{"as_of":"2026-08-07T12:06:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:bd5fb4f7f5daa839c8f9a385fb3c9d352d26e0c31d14b290dcedeb209f9a6cd2","coverage":[{"denominator":40,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":40,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-01T06:08:27.749037Z","state":"measured"},{"denominator":40,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":40,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-07T06:34:17.273281+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2607.22010/citation-record","integrity":"/paper/2607.22010/integrity","json":"/paper/2607.22010/citation-record.json","paper":"/paper/2607.22010"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-01T06:08:23.967824Z","title":"Generalizable speech deepfake detection via meta-learned lora,","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:23.967824Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:787fba5969198bc751a8f227fe6986b91a04bd40ff00bbb96392427a256e4b17","observation_id":"8346fc9b-b14b-4b4d-93f0-06c831f2b3e3","resolution":{"observed_at":"2026-08-01T06:08:23.967824Z","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":"2026-08-01T06:08:24.060046Z","title":"Does Audio Deepfake Detection Generalize?","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:24.060046Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:7f2a0af2a95a31db508eb76eda8da27353da6609cb4dcbe20ec2106ecf29980f","observation_id":"f88dc750-8531-4f21-aea5-34bf052fad41","resolution":{"observed_at":"2026-08-01T06:08:24.060046Z","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":"2026-08-01T06:08:24.133941Z","title":"Domain generalization: A survey,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:24.133941Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:1c81bc093cddd9a695729ba6dfbb8f3aa304abfd3dfa07e28b94cf79adb992d5","observation_id":"2f9c2144-4169-4d6b-a46f-1e4a6dfee18b","resolution":{"observed_at":"2026-08-01T06:08:24.133941Z","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":"2026-08-01T06:08:24.293198Z","title":"Learning to learn: Introduction and overview,","venue":null,"work_id":null,"year":1998},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:24.293198Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:145ddeb892562b3e6ffc84d4dec93e954be11ed6ef79d57ec12f86a37ad7b596","observation_id":"3d7b300c-3cc4-4b6b-bd01-30151fa781cc","resolution":{"observed_at":"2026-08-01T06:08:24.293198Z","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":"2026-08-01T06:08:24.379902Z","title":"Model-agnostic meta-learning for fast adaptation of deep networks,","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:24.379902Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:7441baa2ca24dfb9b044cd9c7c400ecb89fc29b15a77141ac9d5929b9d3abb1d","observation_id":"ce61daa1-c317-4b7b-a63e-1b4ad58f3dc2","resolution":{"observed_at":"2026-08-01T06:08:24.379902Z","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":"2026-08-01T06:08:24.529221Z","title":"Meta-learning approaches for improving detection of unseen speech deepfakes,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:24.529221Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:d97cbd8166f52783b24c7ee125fe5e94e6562c7a3940b89bfc1d01b1854f5008","observation_id":"b4c300ac-de7e-4823-b79e-3f253582e8bd","resolution":{"observed_at":"2026-08-01T06:08:24.529221Z","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":"2026-08-01T06:08:24.650899Z","title":"Learning to generalize: Meta-learning for domain generalization,","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:24.650899Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:c7e6eb4d462a739c2ec6179adbd3bd607824d059fd94dd8130232b59a2439aec","observation_id":"e285f081-c2df-4feb-83f0-58f86b1ee394","resolution":{"observed_at":"2026-08-01T06:08:24.650899Z","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":"2026-08-01T06:08:24.738477Z","title":"LoRA: Low-rank adaptation of large language models,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:24.738477Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:2dbb998e942466f72d9cd4bf9d192f95a329e5ebf42d49860991d9b517da5d18","observation_id":"b3ef6647-6c5c-4b5f-844c-4b41848441eb","resolution":{"observed_at":"2026-08-01T06:08:24.738477Z","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":"2026-08-01T06:08:24.898456Z","title":"Principles of risk minimization for learning theory,","venue":null,"work_id":null,"year":1991},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:24.898456Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:c68f8950ee0e31554438e1712aafd05e8c90283cedfbf0e0a715d87d6d1a3dec","observation_id":"16e0fa0b-511f-4208-a4b7-4c664a4a7cba","resolution":{"observed_at":"2026-08-01T06:08:24.898456Z","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":"2026-08-01T06:08:25.069472Z","title":"New insights and perspectives on the natural gradient method,","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:25.069472Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:abdd4f6e73606a2b51c2acbf58830262f9fbeb7c130f4387661808c91acbc0b7","observation_id":"1536d5b0-b041-4764-afb5-d8b34f2d4a53","resolution":{"observed_at":"2026-08-01T06:08:25.069472Z","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":"2026-08-01T06:08:25.182149Z","title":"The effective rank: A measure of effective dimensionality,","venue":null,"work_id":null,"year":2007},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:25.182149Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:476b5d1c4371e5e7aec5383e9452f97f59916f5285de640c7e82284363c01e8d","observation_id":"9a5b4c76-0847-46d5-a56d-670c94f55d09","resolution":{"observed_at":"2026-08-01T06:08:25.182149Z","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":"2026-08-01T06:08:25.272051Z","title":"Rankme: Assessing the downstream performance of pretrained self-supervised representations by their rank,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:25.272051Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:4369208e99dc2b29bf13690c284db7378b08f606acd10f3c29f3a1693e64f0da","observation_id":"05ea43e6-3fc9-4bd6-80f6-05b7d5100719","resolution":{"observed_at":"2026-08-01T06:08:25.272051Z","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":"2026-08-01T06:08:25.346042Z","title":"Intrinsic dimensionality explains the effectiveness of language model fine-tuning,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:25.346042Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:a04d4abe08c8923e5c34d167d90718a37b7f39924ab24c11885931970c45b91a","observation_id":"fb77a02d-28a2-46c5-a953-210d744a9701","resolution":{"observed_at":"2026-08-01T06:08:25.346042Z","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":"2026-08-01T06:08:25.428912Z","title":"Adaptive budget allocation for parameter-efficient fine-tuning,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:25.428912Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:1b69d1f22d505d14cfaddb22eeaae6ad8d68c1828008d59bc1b0377f6bc64f38","observation_id":"e5e3cef7-152c-4229-b687-4c3d1b903347","resolution":{"observed_at":"2026-08-01T06:08:25.428912Z","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":"2026-08-01T06:08:25.513294Z","title":"Pissa: Principal singular values and singular vectors adaptation of large language models,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:25.513294Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:cd9c898705613dfd602ac3fa431245968d2820b514e63aabc970a211ab280bf4","observation_id":"2ed24da5-bff7-4a00-b8bf-d2697566462c","resolution":{"observed_at":"2026-08-01T06:08:25.513294Z","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":"2026-08-01T06:08:25.606813Z","title":"Lora vs full fine-tuning: An illusion of equivalence,","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:25.606813Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:a5b8bd7ef6d3100925177b229c597070085739240b749ff6c265ceb98c7e3384","observation_id":"b8025ffe-e2cb-4eb0-9ee2-bca45df5845c","resolution":{"observed_at":"2026-08-01T06:08:25.606813Z","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":"2026-08-01T06:08:25.668247Z","title":"Asymmetry in low-rank adapters of foundation models,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:25.668247Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:cb876d9f82851fc7affb2845db1bda37eef0aa07232931d328bd5d3632b625d4","observation_id":"b10510d0-2bcb-4e8b-a537-e3eef8cc1ff4","resolution":{"observed_at":"2026-08-01T06:08:25.668247Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2405.09673","last_updated":"2024-09-20T21:21:56Z","snapshot_observed_at":"2026-08-04T07:20:49.662052Z","submitted_at":"2024-05-15T19:27:45Z","title":"LoRA Learns Less and Forgets Less","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2405.09673","snapshot_observed_at":"2026-08-01T06:08:25.705441Z","title":"Lora learns less and forgets less,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:25.705441Z"},"links":{"cited_paper":"/paper/2405.09673","citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:03a8955593c6d2e9f13413f40ca5067fb2d525dac12013cf37db4351780973cb","observation_id":"8e4cae20-cd6b-4b34-92ad-aa2b500648b6","resolution":{"observed_at":"2026-08-01T06:08:25.705441Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.02347","last_updated":"2024-06-05T06:36:45Z","snapshot_observed_at":"2026-07-06T17:24:57.402088Z","submitted_at":"2024-02-04T05:05:43Z","title":"Riemannian Preconditioned LoRA for Fine-Tuning Foundation Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.02347","snapshot_observed_at":"2026-08-01T06:08:25.824474Z","title":"Riemannian preconditioned lora for fine- tuning foundation models,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:25.824474Z"},"links":{"cited_paper":"/paper/2402.02347","citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:364d5a59d210e7fa4fe56837c7b7b5825f03f4b249d4b03d27158307a3859ea7","observation_id":"c32bdcd8-71e2-4f3c-8759-8a077f3e5d3f","resolution":{"observed_at":"2026-08-01T06:08:25.824474Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2605.01046","last_updated":"2026-05-26T18:00:50Z","snapshot_observed_at":"2026-07-06T23:14:20.123289Z","submitted_at":"2026-05-01T19:20:25Z","title":"Learning in the Fisher Subspace: A Guided Initialization for LoRA Fine-Tuning","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2605.01046","snapshot_observed_at":"2026-08-01T06:08:25.933242Z","title":"Learning in the fisher sub- space: A guided initialization for lora fine-tuning,","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:25.933242Z"},"links":{"cited_paper":"/paper/2605.01046","citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:e0cf71f7a4baefbe0ffecd916335ae6c149600d5a50df084422fb01870f76fc8","observation_id":"6462b643-34b3-4ec0-83ed-d8dea7331b61","resolution":{"observed_at":"2026-08-01T06:08:25.933242Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2603.29824","last_updated":"2026-06-07T17:51:52Z","snapshot_observed_at":"2026-08-06T06:00:26.051111Z","submitted_at":"2026-03-31T14:46:39Z","title":"Curvature-Guided LoRA: Matching Full Fine-Tuning in Function Space","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2603.29824","snapshot_observed_at":"2026-08-01T06:08:26.040122Z","title":"Curvature-guided lora: Steering in the pretrained ntk subspace,","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:26.040122Z"},"links":{"cited_paper":"/paper/2603.29824","citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:99cf4a3b9b49b668cebf09f5f461096527eac3c64bf21ca1e379009a8e955cfa","observation_id":"d0abc3be-3f8d-44ea-b550-8b23de4fde28","resolution":{"observed_at":"2026-08-01T06:08:26.040122Z","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":"2026-08-01T06:08:26.151391Z","title":"Limitations of the empirical fisher approximation for natural gradient descent,","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:26.151391Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:84155b05897a8b2eae7340029c04edbdd78e2a13eb12b8f33618b2c8b9244325","observation_id":"ded19160-1a95-431c-a6bb-f541df0b1284","resolution":{"observed_at":"2026-08-01T06:08:26.151391Z","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":"2026-08-01T06:08:26.274328Z","title":"wav2vec 2.0: A framework for self-supervised learning of speech representations,","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:26.274328Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:7311e0eef10d136d6d443794fbd07621c74ee1c795f108f27e5e02845ddb998e","observation_id":"257afeff-9ef3-4e83-9998-882f971ed813","resolution":{"observed_at":"2026-08-01T06:08:26.274328Z","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":"2026-08-01T06:08:26.354782Z","title":"Aasist: Audio anti-spoofing using integrated spectro-temporal graph attention networks,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:26.354782Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:52198b00de72e012e2bc3f79e0181892e7300b2bd98504762864986fae8d7147","observation_id":"93025460-b42f-4eae-b99f-af03492b1318","resolution":{"observed_at":"2026-08-01T06:08:26.354782Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1412.1193","last_updated":"2020-09-19T15:16:47Z","snapshot_observed_at":"2026-08-06T19:02:23.137226Z","submitted_at":"2014-12-03T05:21:13Z","title":"New insights and perspectives on the natural gradient method","version":11},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1412.1193","snapshot_observed_at":"2026-08-01T06:08:26.493783Z","title":"New insights and perspectives on the natural gradient method,","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:26.493783Z"},"links":{"cited_paper":"/paper/1412.1193","citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:7d332f655ea372e65cd397a03a571aa8af4afbdc8b2e5f1cb7a83759e7011f68","observation_id":"444e57f1-0efa-41be-999d-bddec79102b2","resolution":{"observed_at":"2026-08-01T06:08:26.493783Z","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":"2026-08-01T06:08:26.598802Z","title":"Optimizing neural networks with kronecker-factored approximate curvature,","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:26.598802Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:17b942a982947275e1c205fcae3b9b230511bc865631de5b0d9eafea35b2ee07","observation_id":"a9c9eb30-7aca-4f4a-b17a-1b1f2f43efa1","resolution":{"observed_at":"2026-08-01T06:08:26.598802Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2406.06420","last_updated":"2024-11-05T21:26:58Z","snapshot_observed_at":"2026-07-06T18:28:16.301122Z","submitted_at":"2024-06-10T16:12:32Z","title":"An Improved Empirical Fisher Approximation for Natural Gradient Descent","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2406.06420","snapshot_observed_at":"2026-08-01T06:08:26.702744Z","title":"An improved empirical fisher approximation for natural gradient descent,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:26.702744Z"},"links":{"cited_paper":"/paper/2406.06420","citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:5952539abd6315409127a8f9f4f6b5940a073b935f11d6bf0a4bdbfe535e7866","observation_id":"f6f42f2b-c267-49ee-a1ae-0f09c55cce96","resolution":{"observed_at":"2026-08-01T06:08:26.702744Z","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":"2026-08-01T06:08:26.783478Z","title":"An investigation into neural net optimization via hessian eigenvalue density,","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:26.783478Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:5b873e93cf641c237f751675b128952e3a47b97f493d2b4df34c78234b8dbd8e","observation_id":"eafc7ca4-6d4b-4f54-bf51-97cb4ff09402","resolution":{"observed_at":"2026-08-01T06:08:26.783478Z","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":"2026-08-01T06:08:26.870135Z","title":"Neural tangent kernel: Convergence and generalization in neural networks,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:26.870135Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:640efd06ad32e13ff28c1284c4c385767095918acf12adf250a55ef2243ff682","observation_id":"9dae94ce-8d3c-4e07-a173-a9a11a67bd9d","resolution":{"observed_at":"2026-08-01T06:08:26.870135Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2210.02885","last_updated":"2023-06-26T12:17:11Z","snapshot_observed_at":"2026-08-04T03:38:12.657021Z","submitted_at":"2022-10-05T12:13:04Z","title":"RankMe: Assessing the downstream performance of pretrained self-supervised representations by their rank","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2210.02885","snapshot_observed_at":"2026-08-01T06:08:27.040474Z","title":"RankMe: Assessing the downstream performance of pretrained self- supervised representations by their rank,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:27.040474Z"},"links":{"cited_paper":"/paper/2210.02885","citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:05a4d9cad22ff27e609e1bb189c91283eb111bfb6328685de47091b9ebfc36c8","observation_id":"6c32635c-384c-4f86-af7e-ebeee338c32e","resolution":{"observed_at":"2026-08-01T06:08:27.040474Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.03010","last_updated":"2025-05-15T20:42:47Z","snapshot_observed_at":"2026-08-06T18:55:22.053479Z","submitted_at":"2023-10-04T17:53:53Z","title":"Spectral alignment of stochastic gradient descent for high-dimensional classification tasks","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.03010","snapshot_observed_at":"2026-08-01T06:08:27.109643Z","title":"High-dimensional SGD aligns with emerging outlier eigenspaces,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:27.109643Z"},"links":{"cited_paper":"/paper/2310.03010","citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:a4b2bf6b6e37e39e4bde77a1326e4b88912bab6b786ea8890fa059846a20c442","observation_id":"d5a9fd55-a141-4132-94ce-b54f29e706d5","resolution":{"observed_at":"2026-08-01T06:08:27.109643Z","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":"2026-08-01T06:08:27.184551Z","title":"ASVspoof 2019: Future Horizons in Spoofed and Fake Audio Detection,","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:27.184551Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:c02edc487842437606910ec8b503f4282f55e453e714262c650f5376516a098a","observation_id":"c3dbec53-cd41-4b8b-92cc-8e62eb210573","resolution":{"observed_at":"2026-08-01T06:08:27.184551Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2109.00537","last_updated":"2021-09-01T16:17:31Z","snapshot_observed_at":"2026-07-06T11:43:28.084184Z","submitted_at":"2021-09-01T16:17:31Z","title":"ASVspoof 2021: accelerating progress in spoofed and deepfake speech detection","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2109.00537","snapshot_observed_at":"2026-08-01T06:08:27.261924Z","title":"ASVspoof 2021: accelerating progress in spoofed and deepfake speech detection,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:27.261924Z"},"links":{"cited_paper":"/paper/2109.00537","citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:97b520119c945f56a634b889cc2595a573d0cd9f10c3b0892b400a676b59ea49","observation_id":"e83b3de9-4a11-401f-b740-0bdada43e83a","resolution":{"observed_at":"2026-08-01T06:08:27.261924Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2408.08739","last_updated":"2024-08-16T13:37:20Z","snapshot_observed_at":"2026-07-06T19:01:32.237848Z","submitted_at":"2024-08-16T13:37:20Z","title":"ASVspoof 5: Crowdsourced Speech Data, Deepfakes, and Adversarial Attacks at Scale","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2408.08739","snapshot_observed_at":"2026-08-01T06:08:27.367744Z","title":"ASVspoof 5: Crowdsourced Speech Data, Deepfakes, and Adversarial Attacks at Scale,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:27.367744Z"},"links":{"cited_paper":"/paper/2408.08739","citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:d98fc6f20b9a2f070c8df727d06046611c8ba6fb42b5df3af19332aa517a8cf3","observation_id":"36297aa2-ee53-45c5-8c8a-808525cd3d3c","resolution":{"observed_at":"2026-08-01T06:08:27.367744Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2108.05080","last_updated":"2022-03-01T10:38:07Z","snapshot_observed_at":"2026-08-06T06:30:26.873321Z","submitted_at":"2021-08-11T07:49:36Z","title":"FakeAVCeleb: A Novel Audio-Video Multimodal Deepfake Dataset","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2108.05080","snapshot_observed_at":"2026-08-01T06:08:27.474845Z","title":"FakeA VCeleb: A novel audio-video multimodal deepfake dataset,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:27.474845Z"},"links":{"cited_paper":"/paper/2108.05080","citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:afc394170103d677d1fc87cfee6723fa48fe6ac1a8b4c208180e4fabe08a6019","observation_id":"cb90ba01-fab1-4831-b3bf-919d156dc645","resolution":{"observed_at":"2026-08-01T06:08:27.474845Z","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":"2026-08-01T06:08:27.571914Z","title":"Audio deepfake detection: A survey,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:27.571914Z"},"links":{"citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:cc3d19e767666fadbfa3761e2ac2d5d0c36e5f75efa889405bcb06c5c3d28809","observation_id":"2054ba4f-37bc-48f7-929b-8e2ba982880b","resolution":{"observed_at":"2026-08-01T06:08:27.571914Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2104.01027","last_updated":"2021-09-08T04:12:36Z","snapshot_observed_at":"2026-08-04T09:17:59.235960Z","submitted_at":"2021-04-02T12:53:15Z","title":"Robust wav2vec 2.0: Analyzing Domain Shift in Self-Supervised Pre-Training","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2104.01027","snapshot_observed_at":"2026-08-01T06:08:27.749037Z","title":"Robust wav2vec 2.0: Analyzing domain shift in self-supervised pre-training,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:27.749037Z"},"links":{"cited_paper":"/paper/2104.01027","citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:32ee34d225db9bd9acee85ff1c9564b5c272855b8e1189b7bff815fac8e25673","observation_id":"3b2c8806-5029-448d-a9f0-f8d30e158781","resolution":{"observed_at":"2026-08-01T06:08:27.749037Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1806.07572","last_updated":"2020-02-10T08:39:09Z","snapshot_observed_at":"2026-07-30T19:50:14.167277Z","submitted_at":"2018-06-20T06:35:46Z","title":"Neural Tangent Kernel: Convergence and Generalization in Neural Networks","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1806.07572","snapshot_observed_at":"2026-08-01T06:08:26.977225Z","title":"Available: https://arxiv.org/abs/1806.07572","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":2018,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:26.977225Z"},"links":{"cited_paper":"/paper/1806.07572","citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:c1222bad08fd1246af756c6c2702a1a1fe100ff00d49eea231b9b8c046b40f80","observation_id":"6efd4d0f-16af-4acb-a937-54ddfebc85c9","resolution":{"observed_at":"2026-08-01T06:08:26.977225Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2103.02503","last_updated":"2022-08-12T08:24:43Z","snapshot_observed_at":"2026-08-07T03:55:44.562896Z","submitted_at":"2021-03-03T16:12:22Z","title":"Domain Generalization: A Survey","version":7},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2103.02503","snapshot_observed_at":"2026-08-01T06:08:24.220094Z","title":"Available: https://arxiv.org/abs/2103.02503","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":2021,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:24.220094Z"},"links":{"cited_paper":"/paper/2103.02503","citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:75a9088d95509bf3aef158873854ac5d7fd7f201ec8584462d7d24c49462b58a","observation_id":"9ecae8c7-8b41-48a7-ab47-b5f94b9efde0","resolution":{"observed_at":"2026-08-01T06:08:24.220094Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2308.14970","last_updated":"2023-08-29T01:50:01Z","snapshot_observed_at":"2026-07-06T16:11:31.948429Z","submitted_at":"2023-08-29T01:50:01Z","title":"Audio Deepfake Detection: A Survey","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2308.14970","snapshot_observed_at":"2026-08-01T06:08:27.680708Z","title":"Available: https://arxiv.org/abs/2308.14970","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection","version":1},"reference_index":2023,"source":"pdf_text","source_observed_at":"2026-08-01T06:08:27.680708Z"},"links":{"cited_paper":"/paper/2308.14970","citing_paper":"/paper/2607.22010"},"observation_digest":"sha256:42fcfd58fbb0a44ec61fdd264564cf666928fa00e828cb0004898a2614da4ce9","observation_id":"3e4629c8-d473-409a-8272-d5c11dd0b926","resolution":{"observed_at":"2026-08-01T06:08:27.680708Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2607.22010","last_updated":"2026-07-24T06:15:39Z","latest_version":1,"primary_category":"eess.AS","snapshot_observed_at":"2026-08-04T11:43:46.786247Z","submitted_at":"2026-07-24T06:15:39Z","title":"How Meta-Learning Shapes LoRA Adapter Geometry in Speech Deepfake Detection"},"reference_resolution":{"displayed":40,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":40,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":40},"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":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"thesis":"As of 7 August 2026, this Paper Citation Record lists 40 of 40 outbound references and 0 inbound Pith citation observations for arXiv:2607.22010."}