{"as_of":"2026-08-08T19:35:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:88f72405708a35a720d4e13978d7eed260a35b878f51fd4a9d111dbc4a9f3d1f","coverage":[{"denominator":48,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":48,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-06T19:03:14.907976Z","state":"measured"},{"denominator":49,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":49,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-08T06:32:00.761636+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":"2026-08-02T08:23:08.693861Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2507.07149","snapshot_observed_at":"2026-08-02T08:23:08.693861Z","title":null,"venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.21624","last_updated":"2026-07-07T10:18:25Z","snapshot_observed_at":"2026-08-08T06:52:26.281349Z","submitted_at":"2026-07-07T10:18:25Z","title":"FBLayout: Optimizing Memory Layout for Efficient LLM Finetuning on Mobile GPUs","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-02T08:23:08.693861Z"},"links":{"cited_paper":"/paper/2507.07149","citing_paper":"/paper/2607.21624"},"observation_digest":"sha256:fc9def6d629769e59cecf7c62e894c4b9df19b58ce4ccb4eac940759ccfb86a8","observation_id":"635a5586-8c90-40b4-839a-4cbbfea2d48d","resolution":{"observed_at":"2026-08-02T08:23:08.693861Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2507.07149/citation-record","integrity":"/paper/2507.07149/integrity","json":"/paper/2507.07149/citation-record.json","paper":"/paper/2507.07149"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.687260Z","title":"Vi-map: Infrastructure-assisted real-time hd mapping for autonomous driving","venue":null,"work_id":"f3862117-d548-4491-bf95-09f5b5fa1138","year":2023},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.601662Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:6cf1544eb99161e440c3793024131d73b3771c08d653c1e37a2b6e636e75a065","observation_id":"dc282ab2-dd7d-40df-8495-1adc6e8716c6","resolution":{"observed_at":"2026-08-06T19:03:15.691304Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.673805Z","title":"Robust real-time multi-vehicle collaboration on asynchronous sensors","venue":null,"work_id":"06b5e9bb-db18-424b-8322-d6c2a73c5bbd","year":2023},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.606344Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:d13d5c05c25d0a04ef7608166da07add6351b7ec67f446789db8b7c11157d4c5","observation_id":"5a75b39a-e878-4b81-90e6-9ef582849195","resolution":{"observed_at":"2026-08-06T19:03:15.678043Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.660892Z","title":"Autofed: Heterogeneity-aware federated multimodal learning for robust autonomous driving","venue":null,"work_id":"736f0534-3e84-4bf7-a176-0e8054e2c7ec","year":2023},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.610594Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:69a14eaf09f74e6d977ab0044930e8c76300ca5b7bafe5b4c84596c6f3d2fbb9","observation_id":"46af045c-c7dd-495d-bfc0-56d0474c7e37","resolution":{"observed_at":"2026-08-06T19:03:15.664818Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.648037Z","title":"Scaleflow: Efficient deep vision pipeline with closed-loop scale-adaptive inference","venue":null,"work_id":"dc3a4bbf-2318-4cda-8d9c-44a51eac942c","year":2023},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.615011Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:4af9813d813a6c98355362d690d20c31671c99628b7ebbb2f92c8b6ec7c1e91e","observation_id":"5774dbc7-9320-4926-90f9-7f03be6e3cfe","resolution":{"observed_at":"2026-08-06T19:03:15.652028Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.634562Z","title":"Speech understanding on tiny devices with a learning cache","venue":null,"work_id":"b3d00691-9810-4a5d-98ae-f945ec12b541","year":2024},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.732785Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:a06a2e28ef17aa9048eafa4157f35ab136516bfe15fd62e102874681ce2dcd67","observation_id":"55738a4f-a7a9-4662-96b4-9f57321eaffa","resolution":{"observed_at":"2026-08-06T19:03:15.638614Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.621411Z","title":"Autodroid: Llm-powered task automation in android","venue":null,"work_id":"c828b9f0-a366-469d-8062-391569202fb2","year":2024},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.737415Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:6bcef691ba11b796597e0da61f3cbcb13eef72d1cc3f79ae865395f46498f055","observation_id":"6ffc4ba6-755d-4918-9716-80a43220abb6","resolution":{"observed_at":"2026-08-06T19:03:15.625649Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+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":"2026-08-06T19:03:14.742239Z","title":"Mobilegpt: Augmenting llm with human-like app memory for mobile task automation","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.742239Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:4609eff5d66614686fff28db8a278d3d8030e7f66b9b7cf647a8a077917895d3","observation_id":"fb08c56e-9657-4ddf-bd8e-a82c008359b2","resolution":{"observed_at":"2026-08-06T19:03:14.742239Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.606545Z","title":"Experience: Practical challenges for indoor ar applications","venue":null,"work_id":"836c4cfd-49b9-47c7-b2c4-5180e425a2dd","year":2024},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.746560Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:5337ab1d86edc7382d820bdc3398b21c731788812821cb3eec40097ade23fc2c","observation_id":"8cd800b5-2f51-4579-a977-01148fe30bb1","resolution":{"observed_at":"2026-08-06T19:03:15.612022Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.593220Z","title":"Arise: High-capacity ar offloading inference serving via proactive scheduling","venue":null,"work_id":"a9e197df-e8d1-42da-a8e0-55b14ad9e5a2","year":2024},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.750716Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:e93ab6099b58ba840cc7db1590e04e6927714636141257729e997d8b5005955f","observation_id":"6b24dde8-9999-46a9-aefe-59bb241cb358","resolution":{"observed_at":"2026-08-06T19:03:15.597553Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.579446Z","title":"Band: coordinated multi-dnn inference on heterogeneous mobile processors","venue":null,"work_id":"0e924b4f-5a48-4bd5-bb5d-38b061d069bf","year":2022},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.754906Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:f44a590c0fb35e26d2024ab276e24a30ca98a9998c866be4c3af163e191a386b","observation_id":"bef29838-329a-4b0d-b08f-af52b4d057df","resolution":{"observed_at":"2026-08-06T19:03:15.583858Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.565390Z","title":"Theia: Gaze-driven and perception-aware volumetric content delivery for mixed reality headsets","venue":null,"work_id":"4d353d02-ae16-4e78-ad8c-c958ddcd7fd4","year":2024},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.759134Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:df0b4aaacc2802e65bed06b23680a6795499eed15d6277e8d75b1bde48e2e04b","observation_id":"4af86d22-ecf6-41ea-9167-9ab768414d50","resolution":{"observed_at":"2026-08-06T19:03:15.569697Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.551756Z","title":"Zero-effort cross-domain gesture recognition with wi-fi","venue":null,"work_id":"78044131-67e3-47b7-9ce1-087a66e20ec2","year":2019},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.763289Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:00ee7f34d238ab6cf0e2415c6e525c0f9214af569400bd943b3c0e3bc7415e82","observation_id":"2c407dc4-757b-4480-9c10-7e87daef78e4","resolution":{"observed_at":"2026-08-06T19:03:15.555996Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.538447Z","title":"Rf genesis: Zero-shot generalization of mmwave sensing through simulation-based data synthesis and generative diffusion models","venue":null,"work_id":"bdf66e2c-44e4-4a5a-acba-b68e8fa80966","year":2023},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.767090Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:8cde0c66abba41ee3a7191e2fa6b569e6eb9c253332e2daf8cc8cd0586f803ee","observation_id":"d0dd489b-d726-4ba1-9a81-4b1170acf814","resolution":{"observed_at":"2026-08-06T19:03:15.542663Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.524281Z","title":"Cosmo: contrastive fusion learning with small data for multimodal human activity recognition","venue":null,"work_id":"d3b98d1a-d8d0-482d-99d2-c7d6955e905e","year":2022},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.770945Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:5552d57cc1e6faa43da7b065731cc196e179c252f3e73eb9c6144783f213add7","observation_id":"d258a1f8-e8bd-4c92-b888-db7ec5580aed","resolution":{"observed_at":"2026-08-06T19:03:15.528395Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.510325Z","title":"In USENIX Workshop on Hot Topics in Edge Computing (HotEdge 18), Boston, MA, July 2018","venue":null,"work_id":"8e2ed8d5-8d66-4757-8b93-8f00183f1981","year":2018},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.774691Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:0eb08404ce7c2cb451131e05554c2825e763aaf7b09ea6196470365dafad6258","observation_id":"842da943-a6f9-4667-bdae-a30bf150ac2b","resolution":{"observed_at":"2026-08-06T19:03:15.514880Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.496502Z","title":"On-device training under 256kb memory","venue":null,"work_id":"dcf5a131-fbd2-4f67-a118-30a2e77109a2","year":2022},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.778903Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:47915fced38935c3262d36b6ca8ec7c1c4f8504e8afb37a5bbecae7556db5c30","observation_id":"2c10a7a7-b311-4dde-8c77-d82f6538b838","resolution":{"observed_at":"2026-08-06T19:03:15.500922Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2208.02507","last_updated":"2022-08-04T07:37:07Z","snapshot_observed_at":"2026-07-06T13:38:40.554369Z","submitted_at":"2022-08-04T07:37:07Z","title":"ZeroFL: Efficient On-Device Training for Federated Learning with Local Sparsity","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2208.02507","snapshot_observed_at":"2026-08-06T19:03:14.782996Z","title":"Zerofl: Efficient on-device training for federated learning with local sparsity","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.782996Z"},"links":{"cited_paper":"/paper/2208.02507","citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:698b3c8586fa5d1456eae9921f7c24d43fc3c7949db6a37ab5d83bf2d90c9fe6","observation_id":"973a34d0-c9d2-4504-a9b9-d5bbb2ea1721","resolution":{"observed_at":"2026-08-06T19:03:14.782996Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.482713Z","title":"Spotlight: Optimizing device place- ment for training deep neural networks","venue":null,"work_id":"b528986f-08bc-4b2b-acae-b57c0170175c","year":2018},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.787308Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:96193855c3ad6d448bf30aab0c78bd53d4ff1c72593124fedfc4eaa7d4eee241","observation_id":"619ae321-6dc0-4db0-b80e-08fa799a97c0","resolution":{"observed_at":"2026-08-06T19:03:15.487175Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.468980Z","title":"Actnn: Reducing training memory footprint via 2-bit activation compressed training","venue":null,"work_id":"23fecbc2-c8ff-4a53-884e-34f9f2d4d99c","year":2021},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.791363Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:125f6c8a5a731b39226dbc70574376ba3555818c29ad8692ea410553fccb76d8","observation_id":"372f9455-1c91-4d4d-81c3-fc5fe6109ae5","resolution":{"observed_at":"2026-08-06T19:03:15.472976Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.455374Z","title":"Division: memory efficient training via dual activation precision","venue":null,"work_id":"a44cbedd-adb6-4cdc-b135-597dd9af0863","year":2023},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.795488Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:9443f67f47600e67bf4413487545867cb84727b994ff2f82c1435548954b34df","observation_id":"cfed7143-6c4c-4913-ae9f-7db49ecc20fd","resolution":{"observed_at":"2026-08-06T19:03:15.459680Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.441672Z","title":"Flexpoint: An adaptive numerical format for efficient training of deep neural networks","venue":null,"work_id":"d08dffa8-2455-44bb-b1f8-88ba54f66ab7","year":2017},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.799564Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:8be6f36112babdaa68b4ef90940229dced1fcbfb5be4d900fdacd9bb9ac63daf","observation_id":"4b1ea5e9-27ab-475f-95c2-d3cbd2550452","resolution":{"observed_at":"2026-08-06T19:03:15.446141Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.427576Z","title":"Fractrain: Fractionally squeezing bit savings both temporally and spatially for efficient dnn training","venue":null,"work_id":"6205097a-2046-4bb9-b5fd-03fb340988b5","year":2020},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.803599Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:4dea155788c4c3d9db2b9a14e7aee49bc1567a99dbf94072e5b6b289e7ccbd11","observation_id":"48803f5f-a1cb-4fe3-a21e-60079092b4bf","resolution":{"observed_at":"2026-08-06T19:03:15.431859Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.413714Z","title":"Gact: Activation com- pressed training for generic network architectures","venue":null,"work_id":"f043280f-c398-4f39-810b-7020f9ad7638","year":2022},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.807393Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:9b4e06e37f098628e5fb51de7ccf3416b06dffa2115753d73a15c44a1f7f3030","observation_id":"2f0cdc25-9d95-4396-bd19-62198ed1f953","resolution":{"observed_at":"2026-08-06T19:03:15.418086Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1805.06085","last_updated":"2018-07-17T07:33:19Z","snapshot_observed_at":"2026-07-06T06:39:21.688392Z","submitted_at":"2018-05-16T01:19:43Z","title":"PACT: Parameterized Clipping Activation for Quantized Neural Networks","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1805.06085","snapshot_observed_at":"2026-08-06T19:03:14.811289Z","title":"Pact: Parameterized clipping activation for quantized neural networks.arXiv preprint arXiv:1805.06085, 2018","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.811289Z"},"links":{"cited_paper":"/paper/1805.06085","citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:329ee2cb94ce4f45ab3cc8395dc02402dc87fddbb65017572e400f5191ac4ced","observation_id":"2665bcea-9ea4-4691-a78a-2580113258a5","resolution":{"observed_at":"2026-08-06T19:03:14.811289Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.400522Z","title":"Fixed-point back-propagation training","venue":null,"work_id":"bbbb6a9f-ffb4-42c4-bdd4-95bc7b60c336","year":2020},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.815558Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:0d8d0ca655b91ac1cb2cf9c50b943f2e114bff0ba370416987701b7f41b2ebcb","observation_id":"7dc87a99-b4ee-471f-b191-628768754234","resolution":{"observed_at":"2026-08-06T19:03:15.404639Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.385793Z","title":"Dynaspa: Exploiting spatial sparsity for efficient dynamic dnn inference on devices","venue":null,"work_id":"9afa5088-ac61-434f-99c2-9274f339b365","year":2024},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.819443Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:53040795e0d6951db42af3b84c6100ee09f6fdd083eb9faac339356ceeba5bbb","observation_id":"fa0a1d20-0916-40c4-b559-82b6465b21e6","resolution":{"observed_at":"2026-08-06T19:03:15.391107Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.371138Z","title":"Memory-efficient dnn training on mobile devices","venue":null,"work_id":"50c3c023-7131-40a3-b0fe-eea936e561de","year":2022},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.823612Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:d6d480662b949eff2abb9c831948efa0e839101c022cfa6b24c15e71b615cd58","observation_id":"81ef9d7d-abbb-421a-9029-b732ecbe5bae","resolution":{"observed_at":"2026-08-06T19:03:15.375434Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.357352Z","title":"Poet: Training neural networks on tiny devices with integrated rematerialization and paging","venue":null,"work_id":"29121cf2-4463-4f60-88be-46a90072e8cd","year":2022},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.827654Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:8dfca932422be3b9fd6bd3023d08490c7ce80f5ca1bf5ecb62c174f302436596","observation_id":"71dec313-57b5-4502-8027-ecc1d23f8084","resolution":{"observed_at":"2026-08-06T19:03:15.361757Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.342751Z","title":"Elastictrainer: Speeding up on-device training with runtime elastic tensor selection","venue":null,"work_id":"c0c73e1c-0d58-462b-8c59-c73df6873161","year":2023},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.831539Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:89d3e2a28e838438db6dfceef4c5f7a607e1731fb3da6709d3bf2d6641491601","observation_id":"60d27467-1a38-4464-9cd0-27c027ac56e6","resolution":{"observed_at":"2026-08-06T19:03:15.347320Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.328231Z","title":"Nvidia cub, 2024","venue":null,"work_id":"e70aebe9-7472-4c5d-af9d-201baa62817e","year":2024},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.835529Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:e20f8bc3024db34833b6aac628efb94bdba738c0fe2e36f272e6f92ea522a864","observation_id":"dacaa99f-92f0-4402-b86f-3867dee9b0dd","resolution":{"observed_at":"2026-08-06T19:03:15.332315Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.313360Z","title":"Identity mappings in deep residual networks","venue":null,"work_id":"7c9ed91d-0f16-4b02-92f5-6f55ba82ba31","year":2016},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.839302Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:6803a4c5ae17cc95612e18e3d3f7e7419cfc3d7c2deba9441c0ba4f8e4ba79e4","observation_id":"74ef4342-b013-48a6-b5fa-f5709f5267ac","resolution":{"observed_at":"2026-08-06T19:03:15.317916Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1907.11692","last_updated":"2019-07-26T17:48:29Z","snapshot_observed_at":"2026-07-31T22:31:37.910868Z","submitted_at":"2019-07-26T17:48:29Z","title":"RoBERTa: A Robustly Optimized BERT Pretraining Approach","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1907.11692","snapshot_observed_at":"2026-08-06T19:03:14.842892Z","title":"Roberta: A robustly optimized bert pretraining approach","venue":null,"work_id":null,"year":1907},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.842892Z"},"links":{"cited_paper":"/paper/1907.11692","citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:99beae1e4e0fbc89855135b3b3b587d8e0500983c60bfb023e1e73a4b2301d70","observation_id":"01bbd465-1762-4d0c-bcd4-027fecc72f35","resolution":{"observed_at":"2026-08-06T19:03:14.842892Z","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":"2026-08-06T19:03:14.846944Z","title":"Language models are unsupervised multitask learners","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.846944Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:4115e9b4658e593fd873716ad985958f2b1b2f82165cfecdb516ae700979ea66","observation_id":"8f789fe3-49cd-4b40-82dd-5ba8e878b30f","resolution":{"observed_at":"2026-08-06T19:03:14.846944Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2106.09685","last_updated":"2021-10-16T18:40:34Z","snapshot_observed_at":"2026-08-07T07:43:16.294957Z","submitted_at":"2021-06-17T17:37:18Z","title":"LoRA: Low-Rank Adaptation of Large Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2106.09685","snapshot_observed_at":"2026-08-06T19:03:14.850879Z","title":"Lora: Low-rank adaptation of large language models","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.850879Z"},"links":{"cited_paper":"/paper/2106.09685","citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:b595e069d0d56df472a98dd6d5473d7ddab292042a298adf69a7c7327453afe0","observation_id":"94039612-c6d1-4265-9c8c-25e6ceaccb27","resolution":{"observed_at":"2026-08-06T19:03:14.850879Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.288143Z","title":"Throughput of native arithmetic instructions, 2024","venue":null,"work_id":"10f9b646-953c-4dab-aea8-4db8931016b8","year":2024},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.854858Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:9bacf743edea645e4a20534b3b50b52ba3b57eb13733c41d23e5775da224d302","observation_id":"49ead13a-ea02-4504-9441-0b810b4a4e66","resolution":{"observed_at":"2026-08-06T19:03:15.292782Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.273351Z","title":"Instruction throughput and latency, 2024","venue":null,"work_id":"b3b650a1-60c4-41c1-8244-c14a744a811e","year":2024},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.858961Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:e7e571645f6a0b95e9e010ca30bc306f831a0deac02cea980b78898d15034824","observation_id":"564ac629-9d5f-4d28-aae3-afcd264e88ed","resolution":{"observed_at":"2026-08-06T19:03:15.278574Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1603.04467","last_updated":"2016-03-16T16:57:12Z","snapshot_observed_at":"2026-07-06T04:49:28.300758Z","submitted_at":"2016-03-14T20:50:20Z","title":"TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1603.04467","snapshot_observed_at":"2026-08-06T19:03:14.862745Z","title":"Tensorflow: Large-scale machine learning on heterogeneous distributed systems","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.862745Z"},"links":{"cited_paper":"/paper/1603.04467","citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:a85337744b11052517c9261afa304dff8826c43e41e6d129116ff929be3d66aa","observation_id":"6fa0b337-55c8-49b3-9388-abee6a3c395f","resolution":{"observed_at":"2026-08-06T19:03:14.862745Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.259187Z","title":"The multiobjective multidimensional knapsack problem: a survey and a new approach","venue":null,"work_id":"78901522-421b-4cbe-b9da-0565c388f096","year":2012},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.866891Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:9f79c8cff7e8058a34bff9db06ec540f43a632faaa3acd7ff597669f52a68d1b","observation_id":"c6bc9baa-07c4-47f2-afd6-9dd85fedec37","resolution":{"observed_at":"2026-08-06T19:03:15.263921Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.245486Z","title":"Canadian institute for advanced research, 10 classes,","venue":null,"work_id":"455f6fd4-7195-4cef-b24d-0ce17551763e","year":null},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.870734Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:000bc6b2989c21492ecb4fe9a0a18e3b1174c2666947d383ea00c5009a916905","observation_id":"073de215-44ea-4c29-bd78-bc5202573de7","resolution":{"observed_at":"2026-08-06T19:03:15.249646Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.231087Z","title":"Canadian institute for advanced research, 100 classes,","venue":null,"work_id":"b27ef0f2-cb0d-49bb-b7fa-dc889dc5ceb3","year":null},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.874907Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:916f2ec9c277b2ddda2b67f85cb65058facf985a3ab84effdd3d6cf8d8dfde13","observation_id":"f17c98c2-ed1d-412e-ab46-6183e005cd44","resolution":{"observed_at":"2026-08-06T19:03:15.235565Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.215671Z","title":"Imagenet: A large-scale hierarchical image database,","venue":null,"work_id":"f9fd97b5-8aa2-448a-9442-97eae96a113a","year":null},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.878880Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:4ba3c46b617e872ddb6ba5b0204d36965d2f8551a8ed7f4625196931571b182d","observation_id":"baf7a4c6-ea90-432f-b42a-24a8d3da3ef5","resolution":{"observed_at":"2026-08-06T19:03:15.220482Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.201666Z","title":"Semantic textual similarity,","venue":null,"work_id":"570f8f70-55a3-42e1-b009-d182313b9910","year":null},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.883248Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:3472699f3113e9db0c9562d7dbf98b306a4d5ff8ceaaa894944f4dca97b54b89","observation_id":"4bbb7734-eb8b-400c-88d3-960b4b12cc3c","resolution":{"observed_at":"2026-08-06T19:03:15.205964Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.185665Z","title":"Microsoft research paraphrase corpus,","venue":null,"work_id":"87932184-eb93-4a9e-861b-775b538d33d6","year":null},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.887111Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:0d2cbda3ca484c2a095654e08b3373bbb5cb3835ebe8a17fb5d24ed8a17ed9c6","observation_id":"a80956cb-42b2-4d00-ba6a-00e4fd8bfd89","resolution":{"observed_at":"2026-08-06T19:03:15.190554Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.170076Z","title":"The stanford sentiment treebank,","venue":null,"work_id":"9f7e2ad1-0330-4f67-ba5b-c9d634154e6a","year":null},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.891341Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:5fe69b9c5aa7942c3e2d03696ceba19cdc4f66023c57e5f6734ee4da08abd2d3","observation_id":"c828255c-d009-4960-9252-27b69df83ac8","resolution":{"observed_at":"2026-08-06T19:03:15.174569Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.155559Z","title":"End-to-end nlg challenge,","venue":null,"work_id":"6f593c1c-5f5a-4b06-8fac-0590107c6fd7","year":null},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.895723Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:a41ffc31d3cef9636886331945efc8c136b6588253830bd88caecb8ed4ddb854","observation_id":"6bc568b2-c008-40d4-8fda-2dc0481d35fd","resolution":{"observed_at":"2026-08-06T19:03:15.159970Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.141074Z","title":"Creating training corpora for nlg micro-planners,","venue":null,"work_id":"c05f1ad6-f127-4a97-a7f9-ea4d5efc317d","year":null},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.899574Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:7a33efa70ad0a7c0776f2033235484fdaebf727c8d11ccb23148cad1e8fe80a9","observation_id":"76dfb396-d2f9-404c-a31d-04fbacd6b014","resolution":{"observed_at":"2026-08-06T19:03:15.145611Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.126777Z","title":"Towards unified int8 training for convolutional neural network","venue":null,"work_id":"f2eb4286-4557-44b9-8351-7282d1c986ca","year":1969},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.903536Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:35c549b5ff4d82bf609965ad397dc33729b8143ed1712e71a8f01e3c4be04228","observation_id":"24cb2ed6-63ba-4b0b-8e44-cbf1b65cb704","resolution":{"observed_at":"2026-08-06T19:03:15.131224Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T19:03:15.109575Z","title":"Re- source management with deep reinforcement learning","venue":null,"work_id":"f8bcd637-d376-48da-b598-167d7c7c6af4","year":2016},"citing_paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-06T19:03:14.907976Z"},"links":{"citing_paper":"/paper/2507.07149"},"observation_digest":"sha256:4e2137bb968c1f1018a7502edc10c4d0919bbd02f50964f013b99a871ebc251e","observation_id":"5c06b368-ee71-42ed-acf8-a06a99903e87","resolution":{"observed_at":"2026-08-06T19:03:15.116114Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-08T06:32:00.761636+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2507.07149","last_updated":"2025-07-09T08:59:30Z","latest_version":1,"primary_category":"cs.NI","snapshot_observed_at":"2026-08-06T18:55:03.675899Z","submitted_at":"2025-07-09T08:59:30Z","title":"DAF: An Efficient End-to-End Dynamic Activation Framework for on-Device DNN Training"},"reference_resolution":{"displayed":48,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":7,"verified_exact":0,"verified_fuzzy":41},"total_outbound_references":48},"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":"2026-08-08T06:32:00.761636+00:00","source":"crossref"},{"observed_at":"2026-08-08T06:31:55.24221+00:00","source":"retraction_watch"}],"thesis":"As of 8 August 2026, this Paper Citation Record lists 48 of 48 outbound references and 1 inbound Pith citation observation for arXiv:2507.07149."}