{"as_of":"2026-08-19T02:11:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:20b536f90408f4f59aa774830733a6867b5743fb52b437054de858ec96a76f52","coverage":[{"denominator":45,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":45,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-16T05:31:27.452275Z","state":"measured"},{"denominator":46,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":46,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-18T06:34:40.430872+00:00","state":"measured"},{"denominator":1,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":1,"source":"paper_references, paper_reference_links","source_observed_at":"2026-06-29T01:39:19.183198Z","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":"2026-07-01T18:45:58.992084Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"cited_work":{"arxiv_id":"2504.20570","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2504.20570","snapshot_observed_at":"2026-07-01T18:45:58.992084Z","title":"ReCIT: Reconstructing full private data from gradi- ent in parameter-efficient fine-tuning of large language models.arXiv preprint arXiv:2504.20570, 2025","venue":null,"work_id":"7d91bdb6-e8c7-4f65-a9c2-815d0cfca713","year":2025},"citing_paper":{"arxiv_id":"2606.27511","last_updated":"2026-06-25T19:53:37Z","snapshot_observed_at":"2026-08-18T03:01:30.692631Z","submitted_at":"2026-06-25T19:53:37Z","title":"When the Aggregator Cheats: Data-Free Backdoors in Federated LLM-based QA Systems","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-06-29T01:39:19.183198Z"},"links":{"cited_paper":"/paper/2504.20570","citing_paper":"/paper/2606.27511"},"observation_digest":"sha256:0d6d77573d0401c232ad8b36c02559033ee0b72e0b216446e6751c8476ba2b3f","observation_id":"bc853c95-9812-4170-adca-b26077590f1b","resolution":{"observed_at":"2026-07-01T18:45:58.994055Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}}],"links":{"evidence":"/evidence","html":"/paper/2504.20570/citation-record","integrity":"/paper/2504.20570/integrity","json":"/paper/2504.20570/citation-record.json","paper":"/paper/2504.20570"},"outbound":[{"citation":{"cited_paper":{"arxiv_id":"2303.08774","last_updated":"2024-03-04T06:01:33Z","snapshot_observed_at":"2026-08-17T09:58:46.058102Z","submitted_at":"2023-03-15T17:15:04Z","title":"GPT-4 Technical Report","version":6},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2303.08774","snapshot_observed_at":"2026-08-16T05:31:27.243720Z","title":"Gpt-4 technical report,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.243720Z"},"links":{"cited_paper":"/paper/2303.08774","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:baacebc321000d46067d4465f410d5246f49376a1ff2c7a95e8f25ebf2ac866c","observation_id":"5ac832ea-21af-4a70-85a6-8d7721f9e4a2","resolution":{"observed_at":"2026-08-16T05:31:27.243720Z","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":"2026-08-16T05:31:27.249363Z","title":"Attention is all you need,","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.249363Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:238b0febc4cbefb1bf82a298e20b8f2e39712d9e124a4db7327156872c904b1f","observation_id":"0c7964ef-7421-4708-b2e4-30215180974d","resolution":{"observed_at":"2026-08-16T05:31:27.249363Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2303.18223","last_updated":"2026-03-18T05:34:39Z","snapshot_observed_at":"2026-08-14T10:40:26.323157Z","submitted_at":"2023-03-31T17:28:46Z","title":"A Survey of Large Language Models","version":19},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2303.18223","snapshot_observed_at":"2026-08-16T05:31:27.254245Z","title":"A survey of large language models,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.254245Z"},"links":{"cited_paper":"/paper/2303.18223","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:4c6858b6fcb4b5a97af47f8f1a36c3cf3813e4df6d7a0a00678fc0170ef5db60","observation_id":"779f8801-81f8-4cc9-baef-35c55d33dc4a","resolution":{"observed_at":"2026-08-16T05:31:27.254245Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2001.08361","last_updated":"2020-01-23T03:59:20Z","snapshot_observed_at":"2026-08-13T17:41:53.092611Z","submitted_at":"2020-01-23T03:59:20Z","title":"Scaling Laws for Neural Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2001.08361","snapshot_observed_at":"2026-08-16T05:31:27.259573Z","title":"Scaling laws for neural language models,","venue":null,"work_id":null,"year":2001},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.259573Z"},"links":{"cited_paper":"/paper/2001.08361","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:680697e2c28a2432af1eb49b2b1202c474a713db55a3811f8b493b074f1e5cad","observation_id":"898168e2-98b7-4c05-9189-f45e160b2cab","resolution":{"observed_at":"2026-08-16T05:31:27.259573Z","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":"2026-08-16T05:31:27.264640Z","title":"Few-shot parameter-efficient fine-tuning is better and cheaper than in-context learning,","venue":null,"work_id":null,"year":1950},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.264640Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:0c43397f64e9b1e9fbaaf2afcdf58955ad914f95d777fd073332b69c652cbad9","observation_id":"06fc9b4d-6600-4fea-902f-8ee42f07ca78","resolution":{"observed_at":"2026-08-16T05:31:27.264640Z","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":"2026-08-16T05:31:27.269335Z","title":"Parameter-efficient fine-tuning of large- scale pre-trained language models,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.269335Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:11a5ba4b8219e94a1a05bcd5317036af33b3ed0835fc34ae32661828222b5881","observation_id":"39ecf5e8-9d91-45d8-b726-15b7690a0bf8","resolution":{"observed_at":"2026-08-16T05:31:27.269335Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2403.14608","last_updated":"2024-09-16T02:54:50Z","snapshot_observed_at":"2026-08-04T09:07:42.158421Z","submitted_at":"2024-03-21T17:55:50Z","title":"Parameter-Efficient Fine-Tuning for Large Models: A Comprehensive Survey","version":7},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.14608","snapshot_observed_at":"2026-08-16T05:31:27.274665Z","title":"Parameter-efficient fine-tuning for large models: A comprehensive survey,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.274665Z"},"links":{"cited_paper":"/paper/2403.14608","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:69a1696f868aaf8246497c8a2f0749c54a60d2cbc2b282583e9bd1cda79e1991","observation_id":"44ad381a-bbf7-4fc6-9f95-c98be9b1711d","resolution":{"observed_at":"2026-08-16T05:31:27.274665Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2106.09685","last_updated":"2021-10-16T18:40:34Z","snapshot_observed_at":"2026-08-17T18:04:53.578114Z","submitted_at":"2021-06-17T17:37:18Z","title":"LoRA: Low-Rank Adaptation of Large Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2106.09685","snapshot_observed_at":"2026-08-16T05:31:27.279505Z","title":"Lora: Low-rank adaptation of large language models,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.279505Z"},"links":{"cited_paper":"/paper/2106.09685","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:16807a95e4d1bf09f17155390a64d4c279052bcefa6da7393cb9a5aeadd6384b","observation_id":"0eab4788-4812-40f9-9e24-d97d55280890","resolution":{"observed_at":"2026-08-16T05:31:27.279505Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2302.04870","last_updated":"2023-02-09T18:59:55Z","snapshot_observed_at":"2026-08-18T14:30:42.666081Z","submitted_at":"2023-02-09T18:59:55Z","title":"Offsite-Tuning: Transfer Learning without Full Model","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2302.04870","snapshot_observed_at":"2026-08-16T05:31:27.284044Z","title":"Offsite-tuning: Transfer learning without full model,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.284044Z"},"links":{"cited_paper":"/paper/2302.04870","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:68d27e0a5945cc95ae0a1f4e530f08c3e8a3ad1a6ab3b0424ea4f6d6741c1b6e","observation_id":"b12c2171-8902-4417-8077-a83e47f51525","resolution":{"observed_at":"2026-08-16T05:31:27.284044Z","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":"2026-08-16T05:31:28.202782Z","title":"Fedlegal: The first real-world federated learning benchmark for legal nlp,","venue":null,"work_id":"0d91941e-c5a5-4c2e-9719-d8192239d6eb","year":2023},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.288879Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:79d120aee8c3ae2a3f64aa62601287f19fe70d81ca039accee08734f7b30292b","observation_id":"b4bd2977-1609-42bf-914b-1e9c7a8427dd","resolution":{"observed_at":"2026-08-16T05:31:28.207407Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2403.12313","last_updated":"2024-03-18T23:20:08Z","snapshot_observed_at":"2026-08-18T12:54:54.840442Z","submitted_at":"2024-03-18T23:20:08Z","title":"Improving LoRA in Privacy-preserving Federated Learning","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.12313","snapshot_observed_at":"2026-08-16T05:31:27.293628Z","title":"Improving lora in privacy-preserving federated learning,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.293628Z"},"links":{"cited_paper":"/paper/2403.12313","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:a360345a6b601904449ca59d17fbadfb2efb601e6fbf9efd621e939443a2de83","observation_id":"761fdabe-59f9-4a60-973b-e9d5bdf1e801","resolution":{"observed_at":"2026-08-16T05:31:27.293628Z","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":"2026-08-16T05:31:27.299066Z","title":"Deep leakage from gradients,","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.299066Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:93697897bcda07caf31fa9513051edd6bb964536614f981aa3e47f1e92736ec3","observation_id":"62ae2dcd-9b55-44e2-8b90-c730489b37cd","resolution":{"observed_at":"2026-08-16T05:31:27.299066Z","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":"2026-08-16T05:31:27.303694Z","title":"Lamp: Extracting text from gradients with language model priors,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.303694Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:3d6c761be5c73223d5376a3ef1f81233eadde63744cb4c7ad0fb0479cb434001","observation_id":"09bce3d0-7db8-40e0-9456-bd5eb420c7af","resolution":{"observed_at":"2026-08-16T05:31:27.303694Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2405.15586","last_updated":"2024-11-13T08:17:38Z","snapshot_observed_at":"2026-08-16T13:49:43.725943Z","submitted_at":"2024-05-24T14:14:24Z","title":"DAGER: Exact Gradient Inversion for Large Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2405.15586","snapshot_observed_at":"2026-08-16T05:31:27.308484Z","title":"Dager: Exact gradient inversion for large language models,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.308484Z"},"links":{"cited_paper":"/paper/2405.15586","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:f4794eda62b926d1edcac7d8cb925bb74059ef4b6576f05066dbea3e6431557a","observation_id":"0988facd-9c86-494a-8cf4-9f7197967ab5","resolution":{"observed_at":"2026-08-16T05:31:27.308484Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2202.07646","last_updated":"2023-03-06T06:28:18Z","snapshot_observed_at":"2026-08-15T14:36:26.439846Z","submitted_at":"2022-02-15T18:48:31Z","title":"Quantifying Memorization Across Neural Language Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2202.07646","snapshot_observed_at":"2026-08-16T05:31:27.313285Z","title":"Quantifying memorization across neural language models,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.313285Z"},"links":{"cited_paper":"/paper/2202.07646","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:68f3a6c5d7090fc8de5e8c3294f803fbd9fc3421616c4dd4736397f17ea41f48","observation_id":"0c910254-7e02-42bc-81c2-bd93dc6326d4","resolution":{"observed_at":"2026-08-16T05:31:27.313285Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.03010","last_updated":"2023-05-04T17:31:41Z","snapshot_observed_at":"2026-08-16T15:35:34.137508Z","submitted_at":"2023-05-04T17:31:41Z","title":"Sentence Embedding Leaks More Information than You Expect: Generative Embedding Inversion Attack to Recover the Whole Sentence","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.03010","snapshot_observed_at":"2026-08-16T05:31:27.318059Z","title":"Sentence embedding leaks more information than you expect: Generative embedding inversion attack to recover the whole sentence,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.318059Z"},"links":{"cited_paper":"/paper/2305.03010","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:2658d9e7031cf716234c5674cc2ae222b3789a26f66a2b7097e0726d255faace","observation_id":"01508bbe-f781-4785-b3d3-cda9607f59e0","resolution":{"observed_at":"2026-08-16T05:31:27.318059Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2201.12675","last_updated":"2023-05-31T16:05:45Z","snapshot_observed_at":"2026-08-16T17:24:58.734972Z","submitted_at":"2022-01-29T22:38:21Z","title":"Decepticons: Corrupted Transformers Breach Privacy in Federated Learning for Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2201.12675","snapshot_observed_at":"2026-08-16T05:31:27.322737Z","title":"Decepticons: Corrupted transformers breach pri- vacy in federated learning for language models,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.322737Z"},"links":{"cited_paper":"/paper/2201.12675","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:66b640d967a4d6d32622e73eac87b81550450f28268199569d9749df1d402c27","observation_id":"2af6dd09-c9ff-4f31-b42e-686036d5a472","resolution":{"observed_at":"2026-08-16T05:31:27.322737Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2403.00871","last_updated":"2024-03-01T06:15:07Z","snapshot_observed_at":"2026-08-16T14:13:52.437929Z","submitted_at":"2024-03-01T06:15:07Z","title":"Teach LLMs to Phish: Stealing Private Information from Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.00871","snapshot_observed_at":"2026-08-16T05:31:27.327680Z","title":"Teach llms to phish: Stealing private information from language mod- els,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.327680Z"},"links":{"cited_paper":"/paper/2403.00871","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:3506493ef7ed228aacd54ba98bdf8231adeb8a0e82cbefa76a4a27dd2b58e129","observation_id":"8990b79c-7c0f-499f-b542-32f47c6eff48","resolution":{"observed_at":"2026-08-16T05:31:27.327680Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2202.00580","last_updated":"2022-06-19T23:21:42Z","snapshot_observed_at":"2026-08-16T17:24:20.567252Z","submitted_at":"2022-02-01T17:26:11Z","title":"Fishing for User Data in Large-Batch Federated Learning via Gradient Magnification","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2202.00580","snapshot_observed_at":"2026-08-16T05:31:27.332892Z","title":"Fishing for user data in large-batch federated learning via gradient magnifica- tion,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.332892Z"},"links":{"cited_paper":"/paper/2202.00580","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:69d06e1b58e3a74ddd8152f94685ee232980b8c18faf46b2fc8c012fedf62416","observation_id":"db54f403-5f5b-4dfd-bf5e-28f6b71f21ad","resolution":{"observed_at":"2026-08-16T05:31:27.332892Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2306.03013","last_updated":"2024-04-15T17:50:38Z","snapshot_observed_at":"2026-08-18T14:54:42.088361Z","submitted_at":"2023-06-05T16:29:54Z","title":"Hiding in Plain Sight: Disguising Data Stealing Attacks in Federated Learning","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2306.03013","snapshot_observed_at":"2026-08-16T05:31:27.337356Z","title":"Hiding in plain sight: Disguising data stealing attacks in federated learning,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.337356Z"},"links":{"cited_paper":"/paper/2306.03013","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:86d83ae7bd6841f5d4585d34fdbd282fc1ad91aaa5899ab993c8baa25455ea68","observation_id":"3ef76844-d0b4-4744-a924-fde326d3d750","resolution":{"observed_at":"2026-08-16T05:31:27.337356Z","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":"2026-08-16T05:31:28.169046Z","title":"Loki: Large-scale data reconstruction attack against fed- erated learning through model manipulation,","venue":null,"work_id":"d5f82678-5d30-48b2-9d19-dcc6ba532cce","year":2024},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.342052Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:c9ca21bd49c19f8828612c60339dd69835f1f9fab9c80c4b04241c6d550e19d3","observation_id":"108ecdc3-8d83-45a4-8ab5-f7c6f1425463","resolution":{"observed_at":"2026-08-16T05:31:28.173916Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2310.06714","last_updated":"2024-02-22T21:19:59Z","snapshot_observed_at":"2026-08-19T01:07:35.051898Z","submitted_at":"2023-10-10T15:41:26Z","title":"Exploring Memorization in Fine-tuned Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.06714","snapshot_observed_at":"2026-08-16T05:31:27.346627Z","title":"Exploring memorization in fine-tuned language models,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.346627Z"},"links":{"cited_paper":"/paper/2310.06714","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:1cf1d044cc92fe1851525bcc277eb91861d1d5e3c7a200755f8e62d361082d95","observation_id":"f47771f0-b41f-4cf3-957b-cdabb5db648f","resolution":{"observed_at":"2026-08-16T05:31:27.346627Z","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":"2026-08-16T05:31:28.153481Z","title":"Be like a goldfish, don’t memorize! mitigating memorization in generative llms,","venue":null,"work_id":"844dc55f-6741-4e89-81ac-3422b61d8b85","year":2024},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.351466Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:cda120e6fa9251fc5d777c0c6c67f63200c930b3f1963e9ce521bdfeddac1196","observation_id":"0e935697-05e9-42a2-bdaf-a69d17b1129f","resolution":{"observed_at":"2026-08-16T05:31:28.158765Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+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":"2026-08-16T05:31:28.137818Z","title":"Analyzing leakage of personally identifiable information in language models,","venue":null,"work_id":"9deef276-53ae-41c5-8eff-2e19feed8fa0","year":2023},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.355848Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:1311d78eebfd16ea4ecc0b16e82b250d5a1c2c656be394acc211fdabd3c95bd0","observation_id":"0345d142-61c7-4eba-bc3d-2cc51535c0ec","resolution":{"observed_at":"2026-08-16T05:31:28.142679Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+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":"2026-08-16T05:31:28.123259Z","title":"Precurious: How innocent pre- trained language models turn into privacy traps,","venue":null,"work_id":"16154afc-0767-496c-b6ef-6093bbde5aa4","year":2024},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.360100Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:4096154e5cf6144b45482d29cec15fb8dc1c483f28d9f66c41ed1e171a6f814f","observation_id":"62606546-bb03-4bf6-a6e0-f94dc11a4456","resolution":{"observed_at":"2026-08-16T05:31:28.127960Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2407.02943","last_updated":"2025-08-23T08:46:24Z","snapshot_observed_at":"2026-08-18T17:38:16.669458Z","submitted_at":"2024-07-03T09:20:04Z","title":"PII-Compass: Guiding LLM training data extraction prompts towards the target PII via grounding","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.02943","snapshot_observed_at":"2026-08-16T05:31:27.364567Z","title":"Pii- compass: Guiding llm training data extraction prompts towards the target pii via grounding,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.364567Z"},"links":{"cited_paper":"/paper/2407.02943","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:a2213c25dac2169c75a017cd16771072c80e708514670f949fb5922658becc35","observation_id":"a20198c7-06b2-46d7-9feb-5bdb89282a99","resolution":{"observed_at":"2026-08-16T05:31:27.364567Z","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":"2026-08-16T05:31:27.369504Z","title":"Parameter-efficient transfer learning for nlp,","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.369504Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:a45ad73f56dcc1316d531620a415782f948d6f467e41d6e7bc189e6e28d063b8","observation_id":"83765166-d5a5-4e3a-8c16-f7e1d0b13cb4","resolution":{"observed_at":"2026-08-16T05:31:27.369504Z","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":"2026-08-16T05:31:28.099886Z","title":"Propile: Probing privacy leakage in large language models,","venue":null,"work_id":"f374ab10-1f19-4ee9-9029-7bf1565592b8","year":2023},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.373983Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:8fc251d4a7b5e2cf8bab76efb0bdd29c50652c17ae46ba188a073d3d6a5fb0c0","observation_id":"6a56aaaf-3277-42e8-b38d-ba2a7ed36052","resolution":{"observed_at":"2026-08-16T05:31:28.104379Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+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":"2026-08-16T05:31:28.085412Z","title":"Learning to reason and memorize with self-notes,","venue":null,"work_id":"f30ec40f-1282-4cd0-9e41-17a46fd24f29","year":2024},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.378399Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:f7aa04be682b08d72a4ecec258e79f257b90e5d19f8795b2fd4146bc0266e621","observation_id":"d6622eb1-283a-4406-a934-e92f27f35d35","resolution":{"observed_at":"2026-08-16T05:31:28.090312Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2201.11903","last_updated":"2023-01-10T23:07:57Z","snapshot_observed_at":"2026-08-13T07:04:41.220509Z","submitted_at":"2022-01-28T02:33:07Z","title":"Chain-of-Thought Prompting Elicits Reasoning in Large Language Models","version":6},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2201.11903","snapshot_observed_at":"2026-08-16T05:31:27.382881Z","title":"Chain of thought prompting elicits reasoning in large language models,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.382881Z"},"links":{"cited_paper":"/paper/2201.11903","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:f537010ee9b0586f99a2166c3fa928d85b286a89c628611cfe2a5685b58d9990","observation_id":"c49f98d8-8fff-4979-8eaa-15d094c5fa2b","resolution":{"observed_at":"2026-08-16T05:31:27.382881Z","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":"2026-08-16T05:31:28.068347Z","title":"Cocktail party attack: Breaking aggregation-based privacy in federated learning using independent component analysis,","venue":null,"work_id":"be7da2c4-70a8-4295-b158-7e6f688e3331","year":2023},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.387459Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:4e2430c87a3cd8eeb29cae78528d265d773142f3983e54bc4ecc49d8cd858d37","observation_id":"449028d0-06fb-4078-a7b0-0b5c6006c16e","resolution":{"observed_at":"2026-08-16T05:31:28.074039Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2403.03945","last_updated":"2024-12-09T10:10:40Z","snapshot_observed_at":"2026-08-18T17:58:43.964598Z","submitted_at":"2024-03-06T18:52:39Z","title":"SPEAR:Exact Gradient Inversion of Batches in Federated Learning","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2403.03945","snapshot_observed_at":"2026-08-16T05:31:27.391858Z","title":"Spear: Exact gradient inversion of batches in federated learning,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.391858Z"},"links":{"cited_paper":"/paper/2403.03945","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:67a8f550c4f65b3d66cc80b0721764c0b9a70083e1fc1961850e32b0f5462066","observation_id":"886c8998-4556-43a2-877e-b20455969fd3","resolution":{"observed_at":"2026-08-16T05:31:27.391858Z","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":"2026-08-16T05:31:28.051135Z","title":null,"venue":null,"work_id":"45e8c87b-e36b-4086-b898-c15134af26ee","year":2012},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.396613Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:b71d6c6a0dd0da5a6ea7856bc16d53ca115215a378a980848529f8fb5c0bb3c7","observation_id":"a053c742-a180-4172-9142-4d45ee39c46f","resolution":{"observed_at":"2026-08-16T05:31:28.056266Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+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":"2026-08-16T05:31:28.035477Z","title":"Stevenson, Oxford dictionary of English","venue":null,"work_id":"9d2a2d93-64b7-4849-8150-9c0556013c0f","year":2010},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.401002Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:5824fec2b2ebf821cad4d008504412552f115de1204dba455476e020e2449de5","observation_id":"30fc6dfc-c649-4c1d-bbb0-9562b3ba797b","resolution":{"observed_at":"2026-08-16T05:31:28.040613Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+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":"2026-08-16T05:31:28.019913Z","title":"The enron corpus: A new dataset for email classification research,","venue":null,"work_id":"f55469c7-97cf-4aff-b3c8-c0b031214ae0","year":2004},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.405792Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:e18ccf1bce274d6a689975a8d85bdef03016a9bf9fe99348ccff6853e51c1281","observation_id":"e9de6b43-c051-4468-93f1-c1a7abb87070","resolution":{"observed_at":"2026-08-16T05:31:28.024871Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1801.07243","last_updated":"2018-09-25T18:55:07Z","snapshot_observed_at":"2026-08-14T19:53:03.553178Z","submitted_at":"2018-01-22T18:58:18Z","title":"Personalizing Dialogue Agents: I have a dog, do you have pets too?","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1801.07243","snapshot_observed_at":"2026-08-16T05:31:27.410198Z","title":"Personalizing dialogue agents: I have a dog, do you have pets too,","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.410198Z"},"links":{"cited_paper":"/paper/1801.07243","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:d33ac4a7a1c54397c301f0b1604c802904ddfa9dc5443964aaf2837adfa2fbc3","observation_id":"e32bdcc5-5997-456d-9f4e-f91be40365f5","resolution":{"observed_at":"2026-08-16T05:31:27.410198Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1806.03822","last_updated":"2018-06-11T06:10:11Z","snapshot_observed_at":"2026-08-14T19:05:25.482398Z","submitted_at":"2018-06-11T06:10:11Z","title":"Know What You Don't Know: Unanswerable Questions for SQuAD","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1806.03822","snapshot_observed_at":"2026-08-16T05:31:27.414858Z","title":"Know what you don’t know: Unanswerable questions for squad,","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.414858Z"},"links":{"cited_paper":"/paper/1806.03822","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:b2a9af17f60e7d2a887c3219d2b12363b2e942ae97b8f03bb4b873a2f7d1829d","observation_id":"43dcf9c8-fdb2-48d1-adb0-f7aaaf1c512b","resolution":{"observed_at":"2026-08-16T05:31:27.414858Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2101.00027","last_updated":"2020-12-31T19:00:10Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2020-12-31T19:00:10Z","title":"The Pile: An 800GB Dataset of Diverse Text for Language Modeling","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2101.00027","snapshot_observed_at":"2026-08-16T05:31:27.419325Z","title":"The pile: An 800gb dataset of diverse text for language modeling,","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.419325Z"},"links":{"cited_paper":"/paper/2101.00027","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:9ca7c8bf0cd75378a6e505a96ea9de2c979351ddb10cc30c167cb396c7a321cf","observation_id":"388acd3c-864b-4e1c-81d6-d5fc03e823a7","resolution":{"observed_at":"2026-08-16T05:31:27.419325Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2211.01786","last_updated":"2023-05-29T16:40:37Z","snapshot_observed_at":"2026-08-17T05:58:22.357239Z","submitted_at":"2022-11-03T13:19:32Z","title":"Crosslingual Generalization through Multitask Finetuning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2211.01786","snapshot_observed_at":"2026-08-16T05:31:27.423966Z","title":"Crosslingual generalization through multitask finetuning,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.423966Z"},"links":{"cited_paper":"/paper/2211.01786","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:0aa05a11740d0418d31c5ebda1526123b607121c7e19b951609b5f0e5940c9b2","observation_id":"3bdadf13-1f82-419b-8700-f74e46c9ea76","resolution":{"observed_at":"2026-08-16T05:31:27.423966Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2307.09288","last_updated":"2023-07-19T17:08:59Z","snapshot_observed_at":"2026-08-07T12:56:43.323460Z","submitted_at":"2023-07-18T14:31:57Z","title":"Llama 2: Open Foundation and Fine-Tuned Chat Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2307.09288","snapshot_observed_at":"2026-08-16T05:31:27.428627Z","title":"Llama 2: Open foundation and fine-tuned chat models,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.428627Z"},"links":{"cited_paper":"/paper/2307.09288","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:204f217855bb42eb67d2f4014281712cb2ef34a6919de2b7f9c816d9602f460b","observation_id":"c241f52c-d2f8-468c-93c3-9ca2564228e3","resolution":{"observed_at":"2026-08-16T05:31:27.428627Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2501.12948","last_updated":"2026-01-04T03:57:36Z","snapshot_observed_at":"2026-08-15T12:33:55.451951Z","submitted_at":"2025-01-22T15:19:35Z","title":"DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2501.12948","snapshot_observed_at":"2026-08-16T05:31:27.433167Z","title":"Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.433167Z"},"links":{"cited_paper":"/paper/2501.12948","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:d719ec712256bc52d87fa8d134af534a3b2e4fab3c987e499d93913a05680fa8","observation_id":"8d7d371b-ebed-459a-aa0b-e166d3961b8e","resolution":{"observed_at":"2026-08-16T05:31:27.433167Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1910.03771","last_updated":"2020-07-14T03:42:34Z","snapshot_observed_at":"2026-07-06T08:27:58.343233Z","submitted_at":"2019-10-09T03:23:22Z","title":"HuggingFace's Transformers: State-of-the-art Natural Language Processing","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1910.03771","snapshot_observed_at":"2026-08-16T05:31:27.437908Z","title":"Huggingface’s transformers: State-of-the-art natural language processing,","venue":null,"work_id":null,"year":1910},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.437908Z"},"links":{"cited_paper":"/paper/1910.03771","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:1106223dcec0169724cc33a70a46735d40f1e9dd8fee3d395e176f50a0d2f5a8","observation_id":"a798bb1e-5408-4816-b6ea-92a882c6e551","resolution":{"observed_at":"2026-08-16T05:31:27.437908Z","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":"2026-08-16T05:31:28.005069Z","title":"Uncovering gradient inversion risks in practical language model training,","venue":null,"work_id":"ba77484e-8170-4599-9069-10dbbef57173","year":2024},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.442749Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:cd0e65fddd3d9ec96f7991f432505418c7237f6c07bbadc4ffa375ab53609679","observation_id":"f6899cff-1656-45a8-bff2-0547018ef616","resolution":{"observed_at":"2026-08-16T05:31:28.009848Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2205.10162","last_updated":"2023-05-08T19:50:56Z","snapshot_observed_at":"2026-08-16T16:58:54.473327Z","submitted_at":"2022-05-20T13:10:43Z","title":"FedAdapter: Efficient Federated Learning for Modern NLP","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2205.10162","snapshot_observed_at":"2026-08-16T05:31:27.447045Z","title":"Fedadapter: Efficient federated learning for modern nlp,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.447045Z"},"links":{"cited_paper":"/paper/2205.10162","citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:a01afb763381102e21ec35e00110c7a3669790f0d01dbe922646bf3f7736bbf8","observation_id":"eed6d884-d244-44e2-bca2-64fc05a7bc44","resolution":{"observed_at":"2026-08-16T05:31:27.447045Z","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":"2026-08-16T05:31:27.987247Z","title":"The updated warranty is valid until December 31, 2027","venue":null,"work_id":"b975f995-33ba-4e76-ba03-b7436813b962","year":2027},"citing_paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models","version":1},"reference_index":2025,"source":"pdf_text","source_observed_at":"2026-08-16T05:31:27.452275Z"},"links":{"citing_paper":"/paper/2504.20570"},"observation_digest":"sha256:4462fdf30e64e581637acb7f68196ca9d5b1a9fb272343dbac010340722384be","observation_id":"50a731f4-7f53-4f12-b36b-4ed457b36326","resolution":{"observed_at":"2026-08-16T05:31:27.994121Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-18T06:34:40.430872+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2504.20570","last_updated":"2025-04-29T09:23:19Z","latest_version":1,"primary_category":"cs.CR","snapshot_observed_at":"2026-08-18T17:46:52.206144Z","submitted_at":"2025-04-29T09:23:19Z","title":"ReCIT: Reconstructing Full Private Data from Gradient in Parameter-Efficient Fine-Tuning of Large Language Models"},"reference_resolution":{"displayed":45,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":33,"verified_exact":0,"verified_fuzzy":12},"total_outbound_references":45},"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":"2026-08-18T06:34:40.430872+00:00","source":"crossref"},{"observed_at":"2026-08-18T06:34:34.496301+00:00","source":"retraction_watch"}],"thesis":"As of 19 August 2026, this Paper Citation Record lists 45 of 45 outbound references and 1 inbound Pith citation observation for arXiv:2504.20570."}