{"as_of":"2026-08-10T23:18:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:23f9e6a594a84b352e10a5dc0772c31d07f7fd28c4a03a9137c6ee96c4992e45","coverage":[{"denominator":0,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":66,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":66,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-10T06:31:04.303077+00:00","state":"measured"},{"denominator":66,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":66,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-10T22:54:52.088349Z","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":"2026-07-04T04:19:34.480528Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2303.16199","last_updated":"2024-09-18T23:54:36Z","snapshot_observed_at":"2026-08-06T06:36:02.994951Z","submitted_at":"2023-03-28T17:59:12Z","title":"LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention","version":3},"reference_index":54,"source":"arxiv_source","source_observed_at":"2026-05-14T23:07:42.245641Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2303.16199"},"observation_digest":"sha256:3d939183111f36f5e01419c82e3e79a99c72d49132ca2fc4866ea3785f793318","observation_id":"2d025f0a-4988-45b2-964e-460c43f7df64","resolution":{"observed_at":"2026-05-14T23:07:42.945611Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2305.14233","last_updated":"2023-05-23T16:49:14Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2023-05-23T16:49:14Z","title":"Enhancing Chat Language Models by Scaling High-quality Instructional Conversations","version":1},"reference_index":166,"source":"arxiv_source","source_observed_at":"2026-05-15T17:25:07.730933Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2305.14233"},"observation_digest":"sha256:943c6b0267d5e58aba60a39015bae566f3177a51f7efb7d582549d26713014c6","observation_id":"60b6ca99-92b4-4522-9a10-28a09e79f809","resolution":{"observed_at":"2026-05-15T17:25:08.152440Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2309.08532","last_updated":"2025-05-01T11:56:52Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2023-09-15T16:50:09Z","title":"EvoPrompt: Connecting LLMs with Evolutionary Algorithms Yields Powerful Prompt Optimizers","version":3},"reference_index":117,"source":"arxiv_source","source_observed_at":"2026-05-16T06:11:49.475825Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2309.08532"},"observation_digest":"sha256:d7aa6593688edcd67af3672067009c45154731945996bc3ab2ce90282455e4ea","observation_id":"5cb4218d-275e-4a0f-8f93-7ad6741f5686","resolution":{"observed_at":"2026-05-16T06:11:49.581743Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_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},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-05-13T11:32:36.738536Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2403.14608"},"observation_digest":"sha256:05ca92636b14f6b96618e8ec783e087eed25cf39330e27d912c9b7d2d6a62233","observation_id":"b349b0ff-b100-4fb7-b390-5e40129a3cff","resolution":{"observed_at":"2026-05-13T11:32:37.143142Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-10T22:54:52.088349Z","title":"P-tuning v2: Prompt tuning can be comparable to fine- tuning universally across scales and tasks","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2501.00457","last_updated":"2024-12-31T14:13:28Z","snapshot_observed_at":"2026-08-10T22:48:21.728272Z","submitted_at":"2024-12-31T14:13:28Z","title":"Differentiable Prompt Learning for Vision Language Models","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-10T22:54:52.088349Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2501.00457"},"observation_digest":"sha256:19b2b48ad1697f543a251e5a84f2f030c8195c1052945abf61eac7f48b96ba78","observation_id":"986378c6-4c94-4366-b2fb-219c43dc95e7","resolution":{"observed_at":"2026-08-10T22:54:52.088349Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-10T22:43:11.702334Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2501.00999","last_updated":"2025-01-06T01:49:09Z","snapshot_observed_at":"2026-08-10T22:35:24.660670Z","submitted_at":"2025-01-02T01:33:58Z","title":"Exploring Information Processing in Large Language Models: Insights from Information Bottleneck Theory","version":2},"reference_index":2024,"source":"pdf_text","source_observed_at":"2026-08-10T22:43:11.702334Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2501.00999"},"observation_digest":"sha256:41e9379845b31ae4f6b2421cb4f50c1abe31f1145c5e9c78a437affc68077fdd","observation_id":"9b32496b-d7be-40f4-be0f-f7c99e42b244","resolution":{"observed_at":"2026-08-10T22:43:11.702334Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-10T22:17:56.127452Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Uni- versally Across Scales and Tasks","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2501.04040","last_updated":"2025-02-09T08:00:36Z","snapshot_observed_at":"2026-08-10T22:47:31.199700Z","submitted_at":"2025-01-03T21:04:49Z","title":"A Survey on Large Language Models with some Insights on their Capabilities and Limitations","version":2},"reference_index":193,"source":"pdf_text","source_observed_at":"2026-08-10T22:17:56.127452Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2501.04040"},"observation_digest":"sha256:da9cf02d5dc2d24fde9a3ff76194776a28649c3a69a0a7696e0b385dc41e0d8a","observation_id":"f6e8ac26-1d70-4ce9-bc97-45e85eeee92d","resolution":{"observed_at":"2026-08-10T22:17:56.127452Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-10T21:51:12.157702Z","title":"arXiv preprint arXiv:2110.07602 (2021)","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2501.04062","last_updated":"2025-01-07T10:39:14Z","snapshot_observed_at":"2026-08-10T21:47:01.576487Z","submitted_at":"2025-01-07T10:39:14Z","title":"ChronoLLM: A Framework for Customizing Large Language Model for Digital Twins generalization based on PyChrono","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-10T21:51:12.157702Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2501.04062"},"observation_digest":"sha256:e80d1ba46024afbada5c8e78409a617a7f53a41c69a9b399d7455ba34d8bd49b","observation_id":"069e3b93-c481-46ef-9e89-64ee2033c155","resolution":{"observed_at":"2026-08-10T21:51:12.157702Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-10T21:44:33.906829Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2501.04161","last_updated":"2025-01-07T22:19:15Z","snapshot_observed_at":"2026-08-10T21:37:47.594763Z","submitted_at":"2025-01-07T22:19:15Z","title":"KGIF: Optimizing Relation-Aware Recommendations with Knowledge Graph Information Fusion","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-10T21:44:33.906829Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2501.04161"},"observation_digest":"sha256:3678d834d512248e1c3747086f26a12c07a07d4417c6f3b5df0ce2208f70d513","observation_id":"8d8e37cb-3444-42da-bee4-3646679f7ea2","resolution":{"observed_at":"2026-08-10T21:44:33.906829Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-10T19:42:22.792842Z","title":"P- Tuning v2: Prompt Tuning Can Be Comparable to Fine -tuning Universally Across Scales and Tasks , March 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2501.09804","last_updated":"2025-01-16T19:23:11Z","snapshot_observed_at":"2026-08-10T20:21:04.164761Z","submitted_at":"2025-01-16T19:23:11Z","title":"Enhancing Generalization in Chain of Thought Reasoning for Smaller Models","version":1},"reference_index":22,"source":"arxiv_source","source_observed_at":"2026-08-10T19:42:22.792842Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2501.09804"},"observation_digest":"sha256:ac7250c9ab106d500af17e09c8e498ead89394e7f4796485e496330abe001fed","observation_id":"da9dc81e-7d42-42cb-abcf-e872de62fe1b","resolution":{"observed_at":"2026-08-10T19:42:22.792842Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-10T19:29:51.387161Z","title":null,"venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2501.10062","last_updated":"2025-07-21T10:51:55Z","snapshot_observed_at":"2026-08-10T19:21:21.209030Z","submitted_at":"2025-01-17T09:27:08Z","title":"OMoE: Diversifying Mixture of Low-Rank Adaptation by Orthogonal Finetuning","version":2},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-10T19:29:51.387161Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2501.10062"},"observation_digest":"sha256:3be9cd05916d6f9de9058e2303a3b1ad673271c545fb9d8482240848a3a4952b","observation_id":"650efc14-e2ab-44dc-b0f0-b6a3ac978b46","resolution":{"observed_at":"2026-08-10T19:29:51.387161Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-10T15:38:02.959155Z","title":"P- tuning v2: Prompt tuning can be comparable to fine- tuning universally across scales and tasks,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2501.13787","last_updated":"2025-01-23T16:04:23Z","snapshot_observed_at":"2026-08-10T19:15:37.342583Z","submitted_at":"2025-01-23T16:04:23Z","title":"Parameter-Efficient Fine-Tuning for Foundation Models","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-10T15:38:02.959155Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2501.13787"},"observation_digest":"sha256:2b7b1f0042b34f9f30ae8bc20ac428330e3461564f84b7240de7b9fa7d94ee2e","observation_id":"da22df74-cfb0-4e61-813e-2e8ec64339d6","resolution":{"observed_at":"2026-08-10T15:38:02.959155Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-10T14:59:48.297273Z","title":"arXiv preprint arXiv:2110.07602","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2501.14713","last_updated":"2025-01-31T17:38:07Z","snapshot_observed_at":"2026-08-10T15:15:05.490321Z","submitted_at":"2025-01-24T18:46:37Z","title":"FlexiGPT: Pruning and Extending Large Language Models with Low-Rank Weight Sharing","version":2},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-10T14:59:48.297273Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2501.14713"},"observation_digest":"sha256:771a6f21c3b0fa827d87f8f56610ea23c4da685cc4779d8c49a40f9fab7c9df0","observation_id":"b0ff9be9-71d9-456f-a40a-8d06e0fb6c8e","resolution":{"observed_at":"2026-08-10T14:59:48.297273Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-09T19:36:17.821607Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks, 2022","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2502.00311","last_updated":"2025-02-01T04:18:28Z","snapshot_observed_at":"2026-08-10T12:39:07.648073Z","submitted_at":"2025-02-01T04:18:28Z","title":"Sparse Gradient Compression for Fine-Tuning Large Language Models","version":1},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-08-09T19:36:17.821607Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2502.00311"},"observation_digest":"sha256:0979ead70b42b162bd22b2b3d50bfec1ffc5743fcfc35f719338748b567a606e","observation_id":"b90e963b-3618-44b8-9275-5b7cd3163ca9","resolution":{"observed_at":"2026-08-09T19:36:17.821607Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-09T17:43:56.131056Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2502.00796","last_updated":"2025-02-02T13:29:44Z","snapshot_observed_at":"2026-08-10T12:39:06.292362Z","submitted_at":"2025-02-02T13:29:44Z","title":"Task-Specific Adaptation with Restricted Model Access","version":1},"reference_index":24,"source":"arxiv_source","source_observed_at":"2026-08-09T17:43:56.131056Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2502.00796"},"observation_digest":"sha256:bc965c1f158625dc87be0ba8042f381a2fcc93bb25a0bce9a47416a6bafe8bea","observation_id":"ed318721-8664-4c3f-be97-4da3dd11195f","resolution":{"observed_at":"2026-08-09T17:43:56.131056Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-09T14:02:48.432870Z","title":"L., Du, Z., Yang, Z., and Tang, J","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2502.01922","last_updated":"2025-02-04T01:42:45Z","snapshot_observed_at":"2026-08-09T20:13:52.432148Z","submitted_at":"2025-02-04T01:42:45Z","title":"LAST SToP For Modeling Asynchronous Time Series","version":1},"reference_index":33,"source":"arxiv_source","source_observed_at":"2026-08-09T14:02:48.432870Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2502.01922"},"observation_digest":"sha256:dd19fb0e5050c2b957def46006c00ff6efbc1ab51704a88beff5cfaee8032f2a","observation_id":"5648458b-0a06-434e-b974-042e1960c98d","resolution":{"observed_at":"2026-08-09T14:02:48.432870Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-09T10:14:21.515991Z","title":"L., Du, Z., Yang, Z., and Tang, J","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2502.03029","last_updated":"2025-06-18T12:24:26Z","snapshot_observed_at":"2026-08-10T16:11:53.988251Z","submitted_at":"2025-02-05T09:31:27Z","title":"On Zero-Initialized Attention: Optimal Prompt and Gating Factor Estimation","version":3},"reference_index":32,"source":"arxiv_source","source_observed_at":"2026-08-09T10:14:21.515991Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2502.03029"},"observation_digest":"sha256:c03622919d35ad1899acd90b23b5c57527ce9d100af0a0941d8b385d5dcf4c8b","observation_id":"622509aa-0c86-482b-91bf-20d1cac6ab3d","resolution":{"observed_at":"2026-08-09T10:14:21.515991Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-08T12:20:08.343457Z","title":"P- tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2502.07855","last_updated":"2025-06-13T12:20:30Z","snapshot_observed_at":"2026-08-08T16:43:44.128026Z","submitted_at":"2025-02-11T14:04:43Z","title":"Vision-Language Models for Edge Networks: A Comprehensive Survey","version":2},"reference_index":69,"source":"pdf_text","source_observed_at":"2026-08-08T12:20:08.343457Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2502.07855"},"observation_digest":"sha256:22ad5d5410d97d42cbcca2b548b69fa7cb2a6c2b03bbccda507048463c7be989","observation_id":"72361f5c-6d6c-45d2-8789-4822f0e44581","resolution":{"observed_at":"2026-08-08T12:20:08.343457Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-07T15:42:38.828658Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2505.14165","last_updated":"2025-05-20T10:18:10Z","snapshot_observed_at":"2026-08-10T01:53:38.247579Z","submitted_at":"2025-05-20T10:18:10Z","title":"PL-FGSA: A Prompt Learning Framework for Fine-Grained Sentiment Analysis Based on MindSpore","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-07T15:42:38.828658Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2505.14165"},"observation_digest":"sha256:15a289e81507434e5bbe166a6e4d5e28131841fd8e3d2c4b42d8619d3615c90d","observation_id":"96b9ea31-28f3-49fb-ab1f-a90b353d5728","resolution":{"observed_at":"2026-08-07T15:42:38.828658Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-07T15:41:09.118021Z","title":"arXiv preprint arXiv:2110.07602","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2505.14350","last_updated":"2025-05-21T04:25:30Z","snapshot_observed_at":"2026-08-09T16:56:51.460763Z","submitted_at":"2025-05-20T13:34:06Z","title":"OSoRA: Output-Dimension and Singular-Value Initialized Low-Rank Adaptation","version":2},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-07T15:41:09.118021Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2505.14350"},"observation_digest":"sha256:f5bd251cbe34239016107cec9b27bf408c1f02fec204a76991d13f37b3a7d1fb","observation_id":"6afa0470-86ed-43e0-9d89-d52df7cd75bc","resolution":{"observed_at":"2026-08-07T15:41:09.118021Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-07T15:43:40.410233Z","title":null,"venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2505.14742","last_updated":"2025-05-29T22:04:36Z","snapshot_observed_at":"2026-08-09T14:33:06.239071Z","submitted_at":"2025-05-20T07:19:36Z","title":"Quaff: Quantized Parameter-Efficient Fine-Tuning under Outlier Spatial Stability Hypothesis","version":2},"reference_index":46,"source":"arxiv_source","source_observed_at":"2026-08-07T15:43:40.410233Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2505.14742"},"observation_digest":"sha256:8b1942abf7760ba13ad4867bf3508abe6c34906545232868a6d2d29d8f4b7081","observation_id":"9c338ff4-93e0-4a9a-86cf-127d4475a1a2","resolution":{"observed_at":"2026-08-07T15:43:40.410233Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-07T14:24:06.627542Z","title":null,"venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2505.19015","last_updated":"2025-08-08T09:55:18Z","snapshot_observed_at":"2026-08-10T09:36:29.870242Z","submitted_at":"2025-05-25T07:37:34Z","title":"Can Multimodal Large Language Models Understand Spatial Relations?","version":2},"reference_index":26,"source":"arxiv_source","source_observed_at":"2026-08-07T14:24:06.627542Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2505.19015"},"observation_digest":"sha256:3e923b6e329eedd10cda5a34f1cb05cc776922379eeab3548339d6291215b433","observation_id":"8fbdc9b1-e232-4dd3-807e-8d2afa261895","resolution":{"observed_at":"2026-08-07T14:24:06.627542Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-07T13:47:18.763370Z","title":"P-tuning v2: Prompt tuning can be com- parable to fine-tuning universally across scales and tasks","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2505.20941","last_updated":"2025-05-27T09:27:16Z","snapshot_observed_at":"2026-08-10T13:51:34.033455Z","submitted_at":"2025-05-27T09:27:16Z","title":"PMA: Towards Parameter-Efficient Point Cloud Understanding via Point Mamba Adapter","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-07T13:47:18.763370Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2505.20941"},"observation_digest":"sha256:06254203afb98bdb0b4fc27b8573a330d20846f0ed9896d650e777abfbaa9dfe","observation_id":"d879fceb-3210-431b-a7d2-189c9cb0e932","resolution":{"observed_at":"2026-08-07T13:47:18.763370Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-07T13:14:06.400170Z","title":null,"venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2505.22447","last_updated":"2025-05-28T15:09:56Z","snapshot_observed_at":"2026-08-09T16:54:23.434235Z","submitted_at":"2025-05-28T15:09:56Z","title":"Privacy-preserving Prompt Personalization in Federated Learning for Multimodal Large Language Models","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-07T13:14:06.400170Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2505.22447"},"observation_digest":"sha256:21d7c97fb1869d7e5579c67708ab26dec272c9fb91f5f58995034f3a9e751140","observation_id":"1dd4391f-a4f8-4116-8c05-ab7c0efc029b","resolution":{"observed_at":"2026-08-07T13:14:06.400170Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-07T12:43:56.805456Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2505.23694","last_updated":"2025-06-01T16:02:18Z","snapshot_observed_at":"2026-08-10T12:38:53.495879Z","submitted_at":"2025-05-29T17:31:26Z","title":"DA-VPT: Semantic-Guided Visual Prompt Tuning for Vision Transformers","version":2},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-07T12:43:56.805456Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2505.23694"},"observation_digest":"sha256:6dea7891401dc9c2c4e313a45c943aff0aa62caa6b665ab7fce8552288a92241","observation_id":"59ffbfd6-90e6-462a-ac1c-d60ac9294238","resolution":{"observed_at":"2026-08-07T12:43:56.805456Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-07T12:09:14.108874Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.00495","last_updated":"2025-05-31T10:27:08Z","snapshot_observed_at":"2026-08-09T08:51:32.281143Z","submitted_at":"2025-05-31T10:27:08Z","title":"FLoE: Fisher-Based Layer Selection for Efficient Sparse Adaptation of Low-Rank Experts","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-07T12:09:14.108874Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2506.00495"},"observation_digest":"sha256:9c9992c2a533bb115c1acab8e53d466a01f9dc769b48417ea94e6cf3f69dad84","observation_id":"63c044d9-1594-43df-a25e-72eaea3b12f8","resolution":{"observed_at":"2026-08-07T12:09:14.108874Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-07T12:05:52.119348Z","title":null,"venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.00743","last_updated":"2025-05-31T23:09:26Z","snapshot_observed_at":"2026-08-10T11:18:14.899210Z","submitted_at":"2025-05-31T23:09:26Z","title":"Assortment of Attention Heads: Accelerating Federated PEFT with Head Pruning and Strategic Client Selection","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-07T12:05:52.119348Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2506.00743"},"observation_digest":"sha256:051eaaaa7bd02c544b049b79027ee99a301b7c0ccf962dc00a715a0deb16cce2","observation_id":"2c06e49b-2c25-4601-ae23-3d070352ab43","resolution":{"observed_at":"2026-08-07T12:05:52.119348Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-07T11:23:47.471962Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.02696","last_updated":"2025-06-03T09:44:28Z","snapshot_observed_at":"2026-08-09T15:46:04.387261Z","submitted_at":"2025-06-03T09:44:28Z","title":"Shaking to Reveal: Perturbation-Based Detection of LLM Hallucinations","version":1},"reference_index":77,"source":"pdf_text","source_observed_at":"2026-08-07T11:23:47.471962Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2506.02696"},"observation_digest":"sha256:66a810aac695770b39271b0364353fcc9fb1704a0804d85a95e0cc699a6f3820","observation_id":"cf273c16-e974-4c52-8d50-b417d4ebbcf9","resolution":{"observed_at":"2026-08-07T11:23:47.471962Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-07T11:20:30.265155Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks.CoRR, abs/2110.07602, 2021","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.02847","last_updated":"2025-06-03T13:16:00Z","snapshot_observed_at":"2026-08-10T19:33:42.673165Z","submitted_at":"2025-06-03T13:16:00Z","title":"CLONE: Customizing LLMs for Efficient Latency-Aware Inference at the Edge","version":1},"reference_index":89,"source":"pdf_text","source_observed_at":"2026-08-07T11:20:30.265155Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2506.02847"},"observation_digest":"sha256:aa26b60a1821387d162e49cf1440a2a0e8a97dbd3367ba5a59728f99431f0413","observation_id":"98c13187-fbe6-45f1-a899-603fbd6808fd","resolution":{"observed_at":"2026-08-07T11:20:30.265155Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-07T05:33:52.580547Z","title":null,"venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.07583","last_updated":"2025-06-09T09:27:00Z","snapshot_observed_at":"2026-08-10T21:19:05.982732Z","submitted_at":"2025-06-09T09:27:00Z","title":"Beyond the Sentence: A Survey on Context-Aware Machine Translation with Large Language Models","version":1},"reference_index":54,"source":"arxiv_source","source_observed_at":"2026-08-07T05:33:52.580547Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2506.07583"},"observation_digest":"sha256:5ffc77a50d38b8d846fb176e3081113c63156246575365e7c8de7fdd8ac721a4","observation_id":"41675e70-fa0f-4496-8008-27eb014273b7","resolution":{"observed_at":"2026-08-07T05:33:52.580547Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2506.13674","last_updated":"2026-04-19T03:53:58Z","snapshot_observed_at":"2026-08-02T17:06:46.370930Z","submitted_at":"2025-06-16T16:30:26Z","title":"PrefixMemory-Tuning: Modernizing Prefix-Tuning by Decoupling the Prefix from Attention","version":3},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-05-19T09:24:04.289799Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2506.13674"},"observation_digest":"sha256:58d201fa24cc97fc9d016d33538530ee3f16566c15eaa3c42aeeb2bc04d2e500","observation_id":"00d6884a-389f-47a9-9a4b-a47bbc0c35aa","resolution":{"observed_at":"2026-05-19T09:27:14.633334Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-07T00:32:08.307818Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.13750","last_updated":"2025-06-16T17:56:22Z","snapshot_observed_at":"2026-08-09T16:56:18.576968Z","submitted_at":"2025-06-16T17:56:22Z","title":"Test3R: Learning to Reconstruct 3D at Test Time","version":1},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-07T00:32:08.307818Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2506.13750"},"observation_digest":"sha256:d61e5d69066a5373f46bb533417f73e3f34562573b5ca2332f95225f3e52027a","observation_id":"712e8857-7bed-48e4-a9ec-92e57aa3073c","resolution":{"observed_at":"2026-08-07T00:32:08.307818Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-06T22:43:56.429128Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.21119","last_updated":"2025-06-26T09:37:15Z","snapshot_observed_at":"2026-08-06T22:30:28.321862Z","submitted_at":"2025-06-26T09:37:15Z","title":"Progtuning: Progressive Fine-tuning Framework for Transformer-based Language Models","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-06T22:43:56.429128Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2506.21119"},"observation_digest":"sha256:181785fad7946ac79f388d719cdf8719f9a810c1c90029b78dca6ab7a71f840f","observation_id":"20309761-7744-4cec-a52f-8d48e61ff41e","resolution":{"observed_at":"2026-08-06T22:43:56.429128Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-06T21:32:23.803598Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2506.23940","last_updated":"2025-07-01T03:26:52Z","snapshot_observed_at":"2026-08-07T22:13:54.563047Z","submitted_at":"2025-06-30T15:07:41Z","title":"Graft: Integrating the Domain Knowledge via Efficient Parameter Synergy for MLLMs","version":2},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-06T21:32:23.803598Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2506.23940"},"observation_digest":"sha256:9a35b973c8b15e7e34366eac4a208b0eefb59ce4b37f28b0d11c0706a82f221c","observation_id":"6fd8c473-3fda-48ae-81fc-ca8731ef683a","resolution":{"observed_at":"2026-08-06T21:32:23.803598Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-06T20:58:59.143355Z","title":null,"venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2507.01438","last_updated":"2025-07-02T07:47:28Z","snapshot_observed_at":"2026-08-10T12:39:07.196333Z","submitted_at":"2025-07-02T07:47:28Z","title":"EdgeLoRA: An Efficient Multi-Tenant LLM Serving System on Edge Devices","version":1},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-06T20:58:59.143355Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2507.01438"},"observation_digest":"sha256:a33dbf960c926f87e2b2c2b217431559b9ad0421f92b6035ef40dd75321a899c","observation_id":"571983aa-256e-4f56-b96c-68a11dfe7ae9","resolution":{"observed_at":"2026-08-06T20:58:59.143355Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-06T20:21:22.152771Z","title":"Liu, et al., ”P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks,” arXiv preprint arXiv:2110.07602, 2021","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2507.03223","last_updated":"2025-07-03T23:44:50Z","snapshot_observed_at":"2026-08-06T23:56:27.012210Z","submitted_at":"2025-07-03T23:44:50Z","title":"SI-Agent: An Agentic Framework for Feedback-Driven Generation and Tuning of Human-Readable System Instructions for Large Language Models","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-06T20:21:22.152771Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2507.03223"},"observation_digest":"sha256:a71f2eb5138cf01fe9d2499acdd22137fafef0dbc8290a0eac95e59f81eff82f","observation_id":"fe6fb1cd-7926-46e0-9556-4034110176e4","resolution":{"observed_at":"2026-08-06T20:21:22.152771Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-06T20:04:13.722971Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2507.03916","last_updated":"2025-07-26T08:13:41Z","snapshot_observed_at":"2026-08-09T05:08:37.215164Z","submitted_at":"2025-07-05T06:16:31Z","title":"Animation Needs Attention: A Holistic Approach to Slides Animation Comprehension with Visual-Language Models","version":3},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-06T20:04:13.722971Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2507.03916"},"observation_digest":"sha256:497f656ab5633e0bb64d809607767f500ad0ab649870e8e39670e3076a239ac4","observation_id":"516f1b29-e1f9-4c7b-89f5-87d015388132","resolution":{"observed_at":"2026-08-06T20:04:13.722971Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-06T18:12:26.565697Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2507.09132","last_updated":"2025-07-12T04:12:24Z","snapshot_observed_at":"2026-08-09T16:56:55.140311Z","submitted_at":"2025-07-12T04:12:24Z","title":"Heterogeneous Graph Prompt Learning via Adaptive Weight Pruning","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-06T18:12:26.565697Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2507.09132"},"observation_digest":"sha256:9a621e32ad9e85a324877f370a3984eb678e06902b8afd4ec501465977cf82fd","observation_id":"23628d81-c06b-4c72-8ecc-2009a00ec602","resolution":{"observed_at":"2026-08-06T18:12:26.565697Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-06T17:09:12.694490Z","title":null,"venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2507.11677","last_updated":"2025-07-15T19:32:04Z","snapshot_observed_at":"2026-08-10T20:51:11.270191Z","submitted_at":"2025-07-15T19:32:04Z","title":"CLAImate: AI-Enabled Climate Change Communication through Personalized and Localized Narrative Visualizations","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-06T17:09:12.694490Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2507.11677"},"observation_digest":"sha256:da516de259d83306defe5b06d064c6e0f6cbe381b73bf945f2e999176aeab5c7","observation_id":"a2c48fea-3174-4aff-b175-84f878d74b52","resolution":{"observed_at":"2026-08-06T17:09:12.694490Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-06T18:01:27.980962Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2507.15864","last_updated":"2025-07-13T07:16:08Z","snapshot_observed_at":"2026-08-10T15:09:38.586360Z","submitted_at":"2025-07-13T07:16:08Z","title":"Adversarial Demonstration Learning for Low-resource NER Using Dual Similarity","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-06T18:01:27.980962Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2507.15864"},"observation_digest":"sha256:48f7d9c01380f8c7aeadfffb112b2f7dbd77fabe0ce3291bafa8c3aad7d4cb9a","observation_id":"47933b35-c329-4c6f-8bac-e1a69b118de9","resolution":{"observed_at":"2026-08-06T18:01:27.980962Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-06T15:32:51.587979Z","title":"P- tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2507.15905","last_updated":"2025-07-21T12:01:04Z","snapshot_observed_at":"2026-08-09T01:30:02.567154Z","submitted_at":"2025-07-21T12:01:04Z","title":"Foundation Models and Transformers for Anomaly Detection: A Survey","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-06T15:32:51.587979Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2507.15905"},"observation_digest":"sha256:e7cdb8f2f91c90ba0e3e5e67edf23e3e5b44bae35c4ad529d4b27272c70b78b2","observation_id":"65dce568-9abe-404f-9b29-ccb3305d24c5","resolution":{"observed_at":"2026-08-06T15:32:51.587979Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-06T00:04:32.371883Z","title":"P- tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2508.04472","last_updated":"2025-08-06T14:19:32Z","snapshot_observed_at":"2026-08-09T06:08:47.407244Z","submitted_at":"2025-08-06T14:19:32Z","title":"Zero-Residual Concept Erasure via Progressive Alignment in Text-to-Image Model","version":1},"reference_index":2021,"source":"pdf_text","source_observed_at":"2026-08-06T00:04:32.371883Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2508.04472"},"observation_digest":"sha256:d8571af7a9240ce71f23994ac31629b47123b354d0db0edbd858305b29fe209d","observation_id":"f48d25c0-f060-4fb6-8667-026af24bb107","resolution":{"observed_at":"2026-08-06T00:04:32.371883Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-05T23:10:14.775430Z","title":null,"venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2508.05987","last_updated":"2025-08-08T03:43:01Z","snapshot_observed_at":"2026-08-09T20:42:24.085057Z","submitted_at":"2025-08-08T03:43:01Z","title":"Adversarial Topic-aware Prompt-tuning for Cross-topic Automated Essay Scoring","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-05T23:10:14.775430Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2508.05987"},"observation_digest":"sha256:371055be4f18bbb5fe382030290d5f638efb69679040aae6e9ecaa5be2bfef72","observation_id":"79e3a429-aee6-46e7-9d5c-ab126569d40a","resolution":{"observed_at":"2026-08-05T23:10:14.775430Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-05T21:45:22.786563Z","title":"P-tuning v2: Prompt tuning can be comparable to fine- tuning universally across scales and tasks,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2508.08095","last_updated":"2025-08-11T15:33:44Z","snapshot_observed_at":"2026-08-07T16:13:39.029257Z","submitted_at":"2025-08-11T15:33:44Z","title":"Dual Information Speech Language Models for Emotional Conversations","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-05T21:45:22.786563Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2508.08095"},"observation_digest":"sha256:43e20c16daa85955ce7dc57fc3619ee9d9745967a498b82096471e68f2c27d86","observation_id":"246a53c1-09e4-4d34-81bb-321ee383cd53","resolution":{"observed_at":"2026-08-05T21:45:22.786563Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-05T05:12:44.230741Z","title":null,"venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2509.05714","last_updated":"2025-09-06T13:26:04Z","snapshot_observed_at":"2026-08-06T23:54:59.571872Z","submitted_at":"2025-09-06T13:26:04Z","title":"Towards Meta-Cognitive Knowledge Editing for Multimodal LLMs","version":1},"reference_index":36,"source":"arxiv_source","source_observed_at":"2026-08-05T05:12:44.230741Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2509.05714"},"observation_digest":"sha256:a8b9dbefb4189bc6d9d23024d45e792cbab5b00fc0d2e0ce3e81073b77ee86bd","observation_id":"d8806181-bed9-46ad-8da7-907b3f5d7ed7","resolution":{"observed_at":"2026-08-05T05:12:44.230741Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-04T23:55:38.963496Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2509.06282","last_updated":"2025-09-08T02:06:37Z","snapshot_observed_at":"2026-08-07T21:45:49.624401Z","submitted_at":"2025-09-08T02:06:37Z","title":"AI-driven Remote Facial Skin Hydration and TEWL Assessment from Selfie Images: A Systematic Solution","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-04T23:55:38.963496Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2509.06282"},"observation_digest":"sha256:e048098d22bb590486579aa51d2461cc40b68171d2a877725d31e871bbff83b3","observation_id":"54b3679d-f14a-4f55-9afc-712946621397","resolution":{"observed_at":"2026-08-04T23:55:38.963496Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-04T15:18:04.834479Z","title":"P- tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2509.20623","last_updated":"2026-06-01T22:34:49Z","snapshot_observed_at":"2026-08-09T16:56:16.107549Z","submitted_at":"2025-09-24T23:58:23Z","title":"Latent Activation Editing: Inference-Time Refinement of Learned Policies for Safer Multirobot Navigation","version":2},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-04T15:18:04.834479Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2509.20623"},"observation_digest":"sha256:d4bc987c2d9952cd55d5e690b420da5bb48d278a013db783c663b26997dbde01","observation_id":"f7a0dd14-83c8-42b5-81f4-48beb1a547bb","resolution":{"observed_at":"2026-08-04T15:18:04.834479Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-04T11:23:50.599087Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2510.05363","last_updated":"2026-06-04T21:50:20Z","snapshot_observed_at":"2026-08-10T11:35:22.942093Z","submitted_at":"2025-10-06T20:41:43Z","title":"MHA-RAG: Improving Efficiency, Accuracy, and Consistency by Encoding Exemplars as Soft Prompts","version":2},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-04T11:23:50.599087Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2510.05363"},"observation_digest":"sha256:1768ece1b013049595ee707932d7b2be6d0eabcd56e469cac7645996128956d1","observation_id":"bf580ff8-8d8b-4563-89b2-e62c63e54c60","resolution":{"observed_at":"2026-08-04T11:23:50.599087Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-15T12:12:51.663341Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2603.09118","last_updated":"2026-03-10T02:54:37Z","snapshot_observed_at":"2026-08-08T06:22:35.382329Z","submitted_at":"2026-03-10T02:54:37Z","title":"Meta-cavity Quantum Electrodynamics","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-07-15T12:12:51.663341Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2603.09118"},"observation_digest":"sha256:0455d9333ad3bc088260f524b14ba77813d73088f5316e22721ade061bc94aaf","observation_id":"9c604c66-1b7c-4579-96a5-4f84752a8ab4","resolution":{"observed_at":"2026-07-15T12:12:51.663341Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2604.09034","last_updated":"2026-04-10T06:52:58Z","snapshot_observed_at":"2026-07-06T22:57:59.555972Z","submitted_at":"2026-04-10T06:52:58Z","title":"The nextAI Solution to the NeurIPS 2023 LLM Efficiency Challenge","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-05-10T17:52:30.463227Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2604.09034"},"observation_digest":"sha256:8e321591520dc4cb61b79225d6af17264dac1c249118c9cbc742a9d9d5d4f6e3","observation_id":"87fa7bf1-4e7c-4a66-9636-8254a3f9378d","resolution":{"observed_at":"2026-05-11T05:55:59.394134Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2604.09532","last_updated":"2026-04-10T17:48:56Z","snapshot_observed_at":"2026-08-05T22:02:44.771312Z","submitted_at":"2026-04-10T17:48:56Z","title":"Seeing is Believing: Robust Vision-Guided Cross-Modal Prompt Learning under Label Noise","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-05-10T17:15:00.583884Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2604.09532"},"observation_digest":"sha256:001be1651b421159aa4065625536288868fd324724871d0182af8abab65f2c2a","observation_id":"a358a198-0bf9-4e7f-8106-33dce3a46733","resolution":{"observed_at":"2026-05-11T07:16:02.274020Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2604.12610","last_updated":"2026-04-14T11:36:29Z","snapshot_observed_at":"2026-08-02T05:17:11.221332Z","submitted_at":"2026-04-14T11:36:29Z","title":"Transforming External Knowledge into Triplets for Enhanced Retrieval in RAG of LLMs","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-05-10T15:42:59.063869Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2604.12610"},"observation_digest":"sha256:5f4ebdab3bf5afe410f5e69b8f83bae7eeaa2f5fa4936e72c2c12c1b281a2470","observation_id":"007153b7-7c67-43ba-bcd2-955a370607c5","resolution":{"observed_at":"2026-05-11T09:56:05.573047Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2604.18124","last_updated":"2026-04-20T11:43:55Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2026-04-20T11:43:55Z","title":"TLoRA: Task-aware Low Rank Adaptation of Large Language Models","version":1},"reference_index":26,"source":"arxiv_source","source_observed_at":"2026-05-10T05:07:10.885133Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2604.18124"},"observation_digest":"sha256:13c31140e9ba70a7a6d86c18269fd15f84c54a25ca4bb60e4a9df1b100407dbf","observation_id":"071f4d27-c8d3-4d75-b1bf-096024853ed6","resolution":{"observed_at":"2026-05-10T09:48:48.190336Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2604.19015","last_updated":"2026-04-21T03:06:24Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2026-04-21T03:06:24Z","title":"FedProxy: Federated Fine-Tuning of LLMs via Proxy SLMs and Heterogeneity-Aware Fusion","version":1},"reference_index":46,"source":"arxiv_source","source_observed_at":"2026-05-10T02:35:40.593397Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2604.19015"},"observation_digest":"sha256:fdb41a247c9036130267c55a2d713364e96c5cb7bc463b5acbf0c6361304382a","observation_id":"1fc255c9-1f44-499b-ade7-c951b23cbab4","resolution":{"observed_at":"2026-05-11T12:56:06.347783Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2605.14055","last_updated":"2026-05-13T19:25:56Z","snapshot_observed_at":"2026-08-01T07:49:53.576982Z","submitted_at":"2026-05-13T19:25:56Z","title":"PEML: Parameter-efficient Multi-Task Learning with Optimized Continuous Prompts","version":1},"reference_index":84,"source":"arxiv_source","source_observed_at":"2026-05-15T05:25:38.967645Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2605.14055"},"observation_digest":"sha256:b34ecfa8dda3d323f2d7539d1ba5677249328f94e4a62bbe09bee5e4284f8f99","observation_id":"14b05ce8-e8a2-436b-a75d-58efe2c1a945","resolution":{"observed_at":"2026-05-15T05:29:48.175330Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2606.00132","last_updated":"2026-05-28T21:22:31Z","snapshot_observed_at":"2026-07-06T23:40:51.363776Z","submitted_at":"2026-05-28T21:22:31Z","title":"Foundation-Preserving Adaptation via Generalized Rayleigh-Quotient Optimization","version":1},"reference_index":47,"source":"arxiv_source","source_observed_at":"2026-06-29T08:39:01.249243Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2606.00132"},"observation_digest":"sha256:0f2d0e2e6a8d4e3e73780f74eb3ea057e8be5c5b3cde094eba720b8fad052013","observation_id":"891fc892-3191-4915-81bc-8be160aaf4c5","resolution":{"observed_at":"2026-06-29T08:43:15.303977Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2606.00776","last_updated":"2026-05-30T15:44:36Z","snapshot_observed_at":"2026-07-06T23:41:24.911421Z","submitted_at":"2026-05-30T15:44:36Z","title":"Latent Diffusion Pretraining for Crystal Property Prediction","version":1},"reference_index":41,"source":"arxiv_source","source_observed_at":"2026-06-28T19:30:38.954152Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2606.00776"},"observation_digest":"sha256:0449dad181c688ca65d6ef106eb6680a33af13156a2d3adb9f9e0642f54c386d","observation_id":"95d7c5db-79a8-4174-a240-45c1701b45d3","resolution":{"observed_at":"2026-06-28T19:32:34.861716Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2606.02624","last_updated":"2026-05-29T12:12:08Z","snapshot_observed_at":"2026-07-06T23:42:58.036516Z","submitted_at":"2026-05-29T12:12:08Z","title":"TadA-Bench: A Million-Variant Benchmark for Future-Round Discovery Toward Agentic Protein Engineering","version":1},"reference_index":10,"source":"arxiv_source","source_observed_at":"2026-06-28T20:01:10.638647Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2606.02624"},"observation_digest":"sha256:46d041b44cc990b121f3a0a8348db46aab6f25f197b1cf8ae851326cb69ab0bc","observation_id":"6005146d-aaa3-47de-aa04-3d9b89a1d15a","resolution":{"observed_at":"2026-06-28T20:12:37.866501Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2606.09125","last_updated":"2026-06-08T07:19:42Z","snapshot_observed_at":"2026-07-06T23:48:30.569726Z","submitted_at":"2026-06-08T07:19:42Z","title":"Unveiling Privacy Risks in Multi-modal Large Language Models: Task-specific Vulnerabilities and Mitigation Challenges","version":1},"reference_index":30,"source":"arxiv_source","source_observed_at":"2026-06-27T16:33:28.848573Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2606.09125"},"observation_digest":"sha256:8471f25bcbefa13d55e9c5b15b13689bf72b12cd6ff4735cff4ca11f8e0388d9","observation_id":"5a46bd96-336f-4cc5-ae02-065db55e53c6","resolution":{"observed_at":"2026-07-03T01:27:31.059492Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2606.09401","last_updated":"2026-06-08T12:21:02Z","snapshot_observed_at":"2026-07-06T23:48:44.159537Z","submitted_at":"2026-06-08T12:21:02Z","title":"Benchmarking Empirical Privacy Protection for Adaptations of Large Language Models","version":1},"reference_index":187,"source":"arxiv_source","source_observed_at":"2026-06-27T17:13:46.335347Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2606.09401"},"observation_digest":"sha256:613f7d1ec460d0782bac3411176d2637d49b55a5a9cdd2f067be86c1f8b35f5e","observation_id":"9b2c5fce-389d-41ec-bdbd-d40036f4d9f7","resolution":{"observed_at":"2026-07-03T00:27:29.531190Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2606.10610","last_updated":"2026-06-09T09:11:33Z","snapshot_observed_at":"2026-08-02T15:21:45.577242Z","submitted_at":"2026-06-09T09:11:33Z","title":"Small Data, Big Noise: Adversarial Training for Robust Parameter-Efficient Fine-Tuning","version":1},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-06-27T13:20:38.336401Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2606.10610"},"observation_digest":"sha256:9c7d7f8be4380abd98b8096321e0ee7a2681f1b2d28960babe0143493a89b4e1","observation_id":"8034e942-66b4-49d8-81f1-632387bb9165","resolution":{"observed_at":"2026-07-03T05:17:40.244105Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2606.20333","last_updated":"2026-06-18T15:04:47Z","snapshot_observed_at":"2026-07-06T23:55:29.991442Z","submitted_at":"2026-06-18T15:04:47Z","title":"SoftSkill: Behavioral Compression for Contextual Adaptation","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-06-26T17:05:35.902167Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2606.20333"},"observation_digest":"sha256:99d23800d903423bdfbb332d9bc816b84e4615f8c260edd855502967fcfe5300","observation_id":"b8e94ae4-4afd-4bd8-8b8a-ad789c82bedd","resolution":{"observed_at":"2026-07-04T04:19:34.482854Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":"2110.07602","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-07-04T04:19:34.480528Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":"ceb357a3-8497-47d1-ae04-4e844ec59db5","year":2021},"citing_paper":{"arxiv_id":"2606.30937","last_updated":"2026-06-29T21:40:10Z","snapshot_observed_at":"2026-08-07T16:41:17.055343Z","submitted_at":"2026-06-29T21:40:10Z","title":"No Adaptation Without Observation: Observability-Constrained Test-Time Prompt Tuning for LiDAR Semantic Segmentation","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-07-01T01:42:08.308441Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2606.30937"},"observation_digest":"sha256:be83990deb65bbf54043c46a9ea525c0640d998dd3a54284be1e3b6843d2d291","observation_id":"e24ee796-9c62-4685-8b6b-dafb5b0b2652","resolution":{"observed_at":"2026-07-01T12:55:43.918179Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-02T05:16:36.094425Z","title":"P-tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.13425","last_updated":"2026-07-15T04:03:24Z","snapshot_observed_at":"2026-08-03T17:12:53.669172Z","submitted_at":"2026-07-15T04:03:24Z","title":"Data-Efficient Adaptation of LLMs via Attention Head Reweighting","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-02T05:16:36.094425Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2607.13425"},"observation_digest":"sha256:54dc171df55df8ffc67474bad48015631fb81e11238ef7666c0bc85353f609ee","observation_id":"5b3c18af-c98d-4cee-b9db-e189de7098a7","resolution":{"observed_at":"2026-08-02T05:16:36.094425Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-02T09:51:03.925992Z","title":"arXiv preprint arXiv:2110.07602 , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.16252","last_updated":"2026-06-27T03:25:36Z","snapshot_observed_at":"2026-08-08T02:31:05.095460Z","submitted_at":"2026-06-27T03:25:36Z","title":"SOS-LoRA: Static Orthogonal-Subspace Low-Rank Adaptation with Fixed Multi-Scale Scaling","version":1},"reference_index":282,"source":"arxiv_source","source_observed_at":"2026-08-02T09:51:03.925992Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2607.16252"},"observation_digest":"sha256:a3d4b709b29e6d7ee01cf0c67131a1959bb518d1e96cd1d28dc15ed52c7fb217","observation_id":"d3bb60b3-0066-496d-9858-6cf0daf00bdc","resolution":{"observed_at":"2026-08-02T09:51:03.925992Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2110.07602","snapshot_observed_at":"2026-08-02T12:15:44.688680Z","title":"arXiv preprint arXiv:2110.07602 , year=","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.24764","last_updated":"2026-06-04T23:55:15Z","snapshot_observed_at":"2026-08-06T17:08:34.614287Z","submitted_at":"2026-06-04T23:55:15Z","title":"GrocLM: Grocery Category Recommendation in E-Commerce with Large Language Models","version":1},"reference_index":3,"source":"arxiv_source","source_observed_at":"2026-08-02T12:15:44.688680Z"},"links":{"cited_paper":"/paper/2110.07602","citing_paper":"/paper/2607.24764"},"observation_digest":"sha256:21224e967a2bbbbdace35ad3f6cf07a2e6a6990ae82c880ed52d987d0e1c4a0c","observation_id":"8f440844-4da6-436e-b737-c91059ac5b05","resolution":{"observed_at":"2026-08-02T12:15:44.688680Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"links":{"evidence":"/evidence","html":"/paper/2110.07602/citation-record","integrity":"/paper/2110.07602/integrity","json":"/paper/2110.07602/citation-record.json","paper":"/paper/2110.07602"},"outbound":[],"paper":{"arxiv_id":"2110.07602","last_updated":"2022-03-20T15:13:08Z","latest_version":3,"primary_category":"cs.CL","snapshot_observed_at":"2026-08-10T14:24:20.137867Z","submitted_at":"2021-10-14T17:58:47Z","title":"P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks"},"reference_resolution":{"displayed":0,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":0,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":0},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"thesis":"As of 10 August 2026, this Paper Citation Record lists 0 of 0 outbound references and 66 inbound Pith citation observations for arXiv:2110.07602."}