{"as_of":"2026-08-13T12:47:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:d372ee6a5f798c0bed9f758b713c4b84dd684f3f5ff2419dc4d2aaab5462dc8b","coverage":[{"denominator":43,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":43,"source":"paper_references, paper_reference_links","source_observed_at":"2026-07-01T05:33:22.798253Z","state":"measured"},{"denominator":43,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":43,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-13T06:32:02.005865+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/2606.31603/citation-record","integrity":"/paper/2606.31603/integrity","json":"/paper/2606.31603/citation-record.json","paper":"/paper/2606.31603"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.485522Z","title":"Blended latent diffusion.ACM transactions on graphics (TOG), 42 (4):1–11, 2023","venue":null,"work_id":"e22f207e-cfcb-4d86-8df2-bd0fb6400db6","year":2023},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:244de5d8c63a0b8ecc80847bbd1586b13a77dae10976d6fb02a1b7a71493bd14","observation_id":"942359d4-a5dc-4686-b4e7-26be0d0b1091","resolution":{"observed_at":"2026-07-06T21:32:56.486965Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.512939Z","title":"Flux.1 fill [dev]","venue":null,"work_id":"0f402f21-1dba-4b59-9b3d-c7a42469d0ee","year":2024},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:d33f3f36f53baabb9e48f5a7995c1a41082ced666f2f96b4263247f5a307ddff","observation_id":"e96677ad-e6de-47fa-9e26-14a68b1349bf","resolution":{"observed_at":"2026-07-06T21:32:56.514272Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.455568Z","title":"The cityscapes dataset for semantic urban scene understanding","venue":null,"work_id":"56b278d7-b9bb-405d-8af7-f379712606f0","year":2016},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:ae997b7fa2a53399434cd448b78342c2e74e1ced3cd4b29cea60753006e9f771","observation_id":"295ef5f2-14d5-4fb0-a01e-4a368bf8cf4f","resolution":{"observed_at":"2026-07-06T21:32:56.457053Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.491692Z","title":"Class-balanced loss based on effective number of samples","venue":null,"work_id":"343d60d5-be13-4233-997b-802bdd07b42c","year":null},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:dbdf126a0b6f5508ebeb86ce84deae7a463b3f34c44c25f419790b60f7a8396f","observation_id":"c7fa831e-8ef3-4de1-989d-e9208ca07657","resolution":{"observed_at":"2026-07-06T21:32:56.493091Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.497930Z","title":"An image is worth 16x16 words: Transformers for image recognition at scale","venue":null,"work_id":"8c69f67e-6d93-4262-b97f-49daccac5b75","year":2021},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:daf1ef01a4cff4827c011fcb2f2e8325c4d338fddb1ac77e51965c9d332f2f21","observation_id":"c2c1c9b3-e35a-40c2-9541-7e73d54eea1a","resolution":{"observed_at":"2026-07-06T21:32:56.499417Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.464972Z","title":"Active learning by labeling features","venue":null,"work_id":"6651f3ee-f7ab-4d22-8710-5e80ceeb8563","year":2009},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:486738e36993d0df8b0c9baac5c7f3e6d7bf986b676c7c3d900cf0a189e948e2","observation_id":"2c46a83c-0db1-4e38-abde-913b6731f6fb","resolution":{"observed_at":"2026-07-06T21:32:56.466412Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.469243Z","title":"Data augmentation for object detec- tion via controllable diffusion models","venue":null,"work_id":"7586a154-88dd-4a38-bc61-d7f65f066ca5","year":2024},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:9abde2aec4a4a77a45d3900a9b75b2ec6ef337bdfd61644f127f7ca9102234d5","observation_id":"978f4933-e506-48b2-8308-3e096403499f","resolution":{"observed_at":"2026-07-06T21:32:56.470638Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.462800Z","title":null,"venue":null,"work_id":"6ac0b6a3-9cea-4e7e-986c-947cc5c9669a","year":2020},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:87b0580727ead061640a50101373e055c087d6ff54b4e65c70f71e6d4109a224","observation_id":"13a91515-3f4b-4b0a-86e6-271c216da455","resolution":{"observed_at":"2026-07-06T21:32:56.464074Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.457889Z","title":"Dropout as a bayesian approximation: Representing model uncertainty in deep learning","venue":null,"work_id":"0b1ac74f-b55b-4db9-8002-80b4e04422e1","year":2016},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:fc27ca1e337f9e1adc726e7e19a9db1569abaf424602022f2cc62fbf6450880a","observation_id":"90e35923-b866-4e35-bebf-9d1c5de2ac12","resolution":{"observed_at":"2026-07-06T21:32:56.459383Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.508613Z","title":"Deep bayesian active learning with image data","venue":null,"work_id":"9a1b4ee6-2a0e-47ea-a759-ff0be33de8af","year":null},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:8c8a68f4d33f2a8a1e9a258f5a630f15945b082a57e45e98bc63c271ee51e3b8","observation_id":"e43ec385-2702-4d3e-9fe9-ef160f1f60ca","resolution":{"observed_at":"2026-07-06T21:32:56.510129Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.450074Z","title":"Denoising dif- fusion probabilistic models.Advances in neural information processing systems, 33:6840–6851","venue":null,"work_id":"45e032a6-e71b-4525-83cc-6d1e7dd474d8","year":2020},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:1e7edda37cdadb69c91252a7ad7aad402be7492ae3f0d65ed692ed88bdf306fe","observation_id":"8ec1b2b0-9d86-461b-aa57-77873cd30704","resolution":{"observed_at":"2026-07-06T21:32:56.451610Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.453252Z","title":"Diffusemix: Label- preserving data augmentation with diffusion models","venue":null,"work_id":"e891abcd-3372-47dd-9c83-bd016805ae02","year":2024},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:37109d140a8fc2071be1cd5201967d189cdc1a6c4726c1d770b69e154347d462","observation_id":"c50cd004-6a0f-495a-9109-b10de59bfeca","resolution":{"observed_at":"2026-07-06T21:32:56.454776Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-09T06:26:03.107956Z","title":"Computer vision for autonomous vehicles: Prob- lems, datasets and state of the art.Foundations and Trends in Computer Graphics and Vision, 12(1-3):1–308, 2020","venue":null,"work_id":"9cbc0e19-6bc8-4fa0-8664-bb66f1d12147","year":2020},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:9a24520d10826d4f69aa0ff680d0b16d280f96f9aa2d8af1a2e4b9e81f243627","observation_id":"28c52d51-256d-43c6-b09f-037076516d6b","resolution":{"observed_at":"2026-07-06T21:32:56.440642Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2503.09221","last_updated":"2025-03-12T10:09:27Z","snapshot_observed_at":"2026-08-07T17:10:35.719027Z","submitted_at":"2025-03-12T10:09:27Z","title":"Active Learning Inspired ControlNet Guidance for Augmenting Semantic Segmentation Datasets","version":1},"cited_work":{"arxiv_id":"2503.09221","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2503.09221","snapshot_observed_at":"2026-07-01T10:25:41.627287Z","title":"Available: https://arxiv.org/abs/2503.09221","venue":null,"work_id":"623cdd16-b3ed-4c9b-ac3a-eece50741d57","year":2025},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"cited_paper":"/paper/2503.09221","citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:faa62356f7023edb6789d2fc8a385269012d7b1fa8746e230cfe3d3373f83c6c","observation_id":"a096a28f-00a4-4a82-8ed0-a934bfb33059","resolution":{"observed_at":"2026-07-01T10:25:41.629046Z","resolver_source":"arxiv_id","status":"metadata_mismatch"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.434868Z","title":"Dataset enhancement with instance-level augmentations","venue":null,"work_id":"d0296828-841c-403d-995f-a239fead6b3f","year":2024},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:95ae0820a367c833a95f2df875e9f35aa92f87f929206abdd355ceaac55b309b","observation_id":"069ef3a2-a286-470b-b666-84ad8de0aee8","resolution":{"observed_at":"2026-07-06T21:32:56.436419Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.425510Z","title":"Tracer: Extreme attention guided salient object tracing network (stu- dent abstract)","venue":null,"work_id":"938e25a9-5af8-47d1-b66f-cb29dfa2ad43","year":2022},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:951f71bd1da60b9e2ca4ed2413eb3a8a8234ae7e827781fca835347fb4d55be6","observation_id":"db1d3ce5-f25d-45c7-8b2e-5ed87baaf048","resolution":{"observed_at":"2026-07-06T21:32:56.428166Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.432547Z","title":"Blip: Bootstrapping language-image pre-training for unified vision-language understanding and generation","venue":null,"work_id":"ee1757e7-174b-4ff4-bf4c-533078ae83b5","year":2022},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:4c4d29e5b1be5616f901d8c00a5603e73874d867046201a80d0467f6059c8fcb","observation_id":"f4d9e797-2a1b-4be6-af81-2c629e1c2297","resolution":{"observed_at":"2026-07-06T21:32:56.434142Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.411328Z","title":"A simple background augmentation method for object detection with diffusion model","venue":null,"work_id":"35308eaa-50d2-4746-ab17-233d752f2d8f","year":2024},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:359fd1044b80847311eb752384c524b849b2fc594a6682d5874da98ae8fff0ae","observation_id":"582b883f-c547-43b4-b547-2e6f52ec594a","resolution":{"observed_at":"2026-07-06T21:32:56.412842Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.471230Z","title":"Focal loss for dense object detection","venue":null,"work_id":"fe05e4f0-df94-40cd-bfb3-7ce9a27f756d","year":2017},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:13024fdb555ea7ad3b85e788c6be4c3ee6c687e1190b73763d66e1ce3148fd93","observation_id":"7de05d22-507c-4d96-94c2-d48cbbdf1c99","resolution":{"observed_at":"2026-07-06T21:32:56.472555Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.441440Z","title":"Decoupled weight decay regularization","venue":null,"work_id":"73852520-f51b-45be-8647-2d209ee937e8","year":2019},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:e78e96a88f6522d09832b834670fab80d59a94d1194e0a20a52f0cf7344c63dd","observation_id":"befa1b6f-6927-4b73-96b7-7228e2663b84","resolution":{"observed_at":"2026-07-06T21:32:56.443226Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.406056Z","title":"Repaint: Inpainting using denoising diffusion probabilistic models","venue":null,"work_id":"b41c5e02-081f-438e-9328-c75c426ca54a","year":2022},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:278c3658dbf028b54cc6b270b7f01b84cb275cc0e060449d4d8626280d611a70","observation_id":"a658c0d0-4c8c-4bfc-a498-ad281ee81439","resolution":{"observed_at":"2026-07-06T21:32:56.410525Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.475413Z","title":"Uavid: A semantic segmentation dataset for uav imagery.ISPRS journal of photogrammetry and remote sensing, 165:108–119, 2020","venue":null,"work_id":"7d380a6c-448c-4197-83e4-d0ed1fe1a60b","year":2020},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:240c4246b13669b2619aefb447d23a7c24930f660ff9d86a0ae5803f01902512","observation_id":"8eb2cc59-4782-4e58-8b5a-e1d75f47bf39","resolution":{"observed_at":"2026-07-06T21:32:56.478430Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.481537Z","title":"CEREALS - cost- effective region-based active learning for semantic segmen- 9 tation","venue":null,"work_id":"a69389dd-b8f6-4d76-85fa-be91ff867fbd","year":2018},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:ce88cfc34f125b5c3b84d048269e6ff0bf62ab2962c3d800fbf67fb4a8f243e2","observation_id":"1f799cef-72e7-4505-8ffe-1381bfa7c5fe","resolution":{"observed_at":"2026-07-06T21:32:56.482960Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.467045Z","title":"Dataset diffusion: Diffusion-based synthetic data generation for pixel-level semantic segmentation.Advances in Neural Information Processing Systems, 36:76872–76892, 2023","venue":null,"work_id":"ae797ff4-e91b-4154-b9f9-f625db2c5772","year":2023},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:b4aab902320101f4890ef41ea0fab7beb55b0d006f490ca6c0d1a6af73a9533a","observation_id":"413b9e11-2abb-476b-bcb9-aee4bb9ec9fa","resolution":{"observed_at":"2026-07-06T21:32:56.468613Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.473280Z","title":"Uncertainty-aware controlnet: Bridging domain gaps with synthetic image generation","venue":null,"work_id":"8f4e9c70-e61d-4419-8d8c-2ff0e9cd3f5f","year":2025},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:c8032ec279148eb830f274521357bf49aa5e353553fe497d04279e8c9d0a9ad4","observation_id":"cf8bd2db-a759-4958-a718-7663ca8945a4","resolution":{"observed_at":"2026-07-06T21:32:56.474759Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.446519Z","title":null,"venue":null,"work_id":"01958dae-937d-4715-a318-07ca4018120a","year":2024},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:90fe84bacff7e641a536e9e056d47107700754caf8b9cc0a93fa75e3dc968860","observation_id":"9eaf119d-e5c9-4061-9fec-2c6d402a80e7","resolution":{"observed_at":"2026-07-06T21:32:56.448272Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-08T05:14:35.715460Z","title":"Scalable diffusion models with transformers","venue":null,"work_id":"292e4202-5927-473f-b73e-1d6883498f14","year":null},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:e5693eebf477c23b329fbd90ca7bdc79029d2088f0f233ec84474c1a4c6c5398","observation_id":"f594b937-1474-469d-8436-6a5d0ca6de7a","resolution":{"observed_at":"2026-07-06T21:32:56.445529Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.429196Z","title":"Sdxl: Improving latent diffusion models for high-resolution image synthesis","venue":null,"work_id":"fb596b26-90f9-4e6b-ac23-728ca50193a9","year":2024},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:d02699bf00a6774e30edfec52ea2aa61c134531e29c2802627b7884e1836ad6c","observation_id":"7cecbb93-f0ad-4dd1-b042-bef0256bb668","resolution":{"observed_at":"2026-07-06T21:32:56.431225Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.483570Z","title":"High-resolution image synthesis with latent diffusion models","venue":null,"work_id":"778e4f71-2175-43e9-b453-05f1f13d6e71","year":2022},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:d9e9231b8d930e381d8070f1526e7c4ff38a95b81bfc481defcd93b6a152e2d6","observation_id":"a8efbc24-da02-4954-9c75-607e13b62402","resolution":{"observed_at":"2026-07-06T21:32:56.484931Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.460056Z","title":"Training region-based object detectors with online hard ex- ample mining","venue":null,"work_id":"0451f76d-78c9-46e6-ba82-344924b3fb57","year":null},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:98916c6b3d9eb08c1f0272b63ecd352bea6bb91a6d58d77c4eae35c3120a4339","observation_id":"baff80fd-8db6-489a-8ae5-6c9a92f84c8d","resolution":{"observed_at":"2026-07-06T21:32:56.461487Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.417095Z","title":"Effective data augmentation with diffusion models","venue":null,"work_id":"1c7151c8-2567-47b1-9c0e-34dda8f7b8af","year":2024},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:eaa25315e49cdb57ce091507e784f4a45ad1ed2281cfc24b02c4b5cd5e9876b4","observation_id":"1bd9ff57-80bf-4bff-84a2-95b9748ac0d8","resolution":{"observed_at":"2026-07-06T21:32:56.419715Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.420505Z","title":"Datasetdm: Synthesizing data with perception annota- tions using diffusion models.Advances in Neural Informa- tion Processing Systems, 36:54683–54695, 2023","venue":null,"work_id":"57b17d77-b737-4f55-b174-cf6cb01ed78c","year":2023},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:2f73a63239ecf2216b85072e0277c93ed6a8328aabd1709d2d205411ff57dcb8","observation_id":"bce0d741-911e-4295-8de2-1f766db2ec47","resolution":{"observed_at":"2026-07-06T21:32:56.423867Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.504377Z","title":"Diffumask: Synthesizing images with pixel-level annotations for semantic segmentation using dif- fusion models","venue":null,"work_id":"9f9ba931-85a7-4c21-ae4a-7ed5f6385306","year":null},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:f5ea8e9f95e633d3bccd06fdd874f43073bf86aeeced22a2e7decc9c72790428","observation_id":"68d6e1a6-aa0e-47e1-bdff-f583beb3fe95","resolution":{"observed_at":"2026-07-06T21:32:56.505805Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.506447Z","title":"Segformer: Simple and efficient design for semantic segmentation with transform- ers.Advances in neural information processing systems, 34: 12077–12090","venue":null,"work_id":"17c041e1-683f-4b33-a4d1-b50e3aeea25f","year":2021},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:9404ee6ffb13b7c93c8c5d1cad0d33256233009bad1717b7c74967049e8886b2","observation_id":"9e73befe-885f-49d0-92e3-4373f3ef87cd","resolution":{"observed_at":"2026-07-06T21:32:56.507986Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.500036Z","title":"Freemask: Synthetic images with dense annotations make stronger segmentation models.Advances in Neural Information Processing Systems, 36:18659–18675,","venue":null,"work_id":"af0b8d25-7428-4027-8557-1e8664d8af2e","year":null},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:7c554dbdc7a8e9c18aea98b796ec70faf36d1bf211b4206ca9ee1bd83376ca4f","observation_id":"02e6262b-39f8-43d3-b410-f0195bc43cd6","resolution":{"observed_at":"2026-07-06T21:32:56.501592Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.414292Z","title":"Learning loss for ac- tive learning","venue":null,"work_id":"d4c2cde5-0366-4c8d-b1f2-4c76569b58f2","year":null},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:fda6ec4592acb051e395ab7cf01a90f4c8a1b7d81dfa5c8464994af407f148d9","observation_id":"9dfa500c-4ef1-42b6-b071-c52fd146e5f3","resolution":{"observed_at":"2026-07-06T21:32:56.416190Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.502245Z","title":"Bdd100k: A diverse driving dataset for heterogeneous multitask learning","venue":null,"work_id":"05d8055d-a094-42ab-bf31-e9a165718ca6","year":2020},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:af6fa4ee4890a4f02ac167e9da5a12df6cc6c100ff9fc59e2ce11abf5617efb9","observation_id":"1a387c71-0188-421f-8b28-1871ee4272b8","resolution":{"observed_at":"2026-07-06T21:32:56.503747Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.493797Z","title":"Adding conditional control to text-to-image diffusion models","venue":null,"work_id":"a38e769e-f67b-4670-9047-46307be7f23e","year":2023},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:e478c7ddf2b90be5ffcc2ae3b9b0bbacda66cde8e1299150a6ef3636a4020908","observation_id":"1fdaaad7-e913-4507-8939-0f3aa2a469df","resolution":{"observed_at":"2026-07-06T21:32:56.495199Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.495893Z","title":"X-paste: Revisiting scalable copy-paste for instance segmentation us- ing CLIP and StableDiffusion","venue":null,"work_id":"0609af49-dfbc-44fb-a8ee-ec09990a0ce7","year":2023},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:7b4d934661a178bfda25e6f89e89ea1392e57d495cbc259796c719f7cdfc842b","observation_id":"5429818c-063f-4459-9f69-cb62e5125b62","resolution":{"observed_at":"2026-07-06T21:32:56.497304Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.487597Z","title":"Recon: Region-controllable data augmentation with rectification and alignment for object detection.Advances in Neural Information Processing Systems, 38:74897–74926,","venue":null,"work_id":"61200d45-9522-463c-9338-620fc3e8c085","year":null},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:9e807350010766b08cd14b23e5b990568615f13bda9c8b5936a3f25463447a05","observation_id":"0ea9e758-ae00-42b3-b232-bd6453fe35b5","resolution":{"observed_at":"2026-07-06T21:32:56.489063Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.479200Z","title":"The columns show the original image, the predictive-entropy map, the preserved uncertain region, and the final augmented image after inpainting and paste-back","venue":null,"work_id":"1e17de26-f843-4969-89e1-106e97b44179","year":null},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:5263ccc9fde86606fe75cea626843f3c8464526607ce4cbc6a0e75bb045049ae","observation_id":"08190e54-c583-452b-82f4-c2081ec1a806","resolution":{"observed_at":"2026-07-06T21:32:56.480889Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.489693Z","title":"photorealistic, ultra-detailed, 4K high- resolution, sharp focus, high quality, in the style of the Cityscapes dataset","venue":null,"work_id":"25bc6805-0144-4a02-a992-cbba59ef29a1","year":null},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:349130531363fa54c05190958ab64b3316698dde7a1d633d1a694c9b7a3543f1","observation_id":"2621dd72-d9d4-4829-a94a-b118499f835d","resolution":{"observed_at":"2026-07-06T21:32:56.491098Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-07-06T21:32:56.510758Z","title":"To apply this to densely labeled semantic segmentation use-cases like Cityscapes [3], we assign each semantic class to either fore- ground or background","venue":null,"work_id":"446281a4-8f33-45eb-b5a2-228770d89571","year":null},"citing_paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-07-01T05:33:22.798253Z"},"links":{"citing_paper":"/paper/2606.31603"},"observation_digest":"sha256:9ea0e0a0c996d0eee173ce65d0f7ce6a8efd68f93131cb98432ac82579646ea9","observation_id":"f4b158c2-63dd-46f3-ad0f-6c960274fee5","resolution":{"observed_at":"2026-07-06T21:32:56.512277Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2606.31603","last_updated":"2026-06-30T12:51:49Z","latest_version":1,"primary_category":"cs.CV","snapshot_observed_at":"2026-08-06T13:18:38.902593Z","submitted_at":"2026-06-30T12:51:49Z","title":"Preserve the Hard, Regenerate the Rest: Uncertainty-Guided Synthetic Training Data Augmentation with Diffusion Models"},"reference_resolution":{"displayed":43,"state_counts":{"malformed_identifier":0,"metadata_mismatch":1,"parse_uncertain":0,"unresolved":2,"verified_exact":0,"verified_fuzzy":40},"total_outbound_references":43},"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":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"thesis":"As of 13 August 2026, this Paper Citation Record lists 43 of 43 outbound references and 0 inbound Pith citation observations for arXiv:2606.31603."}