{"as_of":"2026-08-10T11:22:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:7615c3a355c2ea110e19ce3fde0aba07beae10070d90ae94b52778538dedc0e2","coverage":[{"denominator":29,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":29,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-06T22:28:10.479678Z","state":"measured"},{"denominator":30,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":30,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-10T06:31:04.303077+00:00","state":"measured"},{"denominator":1,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":1,"source":"paper_references, paper_reference_links","source_observed_at":"2026-06-28T23:26:28.158991Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"arxiv_reference","source_observed_at":"2026-06-28T23:32:47.543386Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"cited_work":{"arxiv_id":"2506.21797","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2506.21797","snapshot_observed_at":"2026-06-28T23:32:47.543386Z","title":"arXiv preprint arXiv:2506.21797 , year=","venue":null,"work_id":"b22c2683-9b54-4096-af92-4b3e022c83b2","year":null},"citing_paper":{"arxiv_id":"2606.00183","last_updated":"2026-05-29T14:58:03Z","snapshot_observed_at":"2026-07-06T23:40:56.510371Z","submitted_at":"2026-05-29T14:58:03Z","title":"Agentic Transformers Provably Learn to Search via Reinforcement Learning","version":1},"reference_index":4,"source":"arxiv_source","source_observed_at":"2026-06-28T23:26:28.158991Z"},"links":{"cited_paper":"/paper/2506.21797","citing_paper":"/paper/2606.00183"},"observation_digest":"sha256:81be0f84b5ae88c5971b247a2ae4a2faff265b942bdaf055460b74ddd04ab45f","observation_id":"9a15a667-a0fc-4f17-8218-e6e66ac86e46","resolution":{"observed_at":"2026-06-28T23:32:47.544809Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}}],"links":{"evidence":"/evidence","html":"/paper/2506.21797/citation-record","integrity":"/paper/2506.21797/integrity","json":"/paper/2506.21797/citation-record.json","paper":"/paper/2506.21797"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T22:28:13.286489Z","title":"Neurosymbolic programming","venue":null,"work_id":"61939437-7f7b-42c3-bbf8-a4153c0dd85f","year":2021},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":1,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:07.854831Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:fd2dc696b82a856f4b612d83ed043a727453afc938d478ae9074a9218c4e2bfa","observation_id":"bf27c8bf-c317-4da3-88ad-8b9af94284b9","resolution":{"observed_at":"2026-08-06T22:28:13.383595Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T22:28:07.905411Z","title":"Neurosymbolic ai: The 3 rd wave","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":2,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:07.905411Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:670e29576c4df6ca4f2bc5cdbcbe2431774062420784433f374e03dfc10deb46","observation_id":"d702ac5d-196b-4a68-acd9-553cdc40b417","resolution":{"observed_at":"2026-08-06T22:28:07.905411Z","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":"2026-08-06T22:28:13.098020Z","title":"On the paradox of learning to reason from data","venue":null,"work_id":"32b12b86-ff58-4964-98fa-420b944212ec","year":2023},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":3,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:07.998356Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:025b7eff824381dfe3befc58e86a1a07b543390569b70aef8bbf478f1bb46ce1","observation_id":"c8a34539-ed7f-4322-acce-3f6297d936d2","resolution":{"observed_at":"2026-08-06T22:28:13.168084Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T22:28:12.919468Z","title":"On the planning abilities of large language models-a critical investigation","venue":null,"work_id":"b3e27d37-131a-48dc-9ec7-8df92c05f942","year":2023},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":4,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:08.082373Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:a591a911d8679451a0194cb9d1cf983288ac69940cc3a2c83e432c22fab0cb03","observation_id":"0a651d92-d0c2-46e5-98b1-3237cf7ee774","resolution":{"observed_at":"2026-08-06T22:28:12.983200Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T22:28:08.168743Z","title":"Composing global optimizers to reasoning tasks via algebraic objects in neural nets","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":5,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:08.168743Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:f315f1005993667a3bb8f8c935399624aa636249ba1bb955a7c6a12d57e3b9d0","observation_id":"9ae760af-4783-48ca-9136-b52874e2c79c","resolution":{"observed_at":"2026-08-06T22:28:08.168743Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2201.02177","last_updated":"2022-01-06T18:43:37Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2022-01-06T18:43:37Z","title":"Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2201.02177","snapshot_observed_at":"2026-08-06T22:28:08.268777Z","title":"Grokking: Generalization beyond overfitting on small algorithmic datasets","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":6,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:08.268777Z"},"links":{"cited_paper":"/paper/2201.02177","citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:4d37ae2f960076365d4f6d6de8895059b4ea0b9390f8c07eff447a9c31111dd9","observation_id":"d64d0028-9dc1-49d7-90a4-b40f3d15c8c1","resolution":{"observed_at":"2026-08-06T22:28:08.268777Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2301.02679","last_updated":"2023-01-06T19:00:01Z","snapshot_observed_at":"2026-08-06T08:20:11.326697Z","submitted_at":"2023-01-06T19:00:01Z","title":"Grokking modular arithmetic","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2301.02679","snapshot_observed_at":"2026-08-06T22:28:08.367301Z","title":"Grokking modular arithmetic","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":7,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:08.367301Z"},"links":{"cited_paper":"/paper/2301.02679","citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:5a1ca330b3ffc3671ec478dbd476852817c9c1e78d14a3867b2014b8c8c96297","observation_id":"ae36f4c4-19d6-4e48-b4b7-3221f9b67485","resolution":{"observed_at":"2026-08-06T22:28:08.367301Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2301.05217","last_updated":"2023-10-19T21:25:32Z","snapshot_observed_at":"2026-08-02T10:20:00.635719Z","submitted_at":"2023-01-12T18:56:49Z","title":"Progress measures for grokking via mechanistic interpretability","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2301.05217","snapshot_observed_at":"2026-08-06T22:28:08.484917Z","title":"Progress measures for grokking via mechanistic interpretability, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":8,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:08.484917Z"},"links":{"cited_paper":"/paper/2301.05217","citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:acb70a5463eb66473f116edfa375f10968f52c3ff37d64beed73be382ea09838","observation_id":"32e91a5f-4ed0-4a11-a7ff-d6e700e98d31","resolution":{"observed_at":"2026-08-06T22:28:08.484917Z","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":"2026-08-06T22:28:12.784146Z","title":"On the power of over-parametrization in neural networks with quadratic activation","venue":null,"work_id":"62843357-6d0e-4092-9cea-78109143f8b2","year":2018},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":9,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:08.591195Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:f1647a7d5b65aeedfa8a9762a8feef834331c950fba307ed70f0a78e9fef2708","observation_id":"015d896b-e3a8-4425-baef-f0b5f022592b","resolution":{"observed_at":"2026-08-06T22:28:12.815801Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2402.03804","last_updated":"2024-02-06T08:45:51Z","snapshot_observed_at":"2026-08-10T09:38:26.089602Z","submitted_at":"2024-02-06T08:45:51Z","title":"ReLU$^2$ Wins: Discovering Efficient Activation Functions for Sparse LLMs","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.03804","snapshot_observed_at":"2026-08-06T22:28:08.715111Z","title":"Relu ^2 wins: Discovering efficient activation functions for sparse llms","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":10,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:08.715111Z"},"links":{"cited_paper":"/paper/2402.03804","citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:625e40115a1a299d4a99a7c0a0832015a371ccbf2bb72b89b17435371f4e20c7","observation_id":"40040ca2-e341-43ff-a3dc-123c505b6d54","resolution":{"observed_at":"2026-08-06T22:28:08.715111Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2002.05202","last_updated":"2020-02-12T19:57:13Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2020-02-12T19:57:13Z","title":"GLU Variants Improve Transformer","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2002.05202","snapshot_observed_at":"2026-08-06T22:28:08.810981Z","title":"Glu variants improve transformer","venue":null,"work_id":null,"year":2002},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":11,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:08.810981Z"},"links":{"cited_paper":"/paper/2002.05202","citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:da015fdc94f96d9eef302bac76316bf9564c80e2040be0887aeb8df93cbbb765","observation_id":"2acd4874-2b1e-4c70-b00c-e6563031c584","resolution":{"observed_at":"2026-08-06T22:28:08.810981Z","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":"2026-08-06T22:28:08.899017Z","title":"Searching for efficient transformers for language modeling","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":12,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:08.899017Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:e1105913874907a14f82582f7f8bf0f73f06fa1266775c167fe380b1fa1b35b4","observation_id":"48bebf49-6714-4995-8d22-920f98c7eee0","resolution":{"observed_at":"2026-08-06T22:28:08.899017Z","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":"2026-08-06T22:28:12.629082Z","title":"Mean-field theory of two-layers neural networks: dimension-free bounds and kernel limit","venue":null,"work_id":"46d66d11-1b40-427d-a958-e28e3e0aa35e","year":2019},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":13,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:09.010607Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:cb6cdc40268b227dee128f4be8ca96a33597b988380741cb236f36074def6aed","observation_id":"3bfb7da4-6437-4c11-b10e-580febd26665","resolution":{"observed_at":"2026-08-06T22:28:12.691757Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T22:28:09.073618Z","title":"Optimal transport: old and new, volume 338","venue":null,"work_id":null,"year":2009},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":14,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:09.073618Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:d763266a128fd944684479d0101f663de2646a930ac325ed018d6d1b90e11cd3","observation_id":"0357e1e3-901d-4291-80cd-0f63f5755e1f","resolution":{"observed_at":"2026-08-06T22:28:09.073618Z","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":"2026-08-06T22:28:12.408776Z","title":"Gradient flows: in metric spaces and in the space of probability measures","venue":null,"work_id":"db7e5a8a-7ae9-4686-9a74-ad31116a8258","year":2008},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":15,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:09.166542Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:8c59a59b0a556c33df27b0c60926a5556b5176ca551888b6b0d1b2edd34c21fc","observation_id":"3ceb0446-3916-4a1f-a83e-bb576c27c933","resolution":{"observed_at":"2026-08-06T22:28:12.516464Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T22:28:12.168303Z","title":"Lectures on phase transitions and the renormalization group","venue":null,"work_id":"5e3a2f58-3f65-4e19-90e1-dd3874b984d1","year":2018},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":16,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:09.265138Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:3128cf796c8fd466eff88e82fc528b6d534e7b57292ade34f504eacb31a363c0","observation_id":"a8457cd4-eaf4-4f4e-a50a-d1e82bc8a6d2","resolution":{"observed_at":"2026-08-06T22:28:12.287135Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T22:28:12.025108Z","title":"The exact sample complexity gain from invariances for kernel regression","venue":null,"work_id":"a1753a38-73b4-4074-b15b-841169db6c2c","year":2023},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":17,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:09.352745Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:bf539b8ad93a95f490e4f072072eaba05949ab2b2589bafdbd87be8c5182e49f","observation_id":"2e35e35b-471e-4726-a6ba-19252d123b16","resolution":{"observed_at":"2026-08-06T22:28:12.097378Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2501.00712","last_updated":"2025-08-21T01:12:10Z","snapshot_observed_at":"2026-08-10T05:18:25.466950Z","submitted_at":"2025-01-01T03:23:00Z","title":"Rethinking Addressing in Language Models via Contexualized Equivariant Positional Encoding","version":2},"cited_work":{"arxiv_id":"2501.00712","doi":null,"metadata_source":"pith","pith_arxiv_id":"2501.00712","snapshot_observed_at":"2026-08-06T22:28:10.867387Z","title":"Rethinking Addressing in Language Models via Contexualized Equivariant Positional Encoding","venue":"cs.CL","work_id":"708eeb4d-57de-4fef-bb8a-175951ff51f0","year":2025},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":18,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:09.442763Z"},"links":{"cited_paper":"/paper/2501.00712","citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:c96ddca9847c2df28b716ea8640f1690b8944d1e815a6bcd31c7aff24b2ac0e5","observation_id":"f5be0254-ed5d-4be9-80ca-dab83bae6cdf","resolution":{"observed_at":"2026-08-06T22:28:10.932173Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T22:28:11.852444Z","title":"The parallelism tradeoff: Limitations of log-precision transformers","venue":null,"work_id":"84f7a7a6-e82e-4f9d-b4a3-4ec59f746215","year":2023},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":19,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:09.518895Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:c72a555ac9e6abef5e70ea2c16d607b3cc70489706b8131f6f4155af7e4808f5","observation_id":"382dfe29-9b46-44d8-94cd-2e10232401b4","resolution":{"observed_at":"2026-08-06T22:28:11.924679Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2404.08819","last_updated":"2025-03-05T23:00:57Z","snapshot_observed_at":"2026-07-06T17:59:38.202928Z","submitted_at":"2024-04-12T21:30:06Z","title":"The Illusion of State in State-Space Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2404.08819","snapshot_observed_at":"2026-08-06T22:28:09.640815Z","title":"The illusion of state in state-space models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":20,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:09.640815Z"},"links":{"cited_paper":"/paper/2404.08819","citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:dbafb0ea99d501d31b8a714d74f0ed29bc66825b59f9ad3edff6affca3304427","observation_id":"3c7a0061-f378-43d9-9db3-41810be3b487","resolution":{"observed_at":"2026-08-06T22:28:09.640815Z","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":"2026-08-06T22:28:09.748279Z","title":"Methods of information geometry, volume 191","venue":null,"work_id":null,"year":2000},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:09.748279Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:a77b4d675967ac748cc4dec95e2507d6e0d308542c4e686efc1cc1ba117bd0f6","observation_id":"7d2d3932-1991-493d-bd86-df748f47db61","resolution":{"observed_at":"2026-08-06T22:28:09.748279Z","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":"2026-08-06T22:28:09.809965Z","title":"Stability and generalization","venue":null,"work_id":null,"year":2002},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":22,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:09.809965Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:648bf4d725c07072b16953619988ae95420475773f9c29c89a99028a904bf156","observation_id":"97ce310c-7b0a-46f8-a2dd-13d820ad1abf","resolution":{"observed_at":"2026-08-06T22:28:09.809965Z","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":"2026-08-06T22:28:09.918761Z","title":"Understanding machine learning: From theory to algorithms","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":23,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:09.918761Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:30b771db086ad3c238186af0ffc96e520979bb13787569a55f7ab534f0f6a98c","observation_id":"76cb324d-31c7-4f26-8880-170dec8f7c24","resolution":{"observed_at":"2026-08-06T22:28:09.918761Z","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":"2026-08-06T22:28:11.695232Z","title":"Equivariant architectures for learning in deep weight spaces","venue":null,"work_id":"35fbbfcb-0d6b-44ba-8f43-a93a736a54c3","year":2023},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":24,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:10.023728Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:4e3aa26553e344e219a49a766617750e837a85125d92b8ad9871d4e0cddedaa7","observation_id":"8664f202-bc94-4073-b232-127afd734342","resolution":{"observed_at":"2026-08-06T22:28:11.749757Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T22:28:11.549275Z","title":"Signal processing for implicit neural representations","venue":null,"work_id":"07b31fff-d898-49d9-890b-327b96b86f3d","year":2022},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:10.134867Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:0c49e754be817aca4d1eb8601c4952aa05935b702449f5ab3f755a2bbd9cf74d","observation_id":"0d7372de-2410-4de8-b772-626e6086fcf6","resolution":{"observed_at":"2026-08-06T22:28:11.628059Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2410.05357","last_updated":"2024-12-05T15:08:56Z","snapshot_observed_at":"2026-08-09T08:16:43.856083Z","submitted_at":"2024-10-07T15:55:55Z","title":"Model-GLUE: Democratized LLM Scaling for A Large Model Zoo in the Wild","version":2},"cited_work":{"arxiv_id":"2410.05357","doi":null,"metadata_source":"pith","pith_arxiv_id":"2410.05357","snapshot_observed_at":"2026-08-06T22:28:10.620733Z","title":"Model-GLUE: Democratized LLM Scaling for A Large Model Zoo in the Wild","venue":"cs.LG","work_id":"1c958d9f-cdbf-454d-a7ac-ad3de96a400b","year":2024},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":26,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:10.209205Z"},"links":{"cited_paper":"/paper/2410.05357","citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:702b7b4f62522adf78d5f1f937918ab63fee725c5cd38ef80bb49286d99f0291","observation_id":"ecb81feb-01f6-4a51-b92c-44308d91092e","resolution":{"observed_at":"2026-08-06T22:28:10.685006Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-06T22:28:11.364319Z","title":"Self-consuming generative models go mad","venue":null,"work_id":"e2491551-fc0e-48ad-8312-7a020bff0181","year":2024},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":27,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:10.329599Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:cbe703ad13ba916e321a3a4746b18cbbd3914b298da6a4e7e8136ef6e662da04","observation_id":"3554cc64-c9f5-43de-b4bd-463b5a9fff6b","resolution":{"observed_at":"2026-08-06T22:28:11.436820Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2307.04001","last_updated":"2024-03-07T05:56:19Z","snapshot_observed_at":"2026-07-06T15:51:46.997267Z","submitted_at":"2023-07-08T16:00:59Z","title":"Polynomial Width is Sufficient for Set Representation with High-dimensional Features","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2307.04001","snapshot_observed_at":"2026-08-06T22:28:10.408292Z","title":"Polynomial width is sufficient for set representation with high-dimensional features","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":28,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:10.408292Z"},"links":{"cited_paper":"/paper/2307.04001","citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:70f80735f8962a86cfd4d4e03df5bf1201ee20b10ca2a45eab2fedc26fb64875","observation_id":"85961c29-5fcf-4e07-898b-3f59822e3a97","resolution":{"observed_at":"2026-08-06T22:28:10.408292Z","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":"2026-08-06T22:28:11.238378Z","title":"Low-dimensional invariant embeddings for universal geometric learning","venue":null,"work_id":"5ad91477-e0fa-43ad-b367-9363cc0082d9","year":2024},"citing_paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning","version":2},"reference_index":29,"source":"arxiv_source","source_observed_at":"2026-08-06T22:28:10.479678Z"},"links":{"citing_paper":"/paper/2506.21797"},"observation_digest":"sha256:17c29f8df56b27f6876e8d6d382904db2cde8907ec50f09d40bb546d60c9246e","observation_id":"67f9f50e-2367-4ff8-a7b3-ac741d0730a7","resolution":{"observed_at":"2026-08-06T22:28:11.281524Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-10T06:31:04.303077+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2506.21797","last_updated":"2025-07-01T18:25:12Z","latest_version":2,"primary_category":"cs.LG","snapshot_observed_at":"2026-08-10T09:38:58.479169Z","submitted_at":"2025-06-26T22:40:30Z","title":"Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation for Neurosymbolic Reasoning"},"reference_resolution":{"displayed":29,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":14,"verified_exact":2,"verified_fuzzy":13},"total_outbound_references":29},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-10T06:31:04.303077+00:00","source":"crossref"},{"observed_at":"2026-08-10T06:30:57.382061+00:00","source":"retraction_watch"}],"thesis":"As of 10 August 2026, this Paper Citation Record lists 29 of 29 outbound references and 1 inbound Pith citation observation for arXiv:2506.21797."}