{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:KJCOJTGN2C7VI2XY2H2QDQ3GRG","short_pith_number":"pith:KJCOJTGN","schema_version":"1.0","canonical_sha256":"5244e4cccdd0bf546af8d1f501c36689890729b02bfb82be9f42984f8e3c0798","source":{"kind":"arxiv","id":"2606.20585","version":1},"attestation_state":"computed","paper":{"title":"Turning Intent into Specifications: A Benchmark and an Interactive User-Assistant Agent","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.HC","authors_text":"Akash Srivastava, Hao Wang, Kai Xu, Ligong Han","submitted_at":"2026-05-13T16:07:43Z","abstract_excerpt":"Today's agents are highly effective at implementing well-scoped software design plans, but user intent is often vague and admits multiple equally valid solutions. In this paper, we introduce SpecBench, a new benchmark for evaluating an agent's ability to translate user intent into a structured, executable specification that aligns with user preferences. The agent is given access to past user conversations and may interact with the user for a fixed number of rounds to ask clarifying questions. We find that existing agents exhibit two extreme behaviors: they either (i) struggle to collaborate pr"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.20585","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.HC","submitted_at":"2026-05-13T16:07:43Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"4d1f73fd3f3547b0299c786167b70a07fef49b3c3d477c14cb41f76dd995df33","abstract_canon_sha256":"d1d3c1946cda9619197ad6851fe90e914f35dfa02f1a2db307995088e4f5b8de"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T00:11:50.617744Z","signature_b64":"JDpl504e1VZV4qa3Dd0xnEsie9s1pebhbCMEgVHKAP2iY6SzSdTspD/DErhcvPjiOvT3r4F76E6+84QyoTLMCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5244e4cccdd0bf546af8d1f501c36689890729b02bfb82be9f42984f8e3c0798","last_reissued_at":"2026-06-23T00:11:50.617284Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T00:11:50.617284Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Turning Intent into Specifications: A Benchmark and an Interactive User-Assistant Agent","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.HC","authors_text":"Akash Srivastava, Hao Wang, Kai Xu, Ligong Han","submitted_at":"2026-05-13T16:07:43Z","abstract_excerpt":"Today's agents are highly effective at implementing well-scoped software design plans, but user intent is often vague and admits multiple equally valid solutions. In this paper, we introduce SpecBench, a new benchmark for evaluating an agent's ability to translate user intent into a structured, executable specification that aligns with user preferences. The agent is given access to past user conversations and may interact with the user for a fixed number of rounds to ask clarifying questions. We find that existing agents exhibit two extreme behaviors: they either (i) struggle to collaborate pr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.20585","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2606.20585/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.20585","created_at":"2026-06-23T00:11:50.617342+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.20585v1","created_at":"2026-06-23T00:11:50.617342+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.20585","created_at":"2026-06-23T00:11:50.617342+00:00"},{"alias_kind":"pith_short_12","alias_value":"KJCOJTGN2C7V","created_at":"2026-06-23T00:11:50.617342+00:00"},{"alias_kind":"pith_short_16","alias_value":"KJCOJTGN2C7VI2XY","created_at":"2026-06-23T00:11:50.617342+00:00"},{"alias_kind":"pith_short_8","alias_value":"KJCOJTGN","created_at":"2026-06-23T00:11:50.617342+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KJCOJTGN2C7VI2XY2H2QDQ3GRG","json":"https://pith.science/pith/KJCOJTGN2C7VI2XY2H2QDQ3GRG.json","graph_json":"https://pith.science/api/pith-number/KJCOJTGN2C7VI2XY2H2QDQ3GRG/graph.json","events_json":"https://pith.science/api/pith-number/KJCOJTGN2C7VI2XY2H2QDQ3GRG/events.json","paper":"https://pith.science/paper/KJCOJTGN"},"agent_actions":{"view_html":"https://pith.science/pith/KJCOJTGN2C7VI2XY2H2QDQ3GRG","download_json":"https://pith.science/pith/KJCOJTGN2C7VI2XY2H2QDQ3GRG.json","view_paper":"https://pith.science/paper/KJCOJTGN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.20585&json=true","fetch_graph":"https://pith.science/api/pith-number/KJCOJTGN2C7VI2XY2H2QDQ3GRG/graph.json","fetch_events":"https://pith.science/api/pith-number/KJCOJTGN2C7VI2XY2H2QDQ3GRG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KJCOJTGN2C7VI2XY2H2QDQ3GRG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KJCOJTGN2C7VI2XY2H2QDQ3GRG/action/storage_attestation","attest_author":"https://pith.science/pith/KJCOJTGN2C7VI2XY2H2QDQ3GRG/action/author_attestation","sign_citation":"https://pith.science/pith/KJCOJTGN2C7VI2XY2H2QDQ3GRG/action/citation_signature","submit_replication":"https://pith.science/pith/KJCOJTGN2C7VI2XY2H2QDQ3GRG/action/replication_record"}},"created_at":"2026-06-23T00:11:50.617342+00:00","updated_at":"2026-06-23T00:11:50.617342+00:00"}