{"as_of":"2026-08-19T22:33:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:1be6ecb5a58dbdfbd0a764ed16ce9bc519db68889337e07abe032a2da03cf259","coverage":[{"denominator":20,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":20,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-01T14:28:22.953274Z","state":"measured"},{"denominator":20,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":20,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-19T06:32:44.657259+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"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":[],"links":{"evidence":"/evidence","html":"/paper/2607.19438/citation-record","integrity":"/paper/2607.19438/integrity","json":"/paper/2607.19438/citation-record.json","paper":"/paper/2607.19438"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:21.328408Z","title":"The next big shifts in AI workloads and hyperscaler strategies","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:21.328408Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:c35a938719e0dcf8b7251c419de89d227aa0cb477a4f547c1420ca4d5aefd635","observation_id":"917cbc47-a43e-4734-a7d0-317a0af55f1d","resolution":{"observed_at":"2026-08-01T14:28:21.328408Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:21.405227Z","title":"Gartner predicts 40% of AI data breaches will arise from cross-border GenAI misuse by 2027","venue":null,"work_id":null,"year":2027},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:21.405227Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:a89af62f4229266be99bd865ca0b808df8f3c8eba36065d6064063a83a40f362","observation_id":"ec89ac0a-8a35-4a04-8860-def978c74569","resolution":{"observed_at":"2026-08-01T14:28:21.405227Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:21.465382Z","title":"Artificial intelligence risk management framework: Generative artificial intelligence profile","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:21.465382Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:2da8201b919f710cd0f3394775090f731fa27fe1d2d7a6bcd1ebb4fc666f3e20","observation_id":"b8d16cae-a90e-4947-9fd6-4550add7f33f","resolution":{"observed_at":"2026-08-01T14:28:21.465382Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2407.04292","last_updated":"2025-06-08T15:01:21Z","snapshot_observed_at":"2026-08-16T13:36:53.052293Z","submitted_at":"2024-07-05T06:54:26Z","title":"DaDu-Corki: Algorithm-Architecture Co-Design for Embodied AI-powered Robotic Manipulation","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.04292","snapshot_observed_at":"2026-08-01T14:28:21.557312Z","title":"DaDu-Corki: Algorithm-Architecture Co-Design for Embodied AI-powered Robotic Manipulation","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:21.557312Z"},"links":{"cited_paper":"/paper/2407.04292","citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:f8f33e0de2b3a977f5e27d9b3c026eef4c8a7365c1cbb4ee8e33b310a6c87c72","observation_id":"2bbbdf06-1b63-4ea9-9c99-bd34b524a8be","resolution":{"observed_at":"2026-08-01T14:28:21.557312Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:21.655863Z","title":"Agentic Artificial Intelligence: Architectures, Taxonomies, and Evaluation of Large Language Model Agents.arXiv preprint arXiv:2601.12560, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:21.655863Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:77a50f0b6097f1ae9377bed8db50c10822635c655ecd134e0b4d916446e37a30","observation_id":"2e5f5688-65a4-48bc-94c3-454f67032f0e","resolution":{"observed_at":"2026-08-01T14:28:21.655863Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2505.16508","last_updated":"2025-05-29T08:56:27Z","snapshot_observed_at":"2026-08-19T14:30:32.965900Z","submitted_at":"2025-05-22T10:43:00Z","title":"Edge-First Language Model Inference: Models, Metrics, and Tradeoffs","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2505.16508","snapshot_observed_at":"2026-08-01T14:28:21.783810Z","title":"Edge-First Language Model Inference: Models, Metrics, and Tradeoffs","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:21.783810Z"},"links":{"cited_paper":"/paper/2505.16508","citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:deecb624b5878b603697c7b4a1aac51c40e1cb54393ea96d87993e010d095dfa","observation_id":"afb5a434-8b04-48b2-bb97-fe96bb5fd551","resolution":{"observed_at":"2026-08-01T14:28:21.783810Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:21.840714Z","title":"Anthropic Status Page: Incident History","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:21.840714Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:07e1b36036dddf90cb10d3a7d86c9e8ba06bae112e01fedbac8a272ae1b20e27","observation_id":"cdee5fd7-9da2-4f2f-bb42-f9a18323f73d","resolution":{"observed_at":"2026-08-01T14:28:21.840714Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:21.920879Z","title":"OpenAI Status Page: Incident History","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:21.920879Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:739b87417494bf0bbf6ffa97de276d8aa326d0678ff90d16899576ba49b28fd1","observation_id":"5d296c41-3ad4-4dbd-8630-d5b30fb75276","resolution":{"observed_at":"2026-08-01T14:28:21.920879Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:21.983462Z","title":"The New Economics of Enterprise Technology in an AI World","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:21.983462Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:2d0234d0a5b78b3486a9f0d59eed6975f6a8c108f1e8eb164f0a9feb06d1eecd","observation_id":"47beab02-d644-4242-a39a-9bf05919c3bf","resolution":{"observed_at":"2026-08-01T14:28:21.983462Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:22.039407Z","title":"State of Open Source AI 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:22.039407Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:7de29f45189cf6361724ce5bcf2a7cb93887cb9246e38751833d6b9b275352e3","observation_id":"1ee459cf-325f-4527-8bb4-8bedba2d21ea","resolution":{"observed_at":"2026-08-01T14:28:22.039407Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2508.08531","last_updated":"2025-08-12T00:06:34Z","snapshot_observed_at":"2026-08-13T09:18:07.505822Z","submitted_at":"2025-08-12T00:06:34Z","title":"Profiling Large Language Model Inference on Apple Silicon: A Quantization Perspective","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2508.08531","snapshot_observed_at":"2026-08-01T14:28:22.107077Z","title":"Profiling Large Language Model Inference on Apple Silicon: A Quantization Perspective.arXiv preprint arXiv:2508.08531, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:22.107077Z"},"links":{"cited_paper":"/paper/2508.08531","citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:b360ee16603b363e564e3fc8ef5bcd53af0bca508db9cf5079e27dc05d6906c2","observation_id":"25a12513-3852-4837-af63-0ffead5d081e","resolution":{"observed_at":"2026-08-01T14:28:22.107077Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:22.184070Z","title":"Artificial Intelligence Index Report 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:22.184070Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:4822d4ccda84fb728bc7e788924eacf2f967ca068fdbd51dd64f6e14da37a7fd","observation_id":"c21d9ed7-2c6f-433e-9ef1-34a968b56ec6","resolution":{"observed_at":"2026-08-01T14:28:22.184070Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2607.00501","last_updated":"2026-07-01T06:37:11Z","snapshot_observed_at":"2026-08-13T10:55:39.712297Z","submitted_at":"2026-07-01T06:37:11Z","title":"BaseRT: Best-in-Class LLM Inference on Apple Silicon via Native Metal","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2607.00501","snapshot_observed_at":"2026-08-01T14:28:22.244763Z","title":"BaseRT: Best-in-Class LLM Inference on Apple Silicon via Native Metal.arXiv preprint arXiv:2607.00501, 2026","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:22.244763Z"},"links":{"cited_paper":"/paper/2607.00501","citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:63877b613b9e61c0762afcdd6b1b95345f074f6a8cdbc478b82dee3ec5ecac14","observation_id":"cf982e02-09b5-45f7-b418-2ba08c2490af","resolution":{"observed_at":"2026-08-01T14:28:22.244763Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:22.365220Z","title":"llama.cpp: LLM inference in C/C++","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:22.365220Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:f28216b857a43011750d78df54c2c1b53d35960b2e40ad7ad43ffd5730233e74","observation_id":"48685e4e-9170-4b00-9298-14e01d914fcf","resolution":{"observed_at":"2026-08-01T14:28:22.365220Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:22.480763Z","title":"MLX: Efficient and flexible machine learning on Apple silicon.https://github.com/ml-explore/mlx, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:22.480763Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:5eff6c43e126ce2b6a1d687ddde049db7158d22bd689272821e32b2f205cfc6e","observation_id":"2db8dbf9-6245-46d7-9356-48a5135b5c48","resolution":{"observed_at":"2026-08-01T14:28:22.480763Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:22.588686Z","title":"Apple unleashes M5, the next big leap in AI performance for Apple silicon","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:22.588686Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:65f354938e709ecda240298fed0b8dbcc491eccad1a3dcf012f99d253ae80755","observation_id":"45623c91-08b9-4fcd-9826-7450d398986b","resolution":{"observed_at":"2026-08-01T14:28:22.588686Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:22.644790Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:22.644790Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:de9d234ce28c86a5f040c42e4898f604df78ec10e441cd2a1920727e3b46d0e4","observation_id":"222a9cfa-2be7-4510-b16b-ceb7bac097b2","resolution":{"observed_at":"2026-08-01T14:28:22.644790Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:22.706340Z","title":"Metal Performance Primitives (MPP) Programming Guide","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:22.706340Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:bce8f5e770f3e91327f3e2cbc6f315685f27186d8d880164aaf6433405452f0c","observation_id":"7eb07e12-ee72-4d06-831b-b4c48e1ab3ba","resolution":{"observed_at":"2026-08-01T14:28:22.706340Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:22.845966Z","title":"Discover Metal 4 tensors and Metal performance primitives","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:22.845966Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:d224056ed5cb527286280351a3fe845519c7a0e5adc49e513633777de4b0a893","observation_id":"01767eae-cea1-472b-9131-b58d8712e579","resolution":{"observed_at":"2026-08-01T14:28:22.845966Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T14:28:22.953274Z","title":"Metal Shading Language Specification","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-01T14:28:22.953274Z"},"links":{"citing_paper":"/paper/2607.19438"},"observation_digest":"sha256:502cf1d334cf7fdc3ad3a67e6e808409d86442032cc94bbfb3f46e511bbfdb59","observation_id":"7163ff69-6fc7-41ab-a61b-4a8ff7f07d77","resolution":{"observed_at":"2026-08-01T14:28:22.953274Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2607.19438","last_updated":"2026-07-21T06:42:18Z","latest_version":1,"primary_category":"cs.AR","snapshot_observed_at":"2026-08-16T05:11:31.055992Z","submitted_at":"2026-07-21T06:42:18Z","title":"BaseRT: Advancing Best-in-Class LLM Inference with Apple M5 Neural Accelerators"},"reference_resolution":{"displayed":20,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":20,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":20},"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":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"thesis":"As of 19 August 2026, this Paper Citation Record lists 20 of 20 outbound references and 0 inbound Pith citation observations for arXiv:2607.19438."}