{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:S5TMPXHEHIFX6W75VKEZYC5VMN","short_pith_number":"pith:S5TMPXHE","schema_version":"1.0","canonical_sha256":"9766c7dce43a0b7f5bfdaa899c0bb5634a328f7a756a89ae1bf5a0b458928a48","source":{"kind":"arxiv","id":"2501.17132","version":1},"attestation_state":"computed","paper":{"title":"ASTRAL: Automated Safety Testing of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SE","authors_text":"Aitor Arrieta, Jos\\'e Antonio Parejo, Miriam Ugarte, Pablo Valle, Sergio Segura","submitted_at":"2025-01-28T18:25:11Z","abstract_excerpt":"Large Language Models (LLMs) have recently gained attention due to their ability to understand and generate sophisticated human-like content. However, ensuring their safety is paramount as they might provide harmful and unsafe responses. Existing LLM testing frameworks address various safety-related concerns (e.g., drugs, terrorism, animal abuse) but often face challenges due to unbalanced and obsolete datasets. In this paper, we present ASTRAL, a tool that automates the generation and execution of test cases (i.e., prompts) for testing the safety of LLMs. First, we introduce a novel black-box"},"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":"2501.17132","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.SE","submitted_at":"2025-01-28T18:25:11Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"c24e36ceb3be153bdf1830c9aab4ba8c79886e89d052b81b4644a22fee7104d5","abstract_canon_sha256":"af0599707523b84aa3c397406c522ed088ec09b8386cf09633fadb21e197842b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:06:34.808636Z","signature_b64":"KPT2wtZZJcGL53DC35mQAu8dHszgPgGWML6VxbvVdh7mFg1Q/Bf2OKALTA4rXHcHKDQHG6jRmRgJFrW1DBxTDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9766c7dce43a0b7f5bfdaa899c0bb5634a328f7a756a89ae1bf5a0b458928a48","last_reissued_at":"2026-07-05T10:06:34.808082Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:06:34.808082Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ASTRAL: Automated Safety Testing of Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.SE","authors_text":"Aitor Arrieta, Jos\\'e Antonio Parejo, Miriam Ugarte, Pablo Valle, Sergio Segura","submitted_at":"2025-01-28T18:25:11Z","abstract_excerpt":"Large Language Models (LLMs) have recently gained attention due to their ability to understand and generate sophisticated human-like content. However, ensuring their safety is paramount as they might provide harmful and unsafe responses. Existing LLM testing frameworks address various safety-related concerns (e.g., drugs, terrorism, animal abuse) but often face challenges due to unbalanced and obsolete datasets. In this paper, we present ASTRAL, a tool that automates the generation and execution of test cases (i.e., prompts) for testing the safety of LLMs. First, we introduce a novel black-box"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.17132","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/2501.17132/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":"2501.17132","created_at":"2026-07-05T10:06:34.808139+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.17132v1","created_at":"2026-07-05T10:06:34.808139+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.17132","created_at":"2026-07-05T10:06:34.808139+00:00"},{"alias_kind":"pith_short_12","alias_value":"S5TMPXHEHIFX","created_at":"2026-07-05T10:06:34.808139+00:00"},{"alias_kind":"pith_short_16","alias_value":"S5TMPXHEHIFX6W75","created_at":"2026-07-05T10:06:34.808139+00:00"},{"alias_kind":"pith_short_8","alias_value":"S5TMPXHE","created_at":"2026-07-05T10:06:34.808139+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2507.02799","citing_title":"Is Reasoning All You Need? Probing Bias in the Age of Reasoning Language Models","ref_index":20,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/S5TMPXHEHIFX6W75VKEZYC5VMN","json":"https://pith.science/pith/S5TMPXHEHIFX6W75VKEZYC5VMN.json","graph_json":"https://pith.science/api/pith-number/S5TMPXHEHIFX6W75VKEZYC5VMN/graph.json","events_json":"https://pith.science/api/pith-number/S5TMPXHEHIFX6W75VKEZYC5VMN/events.json","paper":"https://pith.science/paper/S5TMPXHE"},"agent_actions":{"view_html":"https://pith.science/pith/S5TMPXHEHIFX6W75VKEZYC5VMN","download_json":"https://pith.science/pith/S5TMPXHEHIFX6W75VKEZYC5VMN.json","view_paper":"https://pith.science/paper/S5TMPXHE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.17132&json=true","fetch_graph":"https://pith.science/api/pith-number/S5TMPXHEHIFX6W75VKEZYC5VMN/graph.json","fetch_events":"https://pith.science/api/pith-number/S5TMPXHEHIFX6W75VKEZYC5VMN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/S5TMPXHEHIFX6W75VKEZYC5VMN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/S5TMPXHEHIFX6W75VKEZYC5VMN/action/storage_attestation","attest_author":"https://pith.science/pith/S5TMPXHEHIFX6W75VKEZYC5VMN/action/author_attestation","sign_citation":"https://pith.science/pith/S5TMPXHEHIFX6W75VKEZYC5VMN/action/citation_signature","submit_replication":"https://pith.science/pith/S5TMPXHEHIFX6W75VKEZYC5VMN/action/replication_record"}},"created_at":"2026-07-05T10:06:34.808139+00:00","updated_at":"2026-07-05T10:06:34.808139+00:00"}