{"as_of":"2026-08-19T15:46:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:d2849d031e98faf1ec35a263e6f3bc0cb572417b38d8a1f127a54d0ce6f7798f","coverage":[{"denominator":41,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":41,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-10T23:15:24.945850Z","state":"measured"},{"denominator":42,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":42,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-19T06:32:44.657259+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":"2026-08-15T16:13:41.132740Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"pith","source_observed_at":"2026-08-15T16:13:41.293622Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"cited_work":{"arxiv_id":"2412.20796","doi":null,"metadata_source":"pith","pith_arxiv_id":"2412.20796","snapshot_observed_at":"2026-08-15T16:13:41.293622Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","venue":"cs.DC","work_id":"285d82a1-0eec-4621-9baa-b2a2e055aafb","year":2024},"citing_paper":{"arxiv_id":"2509.08418","last_updated":"2025-09-10T09:06:24Z","snapshot_observed_at":"2026-08-17T22:40:19.092678Z","submitted_at":"2025-09-10T09:06:24Z","title":"Facet: highly efficient E(3)-equivariant networks for interatomic potentials","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-15T16:13:41.132740Z"},"links":{"cited_paper":"/paper/2412.20796","citing_paper":"/paper/2509.08418"},"observation_digest":"sha256:9db2c0ba94f0f6e635b98cc3b3623b565dc10114f7aff7765b25941a7e9cc7ea","observation_id":"77f69f3e-07c3-44ef-819b-0549eac65394","resolution":{"observed_at":"2026-08-15T16:13:41.301697Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}}],"links":{"evidence":"/evidence","html":"/paper/2412.20796/citation-record","integrity":"/paper/2412.20796/integrity","json":"/paper/2412.20796/citation-record.json","paper":"/paper/2412.20796"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-10T23:15:25.602542Z","title":"Scalable parallel algorithm for graph neural network interatomic potentials in molecular dynamics simulations,","venue":null,"work_id":"3f0ab943-7bc5-4526-a9c2-c4186a8b096e","year":2024},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.771307Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:ebc8e88faa0958a07acb0ca9a02c81b99d930151c66329f494441b3daa1e1ff2","observation_id":"d994e61f-4f39-4c86-8b4d-d28b6669798f","resolution":{"observed_at":"2026-08-10T23:15:25.607324Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2401.00096","last_updated":"2025-09-04T13:50:21Z","snapshot_observed_at":"2026-08-04T22:35:15.439821Z","submitted_at":"2023-12-29T23:08:59Z","title":"A foundation model for atomistic materials chemistry","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2401.00096","snapshot_observed_at":"2026-08-10T23:15:24.776720Z","title":"A foundation model for atomistic materials chemistry,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.776720Z"},"links":{"cited_paper":"/paper/2401.00096","citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:cae581d3850f9271481464bdffe14b45db85da13803ff4c723a85144f50213e5","observation_id":"5344c197-3b20-4d84-b30b-e3e1f1cd17f1","resolution":{"observed_at":"2026-08-10T23:15:24.776720Z","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":"2026-08-10T23:15:25.587525Z","title":"Chgnet as a pretrained universal neural network potential for charge-informed atomistic modelling,","venue":null,"work_id":"808fbb30-7116-4eda-bbca-039ac7682234","year":2023},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.781336Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:e845ec9e10df203e92b2ed9ff3546e01a4362d4c54c15ce751e4245407d6ed87","observation_id":"c35f6926-aad6-4e48-b783-9d66fc590339","resolution":{"observed_at":"2026-08-10T23:15:25.593060Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.569151Z","title":"Deepmd-kit: A deep learning package for many-body potential energy representation and molecular dynamics,","venue":null,"work_id":"5b3bc4b6-2987-4b10-8d91-c503450f1f82","year":2018},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.785844Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:b3151f16ca4eb91e57f44335322735db3cb3b2f4ed941be23e8e27b162f29b59","observation_id":"8ed59815-37d2-475b-b812-2ed852fe0074","resolution":{"observed_at":"2026-08-10T23:15:25.577481Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.556372Z","title":"Quantum-chemical insights from deep tensor neural networks,","venue":null,"work_id":"f0e8a0ad-a235-40c6-aee4-affdd7778dc4","year":2017},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.790128Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:45c3f64a1e99cdea575f84a3465e188dd7e34f58ffd0712e3ecfbe20291e6f87","observation_id":"ca5974e2-5b28-4734-98fe-b3203181581f","resolution":{"observed_at":"2026-08-10T23:15:25.560818Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.542644Z","title":"Schnet: A continuous-filter convo- lutional neural network for modeling quantum interactions,","venue":null,"work_id":"f8fb1b6e-f500-4ff1-9917-0a22e8b5dba7","year":2017},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.794406Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:d527c79e00367386a2df7e8c1033d096c8727d13760a40c2a8c95517a9f1230d","observation_id":"4356dc9d-dde1-492e-bdb2-0eae3656a9e0","resolution":{"observed_at":"2026-08-10T23:15:25.548386Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.529067Z","title":"Hierarchical modeling of molecular energies using a deep neural network,","venue":null,"work_id":"ce79cccd-a05f-4203-acba-452e35a2569d","year":2018},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.799244Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:545ccc46d233a8831eb4c224331b77d28bf33db0833a3eabd7987d9b3e587845","observation_id":"39451e4b-c965-4562-80f9-5ca36d4adbdd","resolution":{"observed_at":"2026-08-10T23:15:25.533388Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.515517Z","title":"Physnet: A neural network for predicting energies, forces, dipole moments, and partial charges,","venue":null,"work_id":"8104cc44-0a45-4793-a60d-9736f8f87747","year":2019},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.803329Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:8b6efcaf6cff1a002cc327de208420e218ca4fbce9015176a24e089642b54470","observation_id":"ca98cfb7-f6b9-4883-83ff-2f2d3827e4a1","resolution":{"observed_at":"2026-08-10T23:15:25.520258Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.502723Z","title":"Directional message pass- ing for molecular graphs,","venue":null,"work_id":"5ee08286-f0ef-4cb9-8e4c-4f64a2f5e114","year":2020},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.807442Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:6b5c3309506b9ab55a0bfe4b481f3755ad859934c370ddad1963890faee1a4e0","observation_id":"f3ed123d-e979-4039-9f68-b5acb47cc5b8","resolution":{"observed_at":"2026-08-10T23:15:25.507011Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.489510Z","title":"Mace: Higher order equivariant message passing neural networks for fast and accurate force fields,","venue":null,"work_id":"728f8044-f8e1-4b60-a704-62c9b40f4824","year":2022},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.811490Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:013d7e71fe8facd2b1effc5c11ba325b1aed051fb0807cd77c7dac4425a45c36","observation_id":"a81c9880-6eed-4576-871b-8e4415bf8df8","resolution":{"observed_at":"2026-08-10T23:15:25.493830Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.476613Z","title":"Crystal graph convolutional neural networks for an accurate and interpretable prediction of material properties,","venue":null,"work_id":"31ffd954-1641-4b35-a373-248f57c72863","year":2018},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.815577Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:5ea49cc3486c12ee8214ab14d336be047dd684bb2d1e00e294a949b841083a12","observation_id":"ace5d095-9588-4bd1-be61-0e3512da4211","resolution":{"observed_at":"2026-08-10T23:15:25.481161Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.461227Z","title":"Gemnet: Universal di- rectional graph neural networks for molecules,","venue":null,"work_id":"d8e11156-94b3-498f-a65c-fa4ec10f06a7","year":2021},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.820884Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:4b541548727b76b91829b5b229cbcbb67ce6c431cdb5fbb7ddc8f093c125e318","observation_id":"670a8396-7070-4388-8267-2da872a79e88","resolution":{"observed_at":"2026-08-10T23:15:25.465942Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.447168Z","title":"Ani-1: an extensible neural network potential with dft accuracy at force field computational cost,","venue":null,"work_id":"aff2dc05-5f62-40a7-bbda-2e923755768b","year":2017},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.825410Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:36d231b1d663c04a58d28e7d5d27e2bf02d1ad50085940aaff857e1ac587b81c","observation_id":"ded20530-0fd6-4e97-813b-450416d4c9cf","resolution":{"observed_at":"2026-08-10T23:15:25.452161Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.432013Z","title":"Extending the applicability of the ani deep learning molecular potential to sulfur and halogens,","venue":null,"work_id":"7ae7f6fb-ee08-4faa-97dc-72dc29c26e04","year":2020},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.829774Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:bdbe308828ddede710a2db79fbd04651659d8852a0d09e89567e28a64ee11152","observation_id":"e39ae3e0-3b75-4670-b1ad-9ded8e6e127e","resolution":{"observed_at":"2026-08-10T23:15:25.437393Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.418286Z","title":"Generalized neural-network representation of high-dimensional potential-energy surfaces,","venue":null,"work_id":"28df0960-0479-4f73-8b47-d36d520afd27","year":2007},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.834659Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:e2149fea39c7246606474c9469a9645f51560ac29664aee602f6b51cf1977e11","observation_id":"1156a720-1dc0-44b6-b3aa-b0c67644610b","resolution":{"observed_at":"2026-08-10T23:15:25.422684Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.405549Z","title":"Embedded atom neural network poten- tials: Efficient and accurate machine learning with a physically inspired representation,","venue":null,"work_id":"931a2622-89df-424b-96a0-92038677273b","year":2019},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.839025Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:ade35c51c820e3043bb6247e71467862cc7ddcc011667a9e2d2f5376b26c6faf","observation_id":"2ada07c6-5e92-4256-be6e-d4d1e127bf60","resolution":{"observed_at":"2026-08-10T23:15:25.410273Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.392060Z","title":"Equivariant message passing for the prediction of tensorial properties and molecular spectra,","venue":null,"work_id":"c01d795b-1c09-4405-82f5-f73569d26b51","year":2021},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.843302Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:0468c53f8eec412ea5331c390a1c7e74cc7a8de29e4e170c1283e9770fc54170","observation_id":"b8210c44-4685-42d9-b377-d8e07cbe3b47","resolution":{"observed_at":"2026-08-10T23:15:25.396487Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:24.847946Z","title":"E (3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.847946Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:0cea05ab0217199faab2c0adaae5ddbf1e823503fe80c999a669ac222b84fa91","observation_id":"eccdc498-a5fe-4c06-bd2c-ca8afdbd7b2b","resolution":{"observed_at":"2026-08-10T23:15:24.847946Z","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":"2026-08-10T23:15:25.369699Z","title":"Newtonnet: a newtonian message passing network for deep learning of interatomic potentials and forces,","venue":null,"work_id":"5fcd06ba-5030-40ba-8e57-680839ed322b","year":2022},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.852215Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:ed335705e22bc1315ee206b73f100cd319ae0be6e8351d95715f70ca039eeb53","observation_id":"c54bdf03-dc2c-44e1-96fa-3b7716d750bd","resolution":{"observed_at":"2026-08-10T23:15:25.374057Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.355560Z","title":"Spookynet: Learning force fields with electronic degrees of freedom and nonlocal effects,","venue":null,"work_id":"1baff84a-0a38-49b1-aab3-ac34aa21b7fc","year":2021},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.856393Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:c73a2efaab846a3cbc7697ce0eb93191f5e83717f35b382c7b903305f84b2a2d","observation_id":"d674b0d8-15f0-49b7-a95d-03335425ad1a","resolution":{"observed_at":"2026-08-10T23:15:25.360670Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2011.14115","last_updated":"2022-04-05T11:57:23Z","snapshot_observed_at":"2026-08-18T17:37:54.625606Z","submitted_at":"2020-11-28T11:39:42Z","title":"Fast and Uncertainty-Aware Directional Message Passing for Non-Equilibrium Molecules","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2011.14115","snapshot_observed_at":"2026-08-10T23:15:24.860975Z","title":"Fast and uncertainty-aware directional message passing for non-equilibrium molecules,","venue":null,"work_id":null,"year":2011},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.860975Z"},"links":{"cited_paper":"/paper/2011.14115","citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:933b6dc094ea535c0dd4815da5560431fc3cf8622dcce6afd91df83e8b5cfb08","observation_id":"cec64180-8790-4afc-8f84-6b25e4f56520","resolution":{"observed_at":"2026-08-10T23:15:24.860975Z","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":"2026-08-10T23:15:24.865585Z","title":"A universal graph deep learning interatomic potential for the periodic table,","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.865585Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:417a21e03af73b366a8fcef6a3322c7ccc2c540b129a5620199681ee43e65eac","observation_id":"8650c8f2-11d6-45a7-bb01-9d04d901f13c","resolution":{"observed_at":"2026-08-10T23:15:24.865585Z","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":"2026-08-10T23:15:25.329172Z","title":"Atomistic line graph neural network for improved materials property predictions. npj computational materials, 7 (1): 185,","venue":null,"work_id":"59922c9b-7dff-4836-9309-10ae569a734e","year":2021},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.869612Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:edb32b840dbad36ba43c960939261dd80ecebc8c7c64910c02cd9fb4a7d30083","observation_id":"bde588f7-83ed-497c-ab76-48640552173a","resolution":{"observed_at":"2026-08-10T23:15:25.333653Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.315136Z","title":"Graph networks as a universal machine learning framework for molecules and crystals,","venue":null,"work_id":"25765302-5ea4-4fb2-961a-5b2420953552","year":2019},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.873536Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:28ba6f3918d05505af6c900a0c1383dbdb069ca7220b08093ca7c3499e0d68ff","observation_id":"8bebab16-aef7-4400-8e5e-027d567c8cd8","resolution":{"observed_at":"2026-08-10T23:15:25.319374Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.302636Z","title":"Gptff: A high-accuracy out-of-the- box universal ai force field for arbitrary inorganic materials,","venue":null,"work_id":"f8159cdf-f2db-406a-8fe2-a5fc9b39735d","year":2024},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.877944Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:64dea099588b4389e3da3486a9329de264395a9ba250b566f84d2869ceec404c","observation_id":"33a16f76-c6de-4f0f-a1f6-869f2ce06c39","resolution":{"observed_at":"2026-08-10T23:15:25.307162Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.289168Z","title":"A survey on compiler autotuning using machine learning,","venue":null,"work_id":"f1f05e83-44ca-46c9-9d9c-73035b386ab8","year":2018},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.882073Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:a2968e028267687b3f1fe883efefbff5553b899abccd3ec7e2363e5557b623a4","observation_id":"10f6940b-b2bc-4854-b4e6-a9b0f9465dbd","resolution":{"observed_at":"2026-08-10T23:15:25.293580Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.274421Z","title":"and y. chen. 2018. tvm: An automated end-to-end optimizing compiler for deep learning,","venue":null,"work_id":"3be886ee-313e-480f-a620-5c53f8aa2f4b","year":2018},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.886904Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:ff7a842a0172096ac786328ced0de06621b5d927fd1456f309a245164d8e1b8b","observation_id":"adf0298f-9aad-40f3-9ede-76ffe8b71023","resolution":{"observed_at":"2026-08-10T23:15:25.278756Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.258634Z","title":"Cdnet: A real-time and robust crosswalk detection network on jetson nano based on yolov5,","venue":null,"work_id":"4240e11c-bb24-48ac-8696-4e35f4e98314","year":2022},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.891190Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:6fa83447e4d374c57d4ffbe65aeec880ee0e4335d917dad346886d69fb25f688","observation_id":"381da33b-5758-44ce-b380-1d99639e1f33","resolution":{"observed_at":"2026-08-10T23:15:25.264172Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.243603Z","title":"Ansor: Generating {High-Performance} tensor programs for deep learning,","venue":null,"work_id":"bc321040-f35a-4716-9f32-de1598d13c9b","year":2020},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.895286Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:cf510bfca165da39107ff3d348d6cd0428e2942e79231b13aa6b738afee04069","observation_id":"28c157a6-5376-4f83-9b18-6182fc45f32c","resolution":{"observed_at":"2026-08-10T23:15:25.248890Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2301.13062","last_updated":"2023-01-30T17:03:33Z","snapshot_observed_at":"2026-08-19T00:33:37.815933Z","submitted_at":"2023-01-30T17:03:33Z","title":"Operator Fusion in XLA: Analysis and Evaluation","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2301.13062","snapshot_observed_at":"2026-08-10T23:15:24.900162Z","title":"Operator fusion in xla: Analysis and evalua- tion,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.900162Z"},"links":{"cited_paper":"/paper/2301.13062","citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:9f28c7482b752dcc8ecd0bee14af6d2f737f6c58c58d914f1276e6c615676371","observation_id":"4b8698df-43ab-4610-bd81-3649f8b92484","resolution":{"observed_at":"2026-08-10T23:15:24.900162Z","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":"2026-08-10T23:15:25.228085Z","title":"Pytorch: An imperative style, high-performance deep learning library,","venue":null,"work_id":"1958a4aa-9cc7-459c-95a0-4e3b732d21a6","year":2019},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.904582Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:66f3da1796f15b89b0ffab383acb7b905a9ce49c4ad9466e3c25a111c2d87eb5","observation_id":"35f35ac9-abaf-4c49-a302-3b9138c865d4","resolution":{"observed_at":"2026-08-10T23:15:25.232667Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2008.05000","last_updated":"2021-03-15T15:27:59Z","snapshot_observed_at":"2026-08-19T11:55:05.221872Z","submitted_at":"2020-08-11T20:53:50Z","title":"Degree-Quant: Quantization-Aware Training for Graph Neural Networks","version":3},"cited_work":{"arxiv_id":"2008.05000","doi":null,"metadata_source":"pith","pith_arxiv_id":"2008.05000","snapshot_observed_at":"2026-08-10T23:15:25.078519Z","title":"Degree-Quant: Quantization-Aware Training for Graph Neural Networks","venue":"cs.LG","work_id":"8b6a23fd-a329-450d-820b-fd7c2c9ac9ef","year":2020},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.908925Z"},"links":{"cited_paper":"/paper/2008.05000","citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:a5b06d50eb801b43bc958e5bf863fde099ca38e6c822bc1bb4094c0a297fe5ed","observation_id":"7945c91e-e94d-4453-8a1f-45bedaf84fcf","resolution":{"observed_at":"2026-08-10T23:15:25.085369Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.212795Z","title":"Qgtc: accelerating quantized graph neural networks via gpu tensor core,","venue":null,"work_id":"7c761017-01d6-48ae-a60c-535ac27c9b05","year":2022},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.913256Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:4932a56107bd47615b12dfa929f83dd9a2d3ecef8449d3fe525ee2e0394c912c","observation_id":"783540da-91cb-4539-b035-d33d7adcff2b","resolution":{"observed_at":"2026-08-10T23:15:25.218385Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.197765Z","title":"Vq-gnn: A universal framework to scale up graph neural networks using vector quantization,","venue":null,"work_id":"2e009d23-9a42-4ed5-b2aa-b2bc278fb27b","year":2021},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.917170Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:d0c0f01cb25216e9bbebcabea051a3f93edc0aa0357851409b789262f12b0417","observation_id":"0c4d5a18-7918-4ea4-94f1-ffa4dea6fab5","resolution":{"observed_at":"2026-08-10T23:15:25.202220Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.180155Z","title":"Sgquant: Squeezing the last bit on graph neural networks with specialized quantization,","venue":null,"work_id":"b4dc4a17-c5c7-47b3-be47-2c57f33e8fd7","year":2020},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.921174Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:3dffa6bc075be206518783d060cb039614f7ca91db8cea44a4a6a5cd3a688f94","observation_id":"5c05c234-6d7f-46d7-8dfa-21169d39f7c0","resolution":{"observed_at":"2026-08-10T23:15:25.188077Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:24.925411Z","title":"Rlekf: an optimizer for deep potential with ab initio accuracy,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.925411Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:c18394e102ad51c8e04e265dfa4de6d43238b73ce15ad8b3012e6dc4851fc176","observation_id":"0490c001-9141-4dd4-93dc-c941e7e489c3","resolution":{"observed_at":"2026-08-10T23:15:24.925411Z","resolver_source":null,"status":"malformed_identifier"},"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":"2026-08-10T23:15:25.165412Z","title":"Neural network force field training based on reor- ganized layer-wised extended kalman filter,","venue":null,"work_id":"4a56494d-9616-4035-a8fb-8dd9264f0835","year":null},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.929303Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:85e117f277cc389415a159c0d1d34316e40aa76fdeced7a59136383ad4b24c99","observation_id":"97deb40b-fe9c-476e-a195-4104c26b5b9e","resolution":{"observed_at":"2026-08-10T23:15:25.171484Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.151049Z","title":"The mlip package: moment tensor potentials with mpi and active learning,","venue":null,"work_id":"34d14b54-fcb6-43b4-8281-aaa9034f3397","year":null},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.933578Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:65c9e1eaf6fc6cca217fda1729ac15248ee9afaed70200d3e0b872b0ecba108c","observation_id":"0cd44058-2757-403e-87f5-8ec01dc1a11c","resolution":{"observed_at":"2026-08-10T23:15:25.155729Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:25.136605Z","title":"Spectral neighbor analysis method for automated generation of quantum-accurate interatomic potentials,","venue":null,"work_id":"6447438c-f3f4-4423-879f-aef8408d56bd","year":2015},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.941707Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:4c07458a4acc346ef567389fefdf2e8924016f5019252955a71e22af6b338feb","observation_id":"d89560a6-7d27-40d9-9942-86b67fb0ed49","resolution":{"observed_at":"2026-08-10T23:15:25.141784Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-19T06:32:44.657259+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+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":"2026-08-10T23:15:24.945850Z","title":"Training one deepmd model in minutes: a step towards online learning,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.945850Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:e5f1da9582551a0e26525cd82bf764d5a2599cedaae7991d9ee420f25306194f","observation_id":"b7315a1e-1327-4c59-8f0a-da25803494f2","resolution":{"observed_at":"2026-08-10T23:15:24.945850Z","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":"2026-08-10T23:15:24.937526Z","title":"Available: https://dx.doi.org/10.1088/2632-2153/abc9fe","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs","version":2},"reference_index":2020,"source":"pdf_text","source_observed_at":"2026-08-10T23:15:24.937526Z"},"links":{"citing_paper":"/paper/2412.20796"},"observation_digest":"sha256:9e7a357add6682b0296225468cf6b4e6772a19a8db45c30f2ede3a5f4771e0ed","observation_id":"503e9179-91bb-4c3a-8017-f7f343f72fb5","resolution":{"observed_at":"2026-08-10T23:15:24.937526Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2412.20796","last_updated":"2025-03-14T08:01:35Z","latest_version":2,"primary_category":"cs.DC","snapshot_observed_at":"2026-08-18T19:19:28.240904Z","submitted_at":"2024-12-30T08:38:09Z","title":"FastCHGNet: Training one Universal Interatomic Potential to 1.5 Hours with 32 GPUs"},"reference_resolution":{"displayed":41,"state_counts":{"malformed_identifier":1,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":7,"verified_exact":1,"verified_fuzzy":32},"total_outbound_references":41},"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":"2026-08-19T06:32:44.657259+00:00","source":"crossref"},{"observed_at":"2026-08-19T06:32:39.956319+00:00","source":"retraction_watch"}],"thesis":"As of 19 August 2026, this Paper Citation Record lists 41 of 41 outbound references and 1 inbound Pith citation observation for arXiv:2412.20796."}