{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:2ZP3DHFDXFDMGHHHR4JUM5SYQE","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"709a09281af6e96a1737aca2fffd4e9b568fa9d3e03f1d42a40ee30ef0cfcf92","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2026-06-26T17:59:51Z","title_canon_sha256":"86b35ce22cb1d601f16856391af9649bb0c7f70f12cd50124c1711e03e6dc0a1"},"schema_version":"1.0","source":{"id":"2606.28480","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.28480","created_at":"2026-06-30T00:15:15Z"},{"alias_kind":"arxiv_version","alias_value":"2606.28480v1","created_at":"2026-06-30T00:15:15Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.28480","created_at":"2026-06-30T00:15:15Z"},{"alias_kind":"pith_short_12","alias_value":"2ZP3DHFDXFDM","created_at":"2026-06-30T00:15:15Z"},{"alias_kind":"pith_short_16","alias_value":"2ZP3DHFDXFDMGHHH","created_at":"2026-06-30T00:15:15Z"},{"alias_kind":"pith_short_8","alias_value":"2ZP3DHFD","created_at":"2026-06-30T00:15:15Z"}],"graph_snapshots":[{"event_id":"sha256:709fa9f8333757e1b3aff3d19b9392087b69e7d31e8fd044943380d0ac2be690","target":"graph","created_at":"2026-06-30T00:15:15Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2606.28480/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As large language models and harness frameworks continue to advance, agents operating in terminals are increasingly capable of performing a broader range of general computer-use tasks beyond coding. However, existing benchmarks do not adequately evaluate general-purpose terminal computer-use agents (TUAs): general computer-use benchmarks primarily target graphical user interfaces (GUIs), whereas terminal-based benchmarks largely emphasize technical and programming-centric workflows historically native to the shell. We introduce TUA-Bench, a general-purpose benchmark for terminal-use agents. TU","authors_text":"Belinda Zeng, Fanny Yang, Feiyan Zhou, Luyuan Wang, Shoufa Chen, Xiaohui Zhang, Xuan Yang, Yuanfeng Ji, Yuren Cong, Zhiheng Liu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2026-06-26T17:59:51Z","title":"TUA-Bench: A Benchmark for General-Purpose Terminal-Use Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.28480","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:8d20753e708b0b4626cb3a32850d1719eb87173037dbcad3c2b43261fa9ee117","target":"record","created_at":"2026-06-30T00:15:15Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"709a09281af6e96a1737aca2fffd4e9b568fa9d3e03f1d42a40ee30ef0cfcf92","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.SE","submitted_at":"2026-06-26T17:59:51Z","title_canon_sha256":"86b35ce22cb1d601f16856391af9649bb0c7f70f12cd50124c1711e03e6dc0a1"},"schema_version":"1.0","source":{"id":"2606.28480","kind":"arxiv","version":1}},"canonical_sha256":"d65fb19ca3b946c31ce78f134676588116ac2d930e50d1512303e52b59f008e0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d65fb19ca3b946c31ce78f134676588116ac2d930e50d1512303e52b59f008e0","first_computed_at":"2026-06-30T00:15:15.466152Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-30T00:15:15.466152Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7EXMVqN6DgYciIfRlC6qF/UQYVxcN4qME9nvaAlZKAb0KClwZZWB68u2TqZBnw43i9MWmg5ZjrvmliMJjqpGDA==","signature_status":"signed_v1","signed_at":"2026-06-30T00:15:15.466617Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.28480","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8d20753e708b0b4626cb3a32850d1719eb87173037dbcad3c2b43261fa9ee117","sha256:709fa9f8333757e1b3aff3d19b9392087b69e7d31e8fd044943380d0ac2be690"],"state_sha256":"6c0dba20b2fd2957b68933c3f5f893a82771005f5a583b2f806bcd25f99d0996"}