metadata_source":"arxiv_reference","pith_arxiv_id":"2507.05687","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Autotriton: Automatic triton programming with reinforcement learning in llms","venue":"ArXiv.org","work_id":"becdfbca-4b66-443b-8026-5df50fb71f5b","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"cited_paper":"/paper/2507.05687","citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:58d16aff120524585091c6fe8a84ff5858410387ba694b2814cbe8e4c4c62897","observation_id":"2d873a39-a829-4847-aaf2-9e3fae31f37e","resolution":{"observed_at":"2026-05-21T10:20:00.064098Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2603.02637","last_updated":"2026-08-10T14:46:34Z","snapshot_observed_at":"2026-08-17T18:11:14.965517Z","submitted_at":"2026-03-03T06:04:49Z","title":"StitchCUDA: An Automated Multi-Agents End-to-End GPU Programing Framework with Rubric-based Agentic Reinforcement Learning","version":2},"cited_work":{"arxiv_id":"2603.02637","doi":"10.48550/arxiv.2603.02637","metadata_source":"arxiv_reference","pith_arxiv_id":"2603.02637","snapshot_observed_at":"2026-08-11T03:24:38.516909Z","title":"Stitchcuda: An automated multi-agents end-to-end gpu programing framework with rubric-based agentic reinforcement learning","venue":"arXiv (Cornell University)","work_id":"abee1f2b-c82b-432f-9c38-82c8da82b343","year":2026},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"cited_paper":"/paper/2603.02637","citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:ed3a761436fef95c73b0d5fd0984479b4f54cd35586fb85c3dfb6b8354d6de4d","observation_id":"976771a4-57fc-4df7-b6dc-e5c030a340bb","resolution":{"observed_at":"2026-08-11T03:24:38.516909Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2507.14111","last_updated":"2026-08-05T22:03:49Z","snapshot_observed_at":"2026-08-09T23:09:14.608641Z","submitted_at":"2025-07-18T17:43:56Z","title":"CUDA-L1: Improving CUDA Optimization via Contrastive Reinforcement Learning","version":12},"cited_work":{"arxiv_id":"2507.14111","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2507.14111","snapshot_observed_at":"2026-07-14T02:20:17.802860Z","title":"Cuda-l1: Improving cuda optimization via contrastive reinforcement learning","venue":null,"work_id":"30b05e27-97da-46cc-b53c-09991fe1da44","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"cited_paper":"/paper/2507.14111","citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:5c8736f3d3a48bf7e252ccfdc9edd01f09513fa16ff5f02b921018a5e39f7260","observation_id":"f0eac67c-39fc-4253-929c-82acc619deca","resolution":{"observed_at":"2026-07-14T02:20:17.802860Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2512.23236","last_updated":"2026-07-06T22:18:07Z","snapshot_observed_at":"2026-08-14T21:09:28.913440Z","submitted_at":"2025-12-29T06:31:55Z","title":"KernelEvolve: Scaling Agentic Kernel Coding for Heterogeneous AI Accelerators at Meta","version":4},"cited_work":{"arxiv_id":"2512.23236","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2512.23236","snapshot_observed_at":"2026-07-08T02:18:09.597449Z","title":"KernelEvolve: Scaling agentic kernel coding for heterogeneous AI accelerators at meta","venue":null,"work_id":"dc642e50-d43a-48d0-9954-b1729c54cc37","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"cited_paper":"/paper/2512.23236","citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:51bfe2e4983078ef7644275750e0570453fe6800d1b179571666273d45f5e75b","observation_id":"c63b94c3-8612-446c-8794-15ff1b6145bb","resolution":{"observed_at":"2026-07-08T02:18:09.597449Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Accelerating sparse matrix-matrix multiplication with the Ascend AI core","venue":null,"work_id":"ef897113-eb89-43ba-9bcd-7ea9b0b68d74","year":2023},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:0e8ecdfd09e962f6663352bb4734d3a12d1c9f5be5654c540ae5bebc75e94ca5","observation_id":"2af18a5b-006b-4bb0-ab43-843124b2f989","resolution":{"observed_at":"2026-05-21T10:20:00.706481Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2502.10517","last_updated":"2025-02-14T19:30:53Z","snapshot_observed_at":"2026-08-16T10:36:49.395324Z","submitted_at":"2025-02-14T19:30:53Z","title":"KernelBench: Can LLMs Write Efficient GPU Kernels?","version":1},"cited_work":{"arxiv_id":"2502.10517","doi":"10.48550/arxiv.2502.10517","metadata_source":"pith","pith_arxiv_id":"2502.10517","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"KernelBench: Can LLMs Write Efficient GPU Kernels?","venue":"cs.LG","work_id":"ee44d624-1355-4fa0-ad70-c89205f5ce5a","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"cited_paper":"/paper/2502.10517","citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:91e2a6defba0d8e7c52f20c09ee4d64de90ab3c469ecc022724fbb640ef7ba8b","observation_id":"154d02c0-c315-4c6a-a0e5-117bb5dc285b","resolution":{"observed_at":"2026-05-21T10:20:00.068223Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Halide: A language and compiler for optimizing parallelism, locality, and recomputation in image processing pipelines","venue":null,"work_id":"49f48daf-758b-4e95-b6c9-9723b9173fa4","year":2013},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:2dbc59088042ca4adbf3568b16617e38744f3e79bd605216e1ffeb37363b600a","observation_id":"134ecb07-f631-49fa-a30b-5e18cd51663a","resolution":{"observed_at":"2026-05-21T10:20:00.709212Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2506.03524","last_updated":"2025-06-05T03:26:05Z","snapshot_observed_at":"2026-08-13T13:37:23.026438Z","submitted_at":"2025-06-04T03:17:19Z","title":"Seed-Coder: Let the Code Model Curate Data for Itself","version":2},"cited_work":{"arxiv_id":"2506.03524","doi":"10.48550/arxiv.2506.03524","metadata_source":"arxiv_reference","pith_arxiv_id":"2506.03524","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Seed-coder: Let the code model curate data for itself","venue":"ArXiv.org","work_id":"876e3e57-02d4-4ba4-92be-40b91ba89b35","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"cited_paper":"/paper/2506.03524","citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:50f12b895139db62ccf3d7282dde4facffb22182f96c7762445bb4c96007efb0","observation_id":"eaec2863-bbd7-44e4-9fe2-3212a03f0968","resolution":{"observed_at":"2026-05-21T10:20:00.087467Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"FlashAttention-3: Fast and accurate attention with asynchrony and low-precision.The 38th Conference on Neural Information Processing Systems (NeurIPS)","venue":null,"work_id":"d3e1a424-d4dc-4cb9-81d7-830673dc9071","year":2024},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:678a66ab749cca9052b6c5e65013ea99df73c39cd3e93bb78218fda68cc7663e","observation_id":"0eec9851-a374-4168-af5c-0e81ddc376b3","resolution":{"observed_at":"2026-05-21T10:20:00.697665Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"OpenEvolve: An open-source evolutionary coding agent","venue":null,"work_id":"d8d3ea4e-e3dd-4256-b4d4-b54d66555ccd","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:13ab4e9f4dac28481d58dd521b4ef4881d033ceafbb197a537394054520a72c5","observation_id":"74e09198-7c17-4e3e-8500-6bc869f87d39","resolution":{"observed_at":"2026-05-21T10:20:00.700709Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2512.02551","last_updated":"2026-08-05T18:31:31Z","snapshot_observed_at":"2026-08-16T13:58:25.439020Z","submitted_at":"2025-12-02T09:20:15Z","title":"CUDA-L2: Surpassing cuBLAS Performance for Matrix Multiplication through Reinforcement Learning","version":4},"cited_work":{"arxiv_id":"2512.02551","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2512.02551","snapshot_observed_at":"2026-07-14T02:20:20.617597Z","title":"CUDA-L2: Surpassing cuBLAS performance for matrix multiplication through reinforcement learning","venue":null,"work_id":"859c777b-c20f-4eac-be37-519a54eeb8e3","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"cited_paper":"/paper/2512.02551","citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:ccd1d389abe16c781634ae709480b7a52e52f86d5a1c846afc6b8ec601115721","observation_id":"e89d0978-6698-4d07-a706-547c71bd5372","resolution":{"observed_at":"2026-07-14T02:20:20.617597Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Triton: An intermediate language and compiler for tiled neural network computations","venue":null,"work_id":"6df4b53e-5ebc-4b62-967d-f56b8cb7ae5c","year":2019},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:da73ddb097af7ad5b89da330149c377a91b6de284b1f294078abcebed4ad1bed","observation_id":"f69290e5-1204-4dec-ba99-2255c88449a7","resolution":{"observed_at":"2026-05-21T10:20:00.694673Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2507.23194","last_updated":"2025-07-31T02:26:58Z","snapshot_observed_at":"2026-08-09T05:05:37.816527Z","submitted_at":"2025-07-31T02:26:58Z","title":"Geak: Introducing Triton Kernel AI Agent & Evaluation Benchmarks","version":1},"cited_work":{"arxiv_id":"2507.23194","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2507.23194","snapshot_observed_at":"2026-07-04T04:59:36.315299Z","title":"Geak: Introducing triton kernel ai agent & evaluation benchmarks","venue":null,"work_id":"95d3012c-5893-4ec9-8836-8c9dda6a0bda","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"cited_paper":"/paper/2507.23194","citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:27561d2d5aed2145ea9653a9a1412369d4de190be5797b8006e2742ab1543f55","observation_id":"d0d06f4e-cfc5-4d9d-b013-6a62fa2770f5","resolution":{"observed_at":"2026-05-21T10:20:00.009036Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2504.17577","last_updated":"2025-04-27T11:11:02Z","snapshot_observed_at":"2026-08-16T10:34:21.130219Z","submitted_at":"2025-04-24T14:08:49Z","title":"TileLang: A Composable Tiled Programming Model for AI Systems","version":2},"cited_work":{"arxiv_id":"2504.17577","doi":"10.48550/arxiv.2504.17577","metadata_source":"arxiv_reference","pith_arxiv_id":"2504.17577","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"TileLang: A composable tiled programming model for AI systems","venue":"ArXiv.org","work_id":"d23477a7-29b7-478d-b57d-4a841906c2c6","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"cited_paper":"/paper/2504.17577","citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:aac288946bf80f06bc29c386d523a19b7b8fd44a9612cf2795b9507e697977cb","observation_id":"dce00cc6-6238-47aa-8e50-d70f4497d759","resolution":{"observed_at":"2026-05-21T10:20:00.022432Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2509.07506","doi":"10.48550/arxiv.2509.07506","metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Astra: A multi-agent system for GPU kernel performance optimization","venue":"ArXiv.org","work_id":"d2dbaa77-84ac-4089-831b-fbbc4af827dc","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:6ac46feda80428959d7c24565653f0e5361196bdb34a98dd58eba166567319e3","observation_id":"2a607446-d0ad-4a70-b010-b1193257e533","resolution":{"observed_at":"2026-05-21T10:20:00.004601Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2507.17773","last_updated":"2025-07-26T08:05:03Z","snapshot_observed_at":"2026-08-14T16:13:49.234740Z","submitted_at":"2025-07-20T00:58:33Z","title":"MultiKernelBench: A Multi-Platform Benchmark for Kernel Generation","version":2},"cited_work":{"arxiv_id":"2507.17773","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2507.17773","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Multikernelbench: A multi-platform benchmark for kernel generation","venue":null,"work_id":"b1d15b28-fd95-4487-8111-4cc04bf04388","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"cited_paper":"/paper/2507.17773","citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:198fdc939daa1f28603e911dd0cf3757869da2d4d253de35463856c87bb3edd0","observation_id":"f84cb9ff-3ec1-4038-93da-431015150c07","resolution":{"observed_at":"2026-05-21T10:20:00.013601Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2510.17891","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-04T13:39:50.161017Z","title":"TritonRL: Training LLMs to think and code triton without cheating","venue":null,"work_id":"c87b3170-fefe-489f-8564-29a1ad338ad9","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:d68b372f6c8ef16f948d44e3e80b87e23037d5c876c903e3e5200d75d0db1788","observation_id":"e79dd4e9-3ec2-4d87-ad1d-c1919dab66ff","resolution":{"observed_at":"2026-05-21T10:20:00.054808Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Mirage: A multi-level superoptimizer for tensor programs","venue":null,"work_id":"98e9a5ba-c85c-4066-8903-e3193f029429","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:9ce90bc2d436ef220adef2889bc9bcadb10edb21a49fc86f5470251433e5d251","observation_id":"a78a0716-0ba4-48b7-b45b-6b414edfc4c0","resolution":{"observed_at":"2026-05-21T10:20:00.691753Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"TLP: A deep learning- based cost model for tensor program tuning","venue":null,"work_id":"df71fa95-e0e0-4509-bc90-06550d1274c1","year":2023},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:3bdfa470414ef7bc4ee75dab231f13cca745a1c3d48f738d81f546119f12db89","observation_id":"4850ea38-86d4-4f15-bc3c-9444205126c2","resolution":{"observed_at":"2026-05-21T10:20:00.688908Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2505.10911","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-07T12:53:50.187517Z","title":"ReWiND: Language-guided rewards teach robot policies without new demonstrations","venue":null,"work_id":"003c50e4-9384-4d2d-9774-ad5b6d1f4669","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:291d72c6f756ec81fdeeff05b1e437d0a58c9e076be83b91d812b472dffb5248","observation_id":"bb09a0fe-b3ac-4ab8-bbbb-7a615caf0ca5","resolution":{"observed_at":"2026-05-21T10:20:00.073093Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2511.01884","doi":"10.48550/arxiv.2511.01884","metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"Cudaforge: An agent framework with hardware feedback for cuda kernel optimization","venue":"ArXiv.org","work_id":"5b81cc91-ad3c-42d0-9a5a-eeb7ef8ccaac","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:7054594051805fd00c4d00ccc2a88d4620f311c4a462c8f0532707f169340479","observation_id":"d3561414-f557-4352-8140-ae390af48fd3","resolution":{"observed_at":"2026-05-21T10:20:00.040818Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"AKG: Automatic kernel generation for neural processing units using polyhedral transformations","venue":null,"work_id":"2737090d-1373-47c0-a104-28497ef60316","year":2021},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:ea8b5efdaeca632c10b97854859cc1620410366ef25e0998372e325136307bfd","observation_id":"7f268062-034b-4a89-8e67-68ef8d637304","resolution":{"observed_at":"2026-05-21T10:20:00.723819Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Ansor: Generating high-performance tensor programs for deep learning","venue":null,"work_id":"16954835-f523-4696-b500-22f938799581","year":2020},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:8d2db49f1a80c191d2e4ceca2af3279418335ae62a73d15f9f88f0841b8eb3fb","observation_id":"9c91daf4-c080-4538-8629-25c160cd54a0","resolution":{"observed_at":"2026-05-21T10:20:00.677517Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"TenSet: A large-scale program performance dataset for learned tensor compilers","venue":null,"work_id":"eb7b88a3-d2e1-4c97-afd9-9ddcd3463edf","year":2021},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:b085e2c668542894c4d0e05840bbbc0f311fac640cf78817e2b615cedb24c2bf","observation_id":"0465cf72-6a47-4440-90bd-e7c51ad9243f","resolution":{"observed_at":"2026-05-21T10:20:00.686087Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Squeezing operator performance potential for the ascend architecture","venue":null,"work_id":"a3a45d3e-7d8e-4eb1-a10b-ac1b52ade524","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:0cb474f08180a0cd2f181a6dee96f7b1a21270fa7e11fb47b85bc9137226fa32","observation_id":"078ba9e3-af67-407e-a43e-2d15026dc8a4","resolution":{"observed_at":"2026-05-21T10:20:00.662957Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Accelerating model training on Ascend chips: An industrial system for profiling, analysis and optimization","venue":null,"work_id":"3852b3b5-2d2d-4010-976f-bcb5b4a0d841","year":2025},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:1c33fa4aecda92030da93b505c6efc8720db0bc2234d75ac5446c2d4f43208e6","observation_id":"eb412db0-ac5b-4011-bafe-45a8b802c41c","resolution":{"observed_at":"2026-05-21T10:20:00.668616Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"increase tiling","venue":null,"work_id":"4b6cef56-6dc4-4e7d-89c4-f035f7109914","year":2020},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:f25d954c9cce68bcfd2f1f9bf7520f727211fc7816d5ba223faa44a735036286","observation_id":"df1539b0-d0d5-4a56-879a-058f02055268","resolution":{"observed_at":"2026-05-21T10:20:00.683170Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"7- On the Ascend AI Core, this forces a full stall across MTE1/MTE2/MTE3 and the Vector units, preventing overlap between data movement and computation","venue":null,"work_id":"32848710-cfa7-41e4-bbe6-cfa2808f2552","year":null},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:7a2537a9994686afdefcf8b4ea22936b24078876c3ab8fb31300f9c286d7643d","observation_id":"27ba7d62-daf2-4548-a355-5385aa49567a","resolution":{"observed_at":"2026-05-21T10:20:00.674664Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"12- Increasing it to ‘maxDataCount = 8‘ (32 bytes) improves the payload-to-overhead ratio for DMA commands","venue":null,"work_id":"6a94f7a3-d2a9-4045-a981-f44c8eae8202","year":null},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:bf1bb8354bee4d32a269e168ab3d0b1fce66efd0c91af5aa15e0f6d68e79666e","observation_id":"8be0aec8-7e05-45a0-969a-8daad4f63b3c","resolution":{"observed_at":"2026-05-21T10:20:00.671938Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"17\"bottleneck","venue":null,"work_id":"6ecc476c-39a9-422d-9860-fd696841d5e1","year":null},"citing_paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization","version":2},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-05-21T10:16:15.305706Z"},"links":{"citing_paper":"/paper/2603.23566"},"observation_digest":"sha256:0014aeaa7c8aebfadd9efff646891d987e97a5b7b9ad3fbbefc1281fddf5c2da","observation_id":"c03ece47-65c8-4f92-8d16-c57946408282","resolution":{"observed_at":"2026-05-21T10:20:00.680239Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2603.23566","last_updated":"2026-05-16T02:26:58Z","latest_version":2,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-16T22:25:55.062190Z","submitted_at":"2026-03-24T08:54:53Z","title":"AscendOptimizer: Episodic Agent for Ascend NPU Operator Optimization"},"reference_resolution":{"displayed":44,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":0,"verified_exact":21,"verified_fuzzy":23},"total_outbound_references":44},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"thesis":"As of 17 August 2026, this Paper Citation Record lists 44 of 44 outbound references and 2 inbound Pith citation observations for arXiv:2603.23566."}