{"as_of":"2026-08-17T16:29:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:e600debe6028528c0e13a7f40477cad4037016393d3f43f9c6bf65801589e180","coverage":[{"denominator":67,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":67,"source":"paper_references, paper_reference_links","source_observed_at":"2026-07-12T12:29:35.403453Z","state":"measured"},{"denominator":67,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":67,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-17T06:30:58.91139+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/2606.24722/citation-record","integrity":"/paper/2606.24722/integrity","json":"/paper/2606.24722/citation-record.json","paper":"/paper/2606.24722"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"Aethir whitepaper","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":1,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:b47e7f535dd1708f6d041d60e7e267667752db89792136ebc037b259d2885f0d","observation_id":"19f9ff04-e378-46af-b535-aacc2aac2a54","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Akash network: Decentralized cloud infrastructure marketplace","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":2,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:b1426b602d5a3edfbf69db8e9dfbf35856458007987cf9c6231ea9d441237dff","observation_id":"b4946433-7f79-4343-83dd-78c88d351913","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"QSGD : Communication-efficient SGD via gradient quantization and encoding","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":3,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:40f328e96bc5b79c708c1eee2cd0a2dd44b7a44aec16cb272448fcea1bf3a7f7","observation_id":"14e930d0-716a-46ce-bfda-2484d22d3ea5","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Stochastic gradient push for distributed deep learning","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":4,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:e143b48137300f48d55337f8db6bc89d6c64210463de62b7ad2d0529c6eefdc5","observation_id":"0ab0aac6-b262-4569-b06c-3d469f8f32e2","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Greedy layerwise learning can scale to ImageNet","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":5,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:d8ada60a1eb8320bc057fbe320dea88515028609f33a7b3e5a670102144fd4b4","observation_id":"100afbbc-9688-4052-9921-b49f09ae9268","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1407.7906","last_updated":"2014-09-18T13:30:31Z","snapshot_observed_at":"2026-08-14T23:24:56.180622Z","submitted_at":"2014-07-29T23:32:44Z","title":"How Auto-Encoders Could Provide Credit Assignment in Deep Networks via Target Propagation","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1407.7906","snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"How auto-encoders could provide credit assignment in deep networks via target propagation","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":6,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/1407.7906","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:18ef1e307a71da910874f4f9d2a8b500498006d1e37728cffe1990acaa07ae09","observation_id":"9e4d162b-f32b-4db1-ad14-6fea2b35386e","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Greedy layer-wise training of deep networks","venue":null,"work_id":null,"year":2007},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":7,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:2659fc3d4b8c01f1701a583f5534af910e6905e23a1404bed0853e02c3cb3f91","observation_id":"a859c551-f4d1-4ad7-8613-dd6b32adc55a","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2007.14390","last_updated":"2022-03-05T20:30:32Z","snapshot_observed_at":"2026-07-06T09:42:35.058716Z","submitted_at":"2020-07-28T17:59:07Z","title":"Flower: A Friendly Federated Learning Research Framework","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2007.14390","snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"Beutel, Taner Topal, Akhil Mathur, Xinchi Qiu, Titouan Parcollet, and Nicholas D","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":8,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/2007.14390","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:f16d0f53c96c8d02f3a1454e3607535bd54c2c61802a30a9614c89c8528a88a8","observation_id":"247f4aeb-23bd-4916-8580-02635402f24d","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Training transformers together","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":9,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:4674bf32f34d7d18d3520e8de31d8ea92c8d2a3b17a10aa5dffbc7190e533d74","observation_id":"ad48dd8c-258e-4152-a3fb-27a6f1be971f","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2209.01188","last_updated":"2023-03-02T19:33:31Z","snapshot_observed_at":"2026-08-16T16:35:03.571411Z","submitted_at":"2022-09-02T17:38:03Z","title":"Petals: Collaborative Inference and Fine-tuning of Large Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2209.01188","snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"Petals: Collaborative inference and fine-tuning of large models","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":10,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/2209.01188","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:19fce291c25685c6c67362077e25da6cb7f52ba91f9b875458d6fb7ac7841d19","observation_id":"405d78aa-0b88-4617-a7de-1aecc61a7cc5","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Distributed deep learning in open collaborations","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":11,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:4f29ee7e17fb761e271383c5594a37b168eee587dcd441f31dacd01bb5d22abb","observation_id":"79c94076-ebfd-4cdb-906a-d328ff1339c0","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2211.04344","last_updated":"2022-11-05T06:14:44Z","snapshot_observed_at":"2026-08-16T16:18:26.506375Z","submitted_at":"2022-11-05T06:14:44Z","title":"FLock: Defending Malicious Behaviors in Federated Learning with Blockchain","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2211.04344","snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"FLock : Defending malicious behaviors in federated learning with blockchain","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":12,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/2211.04344","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:5423c74f68cd7cb06bbf8e34998aa58a1cd5e9f7269c27eb7c8f805807aeb9af","observation_id":"581f7009-d95a-4aa1-9150-306cd498b727","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2403.10616","last_updated":"2024-03-15T18:26:51Z","snapshot_observed_at":"2026-08-16T14:09:21.078028Z","submitted_at":"2024-03-15T18:26:51Z","title":"DiPaCo: Distributed Path Composition","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.10616","snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"Rusu, et al","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":13,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/2403.10616","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:bad3daf14412f2e5f28a492e6a8d6640ad3ce83f640e13c31234642cef83304b","observation_id":"51646666-7d4f-4119-ba3b-3c88e434897f","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"DiLoCo : Distributed low-communication training of language models","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":14,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:c8ded407d06bffefbd5579eb01dd213f88a0c4c856488ada0880f46ce9dba4f2","observation_id":"103db19b-15ab-43dd-8925-a843cfb19809","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Joint statement on competition in generative AI foundation models and AI products","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":15,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:1fcf2364edb97e3dd3d6d1af527f437c92b1548368125522f1803a1fad77c5e6","observation_id":"6491a8b3-c876-44e0-90ed-f4ebe26a1a39","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"FLock : Federated machine learning on blockchain","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":16,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:3255daac93976a01a9c5e6fbce0ff14aac0860817a9f34fab72aca1a6cd08563","observation_id":"f10e8f5f-d9d7-40aa-a34d-775c37f6cd5c","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2309.01172","last_updated":"2023-09-03T13:27:56Z","snapshot_observed_at":"2026-08-16T15:03:42.094374Z","submitted_at":"2023-09-03T13:27:56Z","title":"FusionAI: Decentralized Training and Deploying LLMs with Massive Consumer-Level GPUs","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2309.01172","snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"FusionAI : Decentralized training and deploying LLMs with massive consumer-level GPUs , 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":17,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/2309.01172","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:87643a57f5cf21d17d63485733beb211fb228a7512db99d55c9d2c2b2a4adb5f","observation_id":"47e69b2b-95f3-4a1d-afc4-e5ac12db2217","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2410.12707","last_updated":"2024-10-16T16:13:19Z","snapshot_observed_at":"2026-08-17T13:00:53.411450Z","submitted_at":"2024-10-16T16:13:19Z","title":"FusionLLM: A Decentralized LLM Training System on Geo-distributed GPUs with Adaptive Compression","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.12707","snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"FusionLLM : A decentralized LLM training system on geo-distributed GPUs with adaptive compression, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":18,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/2410.12707","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:d9579f8909f67415a49d34670ba24dab05b1d3cdd7a21d6d2bbd8be1b794a87a","observation_id":"7ad81ff6-a1fe-4f25-9ec2-390217962095","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"RL swarm: A framework for collaborative reinforcement learning","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":19,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:cc6d093d128b85154c399263b5d3e2af14c9850e558b8492a2234addf1601753","observation_id":"5bfaf590-4352-4e56-ab3f-c55b92b64f94","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2007.13518","last_updated":"2020-11-08T19:34:25Z","snapshot_observed_at":"2026-08-15T04:01:16.573126Z","submitted_at":"2020-07-27T13:02:08Z","title":"FedML: A Research Library and Benchmark for Federated Machine Learning","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2007.13518","snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"FedML : A research library and benchmark for federated machine learning","venue":null,"work_id":null,"year":2007},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":20,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/2007.13518","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:24d1b037661fea97dc1c062ae942c4fcb6dae48d338bc48a537ce0acca2cece4","observation_id":"ec9a9fa0-f6a3-4e81-9a2b-426109a53c66","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"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":"2026-07-12T12:29:35.403453Z","title":"The forward-forward algorithm: Some preliminary investigations","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/2212.13345","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:5e2aadd092ef7101ca48bc12e2ff9b8e2807b348faf04e9f9a9bd48f5cfe4e7b","observation_id":"dcd30402-e6af-4aed-98df-aa5577ee0fff","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Denoising diffusion probabilistic models","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":22,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:3d5ffd89fe53721fcd4f63f65b0d04c2170ab7843921d21a3c280b5377ca2b50","observation_id":"db77157e-2b88-4424-9b1e-bdbfbb079cf3","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Estimation of non-normalized statistical models by score matching","venue":null,"work_id":null,"year":2005},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":23,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:31f6bb687f12803a36474a8fe2c5adac1a53d54d188a3febd98af0af55bf52c4","observation_id":"f5bdaa8d-cd99-4d99-b64a-684f4083ebe2","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Decoupled neural interfaces using synthetic gradients","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":24,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:0102f030b798bba0a53552ce68297951b54759af9be3479e44ef3a302b6597b0","observation_id":"ebdc850d-ab2b-471e-b7ab-9bc46d4a425f","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2412.01152","last_updated":"2024-12-02T05:52:32Z","snapshot_observed_at":"2026-08-16T04:43:09.667320Z","submitted_at":"2024-12-02T05:52:32Z","title":"INTELLECT-1 Technical Report","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2412.01152","snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"INTELLECT-1 technical report","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/2412.01152","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:e0462ad198b5442fd554b2fa50cc96ee815b3f02abb380e250dc2b7f66a273a5","observation_id":"2d469e1d-4c0b-4743-8b6e-15df9c522b53","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2407.07852","last_updated":"2024-07-10T17:13:17Z","snapshot_observed_at":"2026-08-16T13:35:20.755259Z","submitted_at":"2024-07-10T17:13:17Z","title":"OpenDiLoCo: An Open-Source Framework for Globally Distributed Low-Communication Training","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.07852","snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"OpenDiLoCo : An open-source framework for globally distributed low-communication training","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":26,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/2407.07852","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:4ba345467169d4dd61979c3be43bd9969d9444b53fecd2bcc4c6d39daee0015b","observation_id":"b1c41c56-1a82-4196-83d0-fc4d7f4bb7d7","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Stich, and Martin Jaggi","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":27,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:2fcd39872bccc335f034391bc2a9f3d6a3caf85a4d844ffd2d7db0b766dc77f9","observation_id":"ee499d7a-620a-4c15-ad9c-10e508a15c82","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Stich, and Ananda Theertha Suresh","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":28,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:b97a52c54ef3f1ed8ef1e8e7e78e01a9553e1c78067486c2c0c104922df52568","observation_id":"70a5bea9-e85a-4f88-832d-e75f7e751863","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Elucidating the design space of diffusion-based generative models","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":29,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:d0c275ade30d361520a5e968366217548f27c9938a8ec10fd6efd9b524f98677","observation_id":"ff687a76-da4f-476a-b6a2-5de0722b3d69","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Stich, and Martin Jaggi","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":30,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:f5b3fae2b81974cd318c50164fd096c989419462cdd435ff4f2562ee7c5a844c","observation_id":"c4ee33aa-6390-4c5e-ba38-a4787f7cef53","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":null,"venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":31,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:b4cfbf8cd3349b280492dc80d464c6820d24090de815059a7b79005e09517406","observation_id":"1bfa2e38-e433-4d9e-be3f-17f231769263","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Madhyastha, and Mosharaf Chowdhury","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":32,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:8a357f9feabd548ed7df629bd06e71e5da8a4a907cffd247bfae9886dbad7b4a","observation_id":"49b62d11-0540-4409-977e-bba0a919589e","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Madhyastha, and Mosharaf Chowdhury","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":33,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:d6553b42e5fcffa771d2bfa356a7086a7ba8aa17ce1e851c1695ac54962716f8","observation_id":"ef9d1502-f48b-42bc-aebc-618839339cc9","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Deeply-supervised nets","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":34,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:d00d848721e578b448078fd336d2e415e42951d7f2f8d0bdefbde8b8776b157d","observation_id":"ec0234b2-5548-4e7d-8b1a-8ec8e31bf731","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Difference target propagation","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":35,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:fb0ee71812c01aaf720cf0d59f1a2157466d055300e107ce4f1b764392ca7072","observation_id":"009dda4f-9d7d-4ecd-89e4-767b95e248e6","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2208.03306","last_updated":"2022-08-05T17:46:38Z","snapshot_observed_at":"2026-08-16T16:40:53.472876Z","submitted_at":"2022-08-05T17:46:38Z","title":"Branch-Train-Merge: Embarrassingly Parallel Training of Expert Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2208.03306","snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"Branch-train-merge: Embarrassingly parallel training of expert language models, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":36,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/2208.03306","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:8865bdd891e23928fc74cf6be350b1ccdc2092b9d11eb8a62b763933f9fbc33e","observation_id":"cd94ee50-d4a0-4886-b0fe-3a2337f9ecdb","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Federated optimization in heterogeneous networks","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":37,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:512def1b2171f2af5e6752fa8e67756533d699ac82c226b035a8cf5d66d18ea3","observation_id":"a082207c-f7ea-4158-a2b5-6e7dfc517bc8","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Can decentralized algorithms outperform centralized algorithms? a case study for decentralized parallel stochastic gradient descent","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":38,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:f7a35906ce19bf5e9b2c56da03172590808903593daffce1920ff89549f70533","observation_id":"a0de6314-163e-47d6-88d5-80140275f877","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":null,"venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":39,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:42b0dc98cf6f95234a1019e8f197d70982cdeda28b122f17132586845e4647c0","observation_id":"1a638064-016a-49ef-88f6-900c53794dff","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Decentralized diffusion models","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":40,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:a99fe438d3afdfef875a879f8fcc3071ab30a926f61f83ab495648de7c96ab2c","observation_id":"50ed3859-8351-4063-b148-1aab847b72ae","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Brendan McMahan, Eider Moore, Daniel Ramage, Seth Hampson, and Blaise Ag \\\"u era y Arcas","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":41,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:1f743226653a3497644c529a882f703528bc200508df1192c18af5b6e5a943a0","observation_id":"77d1b535-a189-4686-bbae-a272bcaf7edf","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2401.01728","last_updated":"2024-05-23T08:01:25Z","snapshot_observed_at":"2026-08-16T14:30:05.927080Z","submitted_at":"2024-01-03T13:07:07Z","title":"Ravnest: Decentralized Asynchronous Training on Heterogeneous Devices","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2401.01728","snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"Ravnest: Decentralized asynchronous training on heterogeneous devices, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":42,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/2401.01728","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:29c544d0206747f36b0618f3512570e104dc3ea0ea5125095797602c63125b09","observation_id":"164d96db-acbd-4d76-b646-8a68ae54719d","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Dual-use foundation models with widely available model weights","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":43,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:70bff7573426322af19003a89a4267cc71c4604d349c92fb434b70ab67cb9bbb","observation_id":"049724a7-0536-4651-b10d-0423cbb276a2","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"DisTrO : Distributed training over-the-internet","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":44,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:7af0b3d331d566a98255435f1b2fec992c9f0e232b7fcf3f024fd2671507386e","observation_id":"171a32e1-8ad7-42eb-83c8-eda0398cd900","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Democratizing AI : The psyche network architecture","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":45,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:b2a0f784bc54e237f2fcb5619833bd9c4211864c75cd63419b256727cfdd3d1d","observation_id":"312f825e-6596-40c1-8aa0-dbd15c20daaf","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Competition in artificial intelligence infrastructure","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":46,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:0d8c65f0c0bd8371e4df49e5e7409e38184743b012d635fe61bd7ff75c568d5b","observation_id":"0fb77d02-e78a-414e-b7dd-0e66b901a181","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Bittensor: A peer-to-peer intelligence market","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":47,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:fc6f652f6a09980ac265dd85cbd92d3ccfc87f585801909f39e5a422351f94dd","observation_id":"6d9eac8a-9c42-4a19-a0c7-75869b7619f3","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":48,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:f46bdb9d6b78675ff9d53429ebdea873b143c1a980450e9c5d65a859beb14c0a","observation_id":"f63be351-8362-46d4-8cb1-56f82f3b0033","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Agora: A decentralized pipeline-parallel training system","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":49,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:b8bb5d6c42484a625efd6645953128e4c07feae3dc0aa539c538b0b1dac51868","observation_id":"26791764-eff7-4df4-a3b7-3be1c006c59d","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Hogwild!: A lock-free approach to parallelizing stochastic gradient descent","venue":null,"work_id":null,"year":2011},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":50,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:67f405c01593dc60e202d3bb005e4f26787468a9af8ebcee85db35f920ded34e","observation_id":"acb6b80d-8e94-4611-bd0e-cdf857ae384d","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Towards crowdsourced training of large neural networks using decentralized mixture-of-experts","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":51,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:894624c25f60a872aea6a16eb0eb135bac3cd1859c5897c3578ff4dc126f443c","observation_id":"33672b7f-98cc-480b-a73c-86ea3bb10ef7","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Hivemind : Decentralized deep learning in PyTorch","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":52,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:54350859929d6949fbf4d911eb0d5605af74eaa8a50299744d56e2c22f65d0fe","observation_id":"45f6571b-421d-49f4-b711-901622350b63","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Moshpit SGD : Communication-efficient decentralized training on heterogeneous unreliable devices","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":53,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:84191db2ce71cd4f37b419b91b1ed2e1c670f62ab4fc28c98954e5d25994f90a","observation_id":"eb51bc9f-5987-4789-9ed5-bf4f2b42d50b","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"SWARM parallelism: Training large models can be surprisingly communication-efficient","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":54,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:70d3119a93c3ea31eceb8e6461b57817453393ded419eb2d5b2b7cff70f3d006","observation_id":"0863fe62-1918-45c4-bcad-2d469e721d13","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.08797","last_updated":"2024-02-13T21:10:21Z","snapshot_observed_at":"2026-08-16T14:18:50.107583Z","submitted_at":"2024-02-13T21:10:21Z","title":"Computing Power and the Governance of Artificial Intelligence","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.08797","snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"Computing power and the governance of artificial intelligence","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":55,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/2402.08797","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:e1062101545f3464d1efc0b38c4892c0f94c5ea094f1d5d30fcce488f18eaf88","observation_id":"a7166fcb-2d00-4e0a-8745-1069082e6b43","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Diffusionblocks: Block-wise neural network training via diffusion interpretation","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":56,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:7a3e7714223bcb2c4608739367404d3808ec5e9900216069b4c9b8280bb7f2c3","observation_id":"a1fe9e06-7aff-4415-839f-dbe2b5cd6c2d","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Kingma, Abhishek Kumar, Stefano Ermon, and Ben Poole","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":57,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:b437bea99ba830b689b9c90aba873b9567f92c885a5a9f8666b2f222890a008c","observation_id":"349bd4d7-8883-44e2-a099-3e8bee6bb28c","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Artificial intelligence index report 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":58,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:9fa68566e4a8be34880f8421545b0b6d74600337c2c59cd53976d452a0f45df8","observation_id":"a446325e-292d-47a4-873a-e88354b39f97","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":null,"venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":59,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:c47855264dab9c0e3d89199320c72f2efa5ca8d4a08769fc2c4913873a3e3f3e","observation_id":"3c5f6de4-47b4-4c59-9395-e2399c6a456b","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2403.07816","last_updated":"2024-03-12T16:54:58Z","snapshot_observed_at":"2026-08-16T14:10:29.723801Z","submitted_at":"2024-03-12T16:54:58Z","title":"Branch-Train-MiX: Mixing Expert LLMs into a Mixture-of-Experts LLM","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.07816","snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"Branch-train-mix: Mixing expert LLMs into a mixture-of-experts LLM , 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":60,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/2403.07816","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:d7f747cab7d12b19d9ddf0035f4882fad4338773ae3cf497008f0d83eb9b71b2","observation_id":"993953e3-9cb7-4f4c-97e8-670a6f783b74","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"AI foundation models: Technical update report","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":61,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:12eca2b3ef73a6960f4a91cf5ae43a597cfbcdd06da6ceb9672bf861b7986911","observation_id":"3c3238f4-3a9f-4371-8648-360c6f97139a","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Gated linear networks","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":62,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:042350f25299307189b13cf9e29ace26b894babf4099ee054effcca988da0bb8","observation_id":"a12497eb-5c83-4a71-86d0-e7f8e3869127","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"A connection between score matching and denoising autoencoders","venue":null,"work_id":null,"year":2011},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":63,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:364442bed3f4588c566efc9c59d2a9b4fede1adf9ab22de9e55e41bd45b35039","observation_id":"bcbb64b5-30c4-4f2c-9953-40abae18b78c","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Concentrating intelligence: Scaling and market structure in artificial intelligence","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":64,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:dfe88691c5dad2cc0da512d6108d593f3c830a67e0f2f4ff47219329d973dbca","observation_id":"ebf1b449-b27d-4636-a04c-2ea9fba8c7b4","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"PowerSGD : Practical low-rank gradient compression for distributed optimization","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":65,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:73d8f235fbb58c250799e531f4141b504141b023251b2d544121afe06d1781dc","observation_id":"6b430420-9deb-45d0-a759-75574861d9bd","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","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":"2026-07-12T12:29:35.403453Z","title":"Shastry, Suhas Manamohan, Subhadeep Mukherjee, Vibhor Garg, Rajesh Sarveswara, Katharina H \\\"a ndler, Peter Pickkers, N","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":66,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:e9ff5b096706fbd7f4819c48f84a545b167ba6d1f52f9d130b5f532293490414","observation_id":"9f1dd7ec-f7dd-43d0-8f1e-c3e097b6f31c","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2206.01288","last_updated":"2023-06-21T13:59:20Z","snapshot_observed_at":"2026-08-16T16:55:39.421614Z","submitted_at":"2022-06-02T20:19:51Z","title":"Decentralized Training of Foundation Models in Heterogeneous Environments","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2206.01288","snapshot_observed_at":"2026-07-12T12:29:35.403453Z","title":"Decentralized training of foundation models in heterogeneous environments, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain","version":2},"reference_index":67,"source":"arxiv_source","source_observed_at":"2026-07-12T12:29:35.403453Z"},"links":{"cited_paper":"/paper/2206.01288","citing_paper":"/paper/2606.24722"},"observation_digest":"sha256:8623a859dee279116825236fd9fb1ab94dccd2f41025416b9381de93ec0bc7d3","observation_id":"ad08ce43-9ca8-4f73-898d-638b67ba81ad","resolution":{"observed_at":"2026-07-12T12:29:35.403453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2606.24722","last_updated":"2026-07-03T16:29:30Z","latest_version":2,"primary_category":"cs.AI","snapshot_observed_at":"2026-08-10T00:24:45.716202Z","submitted_at":"2026-06-23T15:47:33Z","title":"Decentralised AI Training and Inference with BlockTrain"},"reference_resolution":{"displayed":67,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":67,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":67},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"thesis":"As of 17 August 2026, this Paper Citation Record lists 67 of 67 outbound references and 0 inbound Pith citation observations for arXiv:2606.24722."}