{"as_of":"2026-08-15T22:19:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:601177c5d99bb927c3800b9aa06a2cf501adcaed9c5969e103cf41c481d5c926","coverage":[{"denominator":63,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":63,"source":"paper_references, paper_reference_links","source_observed_at":"2026-07-31T18:39:46.194028Z","state":"measured"},{"denominator":63,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":63,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-15T06:32:42.880941+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2607.28074/citation-record","integrity":"/paper/2607.28074/integrity","json":"/paper/2607.28074/citation-record.json","paper":"/paper/2607.28074"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-07-31T18:39:42.099906Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":1,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:42.099906Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:2b3ebd200682c48096d0188b32dfe2bc5b99d1d96130df852e5e8de48916bef8","observation_id":"137be435-3747-4572-a2ac-c67e63166f40","resolution":{"observed_at":"2026-07-31T18:39:42.099906Z","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":"2026-07-31T18:39:42.220861Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":3,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:42.220861Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:259fc0023549db9cf9f18da87fe4890c3dc600c5521e03d6ae838313a919b6c8","observation_id":"97ae6b53-393f-4d7f-8edf-281dda5ba50e","resolution":{"observed_at":"2026-07-31T18:39:42.220861Z","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":"2026-07-31T18:39:42.280449Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":4,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:42.280449Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:93717b300cde58bb015b173b0ad77268dd2f4aa7cb91329eb2cf3b50889c2588","observation_id":"e9cf8cd9-4bc5-42d6-b3c3-5bbe4b69bf7e","resolution":{"observed_at":"2026-07-31T18:39:42.280449Z","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":"2026-07-31T18:39:42.344044Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":5,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:42.344044Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:52b288d8e186f5651da054d8122245d95a20634fb5fe96fa9b049696eaacd12c","observation_id":"bfd7e3f8-b3f5-472e-ba1e-ef92ea166f1b","resolution":{"observed_at":"2026-07-31T18:39:42.344044Z","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":"2026-07-31T18:39:42.405512Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":6,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:42.405512Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:9e81fc3a77c80ee37e5312afc47e1df82c9a9dc3ca1e71cf81d91fee2266b65d","observation_id":"9f75d6fe-92de-40f6-abc4-a22bf480fbc9","resolution":{"observed_at":"2026-07-31T18:39:42.405512Z","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":"2026-07-31T18:39:42.434974Z","title":"and Del Verme, Manuel and Marty, Tom and Boisvert, L","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":7,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:42.434974Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:a737de77909f79562adeb3f308710220c765ebecc060681d14e2fead7cb5bdc5","observation_id":"b3330d9b-3d34-464e-aa66-07bec67f5b71","resolution":{"observed_at":"2026-07-31T18:39:42.434974Z","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":"2026-07-31T18:39:42.496346Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":8,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:42.496346Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:ddcd9b027c8cade5ffe3dbbf0776b23ab395546670eebc7e6a857ab7f576b309","observation_id":"9227938a-5950-4691-b5eb-9a242c456799","resolution":{"observed_at":"2026-07-31T18:39:42.496346Z","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":"2026-07-31T18:39:42.657579Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":11,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:42.657579Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:7ec5c87c8ba51e87cf4ce80c4496785e6004468e146535d93dda95be21af0b74","observation_id":"e4493f9e-a9e6-47a8-90b5-9a948ff06192","resolution":{"observed_at":"2026-07-31T18:39:42.657579Z","resolver_source":null,"status":"parse_uncertain"},"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":"2026-07-31T18:39:42.707914Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":12,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:42.707914Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:dc3a1bb1fb20d9be9c1e19c426f882da5f19ef7b2cd782d5ff5e87cdaf7a206b","observation_id":"7a21511f-c9d6-4a39-ac16-0fcd77d39eaa","resolution":{"observed_at":"2026-07-31T18:39:42.707914Z","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":"2026-07-31T18:39:42.769560Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":13,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:42.769560Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:04cebc9c4840f8988c25708d1aecca875b5c1c0eb9fbadff32ffa56a0b5ce4e9","observation_id":"6ba95c86-dcd9-4d04-a613-7f0077140cf0","resolution":{"observed_at":"2026-07-31T18:39:42.769560Z","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":"2026-07-31T18:39:42.800314Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":14,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:42.800314Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:65b19cb1ca21b5aa2660e5d1c5f4b894590f9de19b96cd2a98c9c2e687df40ea","observation_id":"3eb602ed-63ed-4338-8303-3290f8803ffc","resolution":{"observed_at":"2026-07-31T18:39:42.800314Z","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":"2026-07-31T18:39:42.860346Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":15,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:42.860346Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:af521faf4efdc9fec05f49c1b6e0c4758f4d05b654d11f9392053a2ce4f2a09a","observation_id":"27e71320-eed2-4bc9-af56-545f557606a1","resolution":{"observed_at":"2026-07-31T18:39:42.860346Z","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":"2026-07-31T18:39:42.921713Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":16,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:42.921713Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:20860bafcbd152f18e04a1b3d028f2b20c216c4cec816eb247e68c185d79ee0f","observation_id":"77cc0770-42ab-4bc7-94bd-8a6572dd1a20","resolution":{"observed_at":"2026-07-31T18:39:42.921713Z","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":"2026-07-31T18:39:42.963042Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":17,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:42.963042Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:e8eec1b981831ca2a116aef7e569307a92a802815d9980b1000c9a2786048a66","observation_id":"93db081d-9218-4b64-b1f8-c7bf8d839718","resolution":{"observed_at":"2026-07-31T18:39:42.963042Z","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":"2026-07-31T18:39:43.034542Z","title":"2026 , note =","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":19,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:43.034542Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:09b5c58f1c00c67051c1275652da80b215a35a5ba62540e44e467e3d5181aa3a","observation_id":"caeaddcc-df89-4601-83ae-618f76169aaa","resolution":{"observed_at":"2026-07-31T18:39:43.034542Z","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":"2026-07-31T18:39:43.082073Z","title":"Understanding","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":20,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:43.082073Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:508ef093d15b2c8777bb084c78197a1c2686129bf9bba466a9f0ad3b4e99f800","observation_id":"2a0bc408-378f-458a-8e10-5a819e6f843e","resolution":{"observed_at":"2026-07-31T18:39:43.082073Z","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":"2026-07-31T18:39:43.137803Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":21,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:43.137803Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:1923b5aedfd0da1aa3637d5b0727276e1f4ac793141161fa8c43aca8da75fe0b","observation_id":"3d8d16a0-cea3-4a52-bc08-7b1193148510","resolution":{"observed_at":"2026-07-31T18:39:43.137803Z","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":"2026-07-31T18:39:43.231445Z","title":", booktitle =","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":23,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:43.231445Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:c1a040a6c654daf6184463e76ac80cc5c61e7aa18d0b568165e87b70c435c07a","observation_id":"f6930151-03b8-4d97-b6ef-1f4b5c3caf8d","resolution":{"observed_at":"2026-07-31T18:39:43.231445Z","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":"2026-07-31T18:39:43.320099Z","title":"and Zhang, Hao and Gonzalez, Joseph E","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":25,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:43.320099Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:a0cc23f157de631628d07295388bab264877c1c4c69fd6e27805198f4aae2764","observation_id":"5ead834c-4801-45c8-84ff-08ac8fa564dc","resolution":{"observed_at":"2026-07-31T18:39:43.320099Z","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":"2026-07-31T18:39:43.384309Z","title":", journal =","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":26,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:43.384309Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:26314cb5af062e3ac5fcce1fb4490743837b0874383d20022006762f0b4f9aee","observation_id":"5b791473-197f-4b28-825b-154fd5368ef5","resolution":{"observed_at":"2026-07-31T18:39:43.384309Z","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":"2026-07-31T18:39:43.444932Z","title":"Advances in Neural Information Processing Systems (NeurIPS) , year =","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":27,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:43.444932Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:3b34569ce188ff3acad34eaed50dfe634252c73d2231bfca82358654737db05d","observation_id":"1c32bc6f-f3c1-4ff3-91c8-f16b760a1d00","resolution":{"observed_at":"2026-07-31T18:39:43.444932Z","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":"2026-07-31T18:39:43.499407Z","title":"International Conference on Machine Learning (ICML) , year =","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":28,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:43.499407Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:757131e62ba72eba4aa29f571d15e7f6bb4f20f126bbb7ac1854026b3735ad4f","observation_id":"478fffa9-b87c-4968-99bf-8eae39fa3bba","resolution":{"observed_at":"2026-07-31T18:39:43.499407Z","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":"2026-07-31T18:39:43.615133Z","title":"Science , volume =","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":30,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:43.615133Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:1e7151241fd697be8100c24f5a72f0548720900847b44cf1eacced61ca46c712","observation_id":"7afd6df2-0b4f-4445-9091-00356041bda3","resolution":{"observed_at":"2026-07-31T18:39:43.615133Z","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":"2026-07-31T18:39:43.678553Z","title":"and Zhang, Hao and Stoica, Ion , booktitle =","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":31,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:43.678553Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:f574b8cabe2fe70d5c1b8ac8f1b6de0591f7a59839063dd7feb6700aae7174ee","observation_id":"6e4310b7-f3b4-48ae-937e-06906fe3bfcf","resolution":{"observed_at":"2026-07-31T18:39:43.678553Z","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":"2026-07-31T18:39:43.769476Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":32,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:43.769476Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:675b246ef9f8ea5072d435dc337f6f917a09c50e4587c4ec3ec3044c5c95cae7","observation_id":"5072cafa-bed6-4574-a569-93fc801b85fc","resolution":{"observed_at":"2026-07-31T18:39:43.769476Z","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":"2026-07-31T18:39:43.855970Z","title":"2024 , howpublished =","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":33,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:43.855970Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:6dba29985c8e3413ce603d0d243579566a285a9e0549561f61fdc30cba6eb103","observation_id":"78100054-879c-437a-b52e-979df92b736e","resolution":{"observed_at":"2026-07-31T18:39:43.855970Z","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":"2026-07-31T18:39:43.913264Z","title":"2025 , howpublished =","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":34,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:43.913264Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:2761d2679f8d0843e1ffaab426759cf4f2696b4dabc8b83ef2a238bae1787723","observation_id":"d40b41aa-bb53-4d70-a307-6ef48c35d523","resolution":{"observed_at":"2026-07-31T18:39:43.913264Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2604.06126","last_updated":"2026-04-07T17:38:15Z","snapshot_observed_at":"2026-08-11T07:48:07.347996Z","submitted_at":"2026-04-07T17:38:15Z","title":"Gym-Anything: Turn any Software into an Agent Environment","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2604.06126","snapshot_observed_at":"2026-07-31T18:39:43.945347Z","title":"Gym-Anything : Turn any software into an agent environment","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":35,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:43.945347Z"},"links":{"cited_paper":"/paper/2604.06126","citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:f1d4913c4d21a659de742d68bce398e2f5daacf0a032b0f93c52a468436ceabf","observation_id":"dad33717-ee1b-4e73-9c3e-6c070d8b1ead","resolution":{"observed_at":"2026-07-31T18:39:43.945347Z","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":"2026-07-31T18:39:44.000362Z","title":"Model context protocol","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":36,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.000362Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:4d55d6eda837c3a9bd45cf566de1e1bd62a797d6e32a99364b300529f7687f07","observation_id":"b5a63d26-a98c-4072-aaed-6fffbef3d932","resolution":{"observed_at":"2026-07-31T18:39:44.000362Z","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":"2026-07-31T18:39:44.064630Z","title":"Fara-7B : An efficient agentic model for computer use","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":37,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.064630Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:6cea9fc1104d156b463bebed58e6ba3a9f5d78b5223a6527f141687586d153db","observation_id":"dc0c6882-bd4d-42a9-a945-17462ab5a9b7","resolution":{"observed_at":"2026-07-31T18:39:44.064630Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2606.20785","last_updated":"2026-07-22T17:47:01Z","snapshot_observed_at":"2026-08-11T15:37:11.056561Z","submitted_at":"2026-06-18T17:53:03Z","title":"Fara-1.5: Scalable Learning Environments for Computer Use Agents","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2606.20785","snapshot_observed_at":"2026-07-31T18:39:44.098849Z","title":"Fara-1.5 : Scalable learning environments for computer use agents","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":38,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.098849Z"},"links":{"cited_paper":"/paper/2606.20785","citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:18966cafcfc12601b74b158b7d3606749975eef110efabe09ae4bfb0d9f520a3","observation_id":"acc004aa-5078-4155-a4ce-8b2af06ac86f","resolution":{"observed_at":"2026-07-31T18:39:44.098849Z","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":"2026-07-31T18:39:44.149400Z","title":"Browserbase: Headless browser infrastructure for AI agents","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":39,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.149400Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:e11e5abbaf06a73c7f809114273dfadbfe17a94271366ade7d86a2c5fd8b0bed","observation_id":"3e7a7788-e82b-4ec0-802e-6956c94d64ee","resolution":{"observed_at":"2026-07-31T18:39:44.149400Z","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":"2026-07-31T18:39:44.200623Z","title":"Inside A irbnb: Adding data to the debate","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":40,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.200623Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:2e6272e683cfd5cadc1ff66d889167a794eae5e1d31eace187b1b4d85bcb788c","observation_id":"d2e7bd58-7b23-4c99-abd8-40783b950a56","resolution":{"observed_at":"2026-07-31T18:39:44.200623Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2412.05467","last_updated":"2025-02-28T16:02:27Z","snapshot_observed_at":"2026-08-13T05:20:59.084759Z","submitted_at":"2024-12-06T23:43:59Z","title":"The BrowserGym Ecosystem for Web Agent Research","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2412.05467","snapshot_observed_at":"2026-07-31T18:39:44.230646Z","title":"Xu, Siva Reddy, Quentin Cappart, Graham Neubig, Ruslan Salakhutdinov, Nicolas Chapados, and Alexandre Lacoste","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":41,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.230646Z"},"links":{"cited_paper":"/paper/2412.05467","citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:8d8cbe6bf504b691ef34e8a3adff6eff9b58978e61c47c1499ee144b9d168fcd","observation_id":"ad60621c-c015-4277-a748-278b51c386c7","resolution":{"observed_at":"2026-07-31T18:39:44.230646Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2501.12948","last_updated":"2026-01-04T03:57:36Z","snapshot_observed_at":"2026-08-15T12:33:55.451951Z","submitted_at":"2025-01-22T15:19:35Z","title":"DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2501.12948","snapshot_observed_at":"2026-07-31T18:39:44.261517Z","title":"DeepSeek-R1 : Incentivizing reasoning capability in LLM s via reinforcement learning","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":42,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.261517Z"},"links":{"cited_paper":"/paper/2501.12948","citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:cafa344f8d64bd06c83da268ac2ac34d466caaa18d4e054e7c17dc528a468ee5","observation_id":"a1538d34-b7b7-45c8-a0de-757271f13838","resolution":{"observed_at":"2026-07-31T18:39:44.261517Z","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":"2026-07-31T18:39:44.291598Z","title":"Mind2Web : Towards a generalist agent for the web","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":43,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.291598Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:51d18a3fc378b26bc2147e87829ee87802c668d659e3d150f50bc968d73567b1","observation_id":"d459fd57-4c47-4146-af16-ae518ed1b407","resolution":{"observed_at":"2026-07-31T18:39:44.291598Z","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":"2026-07-31T18:39:44.380532Z","title":"Emergent complexity and zero-shot transfer via unsupervised environment design","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":44,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.380532Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:860222dc219bfaa099fcdf132988ea9d3e2efc02bbb6c8ecb634ed91c3817d09","observation_id":"c5b930c6-694c-4856-b76e-552c30d4f2f5","resolution":{"observed_at":"2026-07-31T18:39:44.380532Z","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":"2026-07-31T18:39:44.440845Z","title":"Laradji, Manuel Del Verme, Tom Marty, L \\'e o Boisvert, Megh Thakkar, Quentin Cappart, David Vazquez, Nicolas Chapados, and Alexandre Lacoste","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":45,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.440845Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:3839fc44393bee807ebe3fdae5cc7b31262d0cdecf33c761b17e11293bd160d5","observation_id":"3a24864c-c52d-4262-9f1b-5e9d5bbc9ae1","resolution":{"observed_at":"2026-07-31T18:39:44.440845Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2401.13919","last_updated":"2024-06-06T18:37:34Z","snapshot_observed_at":"2026-08-15T01:23:25.524198Z","submitted_at":"2024-01-25T03:33:18Z","title":"WebVoyager: Building an End-to-End Web Agent with Large Multimodal Models","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2401.13919","snapshot_observed_at":"2026-07-31T18:39:44.498942Z","title":"WebVoyager : Building an end-to-end web agent with large multimodal models","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":46,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.498942Z"},"links":{"cited_paper":"/paper/2401.13919","citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:b27bb8ff7b865dd5c1f582fdd038497d71b4c5deb13e63d59aa711f5a723259d","observation_id":"8f18f7f5-f3a4-46a2-81dc-a7ada469be0f","resolution":{"observed_at":"2026-07-31T18:39:44.498942Z","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":"2026-07-31T18:39:44.555932Z","title":"VisualWebArena : Evaluating multimodal agents on realistic visual web tasks","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":47,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.555932Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:210d479fbd7324b01330fa357cc52e820f6fbbec77f496ae8f26b0452c9ade98","observation_id":"516fee1f-beb1-4070-931e-a7e2f9e0d402","resolution":{"observed_at":"2026-07-31T18:39:44.555932Z","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":"2026-07-31T18:39:44.614348Z","title":"Gonzalez, Hao Zhang, and Ion Stoica","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":48,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.614348Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:f35e7e1a4ee53367691ff834165ba361362aa5044f7e58ea37fb011810357824","observation_id":"0919cced-15cb-4e5a-96e9-95685a0d6df0","resolution":{"observed_at":"2026-07-31T18:39:44.614348Z","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":"2026-07-31T18:39:44.668691Z","title":"A practical recipe for training computer-use agents with reinforcement learning","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":49,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.668691Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:b4f87cf91b54ed20df2d816969f00019ee301ee211c7d5ee2bb7b2f758be4be3","observation_id":"963043fd-2d3e-4a6c-bd53-13375940904d","resolution":{"observed_at":"2026-07-31T18:39:44.668691Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2503.20783","last_updated":"2025-10-06T09:30:03Z","snapshot_observed_at":"2026-08-13T12:34:54.476684Z","submitted_at":"2025-03-26T17:59:14Z","title":"Understanding R1-Zero-Like Training: A Critical Perspective","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2503.20783","snapshot_observed_at":"2026-07-31T18:39:44.726861Z","title":"Understanding R1 -zero-like training: A critical perspective","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":50,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.726861Z"},"links":{"cited_paper":"/paper/2503.20783","citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:cbb2bab1b58c3d1a4a71795cb4365406148419cbf1801869096574281a3193d8","observation_id":"20d3f0fb-f289-4b35-9a28-fd46ca266413","resolution":{"observed_at":"2026-07-31T18:39:44.726861Z","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":"2026-07-31T18:39:44.781401Z","title":"Playwright: Fast and reliable end-to-end testing for modern web apps","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":51,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.781401Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:2bd45e77a31ad0c22c21d4f7521712726c7e1edfa3cc2fa8dcb9cad5d94639d3","observation_id":"69c1a6f3-94b3-42fe-ac3a-1fed6b139f4e","resolution":{"observed_at":"2026-07-31T18:39:44.781401Z","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":"2026-07-31T18:39:44.840681Z","title":"Evolving curricula with regret-based environment design","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":52,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.840681Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:5941a970eb7b2aa8c7283ac09167f64e9fd16a892c6762450f65b59511de40ba","observation_id":"fe3b5591-d2ab-4162-a9ad-898ea62a11ed","resolution":{"observed_at":"2026-07-31T18:39:44.840681Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2003.04664","last_updated":"2020-05-28T20:51:40Z","snapshot_observed_at":"2026-08-09T01:09:42.786424Z","submitted_at":"2020-03-10T12:38:31Z","title":"Automatic Curriculum Learning For Deep RL: A Short Survey","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2003.04664","snapshot_observed_at":"2026-07-31T18:39:44.885264Z","title":"Automatic curriculum learning for deep RL : A short survey","venue":null,"work_id":null,"year":2003},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":53,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.885264Z"},"links":{"cited_paper":"/paper/2003.04664","citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:89fad51e6c7ac9260ffbc5b84b0da854556b4d6032dfa0da7634d7b6dfe0af42","observation_id":"32e4cb41-2b30-4429-8eec-1b1de28e5764","resolution":{"observed_at":"2026-07-31T18:39:44.885264Z","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":"2026-07-31T18:39:44.892293Z","title":"Qwen3.5 : Towards native multimodal agents","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":54,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.892293Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:63255bb2b25c93d4a191bd474c7cd0b877c3489e7e5bff841db70d0069504b74","observation_id":"419ac470-8849-4ecf-aa91-ce1325ed6b98","resolution":{"observed_at":"2026-07-31T18:39:44.892293Z","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":"2026-07-31T18:39:44.967348Z","title":"AndroidWorld : A dynamic benchmarking environment for autonomous agents","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":55,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:44.967348Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:e8f99428ff68c1f5eeb47385584c1035138f7767866ed7f9a18cd64ab136ac7d","observation_id":"ac15a7c0-cf7f-4449-be18-ec00a5c15584","resolution":{"observed_at":"2026-07-31T18:39:44.967348Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2604.06240","last_updated":"2026-04-05T06:26:05Z","snapshot_observed_at":"2026-08-14T06:57:56.493901Z","submitted_at":"2026-04-05T06:26:05Z","title":"The Art of Building Verifiers for Computer Use Agents","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2604.06240","snapshot_observed_at":"2026-07-31T18:39:45.020535Z","title":"The art of building verifiers for computer use agents","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":56,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:45.020535Z"},"links":{"cited_paper":"/paper/2604.06240","citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:22191983e9d363fb8c4296e80374229a8bf8b83740babca102b708674de6894e","observation_id":"d20d5580-c069-4781-8fde-270123ca1662","resolution":{"observed_at":"2026-07-31T18:39:45.020535Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2402.03300","last_updated":"2024-04-27T15:25:53Z","snapshot_observed_at":"2026-08-06T14:58:42.911363Z","submitted_at":"2024-02-05T18:55:32Z","title":"DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2402.03300","snapshot_observed_at":"2026-07-31T18:39:45.058207Z","title":null,"venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":57,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:45.058207Z"},"links":{"cited_paper":"/paper/2402.03300","citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:7032dc3a7d5cabae6716a04a4178277c241747434c0bed2c285cc5aa9c0757eb","observation_id":"e7a9fb0e-59e4-4591-a6b1-e821f479999f","resolution":{"observed_at":"2026-07-31T18:39:45.058207Z","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":"2026-07-31T18:39:45.117101Z","title":"A general reinforcement learning algorithm that masters chess, shogi, and go through self-play","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":58,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:45.117101Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:bcf0e9cde18bf3672d3ca1314c08e9c0f8d9ae3bbc6831fe980c5eb5fcea2144","observation_id":"5c00a6d5-d131-48d5-a20d-a0a01aa8733a","resolution":{"observed_at":"2026-07-31T18:39:45.117101Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2605.25624","last_updated":"2026-06-08T08:32:35Z","snapshot_observed_at":"2026-08-04T20:10:29.546282Z","submitted_at":"2026-05-25T09:28:03Z","title":"CUA-Gym: Scaling Verifiable Training Environments and Tasks for Computer-Use Agents","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2605.25624","snapshot_observed_at":"2026-07-31T18:39:45.230847Z","title":"CUA-Gym : Scaling verifiable training environments and tasks for computer-use agents","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":59,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:45.230847Z"},"links":{"cited_paper":"/paper/2605.25624","citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:a33fb34574a8396bade215ec3efb98f2c65c3c94cb4cb80d3e033773440e3e69","observation_id":"fa883f08-0499-4e92-816c-0fb8e8c3d946","resolution":{"observed_at":"2026-07-31T18:39:45.230847Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1901.01753","last_updated":"2019-02-21T01:17:31Z","snapshot_observed_at":"2026-08-14T17:34:20.914747Z","submitted_at":"2019-01-07T11:28:36Z","title":"Paired Open-Ended Trailblazer (POET): Endlessly Generating Increasingly Complex and Diverse Learning Environments and Their Solutions","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1901.01753","snapshot_observed_at":"2026-07-31T18:39:45.339188Z","title":null,"venue":null,"work_id":null,"year":1901},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":60,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:45.339188Z"},"links":{"cited_paper":"/paper/1901.01753","citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:868ee80ffbe59c1c4d6437ef7aed8c5a6505a41efc144a78d5d334204c772e78","observation_id":"65569fd2-e89d-400b-b080-cddc1b053f8d","resolution":{"observed_at":"2026-07-31T18:39:45.339188Z","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":"2026-07-31T18:39:45.431551Z","title":"OSWorld : Benchmarking multimodal agents for open-ended tasks in real computer environments","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":61,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:45.431551Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:ed6681dc194b4196fc28e6f64b9561c8c468d1b8164570c99f55911adb1cb999","observation_id":"5811105c-2a2e-4ef7-a356-3379e8d46408","resolution":{"observed_at":"2026-07-31T18:39:45.431551Z","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":"2026-07-31T18:39:45.594346Z","title":"An illusion of progress? assessing the current state of web agents","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":62,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:45.594346Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:84ca06b740cfc4abbaf509a45b1ddd60e819159d9d1a94a723aee9cc7b02da74","observation_id":"22c232a5-cd89-40e1-8ca9-01799cd91a56","resolution":{"observed_at":"2026-07-31T18:39:45.594346Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2510.17790","last_updated":"2026-05-26T11:29:37Z","snapshot_observed_at":"2026-08-14T08:46:38.704271Z","submitted_at":"2025-10-20T17:48:26Z","title":"UltraCUA: A Foundation Model for Computer Use Agents with Hybrid Action","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2510.17790","snapshot_observed_at":"2026-07-31T18:39:45.654368Z","title":"UltraCUA : A foundation model for computer use agents with hybrid action","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":63,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:45.654368Z"},"links":{"cited_paper":"/paper/2510.17790","citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:8322b018a68469b54246a780e8257d1a04d4b765400a773206a6a9724cf162cc","observation_id":"825fc812-b048-47aa-aff5-3b47b1256cd4","resolution":{"observed_at":"2026-07-31T18:39:45.654368Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2406.12045","last_updated":"2024-06-17T19:33:08Z","snapshot_observed_at":"2026-08-14T14:24:18.501413Z","submitted_at":"2024-06-17T19:33:08Z","title":"$\\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2406.12045","snapshot_observed_at":"2026-07-31T18:39:45.708419Z","title":"-bench: A benchmark for tool-agent-user interaction in real-world domains","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":64,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:45.708419Z"},"links":{"cited_paper":"/paper/2406.12045","citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:cdce5e7db6b19b0139f8b1f7a87fd12d445bd737ea42368ff9031493a40291cc","observation_id":"4e9037c1-14fd-4d0f-95fb-c4a53793daf2","resolution":{"observed_at":"2026-07-31T18:39:45.708419Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2308.01825","last_updated":"2023-09-13T03:57:29Z","snapshot_observed_at":"2026-08-13T12:20:57.788212Z","submitted_at":"2023-08-03T15:34:01Z","title":"Scaling Relationship on Learning Mathematical Reasoning with Large Language Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2308.01825","snapshot_observed_at":"2026-07-31T18:39:45.880604Z","title":"Scaling relationship on learning mathematical reasoning with large language models","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":65,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:45.880604Z"},"links":{"cited_paper":"/paper/2308.01825","citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:0e881ee04cb278de769e30c197b8073160e809bd2357f0529fca5b7c80c05f91","observation_id":"7de6adf9-c520-48d0-8673-4f997c448084","resolution":{"observed_at":"2026-07-31T18:39:45.880604Z","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":"2026-07-31T18:39:45.935761Z","title":null,"venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":66,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:45.935761Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:409f279e00ff6a289af8359362ef1b17b514f4db7fdbd1f053b7b58507752483","observation_id":"0402028f-629e-42bf-90b8-e7caa28864cb","resolution":{"observed_at":"2026-07-31T18:39:45.935761Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2601.04126","last_updated":"2026-06-30T13:54:32Z","snapshot_observed_at":"2026-08-14T20:30:49.131858Z","submitted_at":"2026-01-07T17:40:08Z","title":"InfiniteWeb: Scalable Web Environment Synthesis for GUI Agent Training","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2601.04126","snapshot_observed_at":"2026-07-31T18:39:45.982946Z","title":"InfiniteWeb : Scalable web environment synthesis for GUI agent training","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":67,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:45.982946Z"},"links":{"cited_paper":"/paper/2601.04126","citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:a2d4ac36a558ea181960cca38942e6322907d7528fc7635080fa99cfef653557","observation_id":"efab3cbf-6bf2-47ef-ae5c-9bc8a14889c6","resolution":{"observed_at":"2026-07-31T18:39:45.982946Z","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":"2026-07-31T18:39:46.013422Z","title":"Xing, Hao Zhang, Joseph E","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":68,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:46.013422Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:76589eae79d225786a6b8dfebf64cafbf33d826a17d19586ed68cda407b11fb3","observation_id":"38b790c3-8973-41ee-b3f5-9e2bc0e1a695","resolution":{"observed_at":"2026-07-31T18:39:46.013422Z","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":"2026-07-31T18:39:46.087631Z","title":"WebArena-Infinity : Generating browser environments with verifiable tasks at scale","venue":null,"work_id":null,"year":2026},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":69,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:46.087631Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:0089813a8422d2983ef1354eec1e2c259137d8785d6157105323aa879f0cab67","observation_id":"bc4d1e93-a78d-4554-bd6d-5c17be9bf255","resolution":{"observed_at":"2026-07-31T18:39:46.087631Z","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":"2026-07-31T18:39:46.194028Z","title":"Xu, Hao Zhu, Xuhui Zhou, Robert Lo, Abishek Sridhar, Xianyi Cheng, Tianyue Ou, Yonatan Bisk, Daniel Fried, Uri Alon, and Graham Neubig","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale","version":1},"reference_index":70,"source":"arxiv_source","source_observed_at":"2026-07-31T18:39:46.194028Z"},"links":{"citing_paper":"/paper/2607.28074"},"observation_digest":"sha256:cde98bbad702280c6ce1fa35f28f29174652d3b2ffd60ddc73e538072f6f5c18","observation_id":"382e38aa-a57c-44c1-87ff-4f2fd95ecf20","resolution":{"observed_at":"2026-07-31T18:39:46.194028Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2607.28074","last_updated":"2026-07-30T11:48:22Z","latest_version":1,"primary_category":"cs.AI","snapshot_observed_at":"2026-08-13T05:09:24.773850Z","submitted_at":"2026-07-30T11:48:22Z","title":"Echoverse: Deep, Evolving Environments for Training Computer-Use Agents at Scale"},"reference_resolution":{"displayed":63,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":1,"unresolved":62,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":63},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-15T06:32:42.880941+00:00","source":"crossref"},{"observed_at":"2026-08-15T06:32:39.529945+00:00","source":"retraction_watch"}],"thesis":"As of 15 August 2026, this Paper Citation Record lists 63 of 63 outbound references and 0 inbound Pith citation observations for arXiv:2607.28074."}