{"as_of":"2026-08-21T11:11:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:f7df235b2bd4b3cd2c79af8c5dd097e1e0f9523c71915890bc95d9dd8a0b6283","coverage":[{"denominator":52,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":52,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-10T14:26:25.962597Z","state":"measured"},{"denominator":52,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":52,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-21T06:32:19.484+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/2501.15403/citation-record","integrity":"/paper/2501.15403/integrity","json":"/paper/2501.15403/citation-record.json","paper":"/paper/2501.15403"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:26.551568Z","title":"& Fitch, A","venue":null,"work_id":"f61bdd3a-96c3-4fae-88de-d01aad3d23e6","year":2024},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.794117Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:0c70cd9a6724e1c19030cd72823475b33c38a3c11ce404181109563a2d30fb28","observation_id":"788d188f-1db3-4ca6-8fc8-970f4989ff9a","resolution":{"observed_at":"2026-08-10T14:26:26.555272Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2212.13345","last_updated":"2022-12-27T02:54:46Z","snapshot_observed_at":"2026-08-16T16:05:54.047188Z","submitted_at":"2022-12-27T02:54:46Z","title":"The Forward-Forward Algorithm: Some Preliminary Investigations","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2212.13345","snapshot_observed_at":"2026-08-10T14:26:25.798193Z","title":"The Forward-Forward Algorithm: Some Preliminary Investigations 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.798193Z"},"links":{"cited_paper":"/paper/2212.13345","citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:4dc2af012746b8c904433cde0570f96c673655bdfd7ecdb4dcf63a8ea5fcd7df","observation_id":"d010a68c-6f01-4516-a8fa-c9d13e55e046","resolution":{"observed_at":"2026-08-10T14:26:25.798193Z","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":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:26.542631Z","title":"P., Santoro, A., Marris, L., Akerman, C","venue":null,"work_id":"9d0743ef-610a-474e-90fa-52789896d044","year":2020},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.802088Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:f8281ff8d7b6421133e4a0cc1f60b5c8f9a786ac73f20da5da386cb235993c9e","observation_id":"1f7c5ba6-36cf-4597-a217-9ea8a92f7e80","resolution":{"observed_at":"2026-08-10T14:26:26.545947Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.533970Z","title":null,"venue":null,"work_id":"1b401844-563a-4a16-b25c-e2f378146419","year":2018},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.805428Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:1ef86b71ff3933b8f6850a8f50adc001e95842bf13bd1a85699278c3868ece5c","observation_id":"923a423b-5218-4713-9cea-4fc6ad7485a3","resolution":{"observed_at":"2026-08-10T14:26:26.537140Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.525335Z","title":"Science 380, 398–404","venue":null,"work_id":"3162b5ae-698a-4665-91c0-f49791e31ff3","year":2023},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.808728Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:998cc2634026a779fe7bd4874a9f6b35b8eee8a14800cedfbccaad80d362a89b","observation_id":"870d5b29-7bb5-4830-9a6d-6a3ad515bb24","resolution":{"observed_at":"2026-08-10T14:26:26.528293Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.516923Z","title":null,"venue":null,"work_id":"76431b43-e478-4a5f-aac9-0ace203bdcba","year":2020},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.812034Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:3e5119d9f91b6b2ee24a26d13d3cd99ede9b131c2cb7746468d33347328d0074","observation_id":"ee50ff5f-8f06-4deb-88ac-3d02b4dd76b3","resolution":{"observed_at":"2026-08-10T14:26:26.520064Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.506648Z","title":null,"venue":null,"work_id":"76955ba2-ef9f-493d-83e6-59d7a9e4bfc5","year":null},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.815333Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:4d2257c91e18a466f930c1d997941f50e447299c2e2f9e96f879c012864eaf2a","observation_id":"a9f8fb05-1e80-472f-98f8-35decbf9b21a","resolution":{"observed_at":"2026-08-10T14:26:26.510730Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.497698Z","title":"Nature Communications 2023 14:1 14, 1–18","venue":null,"work_id":"4cf152cb-bc08-4ad8-b151-8c0299838a62","year":2023},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.821646Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:20689250f2d035f87a6e80398784344cac5d249e449824c12380d45bbf0c17ea","observation_id":"76821d90-ba6a-481f-b395-96c1dd2dce80","resolution":{"observed_at":"2026-08-10T14:26:26.500910Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.489108Z","title":"& McCaughan, A","venue":null,"work_id":"27d292af-45f7-4eaa-a2cf-ee8c1d6b897f","year":2022},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.824657Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:386d7888c7f1234eacdac92bb11c9588d889ddc5e05b973faed6d3d69dba2daa","observation_id":"f3d88631-074b-4064-9c35-15ba6415be44","resolution":{"observed_at":"2026-08-10T14:26:26.492018Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:25.818530Z","title":"https://www.science.org/doi/10.1126/sciadv.ado8999 (28 July 2024)","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.818530Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:a2448a54067e7972bcc896b8b7f5ad38506c9a8cd2a9562338f1b30131fa6d10","observation_id":"7260fc34-2651-4b23-80bc-d974290bafb6","resolution":{"observed_at":"2026-08-10T14:26:25.818530Z","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":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:26.480699Z","title":null,"venue":null,"work_id":"0ee0bde9-2139-4f23-a601-e11235f6ae81","year":2022},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.827857Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:e62cfd531f8a0bf592f356acd30f14462ef1d44144549a78ce04f1b02e61ec43","observation_id":"db4e7abd-f0f7-4a2a-8f33-a7e735bd6972","resolution":{"observed_at":"2026-08-10T14:26:26.483758Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.472593Z","title":null,"venue":null,"work_id":"7045a810-6f3a-4fe5-961b-7c8284ae0c31","year":2012},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.830937Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:8e62eb7f71dd86da887f2c095c691eb0a6b4702fd7c530858f5d20f65594e903","observation_id":"89f42011-68c4-48e0-b4ee-f98aa6b08b66","resolution":{"observed_at":"2026-08-10T14:26:26.475461Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.463356Z","title":null,"venue":null,"work_id":"20daf277-71e5-456a-82cd-8ba8abf2acee","year":2017},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.833845Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:8b0a06de4911f99d377baaca0c5244a2f2f797b74f9965b36874d144781ec751","observation_id":"2396ed02-007a-4e92-8448-86ca05709062","resolution":{"observed_at":"2026-08-10T14:26:26.466830Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.453836Z","title":"Frontiers in Neural Circuits 0, 53","venue":null,"work_id":"1f47fbfb-874a-48c7-8452-672de6caa466","year":2018},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.836895Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:a0d0e6425ed5db727a1d9d648f3265c21647f8b65bde012756c57936f8c20cb3","observation_id":"0291fed0-45ee-4476-ae18-1c95f9ea2892","resolution":{"observed_at":"2026-08-10T14:26:26.457010Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.444910Z","title":"& Kailath, T","venue":null,"work_id":"2f1c4367-01af-4b1b-b921-c2212dcde9ea","year":1990},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.840155Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:9a87def5d5aff84369e65eb41f80bc8106ea20a83d6f2a83812a3c06f14e974c","observation_id":"703e0733-6dbf-416d-b439-1c664a4e7b69","resolution":{"observed_at":"2026-08-10T14:26:26.448038Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.436035Z","title":"& Koga, M","venue":null,"work_id":"d834942b-297c-4b1c-820a-0b07f2419b3b","year":1990},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.842982Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:6be491c928f4d0a88958f65fb99262df0143674b645669fd5380d0f9598cdf9b","observation_id":"a5b44ae2-306d-48f6-8caa-10508e901c22","resolution":{"observed_at":"2026-08-10T14:26:26.439302Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.426864Z","title":"A Fast Stochastic Error-Descent Algorithm for Supervised Learning and Optimization in Advances in Neural Information Processing Systems 5 (NIPS 1992) (1992), 244–251","venue":null,"work_id":"1b6ff0d3-6215-41ef-b35b-bb79b2102fb8","year":1992},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.845971Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:ec1c44dbadf16f9eba2bb056ade3eaa8b3a70572f8f93d60b016c6f8f6d25a8f","observation_id":"31a38e50-6c31-4fb3-b71b-b6c4c9dfac40","resolution":{"observed_at":"2026-08-10T14:26:26.430405Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.417845Z","title":"& Jabri, M","venue":null,"work_id":"c2ff2b09-c0ee-4f5e-a626-bbd2511ebaac","year":1992},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.849151Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:6a81bbd4fcd2b3e54582e1175bcdebd0e707710851d4ccd87ef573dcd63c9cb4","observation_id":"f50d2042-d700-479b-b746-03755a43706e","resolution":{"observed_at":"2026-08-10T14:26:26.421069Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.408830Z","title":"Alspector, R","venue":null,"work_id":"3f518613-8a2f-4bb8-9c52-54baba7326a0","year":1992},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.852442Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:e1c1992c27c5ccf413532ad8091bfc52d2ccde2492f724adcc1a9416e8c07fa9","observation_id":"14c2d703-6469-4a4e-9b6f-f30276686707","resolution":{"observed_at":"2026-08-10T14:26:26.411959Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.400215Z","title":null,"venue":null,"work_id":"dc8c935c-3516-4392-96dd-c6287a3e9a46","year":1992},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.859634Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:10ca41f1b75fcf573f99f22c4fd2ac7fe4b1ac460a71ba585c9ecf5250d37788","observation_id":"76228c3b-6898-4a99-8c73-8e7233c65ec7","resolution":{"observed_at":"2026-08-10T14:26:26.403300Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.391085Z","title":"& Kanata, Y","venue":null,"work_id":"24e1472c-42ac-440b-927d-d754a3bc2c82","year":1995},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.864893Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:b89c7664c98bbe186d8de512965a2ea5a359399ab964dd57e20135763b49ee54","observation_id":"4e9ceff3-839b-44d2-aa6c-7af36f1ff8f9","resolution":{"observed_at":"2026-08-10T14:26:26.394547Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.382045Z","title":"An analog VLSI recurrent neural network learning a continuous-time trajectory","venue":null,"work_id":"cffd562c-4095-4225-aaaf-0971b343928a","year":1996},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.868613Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:fa0a632c11c6988e7931423a59ee0bfa802bbe7d397f35427b8517f70902ee39","observation_id":"042b7664-594b-40f4-b9d5-36e7d6f99e1a","resolution":{"observed_at":"2026-08-10T14:26:26.385338Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.373002Z","title":"& Fiesler, E","venue":null,"work_id":"1c554559-296d-47d0-a110-1c765124fe39","year":1996},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.871604Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:75f2a92722423d12131c14ef6b46592073ae1c14c799e4b86601711b6d51be9e","observation_id":"96e736e3-0e2f-481a-aef8-b22d5dbe6c50","resolution":{"observed_at":"2026-08-10T14:26:26.376121Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.363188Z","title":"J., Gyurcsik, R","venue":null,"work_id":"71b26e7e-5fbb-445f-b780-f47f06f6c7c2","year":1997},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.874274Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:5ddcdc7a7a4e097a52f3e3d26188029b27f2b484e73db6b2804895cea9c26dba","observation_id":"17265be3-78ec-4f75-a8df-31813b206ac3","resolution":{"observed_at":"2026-08-10T14:26:26.366718Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.354052Z","title":"& Matsumoto, T","venue":null,"work_id":"e0b08b9e-f2d4-4961-8511-6a004853b952","year":1997},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.876989Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:1ed1db43ae35c16e637c31ebd37e948ff9eea6c35a3823f8714b2d164d68a271","observation_id":"b276c8c9-71d7-427f-b81f-4ca22c5705d7","resolution":{"observed_at":"2026-08-10T14:26:26.356882Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.344263Z","title":"Neural Networks in Analog Hardware - Design and Implementation Issues","venue":null,"work_id":"205e25ce-4c34-4aff-b6da-199aa52ddf5d","year":2000},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.879755Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:7fc63ebc390500ca310996bf87b104b91adf4f41561401857922297b6d3c59b1","observation_id":"26f91214-3488-4a09-b138-6e5fff5ee1e8","resolution":{"observed_at":"2026-08-10T14:26:26.347614Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.334484Z","title":"Analog VLSI Autonomous Systems for Learning and Optimization PhD thesis (Caltech, 1994)","venue":null,"work_id":"690464f7-9ebd-4d5c-bcbe-1f48a9fe0ba8","year":1994},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.882592Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:6aa3169302415fb3ce6393c81a1ea58b03ca50c4c03847233553f6110438a200","observation_id":"3957f906-7d82-47ef-8adb-ad17072c507a","resolution":{"observed_at":"2026-08-10T14:26:26.337993Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:25.885229Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.885229Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:2b607090691415d0bd0f2f6962dd37eabceb65f19a0a12d20a3076c15edf5bf1","observation_id":"6a4782a1-d926-497d-b05f-16b7153649aa","resolution":{"observed_at":"2026-08-10T14:26:25.885229Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.00965","last_updated":"2026-06-02T14:21:05Z","snapshot_observed_at":"2026-08-19T00:48:50.193600Z","submitted_at":"2023-10-02T08:12:51Z","title":"Node Perturbation Can Effectively Train Multi-Layer Neural Networks","version":8},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.00965","snapshot_observed_at":"2026-08-10T14:26:25.888258Z","title":"& Ahmad, N","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.888258Z"},"links":{"cited_paper":"/paper/2310.00965","citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:cbfcb7b0399c13859a73ae3c033418345c8de7aa05561b36daf3222186046102","observation_id":"8d100268-1d1f-4f73-a6ac-9ce38e4af91a","resolution":{"observed_at":"2026-08-10T14:26:25.888258Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2210.03310","last_updated":"2023-03-02T03:08:10Z","snapshot_observed_at":"2026-08-16T16:26:13.880878Z","submitted_at":"2022-10-07T03:52:27Z","title":"Scaling Forward Gradient With Local Losses","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2210.03310","snapshot_observed_at":"2026-08-10T14:26:25.891741Z","title":"& Hinton, G","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.891741Z"},"links":{"cited_paper":"/paper/2210.03310","citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:c412a28ef1c0d2f006ec8ca3d0c245b8a98c191fd215358ad019b8dfeb8f6670","observation_id":"6ee81e4e-4fd0-43df-b402-61312782fb8d","resolution":{"observed_at":"2026-08-10T14:26:25.891741Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2308.09858","last_updated":"2023-10-09T18:00:00Z","snapshot_observed_at":"2026-08-16T15:07:24.442835Z","submitted_at":"2023-08-18T23:56:50Z","title":"Tensor-Compressed Back-Propagation-Free Training for (Physics-Informed) Neural Networks","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2308.09858","snapshot_observed_at":"2026-08-10T14:26:25.895044Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.895044Z"},"links":{"cited_paper":"/paper/2308.09858","citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:47f7c34a726d987b1d0785a74e21dbab6c8e88ade356431857a025fc5f0aed3d","observation_id":"ffe2364f-0059-4222-98b6-3b8fd3eb18ff","resolution":{"observed_at":"2026-08-10T14:26:25.895044Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2208.01623","last_updated":"2022-08-02T17:54:17Z","snapshot_observed_at":"2026-08-20T14:48:45.926049Z","submitted_at":"2022-08-02T17:54:17Z","title":"Single chip photonic deep neural network with accelerated training","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2208.01623","snapshot_observed_at":"2026-08-10T14:26:25.898595Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.898595Z"},"links":{"cited_paper":"/paper/2208.01623","citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:61f0c7794e717ca9f00ccef538a8a6fcc48713df7415e55819ef282af677df9d","observation_id":"1c985b14-056e-4358-969b-f321ea7a87b6","resolution":{"observed_at":"2026-08-10T14:26:25.898595Z","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":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:26.325170Z","title":null,"venue":null,"work_id":"4467f632-c5d4-4bf9-8758-2adc02013946","year":1992},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.902149Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:dc6017c5bdd1ece9655d3f22f0f233926a61891f59b4e440fe387070ba331663","observation_id":"8532de4d-8089-4727-adaf-9ea43f4ec8ea","resolution":{"observed_at":"2026-08-10T14:26:26.328295Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.217456Z","title":"P., Cownden, D., Tweed, D","venue":null,"work_id":"0cbd9b75-7019-4c15-9a55-04487118be57","year":null},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.905075Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:20400cc6b6451fe660599b5ef494962693d8ff6671708b39fd1e6193366511b9","observation_id":"440c4199-daf8-451e-91ad-09d805931ce6","resolution":{"observed_at":"2026-08-10T14:26:26.318746Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:25.908364Z","title":null,"venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.908364Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:bb51b852c12b8982dd334adc2edaaa3847670e795f173f0028f024cb70836410","observation_id":"e5adbaea-ca13-4b52-8c58-63ab3ed73454","resolution":{"observed_at":"2026-08-10T14:26:25.908364Z","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":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:26.208704Z","title":"& Memmesheimer, R","venue":null,"work_id":"c5d7b436-3bfd-42e8-9d8e-e634ecf71234","year":2023},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.911405Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:169c88179113c4b1674c4d58ac60d8156cfccd1b9c673794764b41fbba52f218","observation_id":"8a5fada7-39ce-404f-b59b-7278e3610a31","resolution":{"observed_at":"2026-08-10T14:26:26.211803Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1412.6980","last_updated":"2017-01-30T01:27:54Z","snapshot_observed_at":"2026-08-17T19:26:44.032537Z","submitted_at":"2014-12-22T13:54:29Z","title":"Adam: A Method for Stochastic Optimization","version":9},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1412.6980","snapshot_observed_at":"2026-08-10T14:26:25.914564Z","title":null,"venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.914564Z"},"links":{"cited_paper":"/paper/1412.6980","citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:870c7faa6bfe47e585c7aafbf1a75ff19dfa233b23a791a22cbd8f05a9bdee34","observation_id":"4ad7988e-fafc-4ffe-ae54-1280fd154243","resolution":{"observed_at":"2026-08-10T14:26:25.914564Z","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":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:26.199967Z","title":"& Hinton, G","venue":null,"work_id":"511c0f0a-f115-471d-b017-0bdc508976eb","year":2022},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.917855Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:a61587400533b9216eb1058f2b8f42ea1992e3ce1c2cdc3f337e81183e30caa9","observation_id":"276245be-c4e4-42a0-9e16-9f8ea47dcdc7","resolution":{"observed_at":"2026-08-10T14:26:26.202970Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.189090Z","title":null,"venue":null,"work_id":"2259d137-f607-40d9-954e-0ca85ea017ce","year":2024},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.920921Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:498df58fbf78d957f46c319443fe88df483a2fdcd77d0a9c21f6cb9c57787b15","observation_id":"f2e053ff-091b-4a95-aa06-fcc37d743780","resolution":{"observed_at":"2026-08-10T14:26:26.192549Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.178561Z","title":"B., Zhu, Q","venue":null,"work_id":"7ca9c812-31ac-4c6d-97e6-f604384a30b7","year":2006},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.924314Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:2057a5639ad884a80e9d2ddf7f6e663809b33be58a0a67edf0bd88ac61fd6221","observation_id":"211ce802-bf91-46b2-81b0-1ed7a8ffd687","resolution":{"observed_at":"2026-08-10T14:26:26.182373Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.168385Z","title":"How we created neuromorphic engineering","venue":null,"work_id":"367273f1-4ad6-4126-b18a-f0302da628ef","year":2020},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.927302Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:048b52f65f53b6eb6c405a601389a996f8e2d4a26b82449e9096bb44f3426df0","observation_id":"d54cfc81-36f7-4b90-9e5b-58fa8a3390af","resolution":{"observed_at":"2026-08-10T14:26:26.171804Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.157862Z","title":"P., Kim, H., Budhathoki, R","venue":null,"work_id":"12e5ab39-9182-4885-a89d-017914c72344","year":2015},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.930516Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:2e3d37572bd89b932ff3f0045b5cde3cf3def0382b68fd1ee6cae63959d91196","observation_id":"89a1a2e1-e990-4eb8-a549-76fafdf1fceb","resolution":{"observed_at":"2026-08-10T14:26:26.161391Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.147120Z","title":"& Yao, W","venue":null,"work_id":"09091b22-13dd-4947-81d7-b58bb03e3cb4","year":2019},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.934223Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:7ac84b119a6cc5b5183dfe1fefe9253115907137a646704f24a5edea1b9ee7a5","observation_id":"b0858c63-e040-4be3-8eae-4ac51538838d","resolution":{"observed_at":"2026-08-10T14:26:26.150587Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.136862Z","title":null,"venue":null,"work_id":"1edb8baf-599e-4681-b66f-9ffcc719299e","year":2003},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.937221Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:a3a4b754a0e4e15d082809067a9056726a4863b816b493a2960d0dcb8ef775fc","observation_id":"c08c96af-d8df-48b2-b0c2-26120cff958a","resolution":{"observed_at":"2026-08-10T14:26:26.140331Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:25.940272Z","title":null,"venue":null,"work_id":null,"year":2006},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.940272Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:11489d3a4a6b082bd78161e54c815604d34520c488a061a812af1cd030f2b653","observation_id":"fc7c5287-c3b7-4e19-8501-7bedb10837d2","resolution":{"observed_at":"2026-08-10T14:26:25.940272Z","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":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T14:26:26.127791Z","title":"& Seung, H","venue":null,"work_id":"bfa78156-1d6f-4914-91bb-667365e0c77e","year":2005},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.943439Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:ac55c14e14dd241b61e47328bcc7451753da85464eed8c5e258e2dcb05ffe92f","observation_id":"8c4a306c-5299-4e25-8245-396862b89905","resolution":{"observed_at":"2026-08-10T14:26:26.130923Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.118450Z","title":null,"venue":null,"work_id":"aad3f286-805e-440f-8151-286258169335","year":2023},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.946377Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:9b658e885d008a0afdb3f7a63ec0aecf2795af0da6c64595ceafe2bcdc802129","observation_id":"2186c05b-c03f-4918-8e7c-4e6b7937f0a4","resolution":{"observed_at":"2026-08-10T14:26:26.121461Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2407.12516","last_updated":"2024-07-17T12:09:00Z","snapshot_observed_at":"2026-08-16T21:48:56.059350Z","submitted_at":"2024-07-17T12:09:00Z","title":"Online Pseudo-Zeroth-Order Training of Neuromorphic Spiking Neural Networks","version":1},"cited_work":{"arxiv_id":"2407.12516","doi":null,"metadata_source":"pith","pith_arxiv_id":"2407.12516","snapshot_observed_at":"2026-08-10T14:26:26.040741Z","title":"Online Pseudo-Zeroth-Order Training of Neuromorphic Spiking Neural Networks","venue":"cs.NE","work_id":"de91fdf2-b9d4-4fc6-acde-43d145ffbca7","year":2024},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.949500Z"},"links":{"cited_paper":"/paper/2407.12516","citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:219052255f8fae9fa4f53557c083dd9628da0741df1d42526af13181ba5ebf29","observation_id":"2ba89773-aaff-4db9-98eb-0bb241d464ba","resolution":{"observed_at":"2026-08-10T14:26:26.046378Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+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":"2026-08-10T14:26:26.108425Z","title":null,"venue":null,"work_id":"f02815f5-6af8-4558-8b26-62ef422d2985","year":2022},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.952717Z"},"links":{"citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:5d26e69418fcaf935a5a70d643246b584336f9dca79d83355902deb2bcd49542","observation_id":"76ed7de7-314d-4c3d-8b8e-f4afa5c30b95","resolution":{"observed_at":"2026-08-10T14:26:26.111732Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-21T06:32:19.484+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.17333","last_updated":"2024-01-11T13:56:52Z","snapshot_observed_at":"2026-08-16T15:29:07.086507Z","submitted_at":"2023-05-27T02:28:10Z","title":"Fine-Tuning Language Models with Just Forward Passes","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.17333","snapshot_observed_at":"2026-08-10T14:26:25.955777Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.955777Z"},"links":{"cited_paper":"/paper/2305.17333","citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:fcdc280df98c9fbff0ba5443a12c0e9e44ee0ace724bfe365b0f103a8e6c940b","observation_id":"70e296b1-838a-4e7d-a420-a109fb1b3255","resolution":{"observed_at":"2026-08-10T14:26:25.955777Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.02025","last_updated":"2024-03-15T15:28:11Z","snapshot_observed_at":"2026-08-20T14:44:28.906746Z","submitted_at":"2023-10-03T13:05:36Z","title":"DeepZero: Scaling up Zeroth-Order Optimization for Deep Model Training","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.02025","snapshot_observed_at":"2026-08-10T14:26:25.959322Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.959322Z"},"links":{"cited_paper":"/paper/2310.02025","citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:de4df4119022a8c9b545056d27dcca75cd2ab14c80983965f821dd8817fb2cd2","observation_id":"68b40dc8-cfa9-44f4-82a0-b14ea6a8f9a3","resolution":{"observed_at":"2026-08-10T14:26:25.959322Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2403.14608","last_updated":"2024-09-16T02:54:50Z","snapshot_observed_at":"2026-08-04T09:07:42.158421Z","submitted_at":"2024-03-21T17:55:50Z","title":"Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey","version":7},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.14608","snapshot_observed_at":"2026-08-10T14:26:25.962597Z","title":"& Zhang, S","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms","version":1},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-10T14:26:25.962597Z"},"links":{"cited_paper":"/paper/2403.14608","citing_paper":"/paper/2501.15403"},"observation_digest":"sha256:3a3c868f93946e4b66f7c1d519ace1fc10fe08f58857101d90efb33e974357cf","observation_id":"812a9399-e183-469c-a2d7-fef86edc55de","resolution":{"observed_at":"2026-08-10T14:26:25.962597Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2501.15403","last_updated":"2025-01-26T05:11:42Z","latest_version":1,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-16T09:52:32.951242Z","submitted_at":"2025-01-26T05:11:42Z","title":"Scaling of hardware-compatible perturbative training algorithms"},"reference_resolution":{"displayed":52,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":25,"verified_exact":1,"verified_fuzzy":26},"total_outbound_references":52},"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":"2026-08-21T06:32:19.484+00:00","source":"crossref"},{"observed_at":"2026-08-21T06:32:16.066871+00:00","source":"retraction_watch"}],"thesis":"As of 21 August 2026, this Paper Citation Record lists 52 of 52 outbound references and 0 inbound Pith citation observations for arXiv:2501.15403."}