{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:AQFIRS6X5JR6QLVJTX57U25F7J","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":"9ab711a08a7f240816416bf301a87a8741ef0c7530ec6479a7fe6bd239e98b6f","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-12-20T04:31:03Z","title_canon_sha256":"32b1cfb71191991be7a2759b17aca27fd72018dc3a7677412473f617310fb4db"},"schema_version":"1.0","source":{"id":"2412.15557","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.15557","created_at":"2026-06-19T16:11:09Z"},{"alias_kind":"arxiv_version","alias_value":"2412.15557v4","created_at":"2026-06-19T16:11:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.15557","created_at":"2026-06-19T16:11:09Z"},{"alias_kind":"pith_short_12","alias_value":"AQFIRS6X5JR6","created_at":"2026-06-19T16:11:09Z"},{"alias_kind":"pith_short_16","alias_value":"AQFIRS6X5JR6QLVJ","created_at":"2026-06-19T16:11:09Z"},{"alias_kind":"pith_short_8","alias_value":"AQFIRS6X","created_at":"2026-06-19T16:11:09Z"}],"graph_snapshots":[{"event_id":"sha256:7aa681c00f3bb7c0427fc8680184539b86f83b54d1b64e26898fc218ea7e2c83","target":"graph","created_at":"2026-06-19T16:11:09Z","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/2412.15557/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"With the widespread application of LLM-based dialogue systems in daily life, quality assurance has become more important than ever. Recent research has successfully introduced methods to identify unexpected behaviour in single-turn testing scenarios. However, multi-turn interaction is the common real-world usage of dialogue systems, yet testing methods for such interactions remain underexplored. This is largely due to the oracle problem in multi-turn testing, which continues to pose a significant challenge for dialogue system developers and researchers. In this paper, we propose MORTAR, a meta","authors_text":"Aaron Guoxiang Guo, Aldeida Aleti, Chakkrit Tantithamthavorn, Neelofar Neelofar, Tsong Yueh Chen, Yuanyuan Qi","cross_cats":["cs.CL"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-12-20T04:31:03Z","title":"MORTAR: Multi-turn Metamorphic Testing for LLM-based Dialogue Systems"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.15557","kind":"arxiv","version":4},"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":"sha256:6217e83aa632747dd7418c8eb67a71a08d9ede5c5c1b8d0d28ac466dcffe634b","target":"record","created_at":"2026-06-19T16:11:09Z","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":"9ab711a08a7f240816416bf301a87a8741ef0c7530ec6479a7fe6bd239e98b6f","cross_cats_sorted":["cs.CL"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.SE","submitted_at":"2024-12-20T04:31:03Z","title_canon_sha256":"32b1cfb71191991be7a2759b17aca27fd72018dc3a7677412473f617310fb4db"},"schema_version":"1.0","source":{"id":"2412.15557","kind":"arxiv","version":4}},"canonical_sha256":"040a88cbd7ea63e82ea99dfbfa6ba5fa67b7af7b9d05fbe6600fe03cb1963013","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"040a88cbd7ea63e82ea99dfbfa6ba5fa67b7af7b9d05fbe6600fe03cb1963013","first_computed_at":"2026-06-19T16:11:09.423559Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-19T16:11:09.423559Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9052WQrMqHlIjN20LPhnyNZjwk8B5ssMse3QQS204DomUlji4mJI4N5DeidHbVmRmfJtOfKQndVyRP9L4YMeDw==","signature_status":"signed_v1","signed_at":"2026-06-19T16:11:09.423985Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.15557","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6217e83aa632747dd7418c8eb67a71a08d9ede5c5c1b8d0d28ac466dcffe634b","sha256:7aa681c00f3bb7c0427fc8680184539b86f83b54d1b64e26898fc218ea7e2c83"],"state_sha256":"d3ab763f4bfcb8bb506cd546d45680cf74b370f04e42fa1b0716c705e286458d"}