{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:67XYE5HEO3JBTOQW4EKJHWTFRK","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":"e2a4bddf7b5c9ecf72e8aae5b02fb16fe7316725ff13cdf9702f4d779cc8b8a0","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.HC","submitted_at":"2023-09-17T01:42:30Z","title_canon_sha256":"f7b974513d3ba662ad25517454c417f2e27c6b4fa50519b95c25fbc53580887b"},"schema_version":"1.0","source":{"id":"2309.09128","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.09128","created_at":"2026-07-05T08:15:10Z"},{"alias_kind":"arxiv_version","alias_value":"2309.09128v3","created_at":"2026-07-05T08:15:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.09128","created_at":"2026-07-05T08:15:10Z"},{"alias_kind":"pith_short_12","alias_value":"67XYE5HEO3JB","created_at":"2026-07-05T08:15:10Z"},{"alias_kind":"pith_short_16","alias_value":"67XYE5HEO3JBTOQW","created_at":"2026-07-05T08:15:10Z"},{"alias_kind":"pith_short_8","alias_value":"67XYE5HE","created_at":"2026-07-05T08:15:10Z"}],"graph_snapshots":[{"event_id":"sha256:ad7740be0c1b49aabfbd61e5936121388801b05f67aa97ffba0333ec55838209","target":"graph","created_at":"2026-07-05T08:15:10Z","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/2309.09128/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Evaluating outputs of large language models (LLMs) is challenging, requiring making -- and making sense of -- many responses. Yet tools that go beyond basic prompting tend to require knowledge of programming APIs, focus on narrow domains, or are closed-source. We present ChainForge, an open-source visual toolkit for prompt engineering and on-demand hypothesis testing of text generation LLMs. ChainForge provides a graphical interface for comparison of responses across models and prompt variations. Our system was designed to support three tasks: model selection, prompt template design, and hypot","authors_text":"Chelse Swoopes, Elena Glassman, Ian Arawjo, Martin Wattenberg, Priyan Vaithilingam","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.HC","submitted_at":"2023-09-17T01:42:30Z","title":"ChainForge: A Visual Toolkit for Prompt Engineering and LLM Hypothesis Testing"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.09128","kind":"arxiv","version":3},"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:f9ac0e248e7bcbf3197562c905ab452db7bc71ee9fdcba57be80d5c411fb4e94","target":"record","created_at":"2026-07-05T08:15:10Z","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":"e2a4bddf7b5c9ecf72e8aae5b02fb16fe7316725ff13cdf9702f4d779cc8b8a0","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.HC","submitted_at":"2023-09-17T01:42:30Z","title_canon_sha256":"f7b974513d3ba662ad25517454c417f2e27c6b4fa50519b95c25fbc53580887b"},"schema_version":"1.0","source":{"id":"2309.09128","kind":"arxiv","version":3}},"canonical_sha256":"f7ef8274e476d219ba16e11493da658a93ab0c9a8940505770c1986bd3e63b57","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f7ef8274e476d219ba16e11493da658a93ab0c9a8940505770c1986bd3e63b57","first_computed_at":"2026-07-05T08:15:10.996760Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:15:10.996760Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"r3BZD1SLgHcne/Y18pGtdtlFeIH2Kvc1JPqzCSmrabzU7keyhTAWL/rzx8uua5R9i9g3NZq2rsV88By9TfQ/AQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:15:10.997215Z","signed_message":"canonical_sha256_bytes"},"source_id":"2309.09128","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f9ac0e248e7bcbf3197562c905ab452db7bc71ee9fdcba57be80d5c411fb4e94","sha256:ad7740be0c1b49aabfbd61e5936121388801b05f67aa97ffba0333ec55838209"],"state_sha256":"5ed8b8da59a1e931df8b8758d008af74b632e33bddf53335be69d85ae8adb65c"}