{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:5DPCPQT7GCAHMQQ4JEROXCFA65","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":"89cead7d12376cee3aaea8b849a265310bf64e1b1ad66021ff49ac2b03e521d7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-29T17:17:45Z","title_canon_sha256":"601d0f9459c3e03828f05da7d6d4d12b35de3722dbbb3a8e0314f9885c996b6c"},"schema_version":"1.0","source":{"id":"2606.30573","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.30573","created_at":"2026-06-30T02:18:21Z"},{"alias_kind":"arxiv_version","alias_value":"2606.30573v1","created_at":"2026-06-30T02:18:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.30573","created_at":"2026-06-30T02:18:21Z"},{"alias_kind":"pith_short_12","alias_value":"5DPCPQT7GCAH","created_at":"2026-06-30T02:18:21Z"},{"alias_kind":"pith_short_16","alias_value":"5DPCPQT7GCAHMQQ4","created_at":"2026-06-30T02:18:21Z"},{"alias_kind":"pith_short_8","alias_value":"5DPCPQT7","created_at":"2026-06-30T02:18:21Z"}],"graph_snapshots":[{"event_id":"sha256:dfd665ef92f5d3b928a68aa669fa717f46aa4545dfa43fd208faac6af1c6a575","target":"graph","created_at":"2026-06-30T02:18:21Z","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/2606.30573/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce SWE-Interact, a new testbed for evaluating coding agents on multi-turn, interactive, user-driven software engineering tasks. Existing frontier SWE benchmarks typically provide complete requirements upfront and evaluate agents on autonomous implementation. In contrast, SWE-Interact places agents in a realistic developer workflow: a carefully designed user simulator starts with vague or incomplete instructions, progressively reveals requirements, inspects the agent's workspace, and provides targeted feedback, revisions, and new constraints until the full task goal has been handed of","authors_text":"Aakash Sabharwal, Anisha Gunjal, Mohit Raghavendra, Yunzhong He","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-29T17:17:45Z","title":"SWE-INTERACT: Reimagining SWE Benchmarks as User-Driven Long-Horizon Coding Sessions"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.30573","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":"sha256:881a538dfcbb3d168fc70db24538fb4952ec081773d2d7a8e3a56115a8da0be0","target":"record","created_at":"2026-06-30T02:18:21Z","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":"89cead7d12376cee3aaea8b849a265310bf64e1b1ad66021ff49ac2b03e521d7","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-06-29T17:17:45Z","title_canon_sha256":"601d0f9459c3e03828f05da7d6d4d12b35de3722dbbb3a8e0314f9885c996b6c"},"schema_version":"1.0","source":{"id":"2606.30573","kind":"arxiv","version":1}},"canonical_sha256":"e8de27c27f308076421c4922eb88a0f766a8d9f684de62ffcbbe996824555c50","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e8de27c27f308076421c4922eb88a0f766a8d9f684de62ffcbbe996824555c50","first_computed_at":"2026-06-30T02:18:21.037487Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-30T02:18:21.037487Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"FBYcr96CjaE1s1C0JWJ7jwngrThh5+fNfpCoiioBW5Wt2l7UYd4iwq7lMS+5GwqVV7RuYTgN/h10PetyNw3tDg==","signature_status":"signed_v1","signed_at":"2026-06-30T02:18:21.037938Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.30573","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:881a538dfcbb3d168fc70db24538fb4952ec081773d2d7a8e3a56115a8da0be0","sha256:dfd665ef92f5d3b928a68aa669fa717f46aa4545dfa43fd208faac6af1c6a575"],"state_sha256":"fd9990dcdddaba22523a98a7cd29b36142a81a4e97c1f1991e461f332b8c1980"}