{"as_of":"2026-08-17T07:50:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:ae1d72b62a9830c1667656145f06d4e606e5d35a87af911809200fce92a9d126","coverage":[{"denominator":54,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":54,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-16T12:37:16.557697Z","state":"measured"},{"denominator":54,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":54,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-17T06:30:58.91139+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2504.12436/citation-record","integrity":"/paper/2504.12436/integrity","json":"/paper/2504.12436/citation-record.json","paper":"/paper/2504.12436"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.368195Z","title":"Bit- Fit: Simple parameter-efficient fine-tuning for transformer- based masked language-models","venue":null,"work_id":"d974963f-267d-4838-963b-5dd000fb5841","year":2022},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.349106Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:373f324cae6c29d724c40ec904346e923440568c38b0d0b45c0691760b8e2f0b","observation_id":"4b763389-f975-44c6-8757-f3031ccd2729","resolution":{"observed_at":"2026-08-16T12:37:17.372477Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.355464Z","title":"Food-101–mining discriminative components with random forests","venue":null,"work_id":"e17ddd9c-de55-4a18-aefc-57d4db649ee6","year":2014},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.354464Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:ae398d2c744f2b1199a9220346f68f7a821da76a37b2c6a85e07f8a2e61ea140","observation_id":"bd99f8a7-c227-4843-9c33-82c1ffb2fce2","resolution":{"observed_at":"2026-08-16T12:37:17.360148Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2310.18356","last_updated":"2023-10-31T04:21:33Z","snapshot_observed_at":"2026-08-16T14:49:32.109169Z","submitted_at":"2023-10-24T00:47:26Z","title":"LoRAShear: Efficient Large Language Model Structured Pruning and Knowledge Recovery","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.18356","snapshot_observed_at":"2026-08-16T12:37:16.358439Z","title":"Lorashear: Efficient large language model structured pruning and knowledge recovery","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.358439Z"},"links":{"cited_paper":"/paper/2310.18356","citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:ce7c0a687da7cc17dfdcffad6d99e04423bbd15d212e8a1f831d60a67ad7b687","observation_id":"59359242-61ba-4fe8-8d6f-54314ac4521a","resolution":{"observed_at":"2026-08-16T12:37:16.358439Z","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":"2026-08-16T12:37:16.362787Z","title":"Fira: Can we achieve full-rank training of llms under low-rank constraint? arXiv preprint arXiv:2410.01623, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.362787Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:d571be1e7bea3d7aa5a705ddd3f9f84682ef4811de4456745f5b07dd8a75c9d0","observation_id":"8cef6c77-aeb4-48cb-b434-8719efd82337","resolution":{"observed_at":"2026-08-16T12:37:16.362787Z","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":"2026-08-16T12:37:17.340905Z","title":"A survey on deep neural network pruning: Taxonomy, compar- ison, analysis, and recommendations","venue":null,"work_id":"02a917ca-129e-45f0-80a3-f95488f16d8e","year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.366763Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:c5b4fc8f714adc290c98741eb2fbde26b6da08578ef5c751c8681350aa14df4e","observation_id":"28768008-8595-4244-b5bb-19cdc21844b9","resolution":{"observed_at":"2026-08-16T12:37:17.345405Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.326807Z","title":"Reproducible scaling laws for contrastive language-image learning","venue":null,"work_id":"d24e69ff-c5cb-445f-a05a-c24e647e228b","year":2023},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.370842Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:868f2385fd2c85ba169f97aaca81a0a307c8965f63461e2d6e88fba5996a0595","observation_id":"024fa9bf-68bf-4ba3-a8c9-5d6ee8392e0a","resolution":{"observed_at":"2026-08-16T12:37:17.331319Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.311782Z","title":"Describing textures in the wild","venue":null,"work_id":"39524b2c-b5f6-47e9-a9ce-1d22d5576fce","year":2014},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.375203Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:ef4cf35977c88d7a8fcd3324a90923e87f17ca63b1da9c0e945e1a4777f76938","observation_id":"ee4bc03d-5c2c-45b7-a486-a80c76873321","resolution":{"observed_at":"2026-08-16T12:37:17.316956Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.295773Z","title":"Imagenet: A large-scale hierarchical image database","venue":null,"work_id":"70ebd1ba-16ab-4051-bba3-77dac9668ac7","year":2009},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.378985Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:d144e6818d531d7555b671d322d105f01a8201cd636a80dde683c093e1764479","observation_id":"8f32c208-df55-464b-a928-c3ac9ccf7b83","resolution":{"observed_at":"2026-08-16T12:37:17.301390Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.383137Z","title":"Everybody prune now: Structured pruning of llms with only forward passes","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.383137Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:60e700c5dce0549a899d258e83e459b08956c0c2296cf5672e703521cd63f2d1","observation_id":"e70a8c91-d79f-4b0f-ab7f-6492350c5889","resolution":{"observed_at":"2026-08-16T12:37:16.383137Z","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":"2026-08-16T12:37:17.280990Z","title":"Learning gener- ative visual models from few training examples: An incre- mental bayesian approach tested on 101 object categories","venue":null,"work_id":"a4a4adc1-747b-45a0-9cc6-b96d8d3f1374","year":2004},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.387346Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:e22f63d2c2245aa90436da37c952e2fe8823ab1b3e2271c4751dff55e612ea05","observation_id":"c49d0108-3498-47d2-be2d-2cc91021d66f","resolution":{"observed_at":"2026-08-16T12:37:17.285604Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2407.07802","last_updated":"2024-07-10T16:20:53Z","snapshot_observed_at":"2026-08-16T13:35:22.102604Z","submitted_at":"2024-07-10T16:20:53Z","title":"ROSA: Random Subspace Adaptation for Efficient Fine-Tuning","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.07802","snapshot_observed_at":"2026-08-16T12:37:16.391193Z","title":"Rosa: Random sub- space adaptation for efficient fine-tuning","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.391193Z"},"links":{"cited_paper":"/paper/2407.07802","citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:212f970892cc316a71093de6064feb745baeec651ca78c7b9767613bf1c62432","observation_id":"4bd8a2f4-3541-4800-aebe-637c6a775b81","resolution":{"observed_at":"2026-08-16T12:37:16.391193Z","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":"2026-08-16T12:37:17.267228Z","title":"SLTrain: a sparse plus low-rank approach for parameter and memory efficient pretraining","venue":null,"work_id":"759d61f5-56c4-438f-a74c-3d80a65274c8","year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.395770Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:e5ff5129ede75426f7c101a6a59f6cf18a6c46d9d301b86a6c9c9e45a61b38a6","observation_id":"65d56f6e-e081-4ef5-ba98-9b628573cd61","resolution":{"observed_at":"2026-08-16T12:37:17.272505Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.253774Z","title":"Towards a unified view of parameter-efficient transfer learning","venue":null,"work_id":"a834b14a-6dd2-4f52-ac35-3a47cfe5a49d","year":2021},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.400104Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:05c0377bfe98b62067b8a5ff5377877cb8022ca2e607fc1dc9a278ec6c9ac0ed","observation_id":"20343172-f602-468d-bf80-80b6e02fd08c","resolution":{"observed_at":"2026-08-16T12:37:17.257958Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.404299Z","title":"Eurosat: A novel dataset and deep learning benchmark for land use and land cover classification","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.404299Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:67ba5d38cb31a9fc3480afc02bae63cc04f3da660bc8a9955d3c9afed9ec1bb3","observation_id":"db4f19a6-af90-4d86-80f5-e470e59c8966","resolution":{"observed_at":"2026-08-16T12:37:16.404299Z","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":"2026-08-16T12:37:17.228184Z","title":"Parameter-efficient transfer learning for nlp","venue":null,"work_id":"87d52451-4ea2-4dce-93ce-fcc93e65ba9c","year":2019},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.408079Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:c90e94d7bbd784be3ee1c0011f02d77b910afea51147c03f8351fb054106ebe9","observation_id":"cbebcd20-0c56-4d5d-85f1-fc1bae8af2b8","resolution":{"observed_at":"2026-08-16T12:37:17.232404Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.412019Z","title":"LoRA: Low-rank adaptation of large language models","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.412019Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:eb4574b610991a22ba6e005a52096e7554bae20059d428ec43421ee20c7b394d","observation_id":"254ab4a0-b492-4b18-9b1f-606fcc9dee9e","resolution":{"observed_at":"2026-08-16T12:37:16.412019Z","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":"2026-08-16T12:37:17.207529Z","title":"Do sparse brain activ- ity patterns underlie human cognition? NeuroImage, 263: 119633, 2022","venue":null,"work_id":"4b45bb4b-f8b7-42e0-b2a8-01b633d887ef","year":2022},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.416318Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:6b66a397ec702b0c71229de52783a73576ca5d661d997a6abe0ae266d511e81f","observation_id":"124434b1-1723-422d-a5ac-ade9defec8ca","resolution":{"observed_at":"2026-08-16T12:37:17.211762Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.194642Z","title":"Vi- sual prompt tuning","venue":null,"work_id":"f9cf5e31-247e-4261-9719-b97b21750252","year":2022},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.420530Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:04e1bf7fb853d458a59ee4aa7b27c52d6c69f31e0e65f12ea2fff47834655673","observation_id":"db920feb-20a1-4215-b6f9-e4909b263505","resolution":{"observed_at":"2026-08-16T12:37:17.198803Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.182619Z","title":"Adam: A method for stochastic optimization","venue":null,"work_id":"7e6114f9-1950-42f5-8cf1-e268e01ad7f7","year":2015},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.424027Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:9a45ea16ddfb35bbfa0f8370fb1b8ebe20dca08b89bbd2c9011d1fe48ca509e8","observation_id":"3cf784bd-9007-40f5-9658-add475d9f722","resolution":{"observed_at":"2026-08-16T12:37:17.186657Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.170816Z","title":"VeRA: Vector-based random matrix adaptation","venue":null,"work_id":"1f4dfd1a-a7f5-41ee-a4ee-81f795c516fa","year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.427731Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:757884def4b6548a02e9aeda1f62ec4cc5e97fc82ecaaa4f209757deffb3e671","observation_id":"fe107bfb-b69a-4c16-a9b8-686d7da904cf","resolution":{"observed_at":"2026-08-16T12:37:17.175380Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.159259Z","title":"3d object representations for fine-grained categorization","venue":null,"work_id":"fc9392e5-93e5-4f53-a03d-1b42f6f5f0bc","year":2013},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.431226Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:9566ddcd351b1de4769408f8ee786d22124edf78b95cd17e0002910b87b414cf","observation_id":"58639b98-16ae-493c-bf20-bf1783d5b2a3","resolution":{"observed_at":"2026-08-16T12:37:17.163216Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.147797Z","title":"Block pruning for faster transformers","venue":null,"work_id":"081b5129-578c-423b-a291-1ec2125d9d2a","year":2021},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.435781Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:eb5061eb49a4278c2411d0d29b183eba5c65f56b3f3be3af141de447ed387ba9","observation_id":"6ec408eb-8f18-4c57-ae86-a6daae0e79ab","resolution":{"observed_at":"2026-08-16T12:37:17.151695Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.135998Z","title":"Relora: High-rank training through low-rank updates","venue":null,"work_id":"11353c0b-c1ec-4791-bf44-34ae40cd2200","year":2023},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.439409Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:670e411b44550a701f8ef57ffe8de15e7c55e04d62e203ea78db9bfd98028db0","observation_id":"f0b70be3-87a7-4fbc-8f8b-7bb51ed153ef","resolution":{"observed_at":"2026-08-16T12:37:17.139953Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2408.10280","last_updated":"2024-08-27T15:34:49Z","snapshot_observed_at":"2026-08-16T13:25:44.626379Z","submitted_at":"2024-08-18T12:18:56Z","title":"NoRA: Nested Low-Rank Adaptation for Efficient Fine-Tuning Large Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2408.10280","snapshot_observed_at":"2026-08-16T12:37:16.443399Z","title":"Nora: Nested low-rank adaptation for efficient fine-tuning large models","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.443399Z"},"links":{"cited_paper":"/paper/2408.10280","citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:a443a65194c7e9877ed172305f15c6b85931415308ef2ae504c53747d18c8f69","observation_id":"a16c9667-beb4-4b56-acdd-77c1e4bf909c","resolution":{"observed_at":"2026-08-16T12:37:16.443399Z","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":"2026-08-16T12:37:17.122668Z","title":"DoRA: Weight-decomposed low-rank adaptation","venue":null,"work_id":"90a0746b-9614-4df9-919a-259d1517867b","year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.447781Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:271331d12a2c64d3689aa5fa1c15a316e5ab545a69812f9d364e467a3867f180","observation_id":"98547500-9f5c-4ae7-a6ed-056bb61f0f34","resolution":{"observed_at":"2026-08-16T12:37:17.127072Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.110551Z","title":"Segclip: Patch aggregation with learnable centers for open-vocabulary semantic segmentation","venue":null,"work_id":"929b8c2c-fee0-439e-b3d2-861b5dcfa2cb","year":2023},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.451903Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:ee8ddd4e7c27f7b494b0241853abf4e1f9dfbe08caad87f7149a3f47b5e30e74","observation_id":"021329c7-ae0d-432f-9d3b-27104159e046","resolution":{"observed_at":"2026-08-16T12:37:17.114649Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.098447Z","title":"Llm-pruner: On the structural pruning of large language models","venue":null,"work_id":"d27713ea-acfd-4be3-b02c-f12cf3f3bcc7","year":null},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.455596Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:de8f4e51c80d5194cbaacc9945fb100cc52123db40ae025e3fdd18999129415a","observation_id":"d0513266-895b-4120-90b2-584a77f72d3a","resolution":{"observed_at":"2026-08-16T12:37:17.102357Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.086227Z","title":"Fine-grained visual classi- fication of aircraft","venue":null,"work_id":"638466f1-c1ba-4d65-a514-aa91197af85c","year":2013},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.459397Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:06960f7e91d2ff2424d5e35bd0d3ca8f27a087db987c5ee6623987e6a256f9bb","observation_id":"e8d7a183-8440-4091-a942-c728b42a1db1","resolution":{"observed_at":"2026-08-16T12:37:17.090748Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.071307Z","title":"PiSSA: Principal singular values and singular vectors adaptation of large language models","venue":null,"work_id":"c78332c0-7db5-4b46-a04e-df024cf3954d","year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.463255Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:26f6a944039d5c68629c9f9c848249f53b08a66b061c61b33e69464e7b338942","observation_id":"c0f5efc4-0232-4868-a634-9f64cb6fde1b","resolution":{"observed_at":"2026-08-16T12:37:17.075979Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.058749Z","title":"Automated flower classification over a large number of classes","venue":null,"work_id":"a528bb91-bc71-4252-9193-62ef35793b0c","year":2008},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.466835Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:53bf4ddad52bb96a8fb30302a754351553f3e7061911e87980d294aca42cdbfe","observation_id":"1085c22d-531f-43be-bba3-bb6d3c7e075b","resolution":{"observed_at":"2026-08-16T12:37:17.063191Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.045805Z","title":"Lisa: Layerwise importance sampling for memory-efficient large language model fine- tuning","venue":null,"work_id":"dc011b28-2f97-448e-811f-7e89f2cc136a","year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.470336Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:cb100e4bd6776a80d9d5cc119daed01e938ff3ed403a4739ed8b7fe7f88a61af","observation_id":"7679b86b-69b3-4126-ab53-e8a420b5b443","resolution":{"observed_at":"2026-08-16T12:37:17.049969Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.032825Z","title":"Cats and dogs","venue":null,"work_id":"132a70c6-22c0-456c-93db-6fbd711bad20","year":2012},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.473947Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:3dc50333bff42363896912cc8263a6f76ce7c1e90efff69f84bda89594a9c075","observation_id":"6952c1ab-4925-41ab-a765-c488357d75c2","resolution":{"observed_at":"2026-08-16T12:37:17.036771Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:17.018398Z","title":"Learn- ing transferable visual models from natural language super- vision","venue":null,"work_id":"630a8532-c545-42e6-ada6-ef0a9f331587","year":2021},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.477709Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:bb1c213d3a9b4b2fc3d314f34670e4313e8857483801d53153a1f45921059dd7","observation_id":"bf0f5305-ca05-4d3d-9ad0-701b93d45fe0","resolution":{"observed_at":"2026-08-16T12:37:17.022483Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.481519Z","title":"A closer look at the few-shot adaptation of large vision-language models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.481519Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:4d004e1901246e6696a3c18cef53a0a4fbff8205a7e0c07fedc61fcf0f494c9e","observation_id":"4eaedcae-dba4-42a9-970b-59eeb5a9e4e6","resolution":{"observed_at":"2026-08-16T12:37:16.481519Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1212.0402","last_updated":"2012-12-03T14:45:31Z","snapshot_observed_at":"2026-08-16T21:21:44.768787Z","submitted_at":"2012-12-03T14:45:31Z","title":"UCF101: A Dataset of 101 Human Actions Classes From Videos in The Wild","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1212.0402","snapshot_observed_at":"2026-08-16T12:37:16.485290Z","title":"Ucf101: A dataset of 101 human actions classes from videos in the wild","venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.485290Z"},"links":{"cited_paper":"/paper/1212.0402","citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:7939a97e20c4535861f7b8bac3d23a0ff048cd0cd213129fa77cbbb642b7473e","observation_id":"9e56f9a5-52c0-41e5-864b-e2ddfe81452d","resolution":{"observed_at":"2026-08-16T12:37:16.485290Z","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":"2026-08-16T12:37:16.998352Z","title":"A simple and effective pruning approach for large language models","venue":null,"work_id":"570f01b0-fe6c-48cb-ab96-eae3157f524d","year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.489319Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:527d6533d1bc54011bf9e3f561f3c070a7de44810f6f998cda3e2cbc257fa8b4","observation_id":"d17d9ea9-e4ab-4d9b-b95f-a0e2f86b4891","resolution":{"observed_at":"2026-08-16T12:37:17.002210Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2304.08109","last_updated":"2023-04-18T03:08:18Z","snapshot_observed_at":"2026-08-16T15:39:59.153035Z","submitted_at":"2023-04-17T09:36:36Z","title":"A Comparative Study between Full-Parameter and LoRA-based Fine-Tuning on Chinese Instruction Data for Instruction Following Large Language Model","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2304.08109","snapshot_observed_at":"2026-08-16T12:37:16.493023Z","title":"A comparative study between full-parameter and lora-based fine-tuning on chinese instruction data for instruction follow- ing large language model","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.493023Z"},"links":{"cited_paper":"/paper/2304.08109","citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:617b452c77f9b5e07e660e539b0f6bf48011962de236a8049f79db0d85b370ab","observation_id":"0222cb59-b50c-4292-a6d8-faa50af7e058","resolution":{"observed_at":"2026-08-16T12:37:16.493023Z","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":"2026-08-16T12:37:16.985591Z","title":"Training neu- ral networks with fixed sparse masks","venue":null,"work_id":"21f49a4e-b62c-48d8-971b-116cd174e282","year":2021},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.497191Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:81d939be978e1114203a2bc3743ea2642c1729aa39c5be9c13c811c23700da1b","observation_id":"1d763693-e3a0-402a-b255-23fc1274303a","resolution":{"observed_at":"2026-08-16T12:37:16.990866Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.972820Z","title":"Cora: Adapting clip for open-vocabulary detection with region prompting and anchor pre-matching","venue":null,"work_id":"506a0e4e-5e88-43b2-8db3-5585b05a4f7d","year":2023},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.500809Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:3b6d4349a847d11249d7a6e93cf6a0cb63afc71e294857e9e98afc4d35562cb5","observation_id":"bce9bfe2-e0ab-4ff7-9f24-950f3d455c6e","resolution":{"observed_at":"2026-08-16T12:37:16.976893Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.961295Z","title":"A simple model for behav- ioral time scale synaptic plasticity (btsp) provides content ad- dressable memory with binary synapses and one-shot learn- ing","venue":null,"work_id":"89c77a83-03ba-4a3f-8149-66d40429e3b0","year":2025},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.504650Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:8e53a0b4b35c11f2e92c4971b772b9c9e58c6c6e3fe7c948dce48700813fbe59","observation_id":"7970666b-7ae8-44fe-8191-5c094675269d","resolution":{"observed_at":"2026-08-16T12:37:16.965402Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.949724Z","title":"Struc- tured pruning learns compact and accurate models","venue":null,"work_id":"761009e5-ea61-413f-933a-cf595ef8e1b0","year":2022},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.508647Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:2964449eadde3d5a0c9aeee92a718ed1f40e09a347e09a685d69284c018eea87","observation_id":"5b67504d-ee2a-43ee-9a3e-70f0ed0bc60a","resolution":{"observed_at":"2026-08-16T12:37:16.953762Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.937316Z","title":"Sun database: Large-scale scene recognition from abbey to zoo","venue":null,"work_id":"74daf5a5-a2a6-4df2-9563-230d8e9e28f9","year":2010},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.512650Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:af32a91999f39515a15ab0fbb63c237a831182c1f13b9113f5eb4939c3032306","observation_id":"94611f9d-f0a6-4c51-ad3e-12cb9f312c06","resolution":{"observed_at":"2026-08-16T12:37:16.941625Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.925238Z","title":"Raise a child in large language model: Towards effective and generalizable fine-tuning","venue":null,"work_id":"ac0a0108-b038-4676-bedd-5982ab261a8b","year":2021},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.516436Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:372bf44627cdf382e2d47f4f943e59b20889e985d4d4ccbc87afe51d761f1362","observation_id":"16e4f042-4e3e-4587-bce8-0bd5ad2b487a","resolution":{"observed_at":"2026-08-16T12:37:16.929184Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.912542Z","title":"CorDA: Context-oriented decomposition adaptation of large language models for task-aware parameter-efficient fine- tuning","venue":null,"work_id":"dcfe37c8-5676-4f79-b685-7841e6583395","year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.520007Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:7d8ae113197917719857ca0d5c48686a30410dfdf8c6f050ce0e32b0e2f00db3","observation_id":"ad64e988-6f9a-43a5-8bb4-8913b65781da","resolution":{"observed_at":"2026-08-16T12:37:16.916808Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.899939Z","title":"Task residual for tuning vision-language models","venue":null,"work_id":"0f48cba1-fb00-4d11-a30d-ab8ef8cc7109","year":2023},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.523789Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:d1664c5c50d7b5667ab9a41753dec6b242114b8a39445e37020776ab47eeec4c","observation_id":"33c08aae-7bf8-4317-816a-49470c97a9fc","resolution":{"observed_at":"2026-08-16T12:37:16.904018Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.888434Z","title":"Low-rank few-shot adaptation of vision-language models","venue":null,"work_id":"6e191db1-ebf0-4a32-8e75-92ed0472b971","year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.527317Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:8d10226c163e7447968714974e8ddded5cdb6427b193b81a81bbdf1569a8eeab","observation_id":"cf42375a-3359-4a78-b1de-a9883afef9a2","resolution":{"observed_at":"2026-08-16T12:37:16.892254Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.876385Z","title":"Gradient- based parameter selection for efficient fine-tuning","venue":null,"work_id":"d6a0ca4f-2d37-45cf-b7f3-d591c7e58386","year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.530928Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:f72c152ea50e689c7295dcf4db4588a37b7868bd911066df2017ecc37591e95a","observation_id":"2eebf403-26bc-4bd0-891e-2a536312136c","resolution":{"observed_at":"2026-08-16T12:37:16.880405Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.863854Z","title":"Galore: Memory- efficient llm training by gradient low-rank projection","venue":null,"work_id":"8c6b632a-79e5-4c27-8af9-d76d663995e9","year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.534490Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:6e6bf6d66e80078eb0b496f98dd70c43a1e1b0b9c10f7441c9dc4030bd44af16","observation_id":"6aa21d53-3e0c-4aa1-9aac-712dae383aa8","resolution":{"observed_at":"2026-08-16T12:37:16.868378Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.849552Z","title":"Conditional prompt learning for vision-language models","venue":null,"work_id":"703a2f4b-d89e-42ba-b7b4-6b7e7e1228ae","year":2022},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.538015Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:edec9888fe7c348f809d3425fce3a814bc5547f2213f0a008400c81dd3a16f5f","observation_id":"1453bc19-85aa-4904-8c60-ca166b78cff6","resolution":{"observed_at":"2026-08-16T12:37:16.853724Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.833601Z","title":"Learning to prompt for vision-language models","venue":null,"work_id":"1c6382d3-4b69-4425-9b9a-33be50a778ba","year":2022},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.542047Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:48a2863b326b0c9111b1d3df4b5a59fd22c753416f36880a531a3a7833bffb9f","observation_id":"103429c2-f3e4-4776-8c14-fcab2f740559","resolution":{"observed_at":"2026-08-16T12:37:16.838027Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.819194Z","title":"Not all features mat- ter: Enhancing few-shot clip with adaptive prior refinement","venue":null,"work_id":"f111bc46-572f-4751-bef5-0c6bf2dcd039","year":2023},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.545594Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:10bd9708215ab5a6337993211b8765f46ac2db11a1c617de85a847e37b969caa","observation_id":"660e4989-aae3-4f3e-8fe0-735e1f600566","resolution":{"observed_at":"2026-08-16T12:37:16.823183Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.804799Z","title":"The effects of regularization and data augmentation are class de- pendent","venue":null,"work_id":"42610481-85b9-40b4-9e85-638c0f15274e","year":2022},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.549199Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:04a6fe569f4474f24f032fbf726d47858fcafd7e197debad2211ba13f06eb5cb","observation_id":"c1fedb22-778a-4b89-9f1c-fccff7e2353d","resolution":{"observed_at":"2026-08-16T12:37:16.809667Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-16T12:37:16.789196Z","title":"CLIP-Adapter: Better vision-language models with feature adapters","venue":null,"work_id":"4712f5f9-1585-43e2-baeb-d62033d4e1b5","year":2024},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.553106Z"},"links":{"citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:91a1a92628b3be8a44502b818093b00e3a49b26e3fd6d0117cebca9f432dd3fe","observation_id":"6127d82a-d8dd-4f20-816f-2a290ced03a7","resolution":{"observed_at":"2026-08-16T12:37:16.795423Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-17T06:30:58.91139+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2310.11454","last_updated":"2024-01-16T18:59:22Z","snapshot_observed_at":"2026-08-16T14:51:17.392681Z","submitted_at":"2023-10-17T17:59:46Z","title":"VeRA: Vector-based Random Matrix Adaptation","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.11454","snapshot_observed_at":"2026-08-16T12:37:16.557697Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation","version":2},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-16T12:37:16.557697Z"},"links":{"cited_paper":"/paper/2310.11454","citing_paper":"/paper/2504.12436"},"observation_digest":"sha256:433174e828d366dd61a5bfb13c1db1f60c2cd329d086e91f4382b80ec53cdff3","observation_id":"b31530e7-b8a5-4313-8aea-e2087a8e3cb9","resolution":{"observed_at":"2026-08-16T12:37:16.557697Z","resolver_source":null,"status":"malformed_identifier"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2504.12436","last_updated":"2025-08-11T16:17:39Z","latest_version":2,"primary_category":"cs.CV","snapshot_observed_at":"2026-08-16T12:29:15.971848Z","submitted_at":"2025-04-16T19:10:34Z","title":"Sparsity Outperforms Low-Rank Projections in Few-Shot Adaptation"},"reference_resolution":{"displayed":54,"state_counts":{"malformed_identifier":1,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":10,"verified_exact":0,"verified_fuzzy":43},"total_outbound_references":54},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-17T06:30:58.91139+00:00","source":"crossref"},{"observed_at":"2026-08-17T06:30:54.323127+00:00","source":"retraction_watch"}],"thesis":"As of 17 August 2026, this Paper Citation Record lists 54 of 54 outbound references and 0 inbound Pith citation observations for arXiv:2504.12436."}