{"as_of":"2026-08-07T11:53:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:3da20443c89adfff91ee4354c43a821a995d16fe490d198061538b2d778141c8","coverage":[{"denominator":46,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":46,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-06T23:25:10.361903Z","state":"measured"},{"denominator":46,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":46,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-07T06:34:17.273281+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2608.04488/citation-record","integrity":"/paper/2608.04488/integrity","json":"/paper/2608.04488/citation-record.json","paper":"/paper/2608.04488"},"outbound":[{"citation":{"cited_paper":{"arxiv_id":"1706.03762","last_updated":"2023-08-02T00:41:18Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2017-06-12T17:57:34Z","title":"Attention Is All You Need","version":7},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1706.03762","snapshot_observed_at":"2026-08-06T23:25:08.162717Z","title":"Attention is all you need,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:08.162717Z"},"links":{"cited_paper":"/paper/1706.03762","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:b581d422be2239e30dc1e2557d4f8552ed38367484f886d5e2493bc7b8de9783","observation_id":"3fd1d063-c31e-4dc6-951f-8e1a0fd2e0d4","resolution":{"observed_at":"2026-08-06T23:25:08.162717Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1810.04805","last_updated":"2019-05-24T20:37:26Z","snapshot_observed_at":"2026-07-30T09:12:38.100527Z","submitted_at":"2018-10-11T00:50:01Z","title":"BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1810.04805","snapshot_observed_at":"2026-08-06T23:25:08.229945Z","title":"Bert: Pre-training of deep bidirectional transformers for language understanding,","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:08.229945Z"},"links":{"cited_paper":"/paper/1810.04805","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:e11daa7c930708b2652a30adb1e8309d3dd1d91b3c9d09cb6f0eed689e652520","observation_id":"5333ec8e-da53-4bde-9d28-b8244f833407","resolution":{"observed_at":"2026-08-06T23:25:08.229945Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.10435","last_updated":"2023-05-21T10:12:02Z","snapshot_observed_at":"2026-07-06T15:28:53.270447Z","submitted_at":"2023-05-11T19:20:38Z","title":"Generative Pre-trained Transformer: A Comprehensive Review on Enabling Technologies, Potential Applications, Emerging Challenges, and Future Directions","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.10435","snapshot_observed_at":"2026-08-06T23:25:08.295434Z","title":"Generative pre-trained transformer: A comprehensive review on enabling technologies, potential applications, emerging challenges, and future directions,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:08.295434Z"},"links":{"cited_paper":"/paper/2305.10435","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:d2070fc09184713b330906c0f6b6879684d2780ddf6c76add5a5e80bb46e3083","observation_id":"73f10c43-5021-45df-a5c6-0e61e2a89114","resolution":{"observed_at":"2026-08-06T23:25:08.295434Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2203.15556","last_updated":"2022-03-29T13:38:03Z","snapshot_observed_at":"2026-07-06T12:54:11.616335Z","submitted_at":"2022-03-29T13:38:03Z","title":"Training Compute-Optimal Large Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2203.15556","snapshot_observed_at":"2026-08-06T23:25:08.418115Z","title":"Training compute-optimal large language models,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:08.418115Z"},"links":{"cited_paper":"/paper/2203.15556","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:8b38b30ef9886ced84b1fe2d7d424bbd965513906332cd106e521b36bd907d12","observation_id":"9b4cdb1a-4174-44a9-877b-bed67f69f5cf","resolution":{"observed_at":"2026-08-06T23:25:08.418115Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:25:08.487893Z","title":"A cost-benefit analysis of on-premise large language model deployment: Breaking even with commercial llm services,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:08.487893Z"},"links":{"citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:b1025193e12ecae21e4cb81fffac3564501e15e4f6c2b09b6ced31c9bb3974f4","observation_id":"017bb74a-979e-4352-8809-c3a8a129d0a2","resolution":{"observed_at":"2026-08-06T23:25:08.487893Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2312.14972","last_updated":"2024-04-16T19:35:53Z","snapshot_observed_at":"2026-08-07T00:58:02.314231Z","submitted_at":"2023-12-20T19:27:59Z","title":"Scaling Down to Scale Up: A Cost-Benefit Analysis of Replacing OpenAI's LLM with Open Source SLMs in Production","version":3},"cited_work":{"arxiv_id":"2312.14972","doi":null,"metadata_source":"pith","pith_arxiv_id":"2312.14972","snapshot_observed_at":"2026-08-06T23:25:10.880076Z","title":"Scaling Down to Scale Up: A Cost-Benefit Analysis of Replacing OpenAI's LLM with Open Source SLMs in Production","venue":"cs.SE","work_id":"b98d3d67-b122-4986-b397-46d611413fd8","year":2023},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:08.553916Z"},"links":{"cited_paper":"/paper/2312.14972","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:0f5df7be2ee4f2a4e91b30da2f90427a911c134717318d91a2414e3cd9bd1970","observation_id":"fb685d05-ec36-49ab-960d-01c7061947ec","resolution":{"observed_at":"2026-08-06T23:25:10.883750Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2408.04693","last_updated":"2024-08-08T16:26:07Z","snapshot_observed_at":"2026-08-04T02:49:00.624726Z","submitted_at":"2024-08-08T16:26:07Z","title":"Understanding the Performance and Estimating the Cost of LLM Fine-Tuning","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2408.04693","snapshot_observed_at":"2026-08-06T23:25:08.608483Z","title":"Understanding the performance and estimating the cost of llm fine-tuning,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:08.608483Z"},"links":{"cited_paper":"/paper/2408.04693","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:d8cecc43fcbad588385fc976952ab35d37132245dc1e6654cb8e97269ac34444","observation_id":"99770478-6db3-483e-b240-f34bf9c1266c","resolution":{"observed_at":"2026-08-06T23:25:08.608483Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2506.02153","last_updated":"2025-09-15T22:15:00Z","snapshot_observed_at":"2026-08-05T18:56:11.132981Z","submitted_at":"2025-06-02T18:35:16Z","title":"Small Language Models are the Future of Agentic AI","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2506.02153","snapshot_observed_at":"2026-08-06T23:25:08.702584Z","title":"Small language models are the future of agentic ai,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:08.702584Z"},"links":{"cited_paper":"/paper/2506.02153","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:27fa1d2bd6f6ac7d985832cb9a648549893260acd30babd653faf485f1d06ffe","observation_id":"e679d980-e22e-4046-9d29-9b755292079c","resolution":{"observed_at":"2026-08-06T23:25:08.702584Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:25:11.056117Z","title":"A survey on small language models,","venue":null,"work_id":"037bf12b-0f4b-4ba8-b24c-8a0f51e1af17","year":2025},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:08.769242Z"},"links":{"citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:b0edc31c36f45eff08b09736045f4410c8c5acfd46def9075ed1317477e5c08e","observation_id":"80ac4c7d-f9fc-4652-b354-65b15ae69667","resolution":{"observed_at":"2026-08-06T23:25:11.059154Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:25:11.046293Z","title":"It’s not just size that matters: Small language models are also few-shot learners,","venue":null,"work_id":"46d4e0fa-b297-450d-a092-92eb205387f1","year":2021},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:08.861332Z"},"links":{"citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:331f1b46328dffb55838b891df53aab314542ff6c72863e516292143b3583269","observation_id":"9e53ab91-50d8-47b5-9856-d9d236b72474","resolution":{"observed_at":"2026-08-06T23:25:11.049868Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2409.15790","last_updated":"2025-02-26T06:34:55Z","snapshot_observed_at":"2026-08-04T07:43:40.230682Z","submitted_at":"2024-09-24T06:36:56Z","title":"Small Language Models: Survey, Measurements, and Insights","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2409.15790","snapshot_observed_at":"2026-08-06T23:25:08.951439Z","title":"Small language models: Survey, measurements, and insights,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:08.951439Z"},"links":{"cited_paper":"/paper/2409.15790","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:ddb544432076627dea6abc509d7208b6eb3b882835a1baaf678e31c830a771dc","observation_id":"1c09d827-bf61-4389-8d18-4dd95f5aaee3","resolution":{"observed_at":"2026-08-06T23:25:08.951439Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2501.05465","last_updated":"2026-05-14T16:52:31Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2025-01-03T19:53:57Z","title":"Small Language Models (SLMs) Can Still Pack a Punch: A survey (updated 2026)","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2501.05465","snapshot_observed_at":"2026-08-06T23:25:09.044619Z","title":"Small language models (slms) can still pack a punch: A survey,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:09.044619Z"},"links":{"cited_paper":"/paper/2501.05465","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:45794405c8a679b8840f29cf770a3219cfe1db8c6477883cae9f7ec7e1fe74b4","observation_id":"00d746b7-272a-4bc4-8426-66d5484bd501","resolution":{"observed_at":"2026-08-06T23:25:09.044619Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2304.14999","last_updated":"2023-04-28T17:39:49Z","snapshot_observed_at":"2026-07-06T15:21:19.790554Z","submitted_at":"2023-04-28T17:39:49Z","title":"Empirical Analysis of the Strengths and Weaknesses of PEFT Techniques for LLMs","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2304.14999","snapshot_observed_at":"2026-08-06T23:25:09.117752Z","title":"Empirical analysis of the strengths and weaknesses of peft techniques for llms,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:09.117752Z"},"links":{"cited_paper":"/paper/2304.14999","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:a812a60db6a608f12b5f302f9778d5d66cf7eb0f4f2c1b4df2f1fad8e21cb7ce","observation_id":"75aefdf6-d6b6-47b4-a2a1-844689fc5e33","resolution":{"observed_at":"2026-08-06T23:25:09.117752Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2407.18078","last_updated":"2024-07-25T14:36:18Z","snapshot_observed_at":"2026-07-06T18:51:54.168411Z","submitted_at":"2024-07-25T14:36:18Z","title":"PEFT-U: Parameter-Efficient Fine-Tuning for User Personalization","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2407.18078","snapshot_observed_at":"2026-08-06T23:25:09.204667Z","title":"Peft-u: Parameter- efficient fine-tuning for user personalization,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:09.204667Z"},"links":{"cited_paper":"/paper/2407.18078","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:defee92f8bfb5c1b141c47bc9339bdcf0ac71fd1481e5caab7dad75f6af3903c","observation_id":"f245ee31-4d82-442b-a91c-1320a36bed56","resolution":{"observed_at":"2026-08-06T23:25:09.204667Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:25:11.037478Z","title":"Parameter efficient fine tuning: A comprehensive analysis across applications,","venue":null,"work_id":"931494d2-4167-4c03-b638-32cd26482da5","year":null},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:09.255548Z"},"links":{"citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:6f15e86f75ba04cde86e2e2ee1fb01661191d6861e38f206e6641414be40a67c","observation_id":"b6b58bc9-48f4-48ea-8bd6-49bd1db75cca","resolution":{"observed_at":"2026-08-06T23:25:11.040366Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2501.13787","last_updated":"2025-01-23T16:04:23Z","snapshot_observed_at":"2026-08-07T07:30:22.453328Z","submitted_at":"2025-01-23T16:04:23Z","title":"Parameter-Efficient Fine-Tuning for Foundation Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2501.13787","snapshot_observed_at":"2026-08-06T23:25:09.452833Z","title":"Parameter-efficient fine-tuning for foundation models,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:09.452833Z"},"links":{"cited_paper":"/paper/2501.13787","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:2a2d1d94339a4c1ce445c0cc5d7cd7d66689835c677871e5a25c02eb65be490d","observation_id":"fe3a21cf-1a1e-41c2-8f91-06e1b6e4b13d","resolution":{"observed_at":"2026-08-06T23:25:09.452833Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1806.05512","last_updated":"2018-08-26T01:33:02Z","snapshot_observed_at":"2026-08-03T01:46:39.566868Z","submitted_at":"2018-06-14T12:55:35Z","title":"NetScore: Towards Universal Metrics for Large-scale Performance Analysis of Deep Neural Networks for Practical On-Device Edge Usage","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1806.05512","snapshot_observed_at":"2026-08-06T23:25:09.580362Z","title":"Netscore: Towards universal metrics for large-scale performance analysis of deep neural networks for practical on-device edge usage,","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:09.580362Z"},"links":{"cited_paper":"/paper/1806.05512","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:54caf5b5e6fa8e6f32aef850328f4ef9d3fb5225a273caa729667199476e5ee1","observation_id":"3b5e4a6c-55b0-48b6-af5d-2fbdefca121a","resolution":{"observed_at":"2026-08-06T23:25:09.580362Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1903.07209","last_updated":"2019-04-15T16:05:35Z","snapshot_observed_at":"2026-07-06T07:39:50.674241Z","submitted_at":"2019-03-18T00:16:04Z","title":"AttoNets: Compact and Efficient Deep Neural Networks for the Edge via Human-Machine Collaborative Design","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1903.07209","snapshot_observed_at":"2026-08-06T23:25:09.741474Z","title":"Attonets: Compact and efficient deep neural networks for the edge via human-machine collaborative design,","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:09.741474Z"},"links":{"cited_paper":"/paper/1903.07209","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:c14f88ab7513c1551d9a44db03fcd65751bb3fd36906b8fc084ae492584b2976","observation_id":"bea53fa0-7308-406b-9670-e212d8c5e71c","resolution":{"observed_at":"2026-08-06T23:25:09.741474Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2409.18664","last_updated":"2024-09-27T11:50:10Z","snapshot_observed_at":"2026-07-06T19:23:19.829130Z","submitted_at":"2024-09-27T11:50:10Z","title":"How green is continual learning, really? Analyzing the energy consumption in continual training of vision foundation models","version":1},"cited_work":{"arxiv_id":"2409.18664","doi":null,"metadata_source":"pith","pith_arxiv_id":"2409.18664","snapshot_observed_at":"2026-08-06T23:25:10.774825Z","title":"How green is continual learning, really? Analyzing the energy consumption in continual training of vision foundation models","venue":"cs.LG","work_id":"7c5b7801-b2f1-4096-8d47-295852a6c263","year":2024},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:09.867422Z"},"links":{"cited_paper":"/paper/2409.18664","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:de60d341f4d462c75073a717025a8dd43f5a0cd56e2c20bc10a6ff79212c7370","observation_id":"bc8777c5-1160-49ab-a97c-3ec2034f605d","resolution":{"observed_at":"2026-08-06T23:25:10.778139Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:25:10.033991Z","title":"An attention-based feature memory design for energy-efficient continual learning,","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.033991Z"},"links":{"citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:28f4e7fa954567a229e9545dc7aedffce5b6bf88cafe1ba2ac36f71ffff6349f","observation_id":"cacfc435-5a8b-49e8-810d-f3889408fbf0","resolution":{"observed_at":"2026-08-06T23:25:10.033991Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2607.02158","last_updated":"2026-07-02T13:31:29Z","snapshot_observed_at":"2026-08-07T04:20:08.981474Z","submitted_at":"2026-07-02T13:31:29Z","title":"Efficient PEFT Methods with Adaptive Checkpointing for Vision Models and VLMs on Resource Constrained Consumer-GPUs","version":1},"cited_work":{"arxiv_id":"2607.02158","doi":null,"metadata_source":"pith","pith_arxiv_id":"2607.02158","snapshot_observed_at":"2026-08-06T23:25:10.673278Z","title":"Efficient PEFT Methods with Adaptive Checkpointing for Vision Models and VLMs on Resource Constrained Consumer-GPUs","venue":"cs.CV","work_id":"ff90407c-ca3b-4b69-8b21-93597a176606","year":2026},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.097141Z"},"links":{"cited_paper":"/paper/2607.02158","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:316014ee80925846b0072db3ae95b178825b844b3e8019a89ae6952cfe56a421","observation_id":"655aa492-db25-4aba-af02-160cae799fb5","resolution":{"observed_at":"2026-08-06T23:25:10.676677Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:25:10.217852Z","title":"Lora: Low-rank adaptation of large language models,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.217852Z"},"links":{"citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:436c3f6ed25c5f976c9d4b4340365d47c44f8d754076950adcaf026613c3aa0f","observation_id":"ae840744-08d4-4b5d-985d-c520fc8b969d","resolution":{"observed_at":"2026-08-06T23:25:10.217852Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.12354","last_updated":"2024-07-04T18:33:00Z","snapshot_observed_at":"2026-08-03T17:12:16.534784Z","submitted_at":"2024-02-19T18:33:49Z","title":"LoRA+: Efficient Low Rank Adaptation of Large Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.12354","snapshot_observed_at":"2026-08-06T23:25:10.291996Z","title":"Lora+: Efficient low rank adaptation of large models,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.291996Z"},"links":{"cited_paper":"/paper/2402.12354","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:89b2977d20d6dc3e487a843525998c0362386759c83aa6b26abae8f9b1a2671b","observation_id":"44c39bbb-6b75-448a-bedb-4e324a9e3582","resolution":{"observed_at":"2026-08-06T23:25:10.291996Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.14314","last_updated":"2023-05-23T17:50:33Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2023-05-23T17:50:33Z","title":"QLoRA: Efficient Finetuning of Quantized LLMs","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.14314","snapshot_observed_at":"2026-08-06T23:25:10.295446Z","title":"Qlora: Efficient finetuning of quantized llms,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.295446Z"},"links":{"cited_paper":"/paper/2305.14314","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:7d85862ca90146ae1b2a730b85d71d3d2d0be436e392b4e72153425b7803b34e","observation_id":"8417260a-3c1d-4078-9265-c2c2da91fd51","resolution":{"observed_at":"2026-08-06T23:25:10.295446Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:25:11.020376Z","title":"Bitfit: Simple parameter- efficient fine-tuning for transformer-based masked language-models,","venue":null,"work_id":"5c0b5a37-dbd6-4755-95fe-d938e7f49bcd","year":null},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.298983Z"},"links":{"citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:cd75d84599c80005a08b0b5bd90d20da65c1cf945ba4e01ef92b4c09de964386","observation_id":"1428ed85-1590-47b3-b8e0-54a763ca4c4b","resolution":{"observed_at":"2026-08-06T23:25:11.023887Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2411.03350","last_updated":"2024-12-28T09:18:36Z","snapshot_observed_at":"2026-08-04T03:34:32.259687Z","submitted_at":"2024-11-04T04:43:01Z","title":"A Comprehensive Survey of Small Language Models in the Era of Large Language Models: Techniques, Enhancements, Applications, Collaboration with LLMs, and Trustworthiness","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2411.03350","snapshot_observed_at":"2026-08-06T23:25:10.306623Z","title":"A comprehensive survey of small language models in the era of large language models: Techniques, enhancements, applications, collaboration with llms, and trustworthiness,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.306623Z"},"links":{"cited_paper":"/paper/2411.03350","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:0f95e4fe45944c6a5d4d2810a3cb3b5db591c694d53e1750ee0bad6363a5ec8a","observation_id":"cdb6da5a-a3c2-4b90-9f2b-e93652809269","resolution":{"observed_at":"2026-08-06T23:25:10.306623Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2401.02385","last_updated":"2024-06-04T02:05:30Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-01-04T17:54:59Z","title":"TinyLlama: An Open-Source Small Language Model","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2401.02385","snapshot_observed_at":"2026-08-06T23:25:10.309716Z","title":"Tinyllama: An open-source small language model,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.309716Z"},"links":{"cited_paper":"/paper/2401.02385","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:f020be7f70aee00b402c8c9c47237bb5dc128748a5523f8a3f867c2216108fbe","observation_id":"c28cb4b2-d610-4d4e-a4dd-3d0d16e5125a","resolution":{"observed_at":"2026-08-06T23:25:10.309716Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:25:10.313605Z","title":"Qwen3 technical report,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.313605Z"},"links":{"citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:e07e63e819d67debc802c0398b47487b49bc510cee08f2bc4a9d6d08655917b6","observation_id":"4a59ae93-a443-4577-8815-190cfe407a8b","resolution":{"observed_at":"2026-08-06T23:25:10.313605Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.14905","last_updated":"2024-06-27T03:53:46Z","snapshot_observed_at":"2026-08-07T05:13:11.135540Z","submitted_at":"2024-02-22T18:58:55Z","title":"MobileLLM: Optimizing Sub-billion Parameter Language Models for On-Device Use Cases","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.14905","snapshot_observed_at":"2026-08-06T23:25:10.320329Z","title":"Mobilellm: Optimizing sub-billion parameter language models for on-device use cases,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.320329Z"},"links":{"cited_paper":"/paper/2402.14905","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:6a8b8861d92fcab0415f2cd06bb968c3419f5de58fc7c88a9d29515cea55ea31","observation_id":"2df68e60-8d33-47ef-90e0-ca5638b32e6f","resolution":{"observed_at":"2026-08-06T23:25:10.320329Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:25:11.001944Z","title":"Demystifying small language models for edge deployment,","venue":null,"work_id":"048097ec-89b0-48bd-9e0e-f7bbf86c0939","year":2025},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.323608Z"},"links":{"citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:5a321dd757d910f7a9d812f4622854a9bb350a87c1ab93117062fe7c2fce71a4","observation_id":"b45a68ae-0c7c-41d0-82e5-a88cb3a1253c","resolution":{"observed_at":"2026-08-06T23:25:11.006416Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2312.00752","last_updated":"2024-05-31T17:55:27Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2023-12-01T18:01:34Z","title":"Mamba: Linear-Time Sequence Modeling with Selective State Spaces","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2312.00752","snapshot_observed_at":"2026-08-06T23:25:10.326843Z","title":"Mamba: Linear-time sequence modeling with selective state spaces,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.326843Z"},"links":{"cited_paper":"/paper/2312.00752","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:0130576a6d329ddaacf88bca925d7149e3875005065c1a11b3d020c64c388fc7","observation_id":"faf94f5f-7515-4e9f-b336-a41ab59824ac","resolution":{"observed_at":"2026-08-06T23:25:10.326843Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2405.21060","last_updated":"2024-05-31T17:50:01Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-05-31T17:50:01Z","title":"Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2405.21060","snapshot_observed_at":"2026-08-06T23:25:10.329751Z","title":"Transformers are ssms: Generalized models and efficient algorithms through structured state space duality,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.329751Z"},"links":{"cited_paper":"/paper/2405.21060","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:1b2dad24e41b5c44d3088a3395238efcb4bbd0eaec15a2d5dd8a08c3f6ef3def","observation_id":"20c58eb9-7600-4a38-ae2d-ce1625dc3d02","resolution":{"observed_at":"2026-08-06T23:25:10.329751Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1804.07461","last_updated":"2019-02-22T23:53:34Z","snapshot_observed_at":"2026-07-06T06:34:26.609892Z","submitted_at":"2018-04-20T06:35:04Z","title":"GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1804.07461","snapshot_observed_at":"2026-08-06T23:25:10.332835Z","title":"Glue: A multi-task benchmark and analysis platform for natural language understanding,","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.332835Z"},"links":{"cited_paper":"/paper/1804.07461","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:1e9cae9fb883519e7a05f29dc3e3489face6f08774b2a6b8715b17955ccb3d9d","observation_id":"23114d8d-44aa-4dcf-97a9-3062bd7e2306","resolution":{"observed_at":"2026-08-06T23:25:10.332835Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2304.11406","last_updated":"2024-06-05T03:29:31Z","snapshot_observed_at":"2026-07-06T15:18:45.940190Z","submitted_at":"2023-04-22T13:42:04Z","title":"LaMP: When Large Language Models Meet Personalization","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2304.11406","snapshot_observed_at":"2026-08-06T23:25:10.336161Z","title":"Lamp: When large language models meet personalization,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.336161Z"},"links":{"cited_paper":"/paper/2304.11406","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:25e8ecae534e391aee97030439f273efc33b4f1596f14ceea0a1f394ac69517c","observation_id":"ab9a3bbc-3e7c-41c7-9115-830d908cee90","resolution":{"observed_at":"2026-08-06T23:25:10.336161Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1204.0885","last_updated":"2012-04-04T08:17:32Z","snapshot_observed_at":"2026-07-06T02:45:44.227887Z","submitted_at":"2012-04-04T08:17:32Z","title":"PID Parameters Optimization by Using Genetic Algorithm","version":1},"cited_work":{"arxiv_id":"1204.0885","doi":null,"metadata_source":"pith","pith_arxiv_id":"1204.0885","snapshot_observed_at":"2026-08-06T23:25:10.472310Z","title":"PID Parameters Optimization by Using Genetic Algorithm","venue":"cs.SY","work_id":"7862dbfc-6548-4bcb-9aaa-c1504646547e","year":2012},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.339285Z"},"links":{"cited_paper":"/paper/1204.0885","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:a242c69d3cc6ee2615c33516fc60a14ce4eab6d017ca36bd34470d0c697d6feb","observation_id":"b6e87864-6c1c-466a-bb88-1a0d5527a17f","resolution":{"observed_at":"2026-08-06T23:25:10.475733Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1606.05250","last_updated":"2016-10-11T02:42:36Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2016-06-16T16:36:00Z","title":"SQuAD: 100,000+ Questions for Machine Comprehension of Text","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1606.05250","snapshot_observed_at":"2026-08-06T23:25:10.342592Z","title":"Squad: 100,000+ questions for machine comprehension of text,","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.342592Z"},"links":{"cited_paper":"/paper/1606.05250","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:db6456f74c3db65667fb861594095507b244a0df899fb7cbfb9a0ba17c606dd7","observation_id":"9fcc7cce-6438-492d-bac7-96833fe452de","resolution":{"observed_at":"2026-08-06T23:25:10.342592Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2112.09118","last_updated":"2022-08-29T12:17:32Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2021-12-16T18:57:37Z","title":"Unsupervised Dense Information Retrieval with Contrastive Learning","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2112.09118","snapshot_observed_at":"2026-08-06T23:25:10.345832Z","title":"Unsupervised dense information retrieval with contrastive learning,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.345832Z"},"links":{"cited_paper":"/paper/2112.09118","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:7814a88e865144c9e9842acd349f4b9ffe51fef1ee5a7de662d669ac1889f936","observation_id":"96e373bb-070a-4ab7-8d2a-6af312e767f6","resolution":{"observed_at":"2026-08-06T23:25:10.345832Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2410.09016","last_updated":"2025-06-09T06:07:50Z","snapshot_observed_at":"2026-08-05T17:44:50.341807Z","submitted_at":"2024-10-11T17:30:28Z","title":"Parameter-Efficient Fine-Tuning of State Space Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.09016","snapshot_observed_at":"2026-08-06T23:25:10.349083Z","title":"Parameter- efficient fine-tuning of state space models,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.349083Z"},"links":{"cited_paper":"/paper/2410.09016","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:4d01824cc534e7cc85e83082110f5e844a19e34a5af1f39abf0dfa0f24e193dc","observation_id":"18746a5b-0cc3-4967-8c25-683a0500e774","resolution":{"observed_at":"2026-08-06T23:25:10.349083Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1910.03771","last_updated":"2020-07-14T03:42:34Z","snapshot_observed_at":"2026-07-06T08:27:58.343233Z","submitted_at":"2019-10-09T03:23:22Z","title":"HuggingFace's Transformers: State-of-the-art Natural Language Processing","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1910.03771","snapshot_observed_at":"2026-08-06T23:25:10.352431Z","title":"Huggingface’s transformers: State-of-the-art natural language processing,","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.352431Z"},"links":{"cited_paper":"/paper/1910.03771","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:2b9a69bec34f8b47b2e577f7618a688cadf577d08e4fb21ee6c85157d3521c28","observation_id":"1e93dc94-463f-4ef5-93a5-b9c9061b1296","resolution":{"observed_at":"2026-08-06T23:25:10.352431Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2410.13229","last_updated":"2024-12-07T07:27:00Z","snapshot_observed_at":"2026-08-04T22:39:19.035342Z","submitted_at":"2024-10-17T05:32:33Z","title":"Quamba: A Post-Training Quantization Recipe for Selective State Space Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.13229","snapshot_observed_at":"2026-08-06T23:25:10.355759Z","title":"Quamba: A post-training quantization recipe for selective state space models,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.355759Z"},"links":{"cited_paper":"/paper/2410.13229","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:2064421b50094f062786a0a2300faf917f4daaa6ef63df0530965f757990733c","observation_id":"200549dc-d921-4109-8bf5-e377a06fe0fa","resolution":{"observed_at":"2026-08-06T23:25:10.355759Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2411.03855","last_updated":"2025-04-01T08:41:31Z","snapshot_observed_at":"2026-07-06T19:46:06.343310Z","submitted_at":"2024-11-06T11:57:55Z","title":"MambaPEFT: Exploring Parameter-Efficient Fine-Tuning for Mamba","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2411.03855","snapshot_observed_at":"2026-08-06T23:25:10.358968Z","title":"Mambapeft: Exploring parameter-efficient fine-tuning for mamba,","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.358968Z"},"links":{"cited_paper":"/paper/2411.03855","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:28a222f93414d8b9e07733632262cbd6ef327b3c8f927fed24c9244c60129e46","observation_id":"bf345921-9ff4-4f4e-806d-fc0ef3ddcbd8","resolution":{"observed_at":"2026-08-06T23:25:10.358968Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2406.00209","last_updated":"2025-08-29T05:39:13Z","snapshot_observed_at":"2026-08-05T21:14:44.166741Z","submitted_at":"2024-05-31T21:46:23Z","title":"Mamba State-Space Models Are Lyapunov-Stable Learners","version":3},"cited_work":{"arxiv_id":"2406.00209","doi":null,"metadata_source":"pith","pith_arxiv_id":"2406.00209","snapshot_observed_at":"2026-08-06T23:25:10.391417Z","title":"Mamba State-Space Models Are Lyapunov-Stable Learners","venue":"cs.LG","work_id":"2f2e6098-49b2-4265-9071-38f967c3cd4e","year":2024},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.361903Z"},"links":{"cited_paper":"/paper/2406.00209","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:10c0283f88d752e9e0230ef48a5db1587d7969e879a1f8228c584fe1e907a00b","observation_id":"d4ed56a8-ea86-4ed8-ab12-4a921f77b6b4","resolution":{"observed_at":"2026-08-06T23:25:10.397889Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-07T06:34:17.273281+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2106.09685","last_updated":"2021-10-16T18:40:34Z","snapshot_observed_at":"2026-08-07T07:43:16.294957Z","submitted_at":"2021-06-17T17:37:18Z","title":"LoRA: Low-Rank Adaptation of Large Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2106.09685","snapshot_observed_at":"2026-08-06T23:25:10.288646Z","title":"Available: https://arxiv.org/abs/2106.09685","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":2021,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.288646Z"},"links":{"cited_paper":"/paper/2106.09685","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:4368535da7b6f2b5d99b7d5ec29c7ae1ce8488043c082d028aca17b8bfff1d6c","observation_id":"c29453e7-a762-40d3-a575-a0326d844a50","resolution":{"observed_at":"2026-08-06T23:25:10.288646Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T23:25:10.302674Z","title":"Available: https://arxiv.org/abs/2106.10199","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":2022,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.302674Z"},"links":{"citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:6546777097571f5f17d4e2e1dee4486bc61cc2bf660b2c330a83831388737e5f","observation_id":"784e7f57-e3b0-4728-b339-f3a89d55b769","resolution":{"observed_at":"2026-08-06T23:25:10.302674Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2404.13506","last_updated":"2024-04-23T21:28:26Z","snapshot_observed_at":"2026-08-03T04:14:41.015154Z","submitted_at":"2024-04-21T02:26:15Z","title":"Parameter Efficient Fine Tuning: A Comprehensive Analysis Across Applications","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2404.13506","snapshot_observed_at":"2026-08-06T23:25:09.288553Z","title":"Available: https://arxiv.org/abs/2404.13506","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":2024,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:09.288553Z"},"links":{"cited_paper":"/paper/2404.13506","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:0ee7a6c0b6aabd4a04a81eb02b62470a84628da048bdc10c9cdbba9593cf02b6","observation_id":"2930e22a-e666-403e-b8aa-0f8b745da9b9","resolution":{"observed_at":"2026-08-06T23:25:09.288553Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2505.09388","last_updated":"2025-05-14T13:41:34Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2025-05-14T13:41:34Z","title":"Qwen3 Technical Report","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2505.09388","snapshot_observed_at":"2026-08-06T23:25:10.317049Z","title":"Available: https://arxiv.org/abs/2505.09388","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs","version":1},"reference_index":2025,"source":"pdf_text","source_observed_at":"2026-08-06T23:25:10.317049Z"},"links":{"cited_paper":"/paper/2505.09388","citing_paper":"/paper/2608.04488"},"observation_digest":"sha256:0d9725e8d6c1aa011aab581c6dd79fd4a1943e8ec4410889e1c59da260fda9a0","observation_id":"d8aac230-a5cc-4fcd-852d-fa82d34c04c1","resolution":{"observed_at":"2026-08-06T23:25:10.317049Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2608.04488","last_updated":"2026-08-05T06:20:57Z","latest_version":1,"primary_category":"cs.CL","snapshot_observed_at":"2026-08-07T11:42:53.328398Z","submitted_at":"2026-08-05T06:20:57Z","title":"Energy- and Memory-Efficient PEFT Methods for Personalized On-Device SLMs on Consumer GPUs"},"reference_resolution":{"displayed":46,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":36,"verified_exact":5,"verified_fuzzy":5},"total_outbound_references":46},"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":"2026-08-07T06:34:17.273281+00:00","source":"crossref"},{"observed_at":"2026-08-07T06:34:11.927384+00:00","source":"retraction_watch"}],"thesis":"As of 7 August 2026, this Paper Citation Record lists 46 of 46 outbound references and 0 inbound Pith citation observations for arXiv:2608.04488."}