{"as_of":"2026-08-17T01:55:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:51d551c370305de6738a0ce12968b460beec6f14bb1f4d96445c952e36a87297","coverage":[{"denominator":221,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":100,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-16T10:44:46.451764Z","state":"measured"},{"denominator":101,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":101,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-16T06:30:59.297886+00:00","state":"measured"},{"denominator":1,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":1,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-06T15:06:48.172897Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"pith","source_observed_at":"2026-08-06T15:06:50.262608Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"cited_work":{"arxiv_id":"2504.17421","doi":null,"metadata_source":"pith","pith_arxiv_id":"2504.17421","snapshot_observed_at":"2026-08-06T15:06:50.262608Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","venue":"cs.LG","work_id":"b9db4856-266c-439e-8e5e-5f82056bbce5","year":2025},"citing_paper":{"arxiv_id":"2507.16731","last_updated":"2025-07-22T16:13:43Z","snapshot_observed_at":"2026-08-16T02:08:34.366113Z","submitted_at":"2025-07-22T16:13:43Z","title":"Collaborative Inference and Learning between Edge SLMs and Cloud LLMs: A Survey of Algorithms, Execution, and Open Challenges","version":1},"reference_index":90,"source":"pdf_text","source_observed_at":"2026-08-06T15:06:48.172897Z"},"links":{"cited_paper":"/paper/2504.17421","citing_paper":"/paper/2507.16731"},"observation_digest":"sha256:aeeb0f0fbdf34042ec514744cd90e8ac7e6122895a24960076eddcb80cfc0056","observation_id":"c35419c0-6f1e-4660-8682-238d0f362c13","resolution":{"observed_at":"2026-08-06T15:06:50.267056Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}}],"links":{"evidence":"/evidence","html":"/paper/2504.17421/citation-record","integrity":"/paper/2504.17421/integrity","json":"/paper/2504.17421/citation-record.json","paper":"/paper/2504.17421"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.077507Z","title":"Will we run out of data? limits of llm scaling based on human-generated data, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.077507Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:05cf2b67991bc2d3f3418fd07f7da9e97c4632dc42b6aae740a544309c8d52da","observation_id":"ae7a00be-dfab-4723-8ef0-9866f8457b81","resolution":{"observed_at":"2026-08-16T10:44:46.077507Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.082175Z","title":"Advances and open challenges in federated foundation models, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.082175Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:36933386a1b287f313b454249c6cbb8ae43864e8f35d1b0647625d1e8a139a79","observation_id":"07727109-f02b-4531-a945-e919ee30e9c4","resolution":{"observed_at":"2026-08-16T10:44:46.082175Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.086132Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.086132Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:589e459910d0414aac6eeee733153c75dbd32928e1cc06e0420f23779ade592a","observation_id":"31d85340-9411-4ff6-b3c7-2ff91a4035f4","resolution":{"observed_at":"2026-08-16T10:44:46.086132Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.089945Z","title":"Introducing chatgpt","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.089945Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:488552d93e9ecef58e306c9e7259861fd002390c612b9027978e99b98cff2f6e","observation_id":"527d3174-11b7-4fef-8030-b82b01d05bd7","resolution":{"observed_at":"2026-08-16T10:44:46.089945Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.093978Z","title":"Gpt-4 technical report","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.093978Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:a9dc5b1167a1e966f0ec9f6740e08e4384796542ff689701dce59a3931c9a029","observation_id":"af8c0acd-5306-4feb-bc2c-7af90109f607","resolution":{"observed_at":"2026-08-16T10:44:46.093978Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2401.08092","last_updated":"2024-09-23T07:37:34Z","snapshot_observed_at":"2026-08-16T14:27:15.896781Z","submitted_at":"2024-01-16T03:35:26Z","title":"A Survey of Resource-efficient LLM and Multimodal Foundation Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2401.08092","snapshot_observed_at":"2026-08-16T10:44:46.098253Z","title":"A survey of resource-efficient llm and multimodal foundation models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.098253Z"},"links":{"cited_paper":"/paper/2401.08092","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:f1d7821dd2941c8663e9b69790f66ba440f8dbe8c226328910194b77d1ce99eb","observation_id":"70885717-2d2e-45eb-bd85-f5a05ab14dbd","resolution":{"observed_at":"2026-08-16T10:44:46.098253Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2303.17564","last_updated":"2023-12-21T06:21:11Z","snapshot_observed_at":"2026-08-14T12:54:48.492396Z","submitted_at":"2023-03-30T17:30:36Z","title":"BloombergGPT: A Large Language Model for Finance","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2303.17564","snapshot_observed_at":"2026-08-16T10:44:46.102577Z","title":"Bloomberggpt: A large language model for finance","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.102577Z"},"links":{"cited_paper":"/paper/2303.17564","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:833d9741e250faf8a8c66c578ef9a46cdbc30aa6b7cd94608adab50fb51650c0","observation_id":"7b7bb11c-3f88-485b-9ddc-b45ad2aefa6c","resolution":{"observed_at":"2026-08-16T10:44:46.102577Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.106712Z","title":"Deepseek-v3 technical report, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.106712Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:dc76a129110b99ec5d86c0ecb3742d1a41fd0eb87b3faf341a17ec174a6a738b","observation_id":"7c74216d-1016-45a3-bfff-45fa5d4bef14","resolution":{"observed_at":"2026-08-16T10:44:46.106712Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.110505Z","title":"Deep residual learning for image recognition","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.110505Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:3de1f9174c29d97d3e869e935271ba1d7085f2ae6ae6df19aff13ca47c31809d","observation_id":"f5d49f81-cf27-47ff-80f0-f807de58cd04","resolution":{"observed_at":"2026-08-16T10:44:46.110505Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.114047Z","title":"Long short-term memory","venue":null,"work_id":null,"year":1997},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.114047Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:a672bb551d7c0dc3ce14c138e778fe96a248348a8aa2a766ec0bd47f556c4b5c","observation_id":"b7ac0d16-bb52-4bd6-941f-50645fefc05f","resolution":{"observed_at":"2026-08-16T10:44:46.114047Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1412.3555","last_updated":"2014-12-11T06:46:53Z","snapshot_observed_at":"2026-08-13T10:35:27.214652Z","submitted_at":"2014-12-11T06:46:53Z","title":"Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1412.3555","snapshot_observed_at":"2026-08-16T10:44:46.117544Z","title":"Empirical evaluation of gated recurrent neural networks on sequence modeling","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.117544Z"},"links":{"cited_paper":"/paper/1412.3555","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:dc1f7276afec08f589f8512629fbad8fe4f696552d3917b33f32a6427db8c18a","observation_id":"05bedc8c-0d2b-4d19-ab7b-834d5d2416a6","resolution":{"observed_at":"2026-08-16T10:44:46.117544Z","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":"zhinengkefu/2024121","doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:47.567164Z","title":"Hudson, Ehsan Adeli, Russ Altman, Simran Arora, Sydney von Arx, Michael S","venue":null,"work_id":"e46e6c12-a517-4c11-ab68-1dd52042ece2","year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.121847Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:9f447fa99882b8ab9d130a9a60d7abf1a35b30d1fec3a51568d772a722a088c4","observation_id":"e9e4c3a2-f206-4108-81ff-2ac42a1922d8","resolution":{"observed_at":"2026-08-16T10:44:47.573116Z","resolver_source":"raw_fallback","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2306.00890","last_updated":"2023-06-01T16:50:07Z","snapshot_observed_at":"2026-08-09T05:23:28.778360Z","submitted_at":"2023-06-01T16:50:07Z","title":"LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2306.00890","snapshot_observed_at":"2026-08-16T10:44:46.125414Z","title":"Llava-med: Training a large language-and-vision assistant for biomedicine in one day","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.125414Z"},"links":{"cited_paper":"/paper/2306.00890","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:034ebb0f7fbda662662e501170926355709c9eb14f2609aa76b89775907badfa","observation_id":"68f055be-55bc-4fb4-b49e-5855ce1e97e0","resolution":{"observed_at":"2026-08-16T10:44:46.125414Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.129251Z","title":"Biogpt: generative pre-trained transformer for biomedical text generation and mining","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.129251Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:27e0c40c5a4eaa7b99dbd38f851221e1d9ae0b9df48cf17668b475208934b0ea","observation_id":"82cac1bc-6ae8-4799-a838-7966cef55311","resolution":{"observed_at":"2026-08-16T10:44:46.129251Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.132808Z","title":"Biobert: a pre-trained biomedical language representation model for biomedical text mining","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.132808Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:198e413f93285d5c6f43c8791aacefe864c875a8a98bee7c60ec25e04753f4f7","observation_id":"b9014512-4479-4c95-ae2c-a21167ef14d6","resolution":{"observed_at":"2026-08-16T10:44:46.132808Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.136432Z","title":"Phi-2: The surprising power of small language models","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.136432Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:933718aa9d0c106301e187dd7a73196bdfce903a96bc5243559c647e0d6687b8","observation_id":"59b71b4e-61dd-424f-90a2-66564b91019d","resolution":{"observed_at":"2026-08-16T10:44:46.136432Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2404.14219","last_updated":"2024-08-30T21:17:17Z","snapshot_observed_at":"2026-08-16T19:15:03.922358Z","submitted_at":"2024-04-22T14:32:33Z","title":"Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2404.14219","snapshot_observed_at":"2026-08-16T10:44:46.140106Z","title":"Phi-3 technical report: A highly capable language model locally on your phone","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.140106Z"},"links":{"cited_paper":"/paper/2404.14219","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:f6030dee2642a2dff5bec567ace22c79733079b531b001a2039927c31d247c69","observation_id":"d3db265d-71df-46bd-a432-afa0e82f4b68","resolution":{"observed_at":"2026-08-16T10:44:46.140106Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.144106Z","title":"Introducing llama 3.1: Our most capable models to date","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.144106Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:a1cb65158d42b1c2edc04246d01eb599a4c83b0008f67262a7c1c08beb6576d3","observation_id":"c6de4bfd-d438-4fe7-a782-d176b87ec8cb","resolution":{"observed_at":"2026-08-16T10:44:46.144106Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.147996Z","title":"Position: Will we run out of data? limits of llm scaling based on human-generated data","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.147996Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:9ca4431d3a4c491b72dc7f70e98ec15313377971383dd53e14225914b7d0d6ad","observation_id":"0c6e4c52-0611-45b6-9119-64a7cc100f88","resolution":{"observed_at":"2026-08-16T10:44:46.147996Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.151585Z","title":"Communication- efficient learning of deep networks from decentralized data","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.151585Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:6e4a76e9e09d74369de4ca7f186230c4149717750ad6ff6f56d7c51aa86ce84b","observation_id":"f34243cf-7323-4ab0-a22e-dadcf06348ad","resolution":{"observed_at":"2026-08-16T10:44:46.151585Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.155080Z","title":"Federated machine learning: Concept and applications","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.155080Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:6d2418bfeda62e8659763ddf02ef5328fbffb98f653d6efbb96df1c65f1e06bb","observation_id":"bd4229b0-1843-4e8b-9108-4712b7a83b91","resolution":{"observed_at":"2026-08-16T10:44:46.155080Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.158929Z","title":"Regulation (eu) 2016/679 of the european parliament and of the council","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.158929Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:eb46259940f75070318baddad9dda75437e517a17d21a4054c3d4bf3e2074f07","observation_id":"fdf85488-a96e-4866-91a1-c649e5da8534","resolution":{"observed_at":"2026-08-16T10:44:46.158929Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.162479Z","title":"California consumer privacy act (ccpa)","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.162479Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:156221e630290b6eca4ea274b28d27a51dcd174d72347cb75880f92ad6096ef9","observation_id":"bf51ba86-2fe7-450d-afb1-c1a1782c224d","resolution":{"observed_at":"2026-08-16T10:44:46.162479Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.166212Z","title":"Health insurance portability and accountability act of 1996","venue":null,"work_id":null,"year":1996},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.166212Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:3fb361bdcdac4c0703f6a5b61fae891f115ea839cc91498e7ecce5ad580a26d8","observation_id":"ab0c4571-c533-486e-8abc-8c19690dd72a","resolution":{"observed_at":"2026-08-16T10:44:46.166212Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.170572Z","title":"Federated learning for healthcare domain-pipeline, applications and challenges","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.170572Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:93c5daedf29e329ce71645569c997dc77d547a7b017f3ffd522a43483f9ceec4","observation_id":"cce5260c-d642-4aff-aa63-b8556e91582e","resolution":{"observed_at":"2026-08-16T10:44:46.170572Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.174476Z","title":"Review on security of federated learning and its application in healthcare","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.174476Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:acc43ab6dc063bc8c671e9dd02063b61d7b64fb48795a395a94215437d147d1c","observation_id":"befa89dc-6e55-4ccc-a6e3-268998941ffd","resolution":{"observed_at":"2026-08-16T10:44:46.174476Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.178325Z","title":"Deep learning-based classification of mesothelioma improves prediction of patient outcome","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.178325Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:bb80595615a6bd6b50985324e06ee1d2446d08e7b162b171871666da5aea53a0","observation_id":"cc06179b-4f86-408d-b09d-13dd5d26de26","resolution":{"observed_at":"2026-08-16T10:44:46.178325Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2303.12944","last_updated":"2023-02-04T17:58:34Z","snapshot_observed_at":"2026-08-16T15:57:42.626228Z","submitted_at":"2023-02-04T17:58:34Z","title":"Use of Federated Learning and Blockchain towards Securing Financial Services","version":1},"cited_work":{"arxiv_id":"2303.12944","doi":null,"metadata_source":"pith","pith_arxiv_id":"2303.12944","snapshot_observed_at":"2026-08-16T10:44:47.475184Z","title":"Use of Federated Learning and Blockchain towards Securing Financial Services","venue":"cs.CR","work_id":"2ac78403-6662-4432-beb7-b73309434944","year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.182455Z"},"links":{"cited_paper":"/paper/2303.12944","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:f593e590acac85758789f54b0e52c4105b83ac6613f338362f3089fbb3e76e49","observation_id":"fa1acefb-4e18-461a-9b28-56d2b7bd8bea","resolution":{"observed_at":"2026-08-16T10:44:47.479355Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+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":"2026-08-16T10:44:46.186624Z","title":"Efficient and secure federated learning for financial applications","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.186624Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:898031e7d1283de590a7a47d399229d24f970e17707f9ace68ead0bbf20c5459","observation_id":"b04be903-2488-4035-9ea3-1d4aca24b738","resolution":{"observed_at":"2026-08-16T10:44:46.186624Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.190177Z","title":"Machine learning ledger orchestration for drug discovery, 2019","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.190177Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:90fbf874cfc9f4a36095e8a0fa51e6934034998dcb3f39025f700ca5554f73cd","observation_id":"256d069d-0885-4a48-b047-1ff9985fdb9d","resolution":{"observed_at":"2026-08-16T10:44:46.190177Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.193921Z","title":"Protecting intellectual property of large language model-based code generation apis via watermarks","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.193921Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:09af8774532e810ba5b4984f963632ca6c93d2bdb57694632a98d6e0fcb550e3","observation_id":"fb081879-77f1-466f-9a68-54e210dc4157","resolution":{"observed_at":"2026-08-16T10:44:46.193921Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.06853","last_updated":"2024-09-23T14:12:40Z","snapshot_observed_at":"2026-08-16T14:19:43.557517Z","submitted_at":"2024-02-10T01:18:15Z","title":"History, Development, and Principles of Large Language Models-An Introductory Survey","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.06853","snapshot_observed_at":"2026-08-16T10:44:46.197542Z","title":"History, development, and principles of large language models-an introductory survey","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.197542Z"},"links":{"cited_paper":"/paper/2402.06853","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:a2f36dc75107cd390feec694610e6739582836d71d6ceaa5a54f3e8dc2db6a9c","observation_id":"9ab6d9e9-0f38-41a3-8b2d-f43e18a7d63a","resolution":{"observed_at":"2026-08-16T10:44:46.197542Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.201585Z","title":"Llmcarbon: Modeling the end-to-end carbon footprint of large language models, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.201585Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:c627be07043a19b89776daede19a31ceb3990d0bf9f2ba0c7a7ef5479ade350f","observation_id":"a84c1ff4-ebd7-435a-afdb-3f69d751aed0","resolution":{"observed_at":"2026-08-16T10:44:46.201585Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2302.09419","last_updated":"2023-05-01T07:48:05Z","snapshot_observed_at":"2026-08-16T15:54:23.434874Z","submitted_at":"2023-02-18T20:51:09Z","title":"A Comprehensive Survey on Pretrained Foundation Models: A History from BERT to ChatGPT","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2302.09419","snapshot_observed_at":"2026-08-16T10:44:46.205388Z","title":"A comprehensive survey on pretrained foundation models: A history from bert to chatgpt","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.205388Z"},"links":{"cited_paper":"/paper/2302.09419","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:572bb1b863df846538262a1ab429178922065176608993de258e83dd5718ccca","observation_id":"212d1029-726d-4da1-bb87-6875f225b0dd","resolution":{"observed_at":"2026-08-16T10:44:46.205388Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1503.02531","last_updated":"2015-03-09T15:44:49Z","snapshot_observed_at":"2026-08-16T18:00:58.008096Z","submitted_at":"2015-03-09T15:44:49Z","title":"Distilling the Knowledge in a Neural Network","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1503.02531","snapshot_observed_at":"2026-08-16T10:44:46.209671Z","title":"Distilling the knowledge in a neural network","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.209671Z"},"links":{"cited_paper":"/paper/1503.02531","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:bc0c19f7594e2828fc506039a021365452baf1adc79440e18c390e4fb09698fe","observation_id":"bfc85f36-16d2-42c7-b222-6e4f8f77264c","resolution":{"observed_at":"2026-08-16T10:44:46.209671Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.213444Z","title":"Knowledge distillation: A survey","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.213444Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:eaa9a5bf3cf8f1a345bc5f336c8ef4c604c54ec688a75cc7f7b662c2ca9ad6f3","observation_id":"08b9cd34-4037-4c19-96e3-482974b1f47f","resolution":{"observed_at":"2026-08-16T10:44:46.213444Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.13116","last_updated":"2024-10-21T16:22:33Z","snapshot_observed_at":"2026-08-10T17:26:33.432994Z","submitted_at":"2024-02-20T16:17:37Z","title":"A Survey on Knowledge Distillation of Large Language Models","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.13116","snapshot_observed_at":"2026-08-16T10:44:46.216890Z","title":"A survey on knowledge distillation of large language models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.216890Z"},"links":{"cited_paper":"/paper/2402.13116","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:52a24e289171a7f12b8d39e80dcb7500d60b5b01a0e59200f5814c63fb882e53","observation_id":"0d87b8bf-6bc0-4cf8-9865-ac7df6ff5267","resolution":{"observed_at":"2026-08-16T10:44:46.216890Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.220701Z","title":"Survey on knowledge distillation for large language models: Methods, evaluation, and application","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.220701Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:d6154c35c6fc64836e56b593cb62ba81f02708bdf81f4fb19081a62a344818bc","observation_id":"501693a4-87f2-4c75-b476-91e7da505dd6","resolution":{"observed_at":"2026-08-16T10:44:46.220701Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2306.08543","last_updated":"2026-01-31T09:16:35Z","snapshot_observed_at":"2026-08-16T00:37:04.298248Z","submitted_at":"2023-06-14T14:44:03Z","title":"MiniLLM: On-Policy Distillation of Large Language Models","version":6},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2306.08543","snapshot_observed_at":"2026-08-16T10:44:46.224100Z","title":"Minillm: Knowledge distillation of large language models","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.224100Z"},"links":{"cited_paper":"/paper/2306.08543","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:0541b827db5302a30a1028d36ff133fffb581cebc94ffaf3bc2c95fdba8c9cf7","observation_id":"e83767b1-ff40-4b35-9743-d8602ef382d1","resolution":{"observed_at":"2026-08-16T10:44:46.224100Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.227990Z","title":"On-policy distillation of language models: Learning from self-generated mistakes","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.227990Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:41fe4f919096c7a2dba400a83d874e8a83f1f1664850eb60aebefbce4a607ff5","observation_id":"bfbf1d8d-1bf6-4009-a24e-70471d6885d7","resolution":{"observed_at":"2026-08-16T10:44:46.227990Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.231960Z","title":"For distillation, tokens are not all you need","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.231960Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:c1b4ccefb75884bece708cde27a7a9db78547eee492b2373239e435ccdcd0183","observation_id":"25091c87-e6bf-4592-a6be-23d7031f6541","resolution":{"observed_at":"2026-08-16T10:44:46.231960Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.235369Z","title":"Baby llama: knowledge distillation from an ensemble of teachers trained on a small dataset with no performance penalty","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.235369Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:31467f47bcde577798b15abb40bac13f675aab5b21e8b9c65c2af6b2bb36c573","observation_id":"3b407a54-23fb-4bbc-bca7-28af039a29f2","resolution":{"observed_at":"2026-08-16T10:44:46.235369Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2407.19409","last_updated":"2024-07-28T06:10:47Z","snapshot_observed_at":"2026-08-16T13:30:30.406611Z","submitted_at":"2024-07-28T06:10:47Z","title":"LLAVADI: What Matters For Multimodal Large Language Models Distillation","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.19409","snapshot_observed_at":"2026-08-16T10:44:46.239106Z","title":"Llavadi: What matters for multimodal large language models distillation","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.239106Z"},"links":{"cited_paper":"/paper/2407.19409","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:e4372e7972f4f010ee8af621f697eff2227f14376bc57b017a60ef1d8e2d0f93","observation_id":"d0fc7f33-c572-4ede-8942-ae1d0ca3df4c","resolution":{"observed_at":"2026-08-16T10:44:46.239106Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.242858Z","title":"Less is more: Task-aware layer-wise distillation for language model compression","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.242858Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:6bc40879e230d3eb6cdc3df8a12f65c4a8c821c5eab5caa760ebe4d062fa4809","observation_id":"011c436e-9a8b-40ab-a24e-d24962197328","resolution":{"observed_at":"2026-08-16T10:44:46.242858Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2407.16154","last_updated":"2024-07-23T03:47:28Z","snapshot_observed_at":"2026-08-16T13:31:55.562567Z","submitted_at":"2024-07-23T03:47:28Z","title":"DDK: Distilling Domain Knowledge for Efficient Large Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.16154","snapshot_observed_at":"2026-08-16T10:44:46.246457Z","title":"Ddk: Distilling domain knowledge for efficient large language models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.246457Z"},"links":{"cited_paper":"/paper/2407.16154","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:cb374eb2095a92f212ea9bad2200bcb407bc5f9e8d2a631cddc42926df9ff4d7","observation_id":"5a0cc5d4-9173-4538-b088-2c889a83fbdf","resolution":{"observed_at":"2026-08-16T10:44:46.246457Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.250219Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.250219Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:7a25e89a4239fcbe8c9a470dc9129c40dee8745d53662a8d6e223731fcb3b1a9","observation_id":"baab51d5-428b-4f44-841f-ca0e02aeecef","resolution":{"observed_at":"2026-08-16T10:44:46.250219Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.254026Z","title":"Crash: Clustering, removing, and sharing enhance fine-tuning without full large language model","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.254026Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:b93183b85d28986416baa275029b2ce86859ce22642a625cd4d4b4a402c53939","observation_id":"acb1a683-5d2f-4b38-99be-39ac61898fb5","resolution":{"observed_at":"2026-08-16T10:44:46.254026Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.257878Z","title":"Orchestration of emulator assisted mobile edge tuning for ai foundation models: A multi-agent deep reinforcement learning approach, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.257878Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:54afaf0b80eca5905890644ba54ab24f01cec98c44287c010b52a05533a34662","observation_id":"9f821dc6-2f87-4765-98dd-cb3ce3c74dc9","resolution":{"observed_at":"2026-08-16T10:44:46.257878Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2302.04870","last_updated":"2023-02-09T18:59:55Z","snapshot_observed_at":"2026-08-16T15:56:26.510814Z","submitted_at":"2023-02-09T18:59:55Z","title":"Offsite-Tuning: Transfer Learning without Full Model","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2302.04870","snapshot_observed_at":"2026-08-16T10:44:46.261726Z","title":"Offsite-tuning: Transfer learning without full model","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.261726Z"},"links":{"cited_paper":"/paper/2302.04870","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:bfc948ac4090d7daacdb746584e3d8f3009d1e8d2a4f3a10faa2f7cc415b7fdd","observation_id":"f40ceb06-56c4-4ea1-9e0d-097122db45b4","resolution":{"observed_at":"2026-08-16T10:44:46.261726Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.265382Z","title":"Fedpft: Federated proxy fine-tuning of foundation models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.265382Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:b56fc26e8e7b6154c2036d4d42a93ab59b4a6627554cf255175e1443dd665be7","observation_id":"a2876486-338c-44e2-b9a5-48247a362d89","resolution":{"observed_at":"2026-08-16T10:44:46.265382Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1910.03581","last_updated":"2019-10-08T18:00:00Z","snapshot_observed_at":"2026-08-14T20:20:16.604226Z","submitted_at":"2019-10-08T18:00:00Z","title":"FedMD: Heterogenous Federated Learning via Model Distillation","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1910.03581","snapshot_observed_at":"2026-08-16T10:44:46.268594Z","title":"Fedmd: Heterogenous federated learning via model distillation","venue":null,"work_id":null,"year":1910},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.268594Z"},"links":{"cited_paper":"/paper/1910.03581","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:518c5fd57916f990f7d555abf19f70afe46d77be91aba41c1ec110fd2806ab25","observation_id":"67e5378f-63a6-4a09-aa01-2facb45ec082","resolution":{"observed_at":"2026-08-16T10:44:46.268594Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.272871Z","title":"Ensemble attention distillation for privacy-preserving federated learning","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.272871Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:f8b1d39d6766f1f09708a4120f17ef985c10655ba4238666135c557f210eb84f","observation_id":"3a62eff5-5d0d-4af3-800e-f24bbb5a4dec","resolution":{"observed_at":"2026-08-16T10:44:46.272871Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.276375Z","title":"Preserving privacy in federated learning with ensemble cross-domain knowledge distillation","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.276375Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:338dfb4e6fd2ff8d36757e361b93cc409c7d5f5bba87ac8a43f949c2863c58c3","observation_id":"7e4bcab2-6b0c-474b-b714-87e7a6a5e628","resolution":{"observed_at":"2026-08-16T10:44:46.276375Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.280179Z","title":"Data shunt: Collaboration of small and large models for lower costs and better performance","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.280179Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:7920cd117f6449cf36fc9a4b680633fbd038e87b473964141b45d769acc7d560","observation_id":"4ab9d33c-c3f6-4736-b589-19e938d32925","resolution":{"observed_at":"2026-08-16T10:44:46.280179Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.284082Z","title":"Parameterized knowledge transfer for personalized federated learning","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.284082Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:5f284c992c14bbad47ffa9bb375e1a2b5832ea87b91d33a03f07043930b80a83","observation_id":"bf0a6ef3-1e1a-4627-a2dd-4cdea7fb5f8f","resolution":{"observed_at":"2026-08-16T10:44:46.284082Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.287781Z","title":"Ensemble distillation for robust model fusion in federated learning","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.287781Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:9ce96b0cedf3228c051a3a1a8c489aaadea2e264b9c6ab8a55f82a26e85b6599","observation_id":"a5e262d6-e8b5-44e4-b40e-7a9c7609ae28","resolution":{"observed_at":"2026-08-16T10:44:46.287781Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.291496Z","title":"Depth anything: Unleashing the power of large-scale unlabeled data","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.291496Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:54144cd0a78952991701d980ea982bb6a505087df8d55871410a7e8c8477be95","observation_id":"bed5d0a6-22d4-4739-9edf-410889bcb96e","resolution":{"observed_at":"2026-08-16T10:44:46.291496Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.295172Z","title":"Robust federated learning with noisy and heterogeneous clients","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":58,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.295172Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:6e2540d800eb55bb6dd92064f7fd7a3f94821149bdcb93a34dca2392ce470f41","observation_id":"cd7adc2e-ccf4-4dce-85e7-b4578323cd13","resolution":{"observed_at":"2026-08-16T10:44:46.295172Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.298784Z","title":"Fedtgp: Trainable global prototypes with adaptive-margin- enhanced contrastive learning for data and model heterogeneity in federated learning","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.298784Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:a8c357ae428c83de9c6211961d4db84b8093687978dea7e0aec3e5287a15bd31","observation_id":"0eeb8e14-547b-41a4-82d5-c3564d6b240f","resolution":{"observed_at":"2026-08-16T10:44:46.298784Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.302487Z","title":"An upload-efficient scheme for transferring knowledge from a server-side pre-trained generator to clients in heterogeneous federated learning","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":60,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.302487Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:58a6f488ed2249a8688db364fce43f223bdc2b362942c82d1ed6abec3fbbd925","observation_id":"9316be4a-f07b-407b-9d44-2e54592468f0","resolution":{"observed_at":"2026-08-16T10:44:46.302487Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.306152Z","title":"Learning from human educational wisdom: A student-centered knowledge distillation method.IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":61,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.306152Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:735a53ffa01c93cc4dcc7a001c7d56a529e662b76f92bf0ba7498019ad52e269","observation_id":"23c3a549-51d9-4166-9a34-a95158e04955","resolution":{"observed_at":"2026-08-16T10:44:46.306152Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.309919Z","title":"Bert learns to teach: Knowledge distillation with meta learning","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":62,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.309919Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:479dbeaf8e246f6fa893d08632911e9d31ef98f550001c018b75257335ee0bb5","observation_id":"e5b36ebc-94c8-410c-b770-2ccc830933dc","resolution":{"observed_at":"2026-08-16T10:44:46.309919Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.313567Z","title":"Fedgems: Federated learning of larger server models via selective knowledge fusion, 2021","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":63,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.313567Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:d8cb8503864d0e0f687a1d45c2c796194e1ef9a671a94acea7430308a030f07d","observation_id":"2aa8ca11-d740-474b-ae70-58dd9f7cbe4b","resolution":{"observed_at":"2026-08-16T10:44:46.313567Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2302.08888","last_updated":"2023-05-06T02:26:32Z","snapshot_observed_at":"2026-08-16T15:54:36.912671Z","submitted_at":"2023-02-17T14:17:44Z","title":"Multimodal Federated Learning via Contrastive Representation Ensemble","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2302.08888","snapshot_observed_at":"2026-08-16T10:44:46.317246Z","title":"Multimodal federated learning via contrastive representation ensemble","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":64,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.317246Z"},"links":{"cited_paper":"/paper/2302.08888","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:e025a90d2491744d65bf24e51ae16663054971370c8a67d8161fb698a345495b","observation_id":"53593872-91e9-4c34-8cb2-cb02754c28fb","resolution":{"observed_at":"2026-08-16T10:44:46.317246Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.321296Z","title":"Ideal: Query-efficient data-free learning from black-box models","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":65,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.321296Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:a3bebba4be3a0b608a3b2f8d07a763c9da87ebdfa15156b33b3b7714542a009d","observation_id":"9355f99e-8834-46d6-a432-ea3a2f808460","resolution":{"observed_at":"2026-08-16T10:44:46.321296Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.324812Z","title":"Towards data-free model stealing in a hard label setting","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":66,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.324812Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:c114acbdee5bd8e5fe69f5ce7cd87c00fe09d22d60081fcc699f9138d4cefe6e","observation_id":"45537e89-01b9-4769-b292-b9c29fd0bfef","resolution":{"observed_at":"2026-08-16T10:44:46.324812Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.328453Z","title":"Zero-shot knowledge distillation from a decision-based black-box model","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":67,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.328453Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:b4473fb906b40895943073268ca04065eb2decfc5fe97f0de4e1b15589ac22d9","observation_id":"99a43378-b052-4007-9267-08b137970370","resolution":{"observed_at":"2026-08-16T10:44:46.328453Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.332161Z","title":"Bridging the gap between decision and logits in decision- based knowledge distillation for pre-trained language models","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":68,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.332161Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:0361dc42ded173bcd5f058b27de884ceaeaa61971122281839e669a149ee92e0","observation_id":"f36365d1-0b86-405a-bc48-364988448f52","resolution":{"observed_at":"2026-08-16T10:44:46.332161Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.335887Z","title":"Synthetic data distillation enables the extraction of clinical information at scale","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":69,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.335887Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:6cc586733608ea0ba0213377211d473e6fdc2ebe973433884a883b17f54b565b","observation_id":"2dd21afb-196a-4f87-a249-7e8edfdda47f","resolution":{"observed_at":"2026-08-16T10:44:46.335887Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.339545Z","title":"ZeroGen: Efficient zero-shot learning via dataset generation","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":70,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.339545Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:9baaad31efcc7977ba25149543ebb30781aa34f294b402a69be88ca5b6fc83da","observation_id":"8c6c23a4-8ac0-45fe-86ef-393c9747ed5d","resolution":{"observed_at":"2026-08-16T10:44:46.339545Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.343330Z","title":"ProGen: Progressive zero- shot dataset generation via in-context feedback","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":71,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.343330Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:b1372867c3f31521b8ef4f47fe20ae554b6d46e9f210fc7697c255faeb9448b9","observation_id":"c8c8c076-7abd-4f2c-ae65-36a0adcd13c9","resolution":{"observed_at":"2026-08-16T10:44:46.343330Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.347192Z","title":"Generating training data with language models: Towards zero-shot language understanding","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":72,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.347192Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:f127b171d048ed882d925c6ad742d260a9b97d26711a641799be447e26640a73","observation_id":"8a469541-c0e0-47e7-bac9-655870860a6f","resolution":{"observed_at":"2026-08-16T10:44:46.347192Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.350901Z","title":"Self-guided noise-free data generation for efficient zero-shot learning","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":73,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.350901Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:1d053b0dd33affd4f5705e0633d50466af8283440b07af30d71fd94b211bb770","observation_id":"f48b7cfe-8cea-4c1e-a25b-64908c6b7c5e","resolution":{"observed_at":"2026-08-16T10:44:46.350901Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2406.12527","last_updated":"2024-06-18T11:55:05Z","snapshot_observed_at":"2026-08-16T13:41:52.810628Z","submitted_at":"2024-06-18T11:55:05Z","title":"FuseGen: PLM Fusion for Data-generation based Zero-shot Learning","version":1},"cited_work":{"arxiv_id":"2406.12527","doi":null,"metadata_source":"pith","pith_arxiv_id":"2406.12527","snapshot_observed_at":"2026-08-16T10:44:47.347688Z","title":"FuseGen: PLM Fusion for Data-generation based Zero-shot Learning","venue":"cs.CL","work_id":"f63ead86-7e3d-48e9-a43e-3f54930fcac9","year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":74,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.354663Z"},"links":{"cited_paper":"/paper/2406.12527","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:35caba645e5f616a5cffc17cb6eeffd349123e7a97eceddf1c90925506e5c522","observation_id":"75743248-dc6f-44d3-9b33-5134231957cb","resolution":{"observed_at":"2026-08-16T10:44:47.351960Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+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":"2026-08-16T10:44:46.358568Z","title":"Retrieval-based knowledge transfer: An effective approach for extreme large language model compression","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":75,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.358568Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:e81f78b027c2c4186336452ee6053998bec6a97c084f15e8d395ac107c92c00f","observation_id":"0ca54650-e4fd-4abf-9808-bf8551cd04ab","resolution":{"observed_at":"2026-08-16T10:44:46.358568Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.11451","last_updated":"2024-05-08T12:11:00Z","snapshot_observed_at":"2026-08-16T14:51:17.504913Z","submitted_at":"2023-10-17T17:58:34Z","title":"Seeking Neural Nuggets: Knowledge Transfer in Large Language Models from a Parametric Perspective","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.11451","snapshot_observed_at":"2026-08-16T10:44:46.362152Z","title":"Seeking neural nuggets: Knowledge transfer in large language models from a parametric perspective","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":76,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.362152Z"},"links":{"cited_paper":"/paper/2310.11451","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:68aefc4e29ba8628c7b0dcde6aaf9c40a68c7a4f1623c57ceff7b16e05eec5e2","observation_id":"1ee041d7-ac44-4558-b998-134b7c065fed","resolution":{"observed_at":"2026-08-16T10:44:46.362152Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.366082Z","title":"Mutual enhancement of large and small language models with cross-silo knowledge transfer, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":77,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.366082Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:c0cb33248d8b350491030240e5cc22706ae205233880b2e0fbb75ec57886b036","observation_id":"a9ec2050-7224-4056-830e-33fde7f91982","resolution":{"observed_at":"2026-08-16T10:44:46.366082Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.369657Z","title":"Zico Kolter","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":78,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.369657Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:386107d68217380fe2b16f581da751ffd5754fe933ca6cf5182bcb75d6d9c3d1","observation_id":"0c3bf45a-3457-4962-bfe4-25e013b7be68","resolution":{"observed_at":"2026-08-16T10:44:46.369657Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.373198Z","title":"Zeroquant: Efficient and affordable post-training quantization for large-scale transformers","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":79,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.373198Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:887c676d1a729b127b7df4230a123f980e5ddb7dedcf1b69cbc3bb33f2d8bd71","observation_id":"eb166ea0-28ce-416c-a97d-ff972c092f70","resolution":{"observed_at":"2026-08-16T10:44:46.373198Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.376718Z","title":"Initializing models with larger ones, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":80,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.376718Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:87ee3705ea985f8cceb86a9e338c5b7886146ee18d1063901465c797b7cc26d3","observation_id":"86366c57-49d8-4123-9ff7-202ab917a613","resolution":{"observed_at":"2026-08-16T10:44:46.376718Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2109.04641","last_updated":"2021-09-10T03:01:01Z","snapshot_observed_at":"2026-08-16T17:57:38.375696Z","submitted_at":"2021-09-10T03:01:01Z","title":"Learning to Teach with Student Feedback","version":1},"cited_work":{"arxiv_id":"2109.04641","doi":null,"metadata_source":"pith","pith_arxiv_id":"2109.04641","snapshot_observed_at":"2026-08-16T10:44:47.319338Z","title":"Learning to Teach with Student Feedback","venue":"cs.LG","work_id":"ebcec828-8a31-41ee-9cdb-034998e0c7ea","year":2021},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":81,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.380503Z"},"links":{"cited_paper":"/paper/2109.04641","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:a75bec1e343eec14b19304d2b9b651d8b9da58437f880dacd280d5f54b0274b0","observation_id":"89a2ed07-072d-4ec6-aaef-8d04619476b8","resolution":{"observed_at":"2026-08-16T10:44:47.323674Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-16T06:30:59.297886+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+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":"2026-08-16T10:44:46.384437Z","title":"Meta pseudo labels","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":82,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.384437Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:f2626e5844a3281b0d406642b97065de714a805fbc9b25d6d5191bc823817175","observation_id":"653bd436-6f53-43c6-bc0f-4519ce29138c","resolution":{"observed_at":"2026-08-16T10:44:46.384437Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.388105Z","title":"Dual knowledge distillation for bidirectional neural machine translation","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":83,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.388105Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:81e3b1e53b220dac8907cf9b83c77ba8b2143d32f5d82102c9c6dab3b0822d7a","observation_id":"fcafe17f-c28b-4962-99c2-7311316d8390","resolution":{"observed_at":"2026-08-16T10:44:46.388105Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.391962Z","title":"Shadow knowledge distillation: Bridging offline and online knowledge transfer","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":84,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.391962Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:2a95c7b40795c74d8c22f8e955456fee86f93c53f8d2fccd35aade4de22a1139","observation_id":"1e94c6be-7799-44a1-80c9-a93f4ec14fd6","resolution":{"observed_at":"2026-08-16T10:44:46.391962Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.395826Z","title":"Reverse knowledge distillation: Training a large model using a small one for retinal image matching on limited data","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":85,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.395826Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:453ed4e4abe221d943f2d89f8ed2b0714ab5ec058f85d9e50f1f019197216f51","observation_id":"5631268b-b98b-4d40-aa12-e532f4a0fb99","resolution":{"observed_at":"2026-08-16T10:44:46.395826Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.399291Z","title":"Weak-to-strong generalization: Eliciting strong capabilities with weak supervision","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":86,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.399291Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:c334bd47c8acc8a07ff1d405feaf15dde196267c32c9e637ea29c1bf5969963e","observation_id":"13d67b45-d8fe-4b8e-8b78-77de7cca103d","resolution":{"observed_at":"2026-08-16T10:44:46.399291Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2406.02224","last_updated":"2024-12-16T16:13:14Z","snapshot_observed_at":"2026-08-16T13:46:19.809104Z","submitted_at":"2024-06-04T11:36:09Z","title":"FedMKT: Federated Mutual Knowledge Transfer for Large and Small Language Models","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2406.02224","snapshot_observed_at":"2026-08-16T10:44:46.402923Z","title":"Fedmkt: Federated mutual knowledge transfer for large and small language models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":87,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.402923Z"},"links":{"cited_paper":"/paper/2406.02224","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:eea4fa744435af5a1822b503c33a2eafa71f86ceb162eb606116cdc1c4903d72","observation_id":"4ffe2dac-ddb9-426f-bf2a-02594e7649a1","resolution":{"observed_at":"2026-08-16T10:44:46.402923Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.407011Z","title":"Federated learning with gan-based data synthesis for non-iid clients","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":88,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.407011Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:26858bf195743eecb06c77dfd0056e30710bdfb94f7b7cd9c5f90f17db1bdcfc","observation_id":"4095410a-db30-48bc-b56a-10be0e0a3220","resolution":{"observed_at":"2026-08-16T10:44:46.407011Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.411120Z","title":"Gfl: Federated learning on non-iid data via privacy-preserving synthetic data","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":89,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.411120Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:5220d2fc8f46887a6302fa789a6f7d10799871b72288f3e3214ab69f033ff881","observation_id":"365e18fa-c049-4bba-a284-34eaf167b378","resolution":{"observed_at":"2026-08-16T10:44:46.411120Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.415019Z","title":"Harnessing large- language models to generate private synthetic text, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":90,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.415019Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:d3ca8a98cac98694867e339fc6594fa53080cb82cb1ba9032d0e79426a4d139d","observation_id":"054a8ce7-5a98-4b47-a802-060b2d8b3028","resolution":{"observed_at":"2026-08-16T10:44:46.415019Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.418584Z","title":"Privacy-preserving instructions for aligning large language models, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":91,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.418584Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:f876b88838fa0d9e7e8582c3bc1b9b05e4578b68f51de3d955c74bd11bac9e17","observation_id":"0673faaf-d50e-4656-87f7-f2adf60cf415","resolution":{"observed_at":"2026-08-16T10:44:46.418584Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.422209Z","title":"Stable federated learning with dataset condensation","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":92,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.422209Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:437b249884c7d7033890048d5df0c17013295eb9d44a737c18aefe2a0c29c50a","observation_id":"a069455d-a82d-497d-9dd4-58c076cb719d","resolution":{"observed_at":"2026-08-16T10:44:46.422209Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.426199Z","title":"Federated learning via decentralized dataset distillation in resource-constrained edge environments","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":93,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.426199Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:d0697c292cb784a8fcdb7db916adb81c6443a7a7096f25b023faaba4ae874acb","observation_id":"3f8db6b6-0076-4ceb-bec3-f5ff6b18a893","resolution":{"observed_at":"2026-08-16T10:44:46.426199Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2009.07999","last_updated":"2021-06-06T06:55:46Z","snapshot_observed_at":"2026-08-13T09:34:55.152682Z","submitted_at":"2020-09-17T01:14:47Z","title":"Distilled One-Shot Federated Learning","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2009.07999","snapshot_observed_at":"2026-08-16T10:44:46.429615Z","title":"Distilled one-shot federated learning","venue":null,"work_id":null,"year":2009},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":94,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.429615Z"},"links":{"cited_paper":"/paper/2009.07999","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:e916b06b64f81e93dee3da75cb8c8511d4c7f75375ffd7c87776d49ff19ebbf0","observation_id":"1521c53d-a7c0-48aa-99b6-5cfb6c1f123e","resolution":{"observed_at":"2026-08-16T10:44:46.429615Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1811.10959","last_updated":"2020-02-24T23:25:50Z","snapshot_observed_at":"2026-08-15T02:56:10.149790Z","submitted_at":"2018-11-27T13:17:45Z","title":"Dataset Distillation","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1811.10959","snapshot_observed_at":"2026-08-16T10:44:46.433313Z","title":"Dataset distillation","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":95,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.433313Z"},"links":{"cited_paper":"/paper/1811.10959","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:9db4bfc97f0b950e6786494055a98d120fa8c9611a4375f4243b80f45e7a9571","observation_id":"f4570f03-8979-4dfc-80e8-0ef801e708c6","resolution":{"observed_at":"2026-08-16T10:44:46.433313Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2006.05929","last_updated":"2021-03-08T13:31:22Z","snapshot_observed_at":"2026-08-09T10:26:02.821053Z","submitted_at":"2020-06-10T16:30:52Z","title":"Dataset Condensation with Gradient Matching","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2006.05929","snapshot_observed_at":"2026-08-16T10:44:46.437416Z","title":"Dataset condensation with gradient matching","venue":null,"work_id":null,"year":2006},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":96,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.437416Z"},"links":{"cited_paper":"/paper/2006.05929","citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:2bb01fa6d3ab12ca0ddb5ffe6598dd3019bade59ad2cc431ecf0ad8391abc1c8","observation_id":"bca449ac-7084-42ab-bcd7-e365633c9d33","resolution":{"observed_at":"2026-08-16T10:44:46.437416Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.441320Z","title":"Dauphin, and David Lopez-Paz","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":97,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.441320Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:15ca1471ad52435347e9812943a3de5968c3649f10ab8f6b55ce04816875c5ab","observation_id":"7d7faec7-e866-4541-bdf2-38ba4ad635f8","resolution":{"observed_at":"2026-08-16T10:44:46.441320Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.444872Z","title":"Mix2fld: Downlink federated learning after uplink federated distillation with two-way mixup","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":98,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.444872Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:eeac73eecb04d74f555c3a8afcbf75d98e56f49380c6aa6b49a32eae38ac2c9b","observation_id":"2ba250a1-8151-4485-91c2-e4738ae7d0e6","resolution":{"observed_at":"2026-08-16T10:44:46.444872Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.448467Z","title":"Differentially private synthetic data via foundation model APIs 1: Images","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":99,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.448467Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:14b466f7b5244ba7c324880377ea03da5e969a88f1cae4f626b771240c68a5dc","observation_id":"94e8fefb-568d-4595-9d9f-abbeae50b9f7","resolution":{"observed_at":"2026-08-16T10:44:46.448467Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T10:44:46.451764Z","title":"Differentially Private Synthetic Data via Foundation Model APIs 2: Text","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks","version":2},"reference_index":100,"source":"pdf_text","source_observed_at":"2026-08-16T10:44:46.451764Z"},"links":{"citing_paper":"/paper/2504.17421"},"observation_digest":"sha256:62b6bb844b7c53c24b873ecf8cca81be5044b4cf0865d7ee87da5a8b3e38780d","observation_id":"50ddea9a-c945-47ab-9d93-cbaf7ff53d12","resolution":{"observed_at":"2026-08-16T10:44:46.451764Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2504.17421","last_updated":"2026-06-28T12:07:24Z","latest_version":2,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-16T10:38:30.037983Z","submitted_at":"2025-04-24T10:24:35Z","title":"Towards Harnessing the Collaborative Power of Large and Small Models for Domain Tasks"},"reference_resolution":{"displayed":100,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":96,"verified_exact":4,"verified_fuzzy":0},"total_outbound_references":221},"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":"2026-08-16T06:30:59.297886+00:00","source":"crossref"},{"observed_at":"2026-08-16T06:30:54.164669+00:00","source":"retraction_watch"}],"thesis":"As of 17 August 2026, this Paper Citation Record lists 100 of 221 outbound references and 1 inbound Pith citation observation for arXiv:2504.17421."}