{"as_of":"2026-08-10T10:29:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:f55e0a76039e609d74716086b5e491a2f7b3b82a32bac92e4bdf659ffa6bff0b","coverage":[{"denominator":50,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":50,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-09T20:51:50.089984Z","state":"measured"},{"denominator":50,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":50,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-10T06:31:04.303077+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/2501.19287/citation-record","integrity":"/paper/2501.19287/integrity","json":"/paper/2501.19287/citation-record.json","paper":"/paper/2501.19287"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-09T20:51:49.912062Z","title":"write newline","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":1,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.912062Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:47c06c732b3a6b51851c85267a925fe8f5659b67115aac7505c3d4b3d9a39be8","observation_id":"808528f3-9742-4e3e-9e1d-a614c9e90715","resolution":{"observed_at":"2026-08-09T20:51:49.912062Z","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":"2026-08-09T20:51:50.831069Z","title":"Privacy amplification by subsampling: Tight analyses via couplings and divergences","venue":null,"work_id":"1935b700-9c91-470c-bfda-60863f90815f","year":2018},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":2,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.917750Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:9928db5fabae3e6a683515c8731b832f45d6c090445324483f0bb67802937dda","observation_id":"bc9ef88a-7c74-4501-82a9-06a428179f26","resolution":{"observed_at":"2026-08-09T20:51:50.834670Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+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":"2026-08-09T20:51:49.923223Z","title":"Hypothesis testing interpretations and renyi differential privacy","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":3,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.923223Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:1ed78fb0387856949d77b8a984cab42c3e69b3d9e8ec246ca77f86538bb02c8a","observation_id":"76ab7b60-36cb-4f4f-afaa-820a1a89aae4","resolution":{"observed_at":"2026-08-09T20:51:49.923223Z","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":"2026-08-09T20:51:49.927606Z","title":"D., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al","venue":null,"work_id":null,"year":1901},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":4,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.927606Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:80402ddbb384dcad63e2e3aff9719b20b435a87479ff23ac6a887076f896b592","observation_id":"e791d802-0d57-4ef4-a46f-ccf087d57c15","resolution":{"observed_at":"2026-08-09T20:51:49.927606Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2403.06634","last_updated":"2024-07-09T17:44:00Z","snapshot_observed_at":"2026-07-06T17:42:33.242895Z","submitted_at":"2024-03-11T11:46:12Z","title":"Stealing Part of a Production Language Model","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.06634","snapshot_observed_at":"2026-08-09T20:51:49.931701Z","title":"D., Steinke, T., Hayase, J., Cooper, A","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":5,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.931701Z"},"links":{"cited_paper":"/paper/2403.06634","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:53106aa185e191cc9cedcb880db8cacd183b470964811710187285da469b386b","observation_id":"62b9262a-9a08-4344-ae83-fb910a0907e2","resolution":{"observed_at":"2026-08-09T20:51:49.931701Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1810.04805","last_updated":"2019-05-24T20:37:26Z","snapshot_observed_at":"2026-07-30T09:12:38.100527Z","submitted_at":"2018-10-11T00:50:01Z","title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1810.04805","snapshot_observed_at":"2026-08-09T20:51:49.936033Z","title":"Bert: Pre-training of deep bidirectional transformers for language understanding","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":6,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.936033Z"},"links":{"cited_paper":"/paper/1810.04805","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:c7d54ba472d578e55b17f87b1bba5dda512604d13b444fea6cf43badb8c3df8b","observation_id":"bc7d88ac-012a-433f-b4ef-ff114003e696","resolution":{"observed_at":"2026-08-09T20:51:49.936033Z","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":"2026-08-09T20:51:50.807010Z","title":"On the privacy risk of in-context learning","venue":null,"work_id":"42b6ec52-bae2-49b7-9fd5-e6bcb7b41ec5","year":2023},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":7,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.940246Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:f12f5b0b119bcebf40cf67c9e179c8d5ed6368dbd4e01263f09cfab6d7414cbe","observation_id":"d33a6ebd-0e12-49a5-85b8-997294928950","resolution":{"observed_at":"2026-08-09T20:51:50.810504Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+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":"2026-08-09T20:51:50.796056Z","title":"Flocks of stochastic parrots: Differentially private prompt learning for large language models","venue":null,"work_id":"faaf05c0-be4a-41a0-a19f-2f8cacd5de41","year":2024},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":8,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.944260Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:9fae98b10b992776f7ac7936f6e7fafd554f57ad484e6ec9e87f518f363a5395","observation_id":"33d8f48b-f27a-4e42-84f9-5dfd88f9a215","resolution":{"observed_at":"2026-08-09T20:51:50.799448Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2407.21783","last_updated":"2024-11-23T23:27:33Z","snapshot_observed_at":"2026-07-06T18:55:11.576666Z","submitted_at":"2024-07-31T17:54:27Z","title":"The Llama 3 Herd of Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.21783","snapshot_observed_at":"2026-08-09T20:51:49.947783Z","title":"The llama 3 herd of models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":9,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.947783Z"},"links":{"cited_paper":"/paper/2407.21783","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:051d0d2f48e5f9ff48e541551c42ffdbab3452bfd89fade5ab543fed9c8ffbe6","observation_id":"737cc81d-01ec-4883-a0d5-c41a2b90cbed","resolution":{"observed_at":"2026-08-09T20:51:49.947783Z","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":"2026-08-09T20:51:49.951411Z","title":"Differential privacy","venue":null,"work_id":null,"year":2006},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":10,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.951411Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:fff41ad613a77cf0f2b26e5257a8d0af039b2aa3cccec45c46f0a92a0586ce18","observation_id":"136d5d45-bf8f-4f7a-8f60-2245602b7c54","resolution":{"observed_at":"2026-08-09T20:51:49.951411Z","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":"2026-08-09T20:51:50.778925Z","title":"and Feldman, V","venue":null,"work_id":"5ac33568-b02f-4bf2-b5a6-0d3bad3b24ae","year":2018},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":11,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.955024Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:5a97ade372fae031d2ad0296c6fd654a4eabd7983df95454d99e7bcfb14da47d","observation_id":"0189da05-4afb-4b2b-a665-8395d2cfab76","resolution":{"observed_at":"2026-08-09T20:51:50.782338Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+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":"2026-08-09T20:51:49.958570Z","title":"The algorithmic foundations of differential privacy","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":12,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.958570Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:c8e5c7a44aa323f96ab094d76d12428f7b86d3fc88ea3c0905842aae79a79b07","observation_id":"36e0bbfe-e5f2-4f9b-ae5f-f2adb2a4f26c","resolution":{"observed_at":"2026-08-09T20:51:49.958570Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1805.04833","last_updated":"2018-05-13T07:07:08Z","snapshot_observed_at":"2026-07-06T06:38:48.758485Z","submitted_at":"2018-05-13T07:07:08Z","title":"Hierarchical Neural Story Generation","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1805.04833","snapshot_observed_at":"2026-08-09T20:51:49.961841Z","title":"Hierarchical neural story generation","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":13,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.961841Z"},"links":{"cited_paper":"/paper/1805.04833","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:f21cee2663b799b8c926d9248bbeb534dd6c4fa1d4a5693f092fff828fecdf00","observation_id":"79a0372e-e2cb-40fd-8bc5-07c8380447b4","resolution":{"observed_at":"2026-08-09T20:51:49.961841Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2403.15638","last_updated":"2024-04-26T20:24:11Z","snapshot_observed_at":"2026-07-06T17:49:20.056379Z","submitted_at":"2024-03-22T22:27:44Z","title":"Differentially Private Next-Token Prediction of Large Language Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.15638","snapshot_observed_at":"2026-08-09T20:51:49.965171Z","title":"Differentially private next-token prediction of large language models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":14,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.965171Z"},"links":{"cited_paper":"/paper/2403.15638","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:0f1f33444751b1e246926a0e9ca46f4682af69baf19280d71d16e8719bd6b2a8","observation_id":"da358724-4a08-4e93-a6b3-b2832b627140","resolution":{"observed_at":"2026-08-09T20:51:49.965171Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2201.00971","last_updated":"2022-01-04T04:23:38Z","snapshot_observed_at":"2026-08-09T15:20:21.245180Z","submitted_at":"2022-01-04T04:23:38Z","title":"Submix: Practical Private Prediction for Large-Scale Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2201.00971","snapshot_observed_at":"2026-08-09T20:51:49.968645Z","title":"Submix: Practical private prediction for large-scale language models","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":15,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.968645Z"},"links":{"cited_paper":"/paper/2201.00971","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:f7c086949ca812193961efab1868216c49ce641c708d1979e49fa2cb29e3c357","observation_id":"480166f5-8630-4e3c-997a-5ab58a03e5cd","resolution":{"observed_at":"2026-08-09T20:51:49.968645Z","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":"2026-08-09T20:51:49.972152Z","title":"SAMS um corpus: A human-annotated dialogue dataset for abstractive summarization","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":16,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.972152Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:bb45557eb16758e2a0ead098c33bda0565483093ef9b380258af13b92ec37836","observation_id":"e9998b91-fde9-4101-afed-4de0aa3a44c7","resolution":{"observed_at":"2026-08-09T20:51:49.972152Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2411.05818","last_updated":"2024-11-15T16:23:17Z","snapshot_observed_at":"2026-07-06T19:47:34.320753Z","submitted_at":"2024-11-02T12:02:09Z","title":"Open LLMs are Necessary for Current Private Adaptations and Outperform their Closed Alternatives","version":2},"cited_work":{"arxiv_id":"2411.05818","doi":null,"metadata_source":"pith","pith_arxiv_id":"2411.05818","snapshot_observed_at":"2026-08-09T20:51:50.567663Z","title":"Open LLMs are Necessary for Current Private Adaptations and Outperform their Closed Alternatives","venue":"cs.LG","work_id":"6f7a209e-38b8-438b-8fdd-ffcdd2268759","year":2024},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":17,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.975733Z"},"links":{"cited_paper":"/paper/2411.05818","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:5b22f0da12a21b40172752ac1b6b313d9a59c23e2096825d15aee02f295bc5f3","observation_id":"5ad38351-9cd6-455d-be7c-ee506c375c0a","resolution":{"observed_at":"2026-08-09T20:51:50.572004Z","resolver_source":"local_arxiv","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2312.03724","last_updated":"2024-03-17T23:16:41Z","snapshot_observed_at":"2026-07-06T16:57:55.137830Z","submitted_at":"2023-11-27T02:01:10Z","title":"DP-OPT: Make Large Language Model Your Privacy-Preserving Prompt Engineer","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2312.03724","snapshot_observed_at":"2026-08-09T20:51:49.979086Z","title":"T., Zhang, C., Li, Z., Li, B., and Wang, Z","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":18,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.979086Z"},"links":{"cited_paper":"/paper/2312.03724","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:9af6fac31247d4ca1c95b8f33484ddd6b8dec7474f5c899a31facf130d46484a","observation_id":"b1c5c8ca-a98a-43f5-89ea-25bff54e137f","resolution":{"observed_at":"2026-08-09T20:51:49.979086Z","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":"2026-08-09T20:51:50.761526Z","title":"Local differential privacy for sampling","venue":null,"work_id":"6b9becfc-b2c7-463d-96ea-6523fe4b33ea","year":2020},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":19,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.982569Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:76017cf1bc6832ce96093c52d633a4697559b262aa68910bf90ccf0d3bd13e4c","observation_id":"4115e0bd-59b8-4448-9aa8-d023106aefb2","resolution":{"observed_at":"2026-08-09T20:51:50.764808Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2309.12871","last_updated":"2024-12-31T07:56:13Z","snapshot_observed_at":"2026-07-06T16:22:21.746857Z","submitted_at":"2023-09-22T13:52:42Z","title":"AnglE-optimized Text Embeddings","version":9},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2309.12871","snapshot_observed_at":"2026-08-09T20:51:49.986135Z","title":"and Li, J","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":20,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.986135Z"},"links":{"cited_paper":"/paper/2309.12871","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:edc3ee9a5406749f0ac77c7c6ae26e9be34c7cd1bfc6fd5319bdcea8da49baf9","observation_id":"af8f52f4-788c-41f8-920e-314b07ce59ed","resolution":{"observed_at":"2026-08-09T20:51:49.986135Z","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":"2026-08-09T20:51:49.989652Z","title":"Rouge: A package for automatic evaluation of summaries","venue":null,"work_id":null,"year":2004},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.989652Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:83c569f058b554337cffe175a79f9e5c00dc63c126a57287982eacf42695206d","observation_id":"6b79170b-93bf-45f6-8e81-cab6a1ed7dc4","resolution":{"observed_at":"2026-08-09T20:51:49.989652Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2205.13621","last_updated":"2022-09-08T20:40:59Z","snapshot_observed_at":"2026-08-05T13:32:50.363974Z","submitted_at":"2022-05-26T20:50:58Z","title":"Differentially Private Decoding in Large Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2205.13621","snapshot_observed_at":"2026-08-09T20:51:49.992976Z","title":"Differentially private decoding in large language models","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":22,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.992976Z"},"links":{"cited_paper":"/paper/2205.13621","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:18b0f7dcea795dc568865ecd45444863474638e5e7d562e94e67e11d75425514","observation_id":"26260a7d-58ec-4e53-adb7-850b15d06720","resolution":{"observed_at":"2026-08-09T20:51:49.992976Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2108.04106","last_updated":"2022-03-15T06:53:43Z","snapshot_observed_at":"2026-07-06T11:36:47.756091Z","submitted_at":"2021-08-09T15:06:26Z","title":"Noisy Channel Language Model Prompting for Few-Shot Text Classification","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2108.04106","snapshot_observed_at":"2026-08-09T20:51:49.996466Z","title":"Noisy channel language model prompting for few-shot text classification","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":23,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:49.996466Z"},"links":{"cited_paper":"/paper/2108.04106","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:eede2d82d2b8e6b5b94488e654b35ce5b142c0f4e599e840481b77b7fe6b901f","observation_id":"c6871c72-d3f7-45fb-b654-b21749b4067a","resolution":{"observed_at":"2026-08-09T20:51:49.996466Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2202.12837","last_updated":"2022-10-20T14:04:10Z","snapshot_observed_at":"2026-08-01T15:23:39.998892Z","submitted_at":"2022-02-25T17:25:19Z","title":"Rethinking the Role of Demonstrations: What Makes In-Context Learning Work?","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2202.12837","snapshot_observed_at":"2026-08-09T20:51:50.000257Z","title":"Rethinking the role of demonstrations: What makes in-context learning work? arXiv preprint arXiv:2202.12837, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":24,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.000257Z"},"links":{"cited_paper":"/paper/2202.12837","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:407093a8da18e132f8869ee11e544edd3da45ea70d38d8de34182a6129830240","observation_id":"b23baa7b-720e-4b2d-a48f-d37ff560db75","resolution":{"observed_at":"2026-08-09T20:51:50.000257Z","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":"2026-08-09T20:51:50.004091Z","title":"R \\'e nyi differential privacy","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.004091Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:b90ff860a970724235926e8ea629a4fdc9c4617949f217add8fae28f380d7bf9","observation_id":"64896e64-6e73-4177-850b-12e680137dde","resolution":{"observed_at":"2026-08-09T20:51:50.004091Z","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":"2026-08-09T20:51:50.008023Z","title":"Smooth sensitivity and sampling in private data analysis","venue":null,"work_id":null,"year":2007},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":26,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.008023Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:d043e67a1073b628bfffe3da2d0927f59b40023ab7b60575355a50887ef1213d","observation_id":"a72d5904-312e-49af-b510-6a8bd14def6f","resolution":{"observed_at":"2026-08-09T20:51:50.008023Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1706.09254","last_updated":"2017-07-06T11:51:09Z","snapshot_observed_at":"2026-08-10T06:43:44.529765Z","submitted_at":"2017-06-28T12:38:53Z","title":"The E2E Dataset: New Challenges For End-to-End Generation","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1706.09254","snapshot_observed_at":"2026-08-09T20:51:50.012550Z","title":"The e2e dataset: New challenges for end-to-end generation","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":27,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.012550Z"},"links":{"cited_paper":"/paper/1706.09254","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:1b580069d8ddcba15e7b52f665decc67d931a96eb56f05c6af5f5b081d5dcd71","observation_id":"6ca606dd-87be-48b6-a173-ff94c85eb830","resolution":{"observed_at":"2026-08-09T20:51:50.012550Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1610.05755","last_updated":"2017-03-03T18:56:43Z","snapshot_observed_at":"2026-07-06T05:15:05.941444Z","submitted_at":"2016-10-18T19:37:37Z","title":"Semi-supervised Knowledge Transfer for Deep Learning from Private Training Data","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1610.05755","snapshot_observed_at":"2026-08-09T20:51:50.016424Z","title":"Semi-supervised knowledge transfer for deep learning from private training data","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":28,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.016424Z"},"links":{"cited_paper":"/paper/1610.05755","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:0b2edd139d7d6bef9778c723d2da1ede89f1d4f26c5093c3b0594b33556d276f","observation_id":"4c19ae90-d1c4-4a80-b5e6-3264d10035a3","resolution":{"observed_at":"2026-08-09T20:51:50.016424Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1802.08908","last_updated":"2018-02-24T20:39:51Z","snapshot_observed_at":"2026-07-06T06:25:11.708379Z","submitted_at":"2018-02-24T20:39:51Z","title":"Scalable Private Learning with PATE","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1802.08908","snapshot_observed_at":"2026-08-09T20:51:50.020032Z","title":"Scalable private learning with pate","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":29,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.020032Z"},"links":{"cited_paper":"/paper/1802.08908","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:2c3fa1c7ab7e4fd49dd9a920edf01e123f80305bd491cc4a66447120e55c5ffc","observation_id":"d14c54b9-16a9-4816-9d11-e4a2dc7c8355","resolution":{"observed_at":"2026-08-09T20:51:50.020032Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.15008","last_updated":"2023-05-24T10:48:05Z","snapshot_observed_at":"2026-07-06T15:32:25.931739Z","submitted_at":"2023-05-24T10:48:05Z","title":"Are Chatbots Ready for Privacy-Sensitive Applications? An Investigation into Input Regurgitation and Prompt-Induced Sanitization","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.15008","snapshot_observed_at":"2026-08-09T20:51:50.023455Z","title":"Are chatbots ready for privacy-sensitive applications? an investigation into input regurgitation and prompt-induced sanitization","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":30,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.023455Z"},"links":{"cited_paper":"/paper/2305.15008","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:92e975d272757144d54a369ababf458b6da53b13a270bdb7022e3b9f736d1793","observation_id":"b5dda8ba-3e12-4343-89a3-34b4f488d937","resolution":{"observed_at":"2026-08-09T20:51:50.023455Z","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":"2026-08-09T20:51:50.026754Z","title":"Language models are unsupervised multitask learners","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":31,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.026754Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:9ab7c9cf3bd30d4070afede840666f1cb1710daf0a77c8a1c3bacc9b13b37f33","observation_id":"083c4571-95e8-4a03-af42-ca3fd5975631","resolution":{"observed_at":"2026-08-09T20:51:50.026754Z","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":"2026-08-09T20:51:50.029811Z","title":"Membership inference attacks against machine learning models","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":32,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.029811Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:fce1b41bb3c1fed06a0d79435b517f1daed599f42d794192be94812418a9b3e2","observation_id":"e749f19a-149e-48dd-8a2f-544adb963195","resolution":{"observed_at":"2026-08-09T20:51:50.029811Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2210.00597","last_updated":"2022-10-26T16:13:14Z","snapshot_observed_at":"2026-08-08T15:59:14.914409Z","submitted_at":"2022-10-02T18:22:31Z","title":"Composition of Differential Privacy & Privacy Amplification by Subsampling","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2210.00597","snapshot_observed_at":"2026-08-09T20:51:50.032967Z","title":"Composition of differential privacy & privacy amplification by subsampling","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":33,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.032967Z"},"links":{"cited_paper":"/paper/2210.00597","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:0e70d154f6351ce3785be3f18662155daa18b4f0bad08155f0b1bc751f88c614","observation_id":"bff61a7e-38b0-4317-9ebc-76edb1f7f1d6","resolution":{"observed_at":"2026-08-09T20:51:50.032967Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2309.11765","last_updated":"2024-01-28T00:24:10Z","snapshot_observed_at":"2026-07-06T16:21:33.523806Z","submitted_at":"2023-09-21T03:59:00Z","title":"Privacy-Preserving In-Context Learning with Differentially Private Few-Shot Generation","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2309.11765","snapshot_observed_at":"2026-08-09T20:51:50.036302Z","title":"A., Manoel, A., Mireshghallah, F., Lin, Z., Gopi, S., Kulkarni, J., and Sim, R","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":34,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.036302Z"},"links":{"cited_paper":"/paper/2309.11765","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:1be8ed4b4f273dcb727f6de8da6fc4f635a27268e9795fb06e088113f04ff177","observation_id":"6f7dbfb2-a126-4553-b61b-3f1b9b55ea43","resolution":{"observed_at":"2026-08-09T20:51:50.036302Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2408.00118","last_updated":"2024-10-02T15:22:49Z","snapshot_observed_at":"2026-08-02T16:20:09.773989Z","submitted_at":"2024-07-31T19:13:07Z","title":"Gemma 2: Improving Open Language Models at a Practical Size","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2408.00118","snapshot_observed_at":"2026-08-09T20:51:50.039713Z","title":"G., Hardin, C., Bhupatiraju, S., Hussenot, L., Mesnard, T., Shahriari, B., Ram \\'e , A., et al","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":35,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.039713Z"},"links":{"cited_paper":"/paper/2408.00118","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:d0701b05c6ed8a7a5085c178df531f16ce9aecc7fe9dc2e925e86ce6b8186732","observation_id":"b60801c0-2b8d-481a-b3e0-03e70a93dc29","resolution":{"observed_at":"2026-08-09T20:51:50.039713Z","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":"2026-08-09T20:51:50.718092Z","title":"L., and He, H","venue":null,"work_id":"415bdd2a-62e8-4801-ac19-3b467804e6a2","year":2022},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":36,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.044395Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:cce6fd69ee43a6df3f5dd42332a211fb8e0b6b89dad697b00255c7753c39e519","observation_id":"588a4c89-d828-4cfc-bdab-712e4d472d24","resolution":{"observed_at":"2026-08-09T20:51:50.721173Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+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":"2026-08-09T20:51:50.047556Z","title":"and Harremos, P","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":37,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.047556Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:f43339c515f489cedec2f3282c84872ca23912a431b1f0684ee25927e73f90b3","observation_id":"1c1fb981-b1ca-42e9-9816-8aa0ef8ce5e7","resolution":{"observed_at":"2026-08-09T20:51:50.047556Z","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":"2026-08-09T20:51:50.700495Z","title":null,"venue":null,"work_id":"bc5cf625-377b-40b7-b947-08a04cd9720c","year":2000},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":38,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.050680Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:01c23aefcac37756c032a244050c4a2316613495b2dbd22b7627e4ac6eca35e0","observation_id":"319c101c-8b8c-45ad-afca-0f83625dc7cb","resolution":{"observed_at":"2026-08-09T20:51:50.703563Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+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":"2026-08-09T20:51:50.690419Z","title":"Decodingtrust: A comprehensive assessment of trustworthiness in gpt models","venue":null,"work_id":"a8f73dc1-f272-41e4-83ff-3d6be252bab0","year":2023},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":39,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.053947Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:bcd7914c80e2a251dc95f5df3af5185bbf0bbcfc3c19208c9ac4f0a69040cc06","observation_id":"20dcebce-9dc2-40b3-baa4-8022853f09e7","resolution":{"observed_at":"2026-08-09T20:51:50.693920Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+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":"2026-08-09T20:51:50.680206Z","title":"Privacy for free: Posterior sampling and stochastic gradient monte carlo","venue":null,"work_id":"f1874ba3-50ab-4dee-aa69-2760a71d6f97","year":2015},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":40,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.056776Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:bf90f917c80b07803e7ccd7c3718652914dccfc0e488e62af50cfa8dd730085c","observation_id":"9a1efcae-f6af-4b52-9383-f62f23e985d3","resolution":{"observed_at":"2026-08-09T20:51:50.683808Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+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":"2026-08-09T20:51:50.670055Z","title":null,"venue":null,"work_id":"e1777f54-18b6-4839-80da-aa46be6d4bbe","year":2019},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":41,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.060105Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:95b44029bb6d3946c3f78951a6df98ceb93e5480153b91ee85ecf0b33dbc2e4d","observation_id":"10c29113-b853-4bc6-9846-fe46aa9f2ba4","resolution":{"observed_at":"2026-08-09T20:51:50.673241Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2303.03846","last_updated":"2023-03-08T07:37:43Z","snapshot_observed_at":"2026-07-06T14:59:39.238136Z","submitted_at":"2023-03-07T12:24:17Z","title":"Larger language models do in-context learning differently","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2303.03846","snapshot_observed_at":"2026-08-09T20:51:50.063105Z","title":"Larger language models do in-context learning differently","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":42,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.063105Z"},"links":{"cited_paper":"/paper/2303.03846","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:95484b62f8d338590defe2940cba172459909c5fc5e7cb2330e1abe5ce4ee9d7","observation_id":"41ce6bf4-8be9-4249-b8ef-f2f69a95b4bb","resolution":{"observed_at":"2026-08-09T20:51:50.063105Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.01639","last_updated":"2023-09-30T12:33:13Z","snapshot_observed_at":"2026-07-06T15:22:28.700249Z","submitted_at":"2023-05-02T17:52:58Z","title":"Privacy-Preserving In-Context Learning for Large Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.01639","snapshot_observed_at":"2026-08-09T20:51:50.066321Z","title":"T., and Mittal, P","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":43,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.066321Z"},"links":{"cited_paper":"/paper/2305.01639","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:d4b3520d10d49532cfd18fdd4fad177f759a0745c9e33ba8df1dcfdb26f77e94","observation_id":"e12c0c87-814e-4150-94ea-a3d6e59210d2","resolution":{"observed_at":"2026-08-09T20:51:50.066321Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2403.01749","last_updated":"2024-07-23T19:19:02Z","snapshot_observed_at":"2026-07-06T17:38:59.580242Z","submitted_at":"2024-03-04T05:57:50Z","title":"Differentially Private Synthetic Data via Foundation Model APIs 2: Text","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.01749","snapshot_observed_at":"2026-08-09T20:51:50.069981Z","title":"A., Nori, H., Jiang, H., Zhang, H., Lee, Y","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":44,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.069981Z"},"links":{"cited_paper":"/paper/2403.01749","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:ff2ed8a42969cfd0ef497d4e02fb2e25073a61d3be19f994925dc03afdde008a","observation_id":"6e061895-0a09-41a9-bdd5-101b9f5582d6","resolution":{"observed_at":"2026-08-09T20:51:50.069981Z","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":"2312.14335","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-09T20:51:50.388954Z","title":"Context-aware decoding reduces hallucination in query-focused summarization","venue":null,"work_id":"4e230bf6-745b-4471-81da-9185f6f23250","year":2023},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":45,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.073214Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:2645e4ba288994ba0d7ad79adf44ae12e96ed5d3786bd3af1c920d3990b8fbd2","observation_id":"44ba2702-304b-41ee-9e7a-39a41d22ebda","resolution":{"observed_at":"2026-08-09T20:51:50.394144Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+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":"2026-08-09T20:51:50.659652Z","title":"Privacy risk in machine learning: Analyzing the connection to overfitting","venue":null,"work_id":"a4ec9844-f511-4236-89b7-5a84caa8b440","year":2018},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":46,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.076152Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:9131d5ec75cde6ce7b37fbe0a80690fff748df2d27c7c03f21c5969fedbaff84","observation_id":"1017b0a2-8886-49c5-8fb5-47385b984fff","resolution":{"observed_at":"2026-08-09T20:51:50.663130Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1904.09675","last_updated":"2020-02-24T18:59:28Z","snapshot_observed_at":"2026-07-29T15:42:51.774083Z","submitted_at":"2019-04-21T23:08:53Z","title":"BERTScore: Evaluating Text Generation with BERT","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1904.09675","snapshot_observed_at":"2026-08-09T20:51:50.079244Z","title":"Q., and Artzi, Y","venue":null,"work_id":null,"year":1904},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":47,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.079244Z"},"links":{"cited_paper":"/paper/1904.09675","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:9415e24862933ff8a42da6daadba3a2a372b31b8bc336c96d136ce8d41f528c4","observation_id":"64c31f16-3636-4dd7-86e1-38639dfbc252","resolution":{"observed_at":"2026-08-09T20:51:50.079244Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1703.10931","last_updated":"2017-07-16T02:28:14Z","snapshot_observed_at":"2026-08-03T20:30:01.994467Z","submitted_at":"2017-03-31T15:05:45Z","title":"Sentence Simplification with Deep Reinforcement Learning","version":2},"cited_work":{"arxiv_id":"1703.10931","doi":null,"metadata_source":"pith","pith_arxiv_id":"1703.10931","snapshot_observed_at":"2026-08-09T20:51:50.130788Z","title":"Sentence Simplification with Deep Reinforcement Learning","venue":"cs.CL","work_id":"6aae4abb-cb7c-4474-ace4-dce9f2fc6a5b","year":2017},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":48,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.083012Z"},"links":{"cited_paper":"/paper/1703.10931","citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:e4b804f21ddd1ae8d84784cf8a27be6632b57f9f8d2050d9c900817e892185d4","observation_id":"66448881-1e02-4b6b-a7c9-17b2f172b21c","resolution":{"observed_at":"2026-08-09T20:51:50.136676Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+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":"2026-08-09T20:51:50.086716Z","title":"Character-level convolutional networks for text classification","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":49,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.086716Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:b4a644a914ab661859589a744bfcbcd95cf4f148fa3d19fdd32725cc8ae1711a","observation_id":"fd8e435b-ce0a-4183-8575-4ca8f188f370","resolution":{"observed_at":"2026-08-09T20:51:50.086716Z","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":"2026-08-09T20:51:50.641697Z","title":"Calibrate before use: Improving few-shot performance of language models","venue":null,"work_id":"2759ba66-5489-40ed-b50a-2d176142de41","year":2021},"citing_paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs","version":1},"reference_index":50,"source":"arxiv_source","source_observed_at":"2026-08-09T20:51:50.089984Z"},"links":{"citing_paper":"/paper/2501.19287"},"observation_digest":"sha256:7948240668b4c76beada0efe68da86346502157f92bb5fb438d1529f56803b1f","observation_id":"4a758ab8-32b2-4ca4-b1bf-851d0c5c4396","resolution":{"observed_at":"2026-08-09T20:51:50.645821Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2501.19287","last_updated":"2025-01-31T16:48:38Z","latest_version":1,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-10T04:23:33.503634Z","submitted_at":"2025-01-31T16:48:38Z","title":"Differentially Private In-context Learning via Sampling Few-shot Mixed with Zero-shot Outputs"},"reference_resolution":{"displayed":50,"state_counts":{"malformed_identifier":0,"metadata_mismatch":1,"parse_uncertain":0,"unresolved":37,"verified_exact":2,"verified_fuzzy":10},"total_outbound_references":50},"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":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"thesis":"As of 10 August 2026, this Paper Citation Record lists 50 of 50 outbound references and 0 inbound Pith citation observations for arXiv:2501.19287."}