{"as_of":"2026-08-24T03:26:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:e7a30420ddecd02db26372a029e79ea57b225d229040d99c0cfda6586e7af4db","coverage":[{"denominator":58,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":58,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-15T18:20:44.518703Z","state":"measured"},{"denominator":62,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":62,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-23T06:30:58.430688+00:00","state":"measured"},{"denominator":4,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":4,"source":"paper_references, paper_reference_links","source_observed_at":"2026-06-27T16:49:14.243931Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"arxiv_reference","source_observed_at":"2026-07-04T20:50:11.125496Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"cited_work":{"arxiv_id":"2507.18302","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2507.18302","snapshot_observed_at":"2026-07-04T20:50:11.125496Z","title":"Lora-leak: Membership inference at- tacks against lora fine-tuned language models","venue":null,"work_id":"cd9872c9-5626-426f-937a-759d7c9aef98","year":2025},"citing_paper":{"arxiv_id":"2511.14045","last_updated":"2026-05-09T12:37:58Z","snapshot_observed_at":"2026-08-11T14:19:04.295776Z","submitted_at":"2025-11-18T01:51:34Z","title":"Auditing Data Membership in Reinforcement Learning With Verifiable Rewards","version":2},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-05-17T21:37:54.702010Z"},"links":{"cited_paper":"/paper/2507.18302","citing_paper":"/paper/2511.14045"},"observation_digest":"sha256:97f954cb38db4a1364ab3e0a41c8578c3a61f532dafd7938fda33ce812f081c3","observation_id":"ed0c2e36-6812-4b10-baed-39fec7d18a7c","resolution":{"observed_at":"2026-05-17T21:40:17.670528Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"cited_work":{"arxiv_id":"2507.18302","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2507.18302","snapshot_observed_at":"2026-07-04T20:50:11.125496Z","title":"Lora-leak: Membership inference at- tacks against lora fine-tuned language models","venue":null,"work_id":"cd9872c9-5626-426f-937a-759d7c9aef98","year":2025},"citing_paper":{"arxiv_id":"2604.21905","last_updated":"2026-04-23T17:50:23Z","snapshot_observed_at":"2026-07-06T23:08:23.939574Z","submitted_at":"2026-04-23T17:50:23Z","title":"Low-Rank Adaptation Redux for Large Models","version":1},"reference_index":152,"source":"pdf_text","source_observed_at":"2026-05-09T21:48:48.992712Z"},"links":{"cited_paper":"/paper/2507.18302","citing_paper":"/paper/2604.21905"},"observation_digest":"sha256:013d3eb741fff616e2a90e22c24d152466cece8af5bc388c56d6ecb244bdffdc","observation_id":"65820145-dd6b-413f-bd2c-f1b089f19195","resolution":{"observed_at":"2026-05-11T14:26:03.830820Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"cited_work":{"arxiv_id":"2507.18302","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2507.18302","snapshot_observed_at":"2026-07-04T20:50:11.125496Z","title":"Lora-leak: Membership inference at- tacks against lora fine-tuned language models","venue":null,"work_id":"cd9872c9-5626-426f-937a-759d7c9aef98","year":2025},"citing_paper":{"arxiv_id":"2606.09038","last_updated":"2026-06-08T05:10:05Z","snapshot_observed_at":"2026-08-16T14:05:07.005152Z","submitted_at":"2026-06-08T05:10:05Z","title":"Personalization Meets Safety:Mechanisms,Risks,and Mitigations in Personalized LLMs","version":1},"reference_index":141,"source":"pdf_text","source_observed_at":"2026-06-27T16:49:14.243931Z"},"links":{"cited_paper":"/paper/2507.18302","citing_paper":"/paper/2606.09038"},"observation_digest":"sha256:cfaf1c8e153a57c88ea91c67f15a5fcd6f179890c2de87851e05cb4b1dd4903d","observation_id":"4ef54cf3-e6a0-46c1-9102-2b036de5cbb0","resolution":{"observed_at":"2026-07-03T01:07:30.242521Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"cited_work":{"arxiv_id":"2507.18302","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2507.18302","snapshot_observed_at":"2026-07-04T20:50:11.125496Z","title":"Lora-leak: Membership inference at- tacks against lora fine-tuned language models","venue":null,"work_id":"cd9872c9-5626-426f-937a-759d7c9aef98","year":2025},"citing_paper":{"arxiv_id":"2606.26021","last_updated":"2026-06-24T16:42:21Z","snapshot_observed_at":"2026-08-10T18:30:05.673725Z","submitted_at":"2026-06-24T16:42:21Z","title":"Privacy Vulnerabilities of Attention Layers in Tabular Foundation Models and Protection of High-Risk Queries","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-06-25T19:36:48.784638Z"},"links":{"cited_paper":"/paper/2507.18302","citing_paper":"/paper/2606.26021"},"observation_digest":"sha256:e178d4b9cb854ad60a587575c9395b0bd7dca8ebe9120df9caa1552e45c4628d","observation_id":"6c45d6aa-1e6f-42ce-9339-3f75e635e9e0","resolution":{"observed_at":"2026-07-04T20:50:11.127675Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+00:00","source":"retraction_watch"}],"state":"measured"}}],"links":{"evidence":"/evidence","html":"/paper/2507.18302/citation-record","integrity":"/paper/2507.18302/integrity","json":"/paper/2507.18302/citation-record.json","paper":"/paper/2507.18302"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-15T18:20:45.121176Z","title":"Adapting large language models via reading comprehension,","venue":null,"work_id":"18f216db-3bb1-4791-a54e-b4a1fab6d2f8","year":2024},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.326655Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:e40003db95bac42d969d6402795c63301aad3a42ca9cdc6906d25bca3dd66069","observation_id":"9d86be8d-d4d9-46ff-9db3-389b26816e35","resolution":{"observed_at":"2026-08-15T18:20:45.124731Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:45.109558Z","title":"AstroLLaMA: Towards specialized foundation models in astronomy,","venue":null,"work_id":"dd86d9f3-be96-40cc-adc8-0023bb079e70","year":2023},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.331344Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:924af160a2924aa5c58a581225f7466c760869950635ccb38e2781096330d882","observation_id":"0e140187-cf90-4458-b46e-b0cde6c5a8eb","resolution":{"observed_at":"2026-08-15T18:20:45.113032Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.335136Z","title":"Chatgpt,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.335136Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:451c0e43b112f1a37826d7aced019ce4b03e18674235b8bbd06ad68292d055fb","observation_id":"6374a8a1-7680-430b-a5c6-5a1d0a4f8d67","resolution":{"observed_at":"2026-08-15T18:20:44.335136Z","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":"2026-08-15T18:20:44.339528Z","title":"Llama 2: Open foundation and fine-tuned chat models,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.339528Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:08cb07e205c19b4e5942785533c0d267cb8e911ad87b38f771592c2599d19732","observation_id":"daa94060-335c-4297-9758-d2b403e35055","resolution":{"observed_at":"2026-08-15T18:20:44.339528Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2308.12950","last_updated":"2024-01-31T19:47:26Z","snapshot_observed_at":"2026-08-18T03:59:39.242039Z","submitted_at":"2023-08-24T17:39:13Z","title":"Code Llama: Open Foundation Models for Code","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2308.12950","snapshot_observed_at":"2026-08-15T18:20:44.342748Z","title":"Code llama: Open foundation models for code,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.342748Z"},"links":{"cited_paper":"/paper/2308.12950","citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:4597f21791486d36c1bb3a21e370d318372b670c15d2dafee7b5be0174cb5d8f","observation_id":"394a6a4f-59f2-4ad7-a1a0-3136de272e5f","resolution":{"observed_at":"2026-08-15T18:20:44.342748Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2309.06126","last_updated":"2023-09-12T11:02:27Z","snapshot_observed_at":"2026-08-20T06:21:48.163248Z","submitted_at":"2023-09-12T11:02:27Z","title":"AstroLLaMA: Towards Specialized Foundation Models in Astronomy","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2309.06126","snapshot_observed_at":"2026-08-15T18:20:44.347163Z","title":"Astrollama: Towards specialized foundation models in astronomy,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.347163Z"},"links":{"cited_paper":"/paper/2309.06126","citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:159c658d9f3a58cc118f56d937dcc8a51ae836715871b2453d7fedbb371894d3","observation_id":"d4298a12-f5f4-40aa-9368-bf64eb8330c9","resolution":{"observed_at":"2026-08-15T18:20:44.347163Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2403.13372","last_updated":"2024-06-27T22:44:48Z","snapshot_observed_at":"2026-08-17T02:51:06.474773Z","submitted_at":"2024-03-20T08:08:54Z","title":"LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.13372","snapshot_observed_at":"2026-08-15T18:20:44.351563Z","title":"Llamafactory: Unified efficient fine-tuning of 100+ language models,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.351563Z"},"links":{"cited_paper":"/paper/2403.13372","citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:5ff474eeed64dc199a0041db3a5d62c5857b974fb7fd3e5e6d456c08b44053fd","observation_id":"6e2d20f8-55c8-4443-be3d-7aaaabae8bdd","resolution":{"observed_at":"2026-08-15T18:20:44.351563Z","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":"2026-08-15T18:20:44.354967Z","title":"Peft: State-of-the-art parameter-efficient fine-tuning methods,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.354967Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:f0863638a8a3fdde369f798b8008f3d005d4dc4043ca3725083237a3575f5d7a","observation_id":"22dd9660-3c89-491f-b22b-963cc80fefad","resolution":{"observed_at":"2026-08-15T18:20:44.354967Z","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":"2026-08-15T18:20:44.358733Z","title":"LoRA: Low-rank adaptation of large language models,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.358733Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:9e5c7c1736a83758ab224afdaa6723b499f52645642d86022088160ca42ef012","observation_id":"66314519-ed52-4733-bb72-178ea0530f7a","resolution":{"observed_at":"2026-08-15T18:20:44.358733Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2312.00374","last_updated":"2024-09-11T12:48:42Z","snapshot_observed_at":"2026-08-18T18:57:32.884383Z","submitted_at":"2023-12-01T06:36:17Z","title":"The Philosopher's Stone: Trojaning Plugins of Large Language Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2312.00374","snapshot_observed_at":"2026-08-15T18:20:44.362089Z","title":"The philosopher’s stone: Trojaning plugins of large language models,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.362089Z"},"links":{"cited_paper":"/paper/2312.00374","citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:c8371741831356299ac49a6c1f6cd15d856aee384abd08a841c50a129933a3a3","observation_id":"219a856b-0bcd-45f8-b50f-3941cb94833b","resolution":{"observed_at":"2026-08-15T18:20:44.362089Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.14314","last_updated":"2023-05-23T17:50:33Z","snapshot_observed_at":"2026-08-20T16:10:20.713555Z","submitted_at":"2023-05-23T17:50:33Z","title":"QLoRA: Efficient Finetuning of Quantized LLMs","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.14314","snapshot_observed_at":"2026-08-15T18:20:44.365872Z","title":"Qlora: Efficient finetuning of quantized llms,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.365872Z"},"links":{"cited_paper":"/paper/2305.14314","citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:25fa7c2ac2fbe225155b193e829dba452878c9e1762def747997609d366f8d6d","observation_id":"7b291545-8067-4400-81e3-e6de9450506a","resolution":{"observed_at":"2026-08-15T18:20:44.365872Z","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":"2026-08-15T18:20:44.369551Z","title":"Membership inference attacks from first principles,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.369551Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:fceae44040d07a883179b7da98c91e4de8b7bc417f393021b21dcd9ff99549a7","observation_id":"5091b6ca-73f0-4106-ad92-df63795b112d","resolution":{"observed_at":"2026-08-15T18:20:44.369551Z","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":"2026-08-15T18:20:45.065108Z","title":"Last one standing: A comparative analysis of security and privacy of soft prompt tuning, lora, and in-context learning,","venue":null,"work_id":"4a8d0fec-5c8e-453e-a2aa-4e01c7339114","year":2023},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.372576Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:9ff2cac438c6277649a1d3cc52407c1a3215e0979fc52959ef0619adfee14e70","observation_id":"598ff9a6-21ad-4fde-811a-288b03b04ca4","resolution":{"observed_at":"2026-08-15T18:20:45.069678Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:45.053920Z","title":"Precurious: How innocent pre-trained language models turn into privacy traps,","venue":null,"work_id":"af9db87d-0383-41ef-9fef-8d70fe3aa5f9","year":2024},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.375253Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:dabeb0cbd8310e02fabdb0c7d9fb9b719905f942cc0d2a8dd0b0e85ebb54e313","observation_id":"0dc1e0b5-6ded-4273-9199-95317a9667dd","resolution":{"observed_at":"2026-08-15T18:20:45.057427Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.377940Z","title":"Character-level convolutional networks for text classification,","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.377940Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:5239d300169226443a3981b12e77bbbc1fe78b7a99b56326ff5a0308a0c23097","observation_id":"b8dacc21-ee3a-4e80-8397-975efdf45d8f","resolution":{"observed_at":"2026-08-15T18:20:44.377940Z","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":"2026-08-15T18:20:45.036741Z","title":"Openassistant conversations - democra- tizing large language model alignment,","venue":null,"work_id":"3df5f549-936b-4abc-9daf-ccf141983188","year":2024},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.380818Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:c99a77ca61a7ff76c4223f6d2c647eb995b695f008a1e4684143a3bfca778676","observation_id":"1b57c2df-f070-4f61-807c-d5c5d5c70b0a","resolution":{"observed_at":"2026-08-15T18:20:45.040864Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2009.13081","last_updated":"2020-09-28T05:07:51Z","snapshot_observed_at":"2026-08-19T11:12:45.951488Z","submitted_at":"2020-09-28T05:07:51Z","title":"What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2009.13081","snapshot_observed_at":"2026-08-15T18:20:44.383247Z","title":"What disease does this patient have? a large-scale open domain question answering dataset from medical exams,","venue":null,"work_id":null,"year":2009},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.383247Z"},"links":{"cited_paper":"/paper/2009.13081","citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:bb0b6e7faa7ed26ee85e27af368c9440fc44fe18b7645390a4bf8bac89e1a009","observation_id":"02a3889b-4302-4928-a701-79ab504aac3c","resolution":{"observed_at":"2026-08-15T18:20:44.383247Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2104.08305","last_updated":"2021-04-16T18:29:58Z","snapshot_observed_at":"2026-08-18T19:05:06.003659Z","submitted_at":"2021-04-16T18:29:58Z","title":"Membership Inference Attack Susceptibility of Clinical Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2104.08305","snapshot_observed_at":"2026-08-15T18:20:44.386267Z","title":"Membership inference attack susceptibility of clinical language models,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.386267Z"},"links":{"cited_paper":"/paper/2104.08305","citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:af74e4a8d800748d4ce90c3c96b032af95d3e9ff4f8c98310c69dea1bd10566a","observation_id":"6bbfa29f-679f-4868-b35b-45f2fa72831d","resolution":{"observed_at":"2026-08-15T18:20:44.386267Z","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":"2026-08-15T18:20:45.023438Z","title":"Quantifying privacy risks of masked language models using membership inference attacks,","venue":null,"work_id":"12634857-33de-4731-8a9f-3042e34e3599","year":2022},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.389546Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:0db94d9810a57f27cc81eb57be86324ad97d51dc2231ac888cfdffed7466991f","observation_id":"e16b6913-f06d-4884-87a2-182061a93837","resolution":{"observed_at":"2026-08-15T18:20:45.027544Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.392617Z","title":"Extracting training data from large language models,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.392617Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:49f7a41ded70f854e6a272cb819f6461c7203da93b953a2e06b6ce3fccac5f63","observation_id":"83b881b1-2c46-4cfd-9f62-dab592240804","resolution":{"observed_at":"2026-08-15T18:20:44.392617Z","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":"2026-08-15T18:20:45.003446Z","title":"Membership inference attacks against language models via neighbourhood comparison,","venue":null,"work_id":"26908f1a-b453-4959-8dee-588a3a492281","year":2023},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.396280Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:3458103cafe8f2fef2181360d18dfff6f450fc67a7edfa93f87e865fa8a9f063","observation_id":"1b5baa0e-ffde-4d11-9e38-560fa43041c8","resolution":{"observed_at":"2026-08-15T18:20:45.008950Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.993772Z","title":"Membership inference attacks against fine-tuned large language models via self-prompt calibration,","venue":null,"work_id":"b61c2789-7acf-4757-a4ae-8789d7f3060a","year":2024},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.400393Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:c0087f4a4ab5b704945061876b7f3a0e9c663ee0533c20bb77bb8d326835bf54","observation_id":"a162bb96-99db-43a5-99c3-9bb6c9a4d731","resolution":{"observed_at":"2026-08-15T18:20:44.996548Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.983949Z","title":"MoPe: Model perturbation based privacy attacks on language models,","venue":null,"work_id":"1bcab303-b8e8-4f10-b1e1-c4966fcb9a19","year":2023},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.403746Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:fe0a6b6b9af13a6c291be545d7389a0ca7cf14ae1d54025a5698ad4e5926ed24","observation_id":"a3a7e0f2-ac1b-40c9-943a-c84b59bf27f4","resolution":{"observed_at":"2026-08-15T18:20:44.987050Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.407227Z","title":"Detecting pretraining data from large language models,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.407227Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:17017f822474c532a860a7dd2e4aae837a622ce7340fd20ccc2da37369d8084c","observation_id":"b1ff11f5-eebb-47ca-9ea7-7bfa5507c0c8","resolution":{"observed_at":"2026-08-15T18:20:44.407227Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2404.02936","last_updated":"2025-02-12T04:41:34Z","snapshot_observed_at":"2026-08-18T19:04:38.405489Z","submitted_at":"2024-04-03T04:25:01Z","title":"Min-K%++: Improved Baseline for Detecting Pre-Training Data from Large Language Models","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2404.02936","snapshot_observed_at":"2026-08-15T18:20:44.411767Z","title":"Min-k%++: Improved baseline for detecting pre-training data from large language models,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.411767Z"},"links":{"cited_paper":"/paper/2404.02936","citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:f9e76370b060bbe7d0bd0b821c971fb5cd2d69f651042b1e6f223b1e25b20141","observation_id":"a5772ff0-aba3-4206-bc06-6849dea5583f","resolution":{"observed_at":"2026-08-15T18:20:44.411767Z","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":"2026-08-15T18:20:44.967516Z","title":"Pandora’s white-box: Increased training data leakage in open llms,","venue":null,"work_id":"201d1e42-97bc-4662-aaf4-0c39099af283","year":2024},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.415135Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:a7345d85c8b2e6698ac352d1954a8ad8f6b2c779aaa63266fb4241f398e1bef4","observation_id":"06f6ebf4-83cd-48e1-8da9-fba2d4318c7d","resolution":{"observed_at":"2026-08-15T18:20:44.971353Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.955711Z","title":"Membership Infer- ence Attacks Against Machine Learning Models,","venue":null,"work_id":"efb9be32-9d96-4ad9-bdd2-46a8ab7b5f40","year":2017},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.418311Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:994b56b41e43227c4fe20e44dc6139c99ff668d5a11a7dabdebb49c213fe5019","observation_id":"a687b62a-eb36-4575-b377-6670f3a80a24","resolution":{"observed_at":"2026-08-15T18:20:44.959567Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.945551Z","title":"Comprehensive Privacy Anal- ysis of Deep Learning: Passive and Active White-box Inference Attacks against Centralized and Federated Learning,","venue":null,"work_id":"733cf95b-4a48-4bf2-95bf-2879587c3fe9","year":2019},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.422317Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:b758e4b8d21651a297f2ccd558a77716370378c554ffa4ca2b87298d5d3128d4","observation_id":"96a9eec8-ddf0-4467-9188-894c1b514f37","resolution":{"observed_at":"2026-08-15T18:20:44.948923Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.935095Z","title":"Stolen memories: Leveraging model memorization for calibrated White-Box membership inference,","venue":null,"work_id":"ab169776-61e2-4bec-82f6-87ffb4192c28","year":2020},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.425326Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:c9b628f824d21c1414b9f82aebdf673e7046db59aaa265bf99a4676f52ecaa10","observation_id":"0e8ecf64-e3b4-413a-ac92-96cd9acec3ca","resolution":{"observed_at":"2026-08-15T18:20:44.939228Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.428177Z","title":"Privacy risks of securing machine learning models against adversarial examples,","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.428177Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:f1a19d9dfe85f6215e91bcea73eacc07bc23001fbcb15b3c84ac4aa026747693","observation_id":"4da78707-4f4d-4391-9777-6fe8968eb15a","resolution":{"observed_at":"2026-08-15T18:20:44.428177Z","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":"2026-08-15T18:20:44.431687Z","title":"Privacy risk in machine learning: Analyzing the connection to overfitting,","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.431687Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:0b86012ff76382579e74578e1f9eb50016c3271a4d574c116061b33006755911","observation_id":"044924be-112c-4fb5-96ce-0e00dae59998","resolution":{"observed_at":"2026-08-15T18:20:44.431687Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2009.05669","last_updated":"2020-09-11T21:53:50Z","snapshot_observed_at":"2026-08-19T17:43:27.185968Z","submitted_at":"2020-09-11T21:53:50Z","title":"Quantifying Membership Inference Vulnerability via Generalization Gap and Other Model Metrics","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2009.05669","snapshot_observed_at":"2026-08-15T18:20:44.434642Z","title":"Quantifying membership inference vulnerability via generalization gap and other model metrics,","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.434642Z"},"links":{"cited_paper":"/paper/2009.05669","citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:294f105d96c1d58e9c03a69c6a8718c3cc162bb16036713eca88202164d51b17","observation_id":"1fd2e2af-6000-42cd-9c37-e1011d2df828","resolution":{"observed_at":"2026-08-15T18:20:44.434642Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2406.17975","last_updated":"2025-03-07T16:30:07Z","snapshot_observed_at":"2026-08-18T19:05:07.850601Z","submitted_at":"2024-06-25T23:12:07Z","title":"SoK: Membership Inference Attacks on LLMs are Rushing Nowhere (and How to Fix It)","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2406.17975","snapshot_observed_at":"2026-08-15T18:20:44.438196Z","title":"Sok: Membership inference attacks on llms are JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2021 13 rushing nowhere (and how to fix it),","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.438196Z"},"links":{"cited_paper":"/paper/2406.17975","citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:c877754fe65f31fae1a37763606e173c3da2cdfea4f49adedec9bb940aa822fc","observation_id":"3a3bcc6b-dfcf-49d3-8bfd-dec43300e2f9","resolution":{"observed_at":"2026-08-15T18:20:44.438196Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2406.16201","last_updated":"2025-03-30T08:39:32Z","snapshot_observed_at":"2026-08-20T03:43:15.680794Z","submitted_at":"2024-06-23T19:40:11Z","title":"Blind Baselines Beat Membership Inference Attacks for Foundation Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2406.16201","snapshot_observed_at":"2026-08-15T18:20:44.441475Z","title":"Blind baselines beat membership inference attacks for foundation models,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.441475Z"},"links":{"cited_paper":"/paper/2406.16201","citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:e6cc903b2c6d6c39e3388538d4806f890459b4f24fc9eed7f384cffb5173b016","observation_id":"1e627200-e1ba-4339-82b0-69d1b15eab00","resolution":{"observed_at":"2026-08-15T18:20:44.441475Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.07841","last_updated":"2024-09-16T13:18:23Z","snapshot_observed_at":"2026-08-16T14:19:15.347006Z","submitted_at":"2024-02-12T17:52:05Z","title":"Do Membership Inference Attacks Work on Large Language Models?","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.07841","snapshot_observed_at":"2026-08-15T18:20:44.444846Z","title":"Do membership inference attacks work on large language models?","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.444846Z"},"links":{"cited_paper":"/paper/2402.07841","citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:2d2d6ea3e7e44956c97c070c83d0970afc3a13a81ae6e92c4ceb0f514d9bdbab","observation_id":"f9b6cf5c-4721-4837-950f-20132d39562d","resolution":{"observed_at":"2026-08-15T18:20:44.444846Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2205.12506","last_updated":"2022-11-04T03:13:41Z","snapshot_observed_at":"2026-08-19T17:26:07.443170Z","submitted_at":"2022-05-25T05:49:31Z","title":"Memorization in NLP Fine-tuning Methods","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2205.12506","snapshot_observed_at":"2026-08-15T18:20:44.449589Z","title":"Memorization in nlp fine-tuning methods,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.449589Z"},"links":{"cited_paper":"/paper/2205.12506","citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:42b925fc2e1d8ce8104342d96375488d453d9ed540aadae8268ef8d23c0706f9","observation_id":"a5339899-e61b-4d10-a234-143a8cde3ebe","resolution":{"observed_at":"2026-08-15T18:20:44.449589Z","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":"2026-08-14T18:16:28.847993Z","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":"2026-08-15T18:20:44.453036Z","title":"Bert: Pre-training of deep bidirectional transformers for language understanding,","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.453036Z"},"links":{"cited_paper":"/paper/1810.04805","citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:2c2d62af9aa3084f63a76a8bea61d27964c201611812c7e963da29d5bab02bcf","observation_id":"919c6002-5c5d-4a68-848d-cf5c05b19f68","resolution":{"observed_at":"2026-08-15T18:20:44.453036Z","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":"2026-08-15T18:20:44.456606Z","title":"Exploring the limits of transfer learning with a unified text-to-text transformer,","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.456606Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:3bd629ed775e1c670512a6a659c960ced65c1e418c7d3494d78d324dcdb373e0","observation_id":"120bf156-69b2-45eb-b2c0-d97657b50bb2","resolution":{"observed_at":"2026-08-15T18:20:44.456606Z","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":"2026-08-15T18:20:44.460315Z","title":"Language models are unsupervised multitask learners,","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.460315Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:e0ebbe29326ad0f3bf9010756836da4c48ed163eb6fe7aeae5a0faa89506fa38","observation_id":"d4b0faa8-8087-4abb-92f2-c72d3f3b85b6","resolution":{"observed_at":"2026-08-15T18:20:44.460315Z","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":"2026-08-15T18:20:44.463286Z","title":"Pythia: A suite for analyzing large language models across training and scaling,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.463286Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:50eae5bd7e49535b4b9799084e6b00b773fc821c4b704295be2ba1666f834620","observation_id":"ac681358-a0fb-404c-aa1b-1266146b1a7e","resolution":{"observed_at":"2026-08-15T18:20:44.463286Z","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":"2026-08-15T18:20:44.902012Z","title":"Openassistant top-1 conversation threads,","venue":null,"work_id":"d408d0a3-9e0d-44a0-b69d-91f9e9491489","year":2023},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.467169Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:b5a30fa7508836903c4a5cf5e3f6c0a23468a0c4ce3da22af53176e2eb6d9777","observation_id":"66ae483d-ef14-4a07-b859-d43461b465b1","resolution":{"observed_at":"2026-08-15T18:20:44.905368Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.892066Z","title":"Chat markup language,","venue":null,"work_id":"84120f61-029a-42ad-9c41-2daccf80360a","year":2020},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.470658Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:ccbc73818e468134b311f9fb3831cd20b0f2e19861313fc9a78e59bede7c8dcd","observation_id":"20f94b7f-7f7b-4760-a253-67e6997d1a20","resolution":{"observed_at":"2026-08-15T18:20:44.895535Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.880663Z","title":"Tl;dr news dataset,","venue":null,"work_id":"6a318c3b-5c95-461d-a700-ecfcad20350a","year":2023},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.473529Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:01b1d28dd88f661b95c6c165f3b8f75e6dd0dc5318f957c36888c00d0040ce2c","observation_id":"59308650-d4e3-4f18-ad4d-225e88defc84","resolution":{"observed_at":"2026-08-15T18:20:44.883868Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.476479Z","title":"Dropout: A simple way to prevent neural networks from overfitting,","venue":null,"work_id":null,"year":1929},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.476479Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:ddebb2cf94d5b7e292088aed8b9a995ad493174a8bf766c212316f05a9e160de","observation_id":"c764e870-98b9-4ef1-b593-52ff0ec99f17","resolution":{"observed_at":"2026-08-15T18:20:44.476479Z","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":"2026-08-15T18:20:44.862568Z","title":"A simple weight decay can improve gener- alization,","venue":null,"work_id":"4d884a78-acab-4288-998c-a8d11cbed121","year":1991},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.479732Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:956c166ed9a140bea0ffa1da263f0c4445c36798f293ad3d71e1920dcdf4c83e","observation_id":"b1574acc-adaf-49a5-aede-786315810e93","resolution":{"observed_at":"2026-08-15T18:20:44.866439Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.482107Z","title":"Decoupled weight decay regularization,","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.482107Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:d8d24fc4344d5f93d301b220bc29482af4b49b047fce9d505d62c86caca3437b","observation_id":"81ecf22f-6a6b-47d3-aa27-67b230280caa","resolution":{"observed_at":"2026-08-15T18:20:44.482107Z","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":"2026-08-15T18:20:44.846491Z","title":"On the effectiveness of regulariza- tion against membership inference attacks,","venue":null,"work_id":"b1fdaecd-3a5c-4995-981c-f718cf7f8c98","year":2020},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.484642Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:3e6a79d33b195df15b5ee570d18c3d85c920bb12c8e6f4ee981fddff2397c7aa","observation_id":"9f64f8ee-7a91-426b-9054-97e6172e74c1","resolution":{"observed_at":"2026-08-15T18:20:44.849928Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.487127Z","title":"Calibrating noise to sensitivity in private data analysis,","venue":null,"work_id":null,"year":2006},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.487127Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:fdd0e1c1f19a57203ffe1de0c4adf954916d8b17d9d5343967ff1f857034af29","observation_id":"1221aa3c-4ac9-4ed9-aecb-d32d6438ce76","resolution":{"observed_at":"2026-08-15T18:20:44.487127Z","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":"2026-08-15T18:20:44.831653Z","title":"Differentially private fine-tuning of language models,","venue":null,"work_id":"c84f9554-6d03-4355-8c4f-002ccc68600d","year":2022},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.489596Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:37f876101c268426089313308ff01abd526cf184f729e68e9cc044f7fcff8801","observation_id":"e30d2880-ced6-4849-bf5e-266fb87c4079","resolution":{"observed_at":"2026-08-15T18:20:44.834800Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.820336Z","title":"dp-transformers: Training transformer models with differential privacy,","venue":null,"work_id":"12ec8224-f05e-42f2-b906-a88c9161bffa","year":2022},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.492062Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:502f998f8bdbde2258e3c131d76784f1edacb22c6a193ee19af9e88ee57eef95","observation_id":"7516f932-b85d-4204-bf30-e630e7924446","resolution":{"observed_at":"2026-08-15T18:20:44.824445Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.809919Z","title":"Openai api reference,","venue":null,"work_id":"e31c450a-e837-4402-a9f5-86c42ed6098f","year":2024},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.495120Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:e17995674f9165f6cbbd03072832233bb6aba52f774a26a357ef0c0bf4334575","observation_id":"e98e1476-3a69-4698-8574-fc0b4b3fadec","resolution":{"observed_at":"2026-08-15T18:20:44.813002Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.799445Z","title":"Hugging face api inference documentation,","venue":null,"work_id":"15c8c8f5-def2-417f-8310-4668dc175bb6","year":2024},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.498189Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:fdf6325eaacd164d534f57ad4838302dd5a0d687c9a32a1e63c29e7c01043f8e","observation_id":"04d51bb5-c98f-481a-a654-88613b2d3048","resolution":{"observed_at":"2026-08-15T18:20:44.803196Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.788916Z","title":"Label- Only Membership Inference Attacks,","venue":null,"work_id":"9b6ed61f-d1cb-4a52-a330-fc9b892ef461","year":2021},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.501156Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:3da3674532622ff858e80ff6ad385f38e8ade836dd18e1d9eed6fcfcc4885082","observation_id":"fcfaae4f-25d7-4c07-bb10-5aec87da947a","resolution":{"observed_at":"2026-08-15T18:20:44.792621Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.777551Z","title":"Membership Leakage in Label-Only Exposures,","venue":null,"work_id":"e3374271-c054-42a6-9923-5f6eed8d93f6","year":2021},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.504504Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:67030793fe267bb046c53d6463a42e166639110d261706d202fe2b53fd80ebdf","observation_id":"b8b6b759-f2a2-4477-89b3-d0501c6e0f56","resolution":{"observed_at":"2026-08-15T18:20:44.781280Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+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":"2026-08-15T18:20:44.767069Z","title":"Trl: Transformer reinforcement learning,","venue":null,"work_id":"8a50df02-d7c1-4601-85de-836015118031","year":2020},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.507795Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:6caaecdafed8d398d3aeeb7ded91d4f0b5c251b35e385f5bc2c0bae9fdf889f6","observation_id":"e249a577-478a-4f68-b1e8-b821d9639e8a","resolution":{"observed_at":"2026-08-15T18:20:44.770877Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2402.09353","last_updated":"2024-07-09T05:59:16Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-02-14T17:59:34Z","title":"DoRA: Weight-Decomposed Low-Rank Adaptation","version":6},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.09353","snapshot_observed_at":"2026-08-15T18:20:44.510905Z","title":"Dora: Weight-decomposed low-rank adaptation,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.510905Z"},"links":{"cited_paper":"/paper/2402.09353","citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:a65f4e237a38706524ef0c3aa5eb7b523c846fff84c0aaa64cc652ae8343e8ba","observation_id":"a16a664c-a975-45a1-86c1-a91e6c4bfa58","resolution":{"observed_at":"2026-08-15T18:20:44.510905Z","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":"2026-08-15T18:20:44.514315Z","title":"The power of scale for parameter-efficient prompt tuning,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.514315Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:6539c8a142f3f566821ed84a1b5028dd4888e49e792284f989463a2007a5ac45","observation_id":"82943747-7d62-4565-8cb2-de785283f2d9","resolution":{"observed_at":"2026-08-15T18:20:44.514315Z","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":"2026-08-15T18:20:44.746967Z","title":"Few-shot parameter-efficient fine-tuning is better and cheaper than in-context learning,","venue":null,"work_id":"d4b554e8-00a9-4acf-afbf-f5fd4c35f8b9","year":2022},"citing_paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models","version":1},"reference_index":58,"source":"pdf_text","source_observed_at":"2026-08-15T18:20:44.518703Z"},"links":{"citing_paper":"/paper/2507.18302"},"observation_digest":"sha256:1a0ac413c2382f270477b08f4311a861a0e18cbfc62431fb8c7372488f5cfa9f","observation_id":"507ecdf1-6f3b-434a-ae0b-90012b20bb43","resolution":{"observed_at":"2026-08-15T18:20:44.751846Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-23T06:30:58.430688+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2507.18302","last_updated":"2025-07-24T11:18:27Z","latest_version":1,"primary_category":"cs.CR","snapshot_observed_at":"2026-08-19T17:26:25.950003Z","submitted_at":"2025-07-24T11:18:27Z","title":"LoRA-Leak: Membership Inference Attacks Against LoRA Fine-tuned Language Models"},"reference_resolution":{"displayed":58,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":32,"verified_exact":0,"verified_fuzzy":26},"total_outbound_references":58},"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":"2026-08-23T06:30:58.430688+00:00","source":"crossref"},{"observed_at":"2026-08-23T06:30:53.778098+00:00","source":"retraction_watch"}],"thesis":"As of 24 August 2026, this Paper Citation Record lists 58 of 58 outbound references and 4 inbound Pith citation observations for arXiv:2507.18302."}