{"as_of":"2026-08-11T16:53:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:ecbe3317c4761c2b6db66af202cb13ddcb71b4082561801d1fb27045451c670b","coverage":[{"denominator":63,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":63,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-06T05:04:06.586284Z","state":"measured"},{"denominator":63,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":63,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-11T06:34:44.6726+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/2508.02387/citation-record","integrity":"/paper/2508.02387/integrity","json":"/paper/2508.02387/citation-record.json","paper":"/paper/2508.02387"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T05:03:59.047613Z","title":"Deep learning","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-06T05:03:59.047613Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:41c1d4b63597a2f931fff4f7e63e2e32e2559b54acba7785950dc6e21491caf4","observation_id":"f5971ff9-75a8-43f2-95de-f5a03517e482","resolution":{"observed_at":"2026-08-06T05:03:59.047613Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2011.04406","last_updated":"2021-02-20T07:28:12Z","snapshot_observed_at":"2026-08-09T15:31:19.794411Z","submitted_at":"2020-11-09T13:16:02Z","title":"A Survey of Label-noise Representation Learning: Past, Present and Future","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2011.04406","snapshot_observed_at":"2026-08-06T05:03:59.098932Z","title":"A survey of label-noise representation learning: Past, present and future","venue":null,"work_id":null,"year":2011},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-06T05:03:59.098932Z"},"links":{"cited_paper":"/paper/2011.04406","citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:eb114deeecb88ebbbe06dc23dae41f80a807a5b8abfb50735a743deeb0ba3640","observation_id":"6f763e9d-46b2-4e95-a44f-f7c83d4269ac","resolution":{"observed_at":"2026-08-06T05:03:59.098932Z","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":"2026-08-06T05:04:18.333244Z","title":"A closer look at memorization in deep networks","venue":null,"work_id":"55155ab1-5155-43bb-9534-2e4139b99a21","year":2017},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-06T05:03:59.181615Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:1dee726ccc3989d93fc1bbe01f07963a789ac7a092eb0c3e97a465fc2edaf4ae","observation_id":"b2a23565-281e-4436-b6ba-fd3f30363993","resolution":{"observed_at":"2026-08-06T05:04:18.496391Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2312.09390","last_updated":"2023-12-14T23:07:33Z","snapshot_observed_at":"2026-07-06T17:02:09.539730Z","submitted_at":"2023-12-14T23:07:33Z","title":"Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2312.09390","snapshot_observed_at":"2026-08-06T05:03:59.252560Z","title":"Weak-to-strong gener- alization: Eliciting strong capabilities with weak supervision","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-06T05:03:59.252560Z"},"links":{"cited_paper":"/paper/2312.09390","citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:7ed0bdafaf7e27ef6f95cccb0922685c92997464c41e546a24727b5c0653eb9f","observation_id":"3559c722-3b2f-4684-9d9b-e541c7b13d1e","resolution":{"observed_at":"2026-08-06T05:03:59.252560Z","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":"2026-08-06T05:04:18.119619Z","title":"Robust loss functions under label noise for deep neural networks","venue":null,"work_id":"84fc22e1-0812-4733-951f-8625da8c473c","year":2017},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-06T05:03:59.349724Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:a979434639136344d89191140014ee593ec2aaad1326cb694127ba3e185c0ba4","observation_id":"09db8373-0299-4544-a4ff-a3812fc0a588","resolution":{"observed_at":"2026-08-06T05:04:18.167104Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:17.910413Z","title":"Normalized loss functions for deep learning with noisy labels","venue":null,"work_id":"fb3edf36-a486-4f4b-966b-23c994d5eb25","year":2020},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-06T05:03:59.444046Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:ea058a79ce38e8dbfd4a6e07783f02bfe2b3662f46272d8a01bfb27789b5248a","observation_id":"f01ce992-5f43-4ef3-b60c-3fdf7c68ef22","resolution":{"observed_at":"2026-08-06T05:04:18.017532Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:17.681056Z","title":"Asymmetric loss functions for learning with noisy labels","venue":null,"work_id":"fa1b1c71-1419-491e-b2c8-39ef995f0e12","year":2021},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-06T05:03:59.525529Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:1b2960482ac449e1ef2947c11038dfb76dafff6d4f9e0f270578cd8efe3380ac","observation_id":"a79bbe72-a0b8-4c37-9772-2b0dcb4a9ce9","resolution":{"observed_at":"2026-08-06T05:04:17.795559Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:17.446138Z","title":"Label distributionally robust losses for multi-class classification: Consistency, robustness and adaptivity","venue":null,"work_id":"d8786b9e-d6eb-4f2e-98bb-956e83745938","year":2023},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-06T05:03:59.624189Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:dfe3ae8fbbbc004c3b296621cfb7610422d8b8936827c4c3f1838f8b6831b4d2","observation_id":"fe977f56-5b0a-4971-ae61-1c41e4b67402","resolution":{"observed_at":"2026-08-06T05:04:17.556375Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:17.294890Z","title":"Noise tolerance under risk minimization","venue":null,"work_id":"b4fada23-e476-4f8c-8e69-d8e85b4b5577","year":2013},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-06T05:03:59.732164Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:a994c53078044dcc3a9b8695e9029e684480af2078a4ea2558db0f8ed96cf52f","observation_id":"a1b6ba7b-cb1a-4aa1-8316-dbb770349004","resolution":{"observed_at":"2026-08-06T05:04:17.366602Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:17.101437Z","title":"Learning with symmetric label noise: The importance of being unhinged","venue":null,"work_id":"d3eae2d0-a67a-48e6-9452-21f666de8c0c","year":2015},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-06T05:03:59.802008Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:fd3a441c993afc011ccb9878ea4e41619182fd093b0ae106190143a4d418440d","observation_id":"2cd9863a-1340-4d5d-8a76-fd9cd7359594","resolution":{"observed_at":"2026-08-06T05:04:17.204778Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:03:59.881806Z","title":"Generalized cross entropy loss for training deep neural networks with noisy labels","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-06T05:03:59.881806Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:7d3ac8f06a24d956e93eade0dcfb6316ea1ed9f521027940537dbd68c72cdb58","observation_id":"c65b14fe-a282-4142-a7cc-8378d423da86","resolution":{"observed_at":"2026-08-06T05:03:59.881806Z","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":"2026-08-06T05:03:59.982253Z","title":"Symmetric cross entropy for robust learning with noisy labels","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-06T05:03:59.982253Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:7eaff469baf791e87a8bfcb6390d1d22997d7bf75ff8e50ec444182140daa5d6","observation_id":"987a7090-3d69-4577-94f7-d62353e8dc15","resolution":{"observed_at":"2026-08-06T05:03:59.982253Z","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":"2026-08-06T05:04:00.076558Z","title":"Generalized jensen-shannon divergence loss for learning with noisy labels","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:00.076558Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:03661961609f85c37edc3b65bc06064431a8a3f354e7acb88f6654b08b8877e9","observation_id":"c868c94b-65d2-43a1-a24a-4d7bc1724421","resolution":{"observed_at":"2026-08-06T05:04:00.076558Z","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":"2026-08-06T05:04:16.944504Z","title":"Learning with noisy labels via sparse regularization","venue":null,"work_id":"2fe8e1d5-02b6-4a5d-9fad-d9d018b12251","year":2021},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:00.159931Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:8ffb8e2b33948052976596659492458f929c5788d5bd08d5f103b064ac253683","observation_id":"ae9bb9a2-4548-4847-a3a1-ddbb61c11f2a","resolution":{"observed_at":"2026-08-06T05:04:17.007644Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:16.789592Z","title":"Non-negative matrix factorization with sparseness constraints","venue":null,"work_id":"097e4402-c99f-4607-aa29-08ce27ad8a60","year":2004},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:00.243288Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:1098e6c14e4221f2b27eb9b7bfe5d21b6a662cecb3a9f4182052abd49bfbd7e0","observation_id":"53ac095f-c415-417f-b1be-5650d1df8c1f","resolution":{"observed_at":"2026-08-06T05:04:16.858058Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:16.518068Z","title":"Variance-enlarged poisson learning for graph-based semi-supervised learning with extremely sparse labeled data","venue":null,"work_id":"72606f0a-2b06-4e48-a3ec-df076794cc74","year":2024},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:00.309137Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:594c71f227a620b2b1303f80066b4c74a1abfef95c2d2362ca0f00079be0948b","observation_id":"30c4d430-6d1b-4922-9417-1c991feec2a5","resolution":{"observed_at":"2026-08-06T05:04:16.678935Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:16.235986Z","title":"On the consistency of top-k surrogate losses","venue":null,"work_id":"084db6bb-1776-4cd0-abfe-ffa431e3ac72","year":2020},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:00.368006Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:177d2c37c9003584d32b4edeb8c17e85efa7ad2b716a3cc3a7d27797e80b739e","observation_id":"6233810d-bbb1-495e-a148-ac5f33266080","resolution":{"observed_at":"2026-08-06T05:04:16.321762Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:15.866710Z","title":"Convexity, classification, and risk bounds","venue":null,"work_id":"73cef0d6-f8d1-43de-b302-61305e1a75a4","year":2006},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:00.443139Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:6f978e7ec9495629dc1c7864d9506d0790f9bcf2436219bcdf325e317c808378","observation_id":"4ad922bc-d30d-469d-9021-40a423311351","resolution":{"observed_at":"2026-08-06T05:04:16.052285Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:00.515731Z","title":"Focal loss for dense object detection","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:00.515731Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:8b1609bdf0d0ca64f930e2813d86ad5a21db99a470e64db65e3d908e59e810ef","observation_id":"c0096626-23c1-47b2-9947-5854f87867c3","resolution":{"observed_at":"2026-08-06T05:04:00.515731Z","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":"2026-08-06T05:04:15.546621Z","title":"Understanding and improving early stopping for learning with noisy labels","venue":null,"work_id":"7773c603-6d0e-4be2-acdd-7dfd68d71959","year":2021},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:00.584426Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:0ab2c6cddc435fb83cda13664150226ed6afda73419d9b3929247cd43c137cc3","observation_id":"04405e2d-13ee-47ec-87de-443f64628d16","resolution":{"observed_at":"2026-08-06T05:04:15.679570Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:00.646849Z","title":"Co-teaching: Robust training of deep neural networks with extremely noisy labels","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:00.646849Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:d4298d013372203f752d0c9701276dcba4d84f3aeb7b9f246df6def9364a92a0","observation_id":"a09ba658-aac5-4dbf-90b4-d4dd6d5c9aa9","resolution":{"observed_at":"2026-08-06T05:04:00.646849Z","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":"2026-08-06T05:04:00.744664Z","title":"Fixmatch: Simplifying semi- supervised learning with consistency and confidence.Advances in neural information processing systems, 33:596–608, 2020","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:00.744664Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:ba9cff0b05bec3addd36a5501391dfc677e4289e21e25080e53597fe8c2136f6","observation_id":"32650b59-fcd5-4648-80e5-8884dac9c747","resolution":{"observed_at":"2026-08-06T05:04:00.744664Z","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":"2026-08-06T05:04:00.839322Z","title":"mixup: Beyond empirical risk minimization","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:00.839322Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:b32e60e412fee46468ef59a813b069278cf555ba188b63f36d3470972c586373","observation_id":"93472cbd-07aa-44f7-986a-1344a25c792f","resolution":{"observed_at":"2026-08-06T05:04:00.839322Z","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":"2026-08-06T05:04:15.277997Z","title":"Learning multiple layers of features from tiny images","venue":null,"work_id":"9c0a14b0-a4aa-4faf-b495-bf517d1428e7","year":2009},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:00.923952Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:e0f961f5f24412439abb03f38d759993be65164703bf390de5aeda88da34cdb1","observation_id":"9f78e290-9105-4787-a7c1-11f9e5e61a18","resolution":{"observed_at":"2026-08-06T05:04:15.366027Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:00.994209Z","title":"Nlnl: Negative learning for noisy labels","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:00.994209Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:8e7a706ab8a443e1712a2d1e1c8a83baf38b46ebe8c2f3c0f666e688379fa39a","observation_id":"d70f6e95-2588-4dbc-a176-75cf19df8467","resolution":{"observed_at":"2026-08-06T05:04:00.994209Z","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":"2026-08-06T05:04:01.085510Z","title":"Mitigating memorization of noisy labels by clipping the model prediction","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:01.085510Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:f2504c9b84915b81a9d3673ad6f6c06d9c8a29cd076d118bc23e47e1f2aefe2a","observation_id":"7ee4416d-646b-4032-acaf-50db567ca002","resolution":{"observed_at":"2026-08-06T05:04:01.085510Z","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":"2026-08-06T05:04:01.150435Z","title":"Visualizing data using t-sne","venue":null,"work_id":null,"year":2008},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:01.150435Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:8142008f5286fd6f6976a80813100034a45b399e328cd9ce5a4c9ead558a95ce","observation_id":"fbde6a72-850b-43e1-af43-28faf27c9e4d","resolution":{"observed_at":"2026-08-06T05:04:01.150435Z","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":"2026-08-06T05:04:14.858055Z","title":"Learning with noisy labels revisited: A study using real-world human annotations","venue":null,"work_id":"0f3a0d2f-7cdc-4226-9418-5f1b4f43a996","year":2021},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:01.275853Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:c83f43623db7d265bc1baa2910ce813be94853d2d2d75cfc794e4d6f48411ca1","observation_id":"f07ff921-9ecc-4705-afe5-f8900b093673","resolution":{"observed_at":"2026-08-06T05:04:15.033821Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:01.494017Z","title":"Making deep neural networks robust to label noise: A loss correction approach","venue":null,"work_id":null,"year":1944},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:01.494017Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:73b96dada8865547772485c74483ceb9bdaecb6837907df7f871b337cd8e34c5","observation_id":"14841b66-6d7a-4da9-bbfb-c2188391c166","resolution":{"observed_at":"2026-08-06T05:04:01.494017Z","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":"2026-08-06T05:04:01.670622Z","title":"Are anchor points really indispensable in label-noise learning? Advances in neural information processing systems, 32, 2019","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:01.670622Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:33ad34cc900040a240ecded69056b2ff7f6756e7a1550152fe08d032d2c576d5","observation_id":"3c27d60e-343d-4e51-b457-25a98debd00f","resolution":{"observed_at":"2026-08-06T05:04:01.670622Z","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":"2026-08-06T05:04:01.880255Z","title":"Peer loss functions: Learning from noisy labels without knowing noise rates","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:01.880255Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:b3a6a3f7ae61ccd0775cd8cf683007bf56dd56dc5459c2e5a054c67004f66e4d","observation_id":"6b595513-15ee-464d-a53b-91db6174f0b4","resolution":{"observed_at":"2026-08-06T05:04:01.880255Z","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":"2026-08-06T05:04:14.526273Z","title":"When optimizing f-divergence is robust with label noise","venue":null,"work_id":"5a13ef73-c4ee-4b56-9600-34379f6d3a07","year":2021},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:02.106628Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:5e8a4f7b5373439ce7e102167c641696c5f916c203b8f6b22934eb214a36f932","observation_id":"658a2342-6ded-4aa9-94c1-b92c1181086c","resolution":{"observed_at":"2026-08-06T05:04:14.625459Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:02.270976Z","title":"To smooth or not? when label smoothing meets noisy labels","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:02.270976Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:9a10980120db282c7a80ad696a676d4c40ad5ceecaf892db42d2cdad37ba1ffb","observation_id":"336d9f27-36a9-42ba-886a-076b87cd3b33","resolution":{"observed_at":"2026-08-06T05:04:02.270976Z","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":"2026-08-06T05:04:14.122866Z","title":"Provably end-to-end label-noise learning without anchor points","venue":null,"work_id":"1dd28154-8908-4db4-8bc1-efef9ff156aa","year":2021},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:02.453769Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:3f7507acfcacf60413351ef105deaf586f3267185b33bde719b6dbdcf64554c0","observation_id":"2c4c97ed-798c-440f-82d7-222066e95fb7","resolution":{"observed_at":"2026-08-06T05:04:14.355382Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:13.842259Z","title":"Asymmetric loss functions for noise-tolerant learning: Theory and applications","venue":null,"work_id":"445b5267-d5e4-41a2-a482-1fb4404865b5","year":2023},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:02.614825Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:1b9b6a55d5c34cb781d67dddcdee22e4f0572d8a8d621f41c56a0cc4698a36e6","observation_id":"a8f3d67c-b4e8-4cd4-a535-13802e966b37","resolution":{"observed_at":"2026-08-06T05:04:13.935474Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:13.488805Z","title":"How does disagreement help generalization against label corruption? In International Conference on Machine Learning, pages 7164–7173","venue":null,"work_id":"0de4511e-0743-42a6-bef2-781582d88b82","year":2019},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:02.780298Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:25ffcc74baefe998c2437db65c4da883552db7732171fc3a416e9a0dd86fde99","observation_id":"472b8d8f-fa3c-453e-8cf9-67dce4963db5","resolution":{"observed_at":"2026-08-06T05:04:13.655200Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:03.046681Z","title":"Combating noisy labels by agreement: A joint training method with co-regularization","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:03.046681Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:17d3cc45744ddfe46c6a0bcae75fc3255544e578fccee1a0c3cd0b24b32869f1","observation_id":"1470140d-9393-459d-81b8-f666938ecdc8","resolution":{"observed_at":"2026-08-06T05:04:03.046681Z","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":"2026-08-06T05:04:13.196425Z","title":"Early- learning regularization prevents memorization of noisy labels","venue":null,"work_id":"4a8f9753-c741-4347-b9ef-88408a44d29b","year":2020},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:03.264032Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:d2346bfa191dce5769db6031b4aad0d13d756e436605d5576344c3d3553cc4e8","observation_id":"e9098f49-4b47-4dfc-a977-082fd633d25a","resolution":{"observed_at":"2026-08-06T05:04:13.334149Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:12.908894Z","title":"Dividemix: Learning with noisy labels as semi-supervised learning","venue":null,"work_id":"f05c2f45-a60a-4f60-8c8d-cf117212dca4","year":2020},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:03.457412Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:adaf22725901d54c1f219295789399778255da7ae187057d1dfad61269edf46e","observation_id":"2c754198-d790-4a82-a131-a5317de89ea3","resolution":{"observed_at":"2026-08-06T05:04:13.033031Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:12.576592Z","title":"Learning with instance-dependent label noise: A sample sieve approach","venue":null,"work_id":"93227ea4-d138-403f-830f-6cb90540c4ae","year":2021},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:03.620691Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:ef58fd4a43bf43f118df42d999e9da4d97628366572fcbe6ef1e0a7d9033a2b7","observation_id":"94755d0b-876c-4774-b6dd-0fe1942387c2","resolution":{"observed_at":"2026-08-06T05:04:12.739984Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:12.182357Z","title":"A second-order approach to learning with instance- dependent label noise","venue":null,"work_id":"616ac87f-1b5d-48da-a4ae-bd076eb1df8d","year":2021},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:03.775358Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:320808d531aeccbfcd97b2398e8816af4af6218c2dc7241049ef1779c0f90fd8","observation_id":"b26cb3e4-a8c8-4d06-acc3-2b3ec9fe12fc","resolution":{"observed_at":"2026-08-06T05:04:12.340101Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:11.864972Z","title":"Robust training under label noise by over- parameterization","venue":null,"work_id":"f14def62-45c5-4f2a-a72d-e3d85e0205bb","year":2022},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:03.951476Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:0a0c77f6e52cc613121e996f6f8a3d1184f27576e3b0d07072e1a32e61751421","observation_id":"2870e630-cef6-4128-aa1b-e696d6239cbb","resolution":{"observed_at":"2026-08-06T05:04:11.973960Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2307.16889","last_updated":"2023-08-22T06:15:15Z","snapshot_observed_at":"2026-07-06T16:00:50.467537Z","submitted_at":"2023-07-28T10:57:38Z","title":"Rethinking Noisy Label Learning in Real-world Annotation Scenarios from the Noise-type Perspective","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2307.16889","snapshot_observed_at":"2026-08-06T05:04:04.161211Z","title":"Rethinking noisy label learning in real-world annotation scenarios from the noise-type perspective","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:04.161211Z"},"links":{"cited_paper":"/paper/2307.16889","citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:5170623d3df525da14349acc4f2549fb467917a10a7dfee5218c79cebdd35b5f","observation_id":"557e5d61-7484-4a53-a36f-b31865d3f43e","resolution":{"observed_at":"2026-08-06T05:04:04.161211Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1708.02862","last_updated":"2017-08-09T14:59:30Z","snapshot_observed_at":"2026-07-06T05:54:38.912241Z","submitted_at":"2017-08-09T14:59:30Z","title":"WebVision Database: Visual Learning and Understanding from Web Data","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1708.02862","snapshot_observed_at":"2026-08-06T05:04:04.364719Z","title":"Webvision database: Visual learning and understanding from web data","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:04.364719Z"},"links":{"cited_paper":"/paper/1708.02862","citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:56d4a7534378df7392a1810e131b2bb3ac2209c9f89e4d874405c44cd543d57c","observation_id":"ec8bd257-d377-4b71-947a-1e9ddb02993c","resolution":{"observed_at":"2026-08-06T05:04:04.364719Z","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":"arxiv_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T05:04:06.915166Z","title":"Imagenet: A large- scale hierarchical image database","venue":null,"work_id":"a203a760-1d02-4591-aa14-fe1cdbd1e1ed","year":2009},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:04.592249Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:9f88f8f0128dc2f2a6b2e8b11db70a387fb0cf59ea7390204cc116d5a3e7cf5d","observation_id":"e00838de-18c9-4b5f-b697-0c940b1cd342","resolution":{"observed_at":"2026-08-06T05:04:06.977933Z","resolver_source":"arxiv_id_nonexistent","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:04.752634Z","title":"Learning from massive noisy labeled data for image classification","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:04.752634Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:262d57e15bd8aa3c252c83643ea95016c44986f219ae1853415ae388e349ad05","observation_id":"f3beeceb-cdef-4602-a54d-b1f419e4c7ee","resolution":{"observed_at":"2026-08-06T05:04:04.752634Z","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":"2026-08-06T05:04:04.918342Z","title":"Randaugment: Practical automated data augmentation with a reduced search space","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:04.918342Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:403bbdf241f23f9ff2c53dc413815804f007a2b5258a7a33a894a87549542e13","observation_id":"81afbc93-c933-49af-995e-560be43a6c17","resolution":{"observed_at":"2026-08-06T05:04:04.918342Z","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":"2026-08-06T05:04:11.505483Z","title":"mean±std","venue":null,"work_id":"683fa005-8f63-4b16-910f-74ac1f368834","year":2020},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:05.148621Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:378659c906260f6486404a8fc6664e528e60b64b0d5ab782008b325be5530a04","observation_id":"093046a2-97fa-42b5-bd3f-93fd2b7bb248","resolution":{"observed_at":"2026-08-06T05:04:11.671950Z","resolver_source":"raw_fallback","status":"malformed_identifier"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:11.293252Z","title":"Specifically, we provide a simple yet effective method for mitigating label noise with elaborated descriptions and theoretical results","venue":null,"work_id":"a552bb13-aa93-4f9a-a208-b2c1d7cfdb20","year":null},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:05.310707Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:d1588e154439f3917fe5497cd04628d766e987f969842c48588ae9736c0730f2","observation_id":"6508c6c8-21cd-49f3-9923-3f540c184340","resolution":{"observed_at":"2026-08-06T05:04:11.397241Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:11.074702Z","title":"Limitations","venue":null,"work_id":"1fdfa8fe-af71-4115-9df8-f68c18fa1517","year":null},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:05.424318Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:dfb5a55d9a36c1407c93a7e60d60393f22580603524de9e92379f191024c9637","observation_id":"b17cb21e-c62f-451c-b296-6220dff2bfb1","resolution":{"observed_at":"2026-08-06T05:04:11.124626Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:10.868411Z","title":"Guidelines: • The answer NA means that the paper does not include theoretical results","venue":null,"work_id":"b6c26714-4012-4ba8-8561-122b708c2f6d","year":null},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:05.510112Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:f3fc2d93db71bc50d39ade365610fc4e04a9c4d0c909572dcd2cff9fc0350745","observation_id":"c8fd7ad1-56e0-43db-a911-b5fb19c1470c","resolution":{"observed_at":"2026-08-06T05:04:10.963292Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:10.667308Z","title":"Guidelines: • The answer NA means that the paper does not include experiments","venue":null,"work_id":"2b0b2132-a5f5-4d00-ad25-e245cafb8326","year":null},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:05.638900Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:2fd913a8ddf8e61bb7aca03543f8eac906576914a04606b7480b7680d514f1cc","observation_id":"ccabdbb1-2b93-40fe-aede-362a40754bff","resolution":{"observed_at":"2026-08-06T05:04:10.764214Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:10.459653Z","title":"And the datasets are obtained from open source","venue":null,"work_id":"984768aa-aad7-447b-b361-6175a43f3df6","year":null},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:05.714620Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:e6180ecae6d31e9525fdad68cef5c73d856e62dd63c3df15c6c58aa3f10b4cf0","observation_id":"497470c5-b881-412d-9367-f561df9f6bf1","resolution":{"observed_at":"2026-08-06T05:04:10.558466Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:10.003336Z","title":"Guidelines: • The answer NA means that the paper does not include experiments","venue":null,"work_id":"8b76d66b-26a2-4dc7-a55b-10ad9b842856","year":null},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:05.810178Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:b6497178194bc77fee93705efa82d82c3369db294e8adb7ad734ba25132ce70f","observation_id":"5e56663a-5d1d-4311-8978-2e9b0dcee743","resolution":{"observed_at":"2026-08-06T05:04:10.369007Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:08.757193Z","title":"Guidelines: • The answer NA means that the paper does not include experiments","venue":null,"work_id":"d3474902-49e2-40f7-ba78-69f14325138e","year":null},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:05.909258Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:cd8e2fb2d896aba8fc9e3a680fa574e5c17894c4b816ca37a81a1795aaae0125","observation_id":"19c91a79-1487-483b-990f-387aa0c7a912","resolution":{"observed_at":"2026-08-06T05:04:09.348617Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:08.500070Z","title":"All experiments are implemented by PyTorch and are conducted on NVIDIA GeForce RTX 4090","venue":null,"work_id":"a9e3e97d-8e58-45ec-b028-2f6f4f263fff","year":null},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:05.976381Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:ef63cdb505381ee1700a747ef238ab20e2c5c897ea8288d54f16351b248e5427","observation_id":"cb18239b-862e-4dac-8148-0ff7c4197c78","resolution":{"observed_at":"2026-08-06T05:04:08.560678Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:08.296708Z","title":"Guidelines: • The answer NA means that the authors have not reviewed the NeurIPS Code of Ethics","venue":null,"work_id":"b3366d98-cad6-47f0-863a-5bb15d8eb65e","year":null},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:06.069661Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:ec43b7bd6fd2da74931a0758de781e4d66f779459a4588599b88720d433296d5","observation_id":"b4189224-db90-4b85-b137-b4937abe927b","resolution":{"observed_at":"2026-08-06T05:04:08.424698Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:08.131409Z","title":"Guidelines: • The answer NA means that there is no societal impact of the work performed","venue":null,"work_id":"d6ff6333-3a38-4e45-9c29-5977816505a8","year":null},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":58,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:06.167637Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:fb6f6b4aaa13545045f2c1af991e24ff790d3d413f6f2085b237f478d541d3f5","observation_id":"a73316d2-2863-4c2e-8965-8f0f1dd1ebe8","resolution":{"observed_at":"2026-08-06T05:04:08.185749Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:07.960184Z","title":"Guidelines: • The answer NA means that the paper poses no such risks","venue":null,"work_id":"ef30bb3d-0d11-424c-a82a-fd1f42eb00cf","year":null},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:06.238162Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:23315abcf52b7a510192e9b9704f8dcf0f780ee5658aec2fe852e3d08fd51195","observation_id":"cb39a0dd-9ac2-47a0-9e82-be3f0a34b9e8","resolution":{"observed_at":"2026-08-06T05:04:08.028313Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:07.779664Z","title":"Guidelines: • The answer NA means that the paper does not use existing assets","venue":null,"work_id":"470c6366-d9a9-45b4-9941-c9f718e89b1f","year":null},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":60,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:06.314339Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:48bb749adbf3ae60d9e8932741cf2edd00e2bf0efeb722dfcaeb313372e1a580","observation_id":"8130c382-108e-41fd-8ea5-1f5f1ac3718b","resolution":{"observed_at":"2026-08-06T05:04:07.853795Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:07.604573Z","title":"Guidelines: • The answer NA means that the paper does not release new assets","venue":null,"work_id":"4a5f1340-656c-468f-9442-30dde88f0474","year":null},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":61,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:06.395262Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:a261ad5566d3138628f6e0d8db72db72293a55c388ba76e5817a30517a83af6c","observation_id":"1953efd3-1d93-4a3d-927d-2318c5b42c31","resolution":{"observed_at":"2026-08-06T05:04:07.700507Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:07.361611Z","title":"Guidelines: • The answer NA means that the paper does not involve crowdsourcing nor research with human subjects","venue":null,"work_id":"a08b2d9d-3fee-40bf-8f2c-6c2a60fad246","year":null},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":62,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:06.468887Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:f803f076ae0bc4122c69e179ded3f9d45eff8477a812b157fdbd13c01fea59a0","observation_id":"10eaddad-30e7-475c-9d89-a0e46a1de2dd","resolution":{"observed_at":"2026-08-06T05:04:07.456468Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+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":"2026-08-06T05:04:07.163837Z","title":"Guidelines: • The answer NA means that the paper does not involve crowdsourcing nor research with human subjects","venue":null,"work_id":"e0ba4e48-6ecf-4ee6-9d33-b918164b7e42","year":null},"citing_paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise","version":1},"reference_index":63,"source":"pdf_text","source_observed_at":"2026-08-06T05:04:06.586284Z"},"links":{"citing_paper":"/paper/2508.02387"},"observation_digest":"sha256:784f6cb6a2d11305091372aae755bafa89d990cbe5374f475bb4affe2af377a9","observation_id":"4a6a4d11-2e34-4ce0-8236-005db7f14cc2","resolution":{"observed_at":"2026-08-06T05:04:07.253518Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-11T06:34:44.6726+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2508.02387","last_updated":"2025-08-04T13:10:48Z","latest_version":1,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-09T15:31:32.320130Z","submitted_at":"2025-08-04T13:10:48Z","title":"$\\epsilon$-Softmax: Approximating One-Hot Vectors for Mitigating Label Noise"},"reference_resolution":{"displayed":63,"state_counts":{"malformed_identifier":1,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":22,"verified_exact":1,"verified_fuzzy":39},"total_outbound_references":63},"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":"2026-08-11T06:34:44.6726+00:00","source":"crossref"},{"observed_at":"2026-08-11T06:34:36.301508+00:00","source":"retraction_watch"}],"thesis":"As of 11 August 2026, this Paper Citation Record lists 63 of 63 outbound references and 0 inbound Pith citation observations for arXiv:2508.02387."}