{"as_of":"2026-08-12T06:16:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:83a7d915a601bfaeb81715b421c66901915bba782861452a10c9111833672e6a","coverage":[{"denominator":42,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":42,"source":"paper_references, paper_reference_links","source_observed_at":"2026-06-30T01:53:18.755724Z","state":"measured"},{"denominator":42,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":42,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-11T06:34:44.6726+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/2606.29518/citation-record","integrity":"/paper/2606.29518/integrity","json":"/paper/2606.29518/citation-record.json","paper":"/paper/2606.29518"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-30T01:53:18.755724Z","title":"The internet of things: A survey.Computer networks, 54(15):2787–2805, 2010","venue":null,"work_id":null,"year":2010},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:3e03b79aed11b6f3c62e57e96e95727303dc51bde51b4229b998d75ed2a0d202","observation_id":"884bb269-b18e-4a61-8921-80925af1c9e4","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"In-datacenter performance analysis of a tensor processing unit","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:7c0ca1c37390dc1af5db191b339136ba8ad9bbf61a3181a2908f3618806a8bb0","observation_id":"e9fbb792-b365-46b6-a7bc-814f10ca8ecb","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2101.08596","last_updated":"2021-01-21T13:25:58Z","snapshot_observed_at":"2026-08-11T06:57:36.884907Z","submitted_at":"2021-01-21T13:25:58Z","title":"LEAF: A Learnable Frontend for Audio Classification","version":1},"cited_work":{"arxiv_id":"2101.08596","doi":null,"metadata_source":"pith","pith_arxiv_id":"2101.08596","snapshot_observed_at":"2026-07-10T05:46:50.406824Z","title":"ArXiv abs/2101.08596","venue":"cs.SD","work_id":"1152c054-8de4-46e7-8c12-cc832c355f17","year":2021},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"cited_paper":"/paper/2101.08596","citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:2c4240cbc1d160c74d14f5036b1e940c8fd0fa3ca01346a9d0b171a444357032","observation_id":"324211a8-7e0d-4e72-beee-a0403a9f7feb","resolution":{"observed_at":"2026-06-30T03:04:14.342259Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2407.18267","last_updated":"2024-07-17T14:51:15Z","snapshot_observed_at":"2026-07-06T18:52:03.385131Z","submitted_at":"2024-07-17T14:51:15Z","title":"MCU-MixQ: A HW/SW Co-optimized Mixed-precision Neural Network Design Framework for MCUs","version":1},"cited_work":{"arxiv_id":"2407.18267","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2407.18267","snapshot_observed_at":"2026-06-30T03:04:14.344076Z","title":"Mcu-mixq: A hw/sw co-optimized mixed-precision neural network design framework for mcus.arXiv preprint arXiv:2407.18267, 2024","venue":null,"work_id":"1a6d9cae-0a8f-466a-a0db-1076a74e99eb","year":2024},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"cited_paper":"/paper/2407.18267","citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:9c59dcca4009217272567a4ff7affaa4d2c3b6fcf3517f9060b8aa34e71f4d6d","observation_id":"66672c66-65ee-47cb-8364-f85f54999ab4","resolution":{"observed_at":"2026-06-30T03:04:14.347732Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-30T01:53:18.755724Z","title":"Empowering edge intelligence: A comprehensive survey on on-device ai models.ACM Computing Surveys, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:e0d72101cc700b8b4a3aed2fe4bcf8eec896f3c277a1dfa8e9a5f1f83910847c","observation_id":"b7f48354-d73b-4e17-943f-b497f271318f","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Yolo9000: better, faster, stronger","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:a649224bdc0a8c4925ac774d42c66eda43ffe74750704d81bdff5f92b4134b56","observation_id":"0a3e85f6-684d-4536-87c6-4793c8904622","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1602.07360","last_updated":"2016-11-04T21:26:08Z","snapshot_observed_at":"2026-07-06T04:47:10.710916Z","submitted_at":"2016-02-24T00:09:45Z","title":"SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size","version":4},"cited_work":{"arxiv_id":"1602.07360","doi":"10.48550/arxiv.1602.07360","metadata_source":"pith","pith_arxiv_id":"1602.07360","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size","venue":"cs.CV","work_id":"755e91c7-dab5-418e-9316-bda99a153fed","year":2016},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"cited_paper":"/paper/1602.07360","citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:c41206f17e50dcaddb82dbedc3089635ff25615067ad3fbc2655fefba8c8e4f1","observation_id":"cfb13ea2-1cdf-4e6f-91d4-e30f2edc7216","resolution":{"observed_at":"2026-06-30T03:04:14.353479Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1704.04861","last_updated":"2017-04-17T03:57:34Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2017-04-17T03:57:34Z","title":"MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications","version":1},"cited_work":{"arxiv_id":"1704.04861","doi":"10.48550/arxiv.1704.04861","metadata_source":"pith","pith_arxiv_id":"1704.04861","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications","venue":"cs.CV","work_id":"3870239a-c950-4625-bf33-c4f902d14175","year":2017},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"cited_paper":"/paper/1704.04861","citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:3a25aa448fa774ef121a8af0712ab98d2417c37f7e2ac0044e40d6aa0930853d","observation_id":"e577c2e0-7d50-4772-859f-8023851f181a","resolution":{"observed_at":"2026-06-30T03:04:14.336837Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-30T01:53:18.755724Z","title":"Efficientnet: Rethinking model scaling for convolutional neural networks","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:6e5b1bdce4c70334a81ee341034a8c95a977760edbdf852384f4fa51125d2877","observation_id":"fdcb8291-8386-4029-9b11-dd130663af65","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Addressing the issue of processing element under- utilization in general-purpose systolic deep learning accelerators","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:91d79d527644273668f21cea80c3710fbe80e8c73e27a9767e5ce2fbfb01a032","observation_id":"b42c1772-3077-4654-a1bb-9ca87f02d61a","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2403.09026","last_updated":"2024-04-11T23:26:33Z","snapshot_observed_at":"2026-08-11T22:36:10.542083Z","submitted_at":"2024-03-14T01:39:12Z","title":"FlexNN: A Dataflow-aware Flexible Deep Learning Accelerator for Energy-Efficient Edge Devices","version":2},"cited_work":{"arxiv_id":"2403.09026","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2403.09026","snapshot_observed_at":"2026-06-30T03:04:14.320567Z","title":"Flexnn: A dataflow-aware flexible deep learning accelerator for energy-efficient edge devices.arXiv preprint arXiv:2403.09026, 2024","venue":null,"work_id":"eb4c4c92-b04b-48ab-80ad-2145dc5a0374","year":2024},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"cited_paper":"/paper/2403.09026","citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:ed5d0941f90a9003deb1e0a12cc5335a1e9a0f7a397bc4f84f734b42ea9584a9","observation_id":"62e42fcc-a644-4ead-b7e5-3b1ba96cc43f","resolution":{"observed_at":"2026-06-30T03:04:14.323112Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-30T01:53:18.755724Z","title":"A comprehensive survey of energy-efficient computing to enable sustain- able massive iot networks.Alexandria Engineering Journal, 91:12–29, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:7080c48707d7a8caa7430c6378dff337ac3aa67a0f7226e48757c0debe233d24","observation_id":"37f763d1-dac4-45ee-9adb-220991d5e155","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Snnap: Approximate computing on programmable socs via neural acceleration","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:000b346f99bbaa13b32fbc1827a3509269716c50e098970292ffd0e17fd9fa01","observation_id":"90aa7da9-ce59-4acd-aed3-cb9d68d3bf5b","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Neural acceleration for general-purpose approximate programs","venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:2aa6e849af0a3f9310d5f483466d122513360bce53a6343ba1e02a5dd8f47240","observation_id":"35acd8c4-726a-4638-93e2-40ddafbcfb22","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Neural network-based accelerators for transcendental function approximation","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:02a62bff090f2ce0aa2fad0b059fc8657534cf52ba1aa31c74b378035d5a0a27","observation_id":"59dd86e7-09c1-40ba-af80-a08370f4a1b2","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2101.09336","last_updated":"2021-01-22T21:13:46Z","snapshot_observed_at":"2026-08-07T00:40:56.156835Z","submitted_at":"2021-01-22T21:13:46Z","title":"A Comprehensive Survey on Hardware-Aware Neural Architecture Search","version":1},"cited_work":{"arxiv_id":"2101.09336","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2101.09336","snapshot_observed_at":"2026-06-30T03:04:14.310083Z","title":"A comprehensive survey on hardware-aware neural architecture search","venue":null,"work_id":"f31ead40-b718-40e6-83b1-eaed3bbef5ca","year":2021},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"cited_paper":"/paper/2101.09336","citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:4968b34601dd726ffbfda2ab98acd746ae2cd3fc00b1493f6c51030dc05807c7","observation_id":"a50ee451-063f-448f-9d81-8aa7269b1374","resolution":{"observed_at":"2026-06-30T03:04:14.312440Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-30T01:53:18.755724Z","title":"Neural ar- chitecture search: A survey.Journal of Machine Learning Research, 20(55):1–21, 2019","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:59b2fe46abb4861bfe4dc1143edfef5aa8505d55950b9ffe3b6cecc9a411003d","observation_id":"ed36c064-9a6b-412e-a18a-aedd837ea560","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Fbnet: Hardware-aware efficient convnet design via differ- entiable neural architecture search","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:60c75dc56b3c0bc50cac1eae710813abe0c8839cef03db91d1a15d184e9bf468","observation_id":"44738867-1321-48a6-ad9a-4b42147f68df","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1812.00332","last_updated":"2019-02-23T01:36:47Z","snapshot_observed_at":"2026-08-10T08:54:59.663457Z","submitted_at":"2018-12-02T05:29:53Z","title":"ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware","version":2},"cited_work":{"arxiv_id":"1812.00332","doi":"10.48550/arxiv.1812.00332","metadata_source":"pith","pith_arxiv_id":"1812.00332","snapshot_observed_at":"2026-08-05T02:28:24.338817Z","title":"ProxylessNAS: Direct Neural Architecture Search on Target Task and Hardware","venue":"cs.LG","work_id":"1b39088b-aa5f-44b5-8208-43a14b2eee32","year":2018},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"cited_paper":"/paper/1812.00332","citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:be77c737ccd7a9c843dc4401d2b2c9b5e8876b003e97496af7f66d6d0d5f5fa5","observation_id":"d1c643c4-359f-4be4-bb84-e8e86f387d4b","resolution":{"observed_at":"2026-06-30T03:04:14.317352Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-30T01:53:18.755724Z","title":"Memory- efficient patch-based inference for tiny deep learning.Advances in Neural Information Processing Systems, 34:2346–2358, 2021","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:76f12b7e6411eabfbc3febb79384f6ccc376569eeff42100e5456a4cc7858fb8","observation_id":"049969c3-3329-46ec-abff-0549c7453e03","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Pruning vs quantization: Which is better?Advances in neural information processing systems, 36:62414–62427, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:26b5f3c4038868f33a3451d94f1c607cc51c648d16faaeee2c39437d4b7a2dce","observation_id":"c5e4c333-43b9-4dac-ba52-33fd802ba609","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2311.11883","last_updated":"2023-11-20T16:20:13Z","snapshot_observed_at":"2026-08-10T06:35:36.878863Z","submitted_at":"2023-11-20T16:20:13Z","title":"Efficient Neural Networks for Tiny Machine Learning: A Comprehensive Review","version":1},"cited_work":{"arxiv_id":"2311.11883","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2311.11883","snapshot_observed_at":"2026-06-30T03:04:14.329904Z","title":"Efficient neural networks for tiny machine learning: A comprehensive review.arXiv preprint arXiv:2311.11883, 2023","venue":null,"work_id":"9eaa4e60-f761-4704-9779-e5937249fb6a","year":2023},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"cited_paper":"/paper/2311.11883","citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:c7cba22e3eb715d8ac1ed809951a5c4728b79265090d4307283eaa4baf7b8915","observation_id":"cc33adbf-92c3-4943-9203-442fb104fafa","resolution":{"observed_at":"2026-06-30T03:04:14.332330Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2304.10020","last_updated":"2023-04-20T00:17:27Z","snapshot_observed_at":"2026-08-10T13:28:20.995336Z","submitted_at":"2023-04-20T00:17:27Z","title":"A Survey on Deep Neural Network Partition over Cloud, Edge and End Devices","version":1},"cited_work":{"arxiv_id":"2304.10020","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2304.10020","snapshot_observed_at":"2026-06-30T03:04:14.294922Z","title":"A survey on deep neural network partition over cloud, edge and end devices.arXiv preprint arXiv:2304.10020, 2023","venue":null,"work_id":"63379d39-b1c0-48f9-990b-5356dd636815","year":2023},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"cited_paper":"/paper/2304.10020","citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:7b3f8fd02b2327532fef039f90a9ffc642e83ac39a535a68d5eb216881e1b8c7","observation_id":"0863311c-9fe4-4ef0-a8d1-08e88258aa8c","resolution":{"observed_at":"2026-06-30T03:04:14.296869Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-30T01:53:18.755724Z","title":"Survey of deep learning accelerators for edge and emerging computing.Electronics, 13(15):2988, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:932b40ddd733473095aa8d2fa46a0513231689201ba673e9d124ef11a957a60c","observation_id":"2ccd2c87-eccd-4a00-b83e-ddc7274ce065","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Efficient processing of deep neural networks: A tutorial and survey.Proceedings of the IEEE, 105(12):2295–2329, 2017","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:4a61ff669936900d5e606ac083d08b82906ab89e54e0d9fb0b0e95f8831a89a1","observation_id":"c1fc9af3-d914-459c-9f4f-1ebaaa5aa224","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"{SHEPHERD}: Serving{DNNs}in the wild","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:a089668f502a2146e5f64459f670fa6fa7a7935f01518473f43654621e6c77dc","observation_id":"fc82685a-9e43-43a3-b444-c200edf684d4","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Maeri: Enabling flexible dataflow mapping over dnn accelerators via recon- figurable interconnects.ACM Sigplan Notices, 53(2):461–475, 2018","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:c8a0e7b6036181159ec46caf79306f9d6bdf91db104aa6c25903ba38dba7e61f","observation_id":"443452c4-0d6d-401e-a4f2-31ad342cdd85","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Eyeriss: A spatial archi- tecture for energy-efficient dataflow for convolutional neural networks","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:d48d898b6f17bbbf5869235301774e044176cff5c10ffb7fd5a2ad8a1217c9ff","observation_id":"43868bdb-7e32-49e6-ae5e-2c5aa710944b","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"A formalism of dnn accelerator flexibility","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:1b869a6cecc724a8fa21af06a2de89122c741ae6556abff44c003474925d5e97","observation_id":"8e9e70de-3811-44e0-80c2-60dfe28fb16c","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Tpu v4: An optically reconfigurable supercomputer for machine learning with hardware support for embeddings","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:1ec7bf36576cbb9eac7355f1b9c83c52651576d9b7b9e79b0641234bbc3c14a0","observation_id":"00b16100-2c8e-4649-8c2f-a27a2b7334fb","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Approximation by superpositions of a sigmoidal function.Mathematics of control, signals and systems, 2(4):303–314, 1989","venue":null,"work_id":null,"year":1989},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:629399d5ddb87611bbb8bd642fb47c904503ec18a93ef2ebfe51fb3643a3e3b9","observation_id":"cc87a9e7-3d98-45bf-8515-cb6f3bf90afe","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Approximation capabilities of multilayer feedforward networks.Neural networks, 4(2):251–257, 1991","venue":null,"work_id":null,"year":1991},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:5e131c1dfe4cc5a3595cf9d9b51370a9069dca8074d232301c63eacf6ea297bf","observation_id":"cc3a58a3-b078-485c-9eab-c21886796323","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Error bounds for approximations with deep relu networks.Neural networks, 94:103–114, 2017","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:e2b36cee502c16ba109dc62dc90f7fc7ffd9dcfa510f52f46eae6cd29a11eac5","observation_id":"41be1f5a-0ee6-4ccd-87aa-2eff2336f3aa","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Optimal approximation rates for deep relu neural networks on sobolev and besov spaces.Journal of Machine Learning Research, 24(357):1–52, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:c94d415ba57f20676f079468f88f555ba7fc6de21d5b64eb02b867c3d04aa1de","observation_id":"d6170bc4-afe8-443e-93b3-0ce7cb023920","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"The expressive power of neural networks: A view from the width","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:88935fe01301107bb60d0aa5e7806516e4d202f7a8b320cd877d3915f0008ae1","observation_id":"e7a4d602-2f75-490d-bd3d-5bd4892978a4","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Neural networks with small weights and depth-separation barriers.Advances in neural information processing systems, 33:19433–19442, 2020","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:6afcbe532d73f980271f83c9448d579256ed04410c07d93f77f7174f0e6a8d81","observation_id":"29935294-0654-4a02-812a-d1da2c1965e5","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Optimal approximation of piecewise smooth functions using deep relu neural networks.Neural Networks, 108:296–330, 2018","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:d499c41bbf01a512486685876eeecded4f637c0d34b091e27a22f85f440f6be4","observation_id":"f01e097b-0d2e-4c8e-b3e9-99457d86a793","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1806.09055","last_updated":"2019-04-23T06:29:32Z","snapshot_observed_at":"2026-08-09T13:23:20.015254Z","submitted_at":"2018-06-24T00:06:13Z","title":"DARTS: Differentiable Architecture Search","version":2},"cited_work":{"arxiv_id":"1806.09055","doi":null,"metadata_source":"pith","pith_arxiv_id":"1806.09055","snapshot_observed_at":"2026-07-10T21:27:35.708872Z","title":"DARTS: Differentiable Architecture Search","venue":"cs.LG","work_id":"2f7f6e44-42e7-498e-a68f-a01e092a8903","year":2018},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"cited_paper":"/paper/1806.09055","citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:6f96021af973495e0db7942f2905248f8e612abd83db76a363ed5e83adba8555","observation_id":"d1fb4d68-9f33-42a8-a3df-9185d39dd578","resolution":{"observed_at":"2026-06-30T03:04:14.328209Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1909.06035","last_updated":"2020-10-20T06:21:28Z","snapshot_observed_at":"2026-07-06T08:21:25.975119Z","submitted_at":"2019-09-13T05:07:57Z","title":"DARTS+: Improved Differentiable Architecture Search with Early Stopping","version":2},"cited_work":{"arxiv_id":"1909.06035","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"1909.06035","snapshot_observed_at":"2026-06-30T07:34:22.002411Z","title":"DARTS+: Improved Differentiable Architecture Search wit h Early Stopping","venue":null,"work_id":"0a89f610-ba2c-4353-be76-8b568aa4fbd8","year":1909},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"cited_paper":"/paper/1909.06035","citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:ea1524a9b58f0cd02a31c98577573abf136914095046bd06d593653ac9d0e9a2","observation_id":"dcd1304b-d4aa-41b3-80f0-60279a79bfea","resolution":{"observed_at":"2026-06-30T03:04:14.301801Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1909.09656","last_updated":"2020-01-28T14:14:05Z","snapshot_observed_at":"2026-08-11T16:01:22.134125Z","submitted_at":"2019-09-20T18:03:06Z","title":"Understanding and Robustifying Differentiable Architecture Search","version":2},"cited_work":{"arxiv_id":"1909.09656","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"1909.09656","snapshot_observed_at":"2026-06-30T07:34:21.981917Z","title":"Understanding and robustifying differentiable architecture search.arXiv preprint arXiv:1909.09656, 2019","venue":null,"work_id":"cca19896-ef4f-473f-9b9f-fbe952a15965","year":1909},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"cited_paper":"/paper/1909.09656","citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:2d80af91ebb847b1119c195eec8276d86f7a85bc87226119ae2d6a5cb2b9ce6a","observation_id":"49a5ea82-91dc-4598-a2d9-2aeed04abde4","resolution":{"observed_at":"2026-06-30T03:04:14.307125Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-06-30T01:53:18.755724Z","title":"Fair darts: Eliminating unfair advantages in differentiable architecture search","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:64d39edf6bf575b215af33943cf3a5cabd45165348bddf1497145f3d0c85062d","observation_id":"345396c8-e73d-4a96-9324-71f20d921976","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","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":"2026-06-30T01:53:18.755724Z","title":"Ultra-low power dnn accelerators for iot: Resource characterization of the max78000","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-06-30T01:53:18.755724Z"},"links":{"citing_paper":"/paper/2606.29518"},"observation_digest":"sha256:c73d3ec233ae8fbfdbaf73b26e5419568473a870013b1fbd7f55dcaf433b2000","observation_id":"0e39a8c1-a270-449d-8142-0b917882315a","resolution":{"observed_at":"2026-06-30T01:53:18.755724Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2606.29518","last_updated":"2026-06-28T17:27:52Z","latest_version":1,"primary_category":"cs.AR","snapshot_observed_at":"2026-08-06T14:17:14.456799Z","submitted_at":"2026-06-28T17:27:52Z","title":"Harvesting AI Computation at the Edge via Generic Approximation"},"reference_resolution":{"displayed":42,"state_counts":{"malformed_identifier":0,"metadata_mismatch":1,"parse_uncertain":0,"unresolved":30,"verified_exact":11,"verified_fuzzy":0},"total_outbound_references":42},"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":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"thesis":"As of 12 August 2026, this Paper Citation Record lists 42 of 42 outbound references and 0 inbound Pith citation observations for arXiv:2606.29518."}