{"as_of":"2026-08-04T12:33:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:2f0a3013f2a713f84a2bffcba793f47a8589152a9d2bf248e87f84c10cce3cd6","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":"2026-05-23T18:46:54.778974Z","state":"measured"},{"denominator":48,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":48,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-04T06:34:03.388597+00:00","state":"measured"},{"denominator":3,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":3,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-03T11:38:48.846818Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"pith","source_observed_at":"2026-07-04T19:30:07.628677Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"cited_work":{"arxiv_id":"2410.09457","doi":null,"metadata_source":"pith","pith_arxiv_id":"2410.09457","snapshot_observed_at":"2026-07-04T19:30:07.628677Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","venue":"cs.LG","work_id":"1f809390-e451-4ae9-ac72-15c05e1ae513","year":2024},"citing_paper":{"arxiv_id":"2606.25349","last_updated":"2026-07-31T12:18:38Z","snapshot_observed_at":"2026-08-04T12:09:59.977144Z","submitted_at":"2026-06-24T03:33:17Z","title":"General Techniques for Reducing Key-Switching Overhead in Privacy-Preserving Two-Party Transformer Inference","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-06-25T21:17:19.995668Z"},"links":{"cited_paper":"/paper/2410.09457","citing_paper":"/paper/2606.25349"},"observation_digest":"sha256:ac80c4f433630fb50ebcb68f73dca010fcb91b16328e0651abba6bfdcced26e7","observation_id":"3b93877c-3093-429d-896d-02586813534c","resolution":{"observed_at":"2026-07-04T19:30:07.629937Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.09457","snapshot_observed_at":"2026-08-03T02:11:04.005972Z","title":"Power-softmax: Towards secure LLM inference over encrypted data","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2606.25349","last_updated":"2026-07-31T12:18:38Z","snapshot_observed_at":"2026-08-04T12:09:59.977144Z","submitted_at":"2026-06-24T03:33:17Z","title":"General Techniques for Reducing Key-Switching Overhead in Privacy-Preserving Two-Party Transformer Inference","version":2},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-03T02:11:04.005972Z"},"links":{"cited_paper":"/paper/2410.09457","citing_paper":"/paper/2606.25349"},"observation_digest":"sha256:b1ad45aaf757094e812ff62f34b2c514955f02c6a95a8d7752e8033ae267ac4d","observation_id":"d4cb14f7-6fcd-4a37-b19d-e3bb1e92e2c7","resolution":{"observed_at":"2026-08-03T02:11:04.005972Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.09457","snapshot_observed_at":"2026-08-03T11:38:48.846818Z","title":"encryption","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.29221","last_updated":"2026-07-31T09:53:51Z","snapshot_observed_at":"2026-08-04T12:11:12.959004Z","submitted_at":"2026-07-31T09:53:51Z","title":"MOSAIC: Masked Outsourcing of Secure AI Computations","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-03T11:38:48.846818Z"},"links":{"cited_paper":"/paper/2410.09457","citing_paper":"/paper/2607.29221"},"observation_digest":"sha256:c409f0a837806d0e98e2d53a2ab4db19657e3cc933377d6597e2345cf79a34ba","observation_id":"47aa6c0e-6b73-4894-86ca-109ff325d6ca","resolution":{"observed_at":"2026-08-03T11:38:48.846818Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2410.09457/citation-record","integrity":"/paper/2410.09457/integrity","json":"/paper/2410.09457/citation-record.json","paper":"/paper/2410.09457"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.56553/popets-2023-0020","metadata_source":"doi_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"HElayers: A tile tensors framework for large neural networks on encrypted data","venue":null,"work_id":"bdd0093d-4f21-4d9b-921e-ee606095a4da","year":2023},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":1,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:09cb04ce024e4592e78eda395dc34937c43230588263230ef2e66dd6c372eb60","observation_id":"e32a1fcf-4158-457d-9c08-c81cd4fffe43","resolution":{"observed_at":"2026-05-23T18:48:19.445169Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"On the privacy of protocols based on cpa-secure homomorphic encryption","venue":null,"work_id":"9fb3c740-f680-4195-92bf-ceab2360f50b","year":2021},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":2,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:d1721539725a5ab3604142f04f6961f661609c917e72d6857ecee18f112fc2d3","observation_id":"cbd7c0c1-1393-4dd3-ae86-71da382ec9ab","resolution":{"observed_at":"2026-05-23T20:28:26.258397Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"Gpt-neox: Large scale autoregressive language modeling in pytorch, 9 2023","venue":null,"work_id":"ff5fbebe-9376-44b6-b665-91f68fab408f","year":2023},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":3,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:c05998ecc9eae2de10eebd4403eef6d86746d88e0619b084ca8ae63856260031","observation_id":"1be5e58d-beeb-4704-a8be-01896f1111e5","resolution":{"observed_at":"2026-05-23T20:28:26.287853Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"AutoFHE : Automated adaption of CNNs for efficient evaluation over FHE","venue":null,"work_id":"2d645812-173e-4b1c-a70b-56d9d9bd79f3","year":2024},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":4,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:00173eba0d4279fd1803b7f735a00664a63dd6ecb7d587a4ce1c650be77b67cc","observation_id":"56a3a3a5-b8b3-4501-ae1d-dc1dba6968b7","resolution":{"observed_at":"2026-05-24T07:46:05.423963Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1007/978-3-031-16815-4","metadata_source":"doi_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"A Methodology for Training Homomorphic Encryption Friendly Neural Networks","venue":null,"work_id":"c4b6e701-db48-4646-b8d6-0065d54eef9e","year":2022},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":5,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:f185d0e47ac110418b6aa339b65dade7d8c633b297ae14b8b39e46e6e0e1d677","observation_id":"166adb83-81a2-459c-89fe-42f61cff2fca","resolution":{"observed_at":"2026-05-23T18:48:19.449342Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2304.14836","last_updated":"2023-06-11T10:07:52Z","snapshot_observed_at":"2026-08-01T13:30:50.481878Z","submitted_at":"2023-04-26T20:41:37Z","title":"Training Large Scale Polynomial CNNs for E2E Inference over Homomorphic Encryption","version":2},"cited_work":{"arxiv_id":"2304.14836","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2304.14836","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Sensitive Tuning of Large Scale CNNs for E2E Secure Prediction using Homomorphic Encryption","venue":null,"work_id":"06a2ded8-aef0-41ac-9b6c-7fe1560f67ca","year":2023},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":6,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"cited_paper":"/paper/2304.14836","citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:4021460b5dff7069513e1fb48290a149438c721a500de856f940963eb9f253d0","observation_id":"190306c3-1ae4-4322-84f6-254acb9d9f74","resolution":{"observed_at":"2026-05-23T18:48:20.279784Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"Pythia : A suite for analyzing large language models across training and scaling","venue":null,"work_id":"eda28d2e-1d69-46f0-9a3f-3115cd6ef893","year":2023},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":7,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:762c7dcbb335d71fa731c1750452eacb7d02894120b73d6022fb922e250cfad3","observation_id":"fc1f7960-387f-4e17-9b5d-21d0cc6ff01e","resolution":{"observed_at":"2026-05-23T20:28:26.265119Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1145/2633600","metadata_source":"doi_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"(Leveled) Fully Homomorphic Encryption without Bootstrapping","venue":null,"work_id":"82da0529-ecf5-4896-9470-bed07937d94c","year":2014},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":8,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:19a42946b0acfd6e60deff0472ecee870d1e9f120c2180325c6c14fcbd08eb62","observation_id":"49537791-481b-48bf-b3cd-5ca862c794c8","resolution":{"observed_at":"2026-05-23T18:48:19.466897Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-05-21T11:55:04.942573+00:00","source":"crossref_status_cache"},{"observed_at":"2026-05-21T11:55:04.942573+00:00","source":"openalex_status_cache"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2206.00216","last_updated":"2022-06-02T01:06:12Z","snapshot_observed_at":"2026-07-06T13:16:13.214340Z","submitted_at":"2022-06-01T03:49:18Z","title":"THE-X: Privacy-Preserving Transformer Inference with Homomorphic Encryption","version":2},"cited_work":{"arxiv_id":"2206.00216","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2206.00216","snapshot_observed_at":"2026-07-02T09:26:51.572791Z","title":"The-x: Privacy-preserving transformer inference with homomorphic encryption","venue":null,"work_id":"d2d5fbfb-9177-4f01-9627-b3f56652aa82","year":2022},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":9,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"cited_paper":"/paper/2206.00216","citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:71bcea929c675a775107096f11f8928989617cfa635b4decbda3bed8b940427e","observation_id":"b1b252ca-f16d-49a4-a916-3f05259f93dc","resolution":{"observed_at":"2026-05-23T18:48:20.273782Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1007/978-3-319-70694-8","metadata_source":"doi_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-29T17:13:44.418782Z","title":"Homomorphic encryption for arithmetic of approximate numbers","venue":null,"work_id":"d91c809d-9059-4f76-a663-4f3f89352ff1","year":2017},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":10,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:9760e07f5818634c11b37eb36ba0de7770f95c084be5bef2a79d42f9e313bd28","observation_id":"3a1c295d-f6d9-4e35-8d3b-0939c717a1b0","resolution":{"observed_at":"2026-05-23T18:48:19.409958Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"P-nets: Deep polynomial neural networks","venue":null,"work_id":"93bba4aa-2bfe-4916-9322-a81716214edd","year":2020},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":11,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:54fcca752f7b32a8ba83ef9db7c7ac21ebf00eeea87b22558abf982bed4af047","observation_id":"13a5c354-bd69-41dc-b8de-b7f6b960124d","resolution":{"observed_at":"2026-05-23T20:28:26.261750Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2308.09923","last_updated":"2023-08-19T06:26:14Z","snapshot_observed_at":"2026-07-06T16:07:57.164115Z","submitted_at":"2023-08-19T06:26:14Z","title":"East: Efficient and Accurate Secure Transformer Framework for Inference","version":1},"cited_work":{"arxiv_id":"2308.09923","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2308.09923","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"East: Efficient and accurate secure transformer framework for inference","venue":null,"work_id":"d07ce972-6d6f-4d85-9e1b-94e1e7035c02","year":2023},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":12,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"cited_paper":"/paper/2308.09923","citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:809abe7086c1541a36c40720a24654d27454e1a91e112da4f24932c1c570886a","observation_id":"ac6600e7-f29d-4daf-bf07-256d71955cfe","resolution":{"observed_at":"2026-05-23T18:48:20.310523Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1007/978-3-031-34671-2_5","metadata_source":"doi_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Efficient skip connections realization for secure inference on encrypted data","venue":null,"work_id":"d1f2e985-ac45-4558-9db2-99e6ac20c6ef","year":2023},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":13,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:4cc00e651bbccc20d6cca453ae1ff930055e0dae4cf6974bca5835d1d1259b37","observation_id":"1ffa5e97-4cde-4262-884d-effaa567932c","resolution":{"observed_at":"2026-05-23T18:48:19.423596Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"Somewhat Practical Fully Homomorphic Encryption","venue":null,"work_id":"ba03d563-2aa7-478d-bb3f-c11ddf4b3849","year":2012},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":14,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:ac809c51a85e468d24544d87d8374faf4a09ba97f97a5c4a9bba241a50935c6e","observation_id":"be048f5e-57eb-470b-810a-c22d2db2e886","resolution":{"observed_at":"2026-05-23T20:28:26.307108Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"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":"2101.00027","doi":"10.1117/1.jmi.10.6.061104","metadata_source":"pith","pith_arxiv_id":"2101.00027","snapshot_observed_at":"2026-07-11T11:50:26.030339Z","title":"The Pile: An 800GB Dataset of Diverse Text for Language Modeling","venue":"cs.CL","work_id":"9b10667a-da61-4358-aceb-10578234d45d","year":2020},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":15,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"cited_paper":"/paper/2101.00027","citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:ecc2d3b1b978ec7a357cfc91bea5bfe418f8991f0fa7ff849ad001540379bbac","observation_id":"8b916aee-5ac6-4ebc-b167-e0fd32c3bc62","resolution":{"observed_at":"2026-05-23T18:48:20.297343Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"A fully homomorphic encryption scheme","venue":null,"work_id":"875acbe9-7fce-4244-bb31-1cbcd4af38c5","year":2009},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":16,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:85acfd2a7bb9bb38cacaf1bf494765da584bac55f1601a0479e39fa2c502f200","observation_id":"35f15c42-66f3-44ba-959e-55bf3eb1fd7d","resolution":{"observed_at":"2026-05-23T20:28:26.338121Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"Cryptonets: Applying neural networks to encrypted data with high throughput and accuracy","venue":null,"work_id":"cdac951a-c227-4f3c-8b26-6d34cd97723d","year":2016},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":17,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:8e0555959e2b949b67bc2f1004cd9e6eb9bdd920ff88b63c1256b75f5bc9fa26","observation_id":"cc441605-16df-4d79-b2df-5920ae5dcc08","resolution":{"observed_at":"2026-05-23T20:28:26.334287Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-07-08T17:35:10.692810Z","title":"Openwebtext corpus","venue":null,"work_id":"f11a315a-8bd8-4b33-84c2-ae94a4cf50a4","year":2019},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":18,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:24f3e3cc79f91898278bcd3ce461f853ae76d19c5b71e6076e01d223aec6ce30","observation_id":"8aeabb14-84b9-464f-88d3-dd14dd62ff87","resolution":{"observed_at":"2026-05-24T07:46:05.416045Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"Applications of division by convergence","venue":null,"work_id":"dc5ad7f8-0143-4bef-a919-045aca32e5a8","year":1964},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":19,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:f2fbaec24cf7e6067efb2bad2e786bf143ed7152fbad2bad53f50df72983f257","observation_id":"ac4fc503-02d7-4611-8f7d-a4c4ab394dd0","resolution":{"observed_at":"2026-05-23T20:28:26.300458Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"Polynomial activation functions","venue":null,"work_id":"4ec94e34-c566-4392-9574-59bffb0dd231","year":2020},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":20,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:12f4dac5e29875e176bd78fa84f3ffe0ff2cad10fa2dfa13d15fc446544acf0a","observation_id":"5df9f3ab-29e0-4318-8f9b-64efdb5763e0","resolution":{"observed_at":"2026-05-23T20:28:26.313216Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"8605.2020","doi":"10.1109/ijcnn48605.2020","metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-11T11:50:26.030339Z","title":"Bayesian neural networks uncertainty quantification with cubature rules","venue":null,"work_id":"79965938-b771-4e64-aff9-668810f65873","year":2020},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:0c48acec95731ac94bcac45a0b71078b21ccdd3c6b1bc84203d2925da6d80ff0","observation_id":"e8ce126f-5090-4bc0-9273-5deb13214e50","resolution":{"observed_at":"2026-05-23T18:48:19.417461Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"SIGMA : Secure GPT inference with function secret sharing","venue":null,"work_id":"3fdb9edc-cac3-4e37-8253-d9694176ba91","year":2023},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":22,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:ad003a290e24183985094dc061efe0ee4876d8d3d19d879a015f818468d75c82","observation_id":"65500de6-04ec-4b3f-9931-516ae6d7188b","resolution":{"observed_at":"2026-05-23T20:28:26.323003Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2312.04356","last_updated":"2025-01-12T23:49:20Z","snapshot_observed_at":"2026-07-06T16:58:19.652601Z","submitted_at":"2023-12-07T15:23:07Z","title":"NeuJeans: Private Neural Network Inference with Joint Optimization of Convolution and FHE Bootstrapping","version":3},"cited_work":{"arxiv_id":"2312.04356","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2312.04356","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Neujeans: Private neural network inference with joint optimization of convolution and bootstrapping","venue":null,"work_id":"5451fb41-dbad-48b2-a4cf-a7ec7cb72bfc","year":2023},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":23,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"cited_paper":"/paper/2312.04356","citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:197d24307448625223970c58d2c486e159b0d3dc59c9b3463c150cc315bf4129","observation_id":"746b0a7b-3043-4b23-9297-7d51bdd67db6","resolution":{"observed_at":"2026-05-23T18:48:20.341112Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"Low-complexity deep convolutional neural networks on fully homomorphic encryption using multiplexed parallel convolutions","venue":null,"work_id":"cf5114ec-8b24-4bda-b9d4-8eeb6158f3c7","year":2022},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":24,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:91621ebbb9667b7709263d6e5959c68a26fef8e135173d32eb1408ef25fbfe15","observation_id":"43c545a0-1881-45f0-9fc2-2e04528d1528","resolution":{"observed_at":"2026-05-23T20:28:26.327062Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2105.10879","last_updated":"2021-06-14T01:34:47Z","snapshot_observed_at":"2026-07-06T11:11:58.263934Z","submitted_at":"2021-05-23T08:06:37Z","title":"Precise Approximation of Convolutional Neural Networks for Homomorphically Encrypted Data","version":4},"cited_work":{"arxiv_id":"2105.10879","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2105.10879","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Precise approximation of convolutional neural networks for homomorphically encrypted data","venue":null,"work_id":"6a088536-e464-4288-b7a0-4bf61a264f5e","year":2021},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"cited_paper":"/paper/2105.10879","citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:1cf178f1309440f67841a9c106d262f3d08329f699db1a0d7f2444bc43b36dbc","observation_id":"e41b1690-fabc-494e-87fd-79f41ed991ce","resolution":{"observed_at":"2026-05-23T18:48:20.335776Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2310.10349","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Optimized layerwise approximation for efficient private inference on fully homomorphic encryption","venue":null,"work_id":"4da17554-b8a2-4c5c-b585-68cadc3c693f","year":2025},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":26,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:65920866df2a760ebcc91ba11b0d83b5ac9ff3bc05ba7fbf66cfaae5d31ae1c3","observation_id":"61a5a14d-a379-4154-8b19-a2af2d3c7e87","resolution":{"observed_at":"2026-05-23T18:48:20.316688Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1609/aaai.v38i18.29964","metadata_source":"doi_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"MERGE : Fast private text generation","venue":null,"work_id":"2246c367-30aa-4798-9984-8188b152f40c","year":2024},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":27,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:aaefda53103307717ad4d68609bfa0fd4db25d9c2755ee9ba752136d732111bf","observation_id":"ceee5066-4dc9-4d65-b620-bb0dbd5a31f4","resolution":{"observed_at":"2026-05-23T18:48:19.455737Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2305.18396","last_updated":"2023-12-15T02:03:10Z","snapshot_observed_at":"2026-07-06T15:34:59.457879Z","submitted_at":"2023-05-28T13:08:13Z","title":"LLMs Can Understand Encrypted Prompt: Towards Privacy-Computing Friendly Transformers","version":3},"cited_work":{"arxiv_id":"2305.18396","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2305.18396","snapshot_observed_at":"2026-07-02T09:26:51.577413Z","title":"Llms can understand encrypted prompt: Towards privacy-computing friendly transformers","venue":null,"work_id":"30b844b3-45f0-44d6-a26b-761e87d4a864","year":2023},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":28,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"cited_paper":"/paper/2305.18396","citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:54e438529d11c3aa7501dac989b84f78f66d4639bca99182e0790ef70f6d7444","observation_id":"87ecf7a0-7cae-47c9-9bbb-e977a5dba6e2","resolution":{"observed_at":"2026-05-23T18:48:20.325451Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1907.11692","last_updated":"2019-07-26T17:48:29Z","snapshot_observed_at":"2026-07-31T22:31:37.910868Z","submitted_at":"2019-07-26T17:48:29Z","title":"RoBERTa: A Robustly Optimized BERT Pretraining Approach","version":1},"cited_work":{"arxiv_id":"1907.11692","doi":"10.1007/s10489-02203627-9","metadata_source":"pith","pith_arxiv_id":"1907.11692","snapshot_observed_at":"2026-07-11T11:50:26.030339Z","title":"RoBERTa: A Robustly Optimized BERT Pretraining Approach","venue":"cs.CL","work_id":"41fe12c4-e538-4890-a244-480650ed3078","year":2019},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":29,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"cited_paper":"/paper/1907.11692","citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:4cdeabe2b4e533e35b5503ae92a106ae0c17d2b807393c872beec8cef21ca2b6","observation_id":"362e3951-9a8e-48e6-91b1-20f1f8593f70","resolution":{"observed_at":"2026-05-23T18:48:20.268069Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"Financial news classification dataset","venue":null,"work_id":"da7ea188-6de1-44da-9811-89dca36ecad5","year":2022},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":30,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:8f40f9b4dd0fd944b422fbfc2b120a7960d73ed566fa94f3b2d25018483de62c","observation_id":"4638ef28-4364-4600-8455-ec0203b91a8a","resolution":{"observed_at":"2026-05-23T20:28:26.284123Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"fairseq: A fast, extensible toolkit for sequence modeling","venue":null,"work_id":"af9bdc36-654d-47cd-a32d-195ad2a8dcc5","year":2019},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":31,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:9badf01a691e75c3d7642132ac4acf290f6107af98245e1fcdc5a41be3b328cd","observation_id":"7b6c7835-9935-4f22-84b7-70b9be4fe584","resolution":{"observed_at":"2026-05-23T20:28:26.294849Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1706.03762","last_updated":"2023-08-02T00:41:18Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2017-06-12T17:57:34Z","title":"Attention Is All You Need","version":7},"cited_work":{"arxiv_id":"1706.03762","doi":"10.1186/s13550-021-00830-6","metadata_source":"pith","pith_arxiv_id":"1706.03762","snapshot_observed_at":"2026-07-14T22:52:12.858825Z","title":"Attention Is All You Need","venue":"cs.CL","work_id":"baafb5a2-5272-43bc-932f-09fa9ffe5316","year":2017},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":32,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"cited_paper":"/paper/1706.03762","citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:021ee636b8016fc1bc3eaa19d689bdb3dfd1c6707580f7e53bea1e6e42817997","observation_id":"aa02b02d-4624-4935-9e12-a9a7ae6514ee","resolution":{"observed_at":"2026-05-23T18:48:20.330334Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.18653/v1/w19-4808","metadata_source":"doi_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-27T07:10:41.523051Z","title":"Analyzing the structure of attention in a transformer language model","venue":null,"work_id":"84358074-f48a-4ae1-a860-55c39b33603c","year":2019},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":33,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:f28283eeee88de2e262865024631fc7ddc9a4cdec4bf7649b5593a7b2cc57563","observation_id":"f6063eb9-660e-4cfa-9998-20dbb1dc797e","resolution":{"observed_at":"2026-05-23T18:48:19.428597Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1804.07461","last_updated":"2019-02-22T23:53:34Z","snapshot_observed_at":"2026-07-06T06:34:26.609892Z","submitted_at":"2018-04-20T06:35:04Z","title":"GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding","version":3},"cited_work":{"arxiv_id":"1804.07461","doi":null,"metadata_source":"pith","pith_arxiv_id":"1804.07461","snapshot_observed_at":"2026-07-04T09:19:44.134394Z","title":"GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding","venue":"cs.CL","work_id":"1bb6fb0c-482d-43cf-94a8-ed18f72a5563","year":2018},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":34,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"cited_paper":"/paper/1804.07461","citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:3c7a8554515821d43f3a04427e237ac98f9e34d680b87a7261d82049beebd575","observation_id":"975e048e-6ac1-4325-8e0b-0baf754dff07","resolution":{"observed_at":"2026-05-23T18:48:20.284881Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2403.05156","last_updated":"2024-03-14T14:17:57Z","snapshot_observed_at":"2026-07-30T18:58:06.958154Z","submitted_at":"2024-03-08T08:47:48Z","title":"On Protecting the Data Privacy of Large Language Models (LLMs): A Survey","version":2},"cited_work":{"arxiv_id":"2403.05156","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2403.05156","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"On protecting the data privacy of large language models (llms): A survey","venue":null,"work_id":"12eba11e-9e6e-41df-bf14-d502222fa95a","year":2024},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":35,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"cited_paper":"/paper/2403.05156","citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:6130c1bbbd9e049e9075472f9eddfbbc006fb89caeaf489742ddcbb069dccac5","observation_id":"4417f7df-e4c1-4270-bd63-ac19c45ce9b0","resolution":{"observed_at":"2026-05-23T18:48:20.262430Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2024.100211","doi":"10.1016/j.bcra.2024.100211","metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-11T11:50:26.030339Z","title":"Energy -Aware Proof-of-Authority: Blockchain Consensus for Clustered Wireless Sensor Network","venue":null,"work_id":"e9199ec7-9479-4a8f-9030-ced147f24009","year":2024},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":36,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:c74651fa8bde0093e91c7b49689dcd34ee5610763f5525603ab85874b62d6b48","observation_id":"6df805a0-523c-4ad2-baad-bf2646cef941","resolution":{"observed_at":"2026-05-23T18:48:19.474007Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2402.11224","last_updated":"2024-06-07T10:34:32Z","snapshot_observed_at":"2026-07-06T17:31:33.340510Z","submitted_at":"2024-02-17T08:54:25Z","title":"Neural Networks with (Low-Precision) Polynomial Approximations: New Insights and Techniques for Accuracy Improvement","version":2},"cited_work":{"arxiv_id":"2402.11224","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2402.11224","snapshot_observed_at":"2026-06-05T21:23:00.469572Z","title":"Neural networks with (low-precision) polynomial approximations: New insights and techniques for accuracy improvement","venue":null,"work_id":"7cfc553e-493d-4de5-bfee-7e08f5481dc3","year":2024},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":37,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"cited_paper":"/paper/2402.11224","citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:94817a6174fc85df42376d037e1b352aa0554ce2ebd00668c7308a4c26321078","observation_id":"af28243b-b029-4762-b043-3fe2f4c1750b","resolution":{"observed_at":"2026-05-23T18:48:20.256833Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"Secure transformer inference made non-interactive","venue":null,"work_id":"9c235e96-fc55-4a15-8b35-65d3e5ce70ca","year":2024},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":38,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:a7999d2b25112c0857523e6f4584a5a10a1cc6a172cb8c81502977f7b88769d0","observation_id":"fa326743-b741-493b-8575-3d4b33fc9100","resolution":{"observed_at":"2026-05-23T20:28:26.272431Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"6929.2023","doi":"10.1109/dac56929.2023","metadata_source":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-11T11:50:26.030339Z","title":"Primer: Fast private transformer inference on encrypted data","venue":null,"work_id":"0e9b26b6-bcd4-4167-aef3-dd2276f52950","year":2025},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":39,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:17ae4f8763475dfb911299c786642bc50b973a072717c98eff520abe0571541d","observation_id":"754a2c82-d6dd-4486-9649-3c781f308799","resolution":{"observed_at":"2026-05-23T18:48:19.438865Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":"10.1016/j.neucom.2019.06.004","metadata_source":"doi_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-06-30T22:45:06.763446Z","title":"Polynomial activation neural networks: Modeling, stability analysis and coverage bp-training","venue":null,"work_id":"ac76a4f3-cd53-470c-9c6d-216f7c1d331d","year":2019},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":40,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:d9dcaa27e6e3bf652810d20e558cf0c47a5a4789150547e02fffe5762c7d0579","observation_id":"fdb66201-2466-4f7f-9438-5f51786b0006","resolution":{"observed_at":"2026-05-23T18:48:19.479475Z","resolver_source":"doi","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":"Converting transformers to polynomial form for secure inference over homomorphic encryption","venue":null,"work_id":"6355906d-e4a3-47f4-bf56-d112d362ca1d","year":2024},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":41,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:7e55281e7ae7efaa09edabbe16c14934339fc5668abbe8b04e1f3f3581d44ef3","observation_id":"29beb3d7-ce4e-4128-bc02-14c1bd0ad61b","resolution":{"observed_at":"2026-05-23T20:28:26.330651Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-07-09T00:05:48.311803Z","title":"write newline","venue":null,"work_id":"8e5fda61-e601-4df4-8204-015bee341570","year":null},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":42,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:1845dd043801b35c9c088623029f2a8bf9805fe7ae4346726f7a85fb85153b63","observation_id":"27966717-c386-44bd-99f9-661b2d44b26d","resolution":{"observed_at":"2026-05-23T20:28:26.276276Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-07-08T23:55:44.012295Z","title":"@esa (Ref","venue":null,"work_id":"b058608d-98d0-4821-a4ae-403d2b7cd411","year":null},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":43,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:d289986b79c036551026343dd07ee9f72d946cab245287eedb3cd071b302f8d7","observation_id":"a1c38f2a-7613-48aa-8a01-587c471abc7a","resolution":{"observed_at":"2026-05-23T20:28:26.280468Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-07-09T00:05:48.337191Z","title":null,"venue":null,"work_id":"ea79bfb8-d434-45e9-8607-416d3839ec5c","year":null},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":44,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:1060b92703174a3278c155c01d29946f4dbc17c716d741134962f9fb57c015b3","observation_id":"8631747a-ca04-4c85-a001-77283afdf7ee","resolution":{"observed_at":"2026-05-23T20:28:26.268910Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+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":"2026-06-05T21:23:00.469572Z","title":null,"venue":null,"work_id":"d051d6aa-efca-49eb-a66d-8ea01bf21294","year":null},"citing_paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data","version":2},"reference_index":45,"source":"arxiv_source","source_observed_at":"2026-05-23T18:46:54.778974Z"},"links":{"citing_paper":"/paper/2410.09457"},"observation_digest":"sha256:bf99a55be8fb7ea12b729268f0f2f168a78c82731aaa118787ffd5ff142a29be","observation_id":"e98e2964-0354-49cb-b3c9-6e6057eaa51d","resolution":{"observed_at":"2026-05-24T07:46:05.419845Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-04T06:34:03.388597+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2410.09457","last_updated":"2026-05-05T16:59:31Z","latest_version":2,"primary_category":"cs.LG","snapshot_observed_at":"2026-07-06T19:32:15.226489Z","submitted_at":"2024-10-12T09:32:42Z","title":"Power-Softmax: Towards Secure LLM Inference over Encrypted Data"},"reference_resolution":{"displayed":45,"state_counts":{"malformed_identifier":0,"metadata_mismatch":3,"parse_uncertain":0,"unresolved":2,"verified_exact":21,"verified_fuzzy":19},"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":"2026-08-04T06:34:03.388597+00:00","source":"crossref"},{"observed_at":"2026-08-04T06:33:57.428241+00:00","source":"retraction_watch"}],"thesis":"As of 4 August 2026, this Paper Citation Record lists 45 of 45 outbound references and 3 inbound Pith citation observations for arXiv:2410.09457."}