{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:GLRDM33FOTV5ZQ246X33NIK7LR","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":"4e2dc355bc688fe37131ba64f654c134bf3ad8dc3f4cc76259e78cf288a54102","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-26T15:23:47Z","title_canon_sha256":"792f02ffd88392af35ea5bcf5534dba921b38513b5d6e575519d87b3ba36f9c9"},"schema_version":"1.0","source":{"id":"2404.17475","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.17475","created_at":"2026-07-05T08:54:49Z"},{"alias_kind":"arxiv_version","alias_value":"2404.17475v2","created_at":"2026-07-05T08:54:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.17475","created_at":"2026-07-05T08:54:49Z"},{"alias_kind":"pith_short_12","alias_value":"GLRDM33FOTV5","created_at":"2026-07-05T08:54:49Z"},{"alias_kind":"pith_short_16","alias_value":"GLRDM33FOTV5ZQ24","created_at":"2026-07-05T08:54:49Z"},{"alias_kind":"pith_short_8","alias_value":"GLRDM33F","created_at":"2026-07-05T08:54:49Z"}],"graph_snapshots":[{"event_id":"sha256:cf74ef0645bde82c584a4d781879a524fd344e1fe2164bbffa2a592ed8ac53c8","target":"graph","created_at":"2026-07-05T08:54:49Z","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/2404.17475/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Counterfactual text generation aims to minimally change a text, such that it is classified differently. Judging advancements in method development for counterfactual text generation is hindered by a non-uniform usage of data sets and metrics in related work. We propose CEval, a benchmark for comparing counterfactual text generation methods. CEval unifies counterfactual and text quality metrics, includes common counterfactual datasets with human annotations, standard baselines (MICE, GDBA, CREST) and the open-source language model LLAMA-2. Our experiments found no perfect method for generating ","authors_text":"Christin Seifert, J\\\"org Schl\\\"otterer, Van Bach Nguyen","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-26T15:23:47Z","title":"CEval: A Benchmark for Evaluating Counterfactual Text Generation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.17475","kind":"arxiv","version":2},"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":"sha256:503abb96579d6c9fb038ab7b0954fb31e0ec3d1e387d4bede36c9552c01ca7c2","target":"record","created_at":"2026-07-05T08:54:49Z","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":"4e2dc355bc688fe37131ba64f654c134bf3ad8dc3f4cc76259e78cf288a54102","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-26T15:23:47Z","title_canon_sha256":"792f02ffd88392af35ea5bcf5534dba921b38513b5d6e575519d87b3ba36f9c9"},"schema_version":"1.0","source":{"id":"2404.17475","kind":"arxiv","version":2}},"canonical_sha256":"32e2366f6574ebdcc35cf5f7b6a15f5c560d634b7c763e00d18cecf2b6c5c1fc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"32e2366f6574ebdcc35cf5f7b6a15f5c560d634b7c763e00d18cecf2b6c5c1fc","first_computed_at":"2026-07-05T08:54:49.775154Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:54:49.775154Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7KK8irq2Oph5EGo8dVSsu4J/M9ZG8qOR6ehH18pZcMCsiqo3ypjG0TWxVMWvQY38HSIdE+4C++nqLJsbKIYzDQ==","signature_status":"signed_v1","signed_at":"2026-07-05T08:54:49.775644Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.17475","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:503abb96579d6c9fb038ab7b0954fb31e0ec3d1e387d4bede36c9552c01ca7c2","sha256:cf74ef0645bde82c584a4d781879a524fd344e1fe2164bbffa2a592ed8ac53c8"],"state_sha256":"60978ef07ebc5dfcc18172d925fc76c6c136b9881a1a05a2ee0ca8d70b3df8e6"}