{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:6FVGTDQS54ZABDYIUPSR5KOFEZ","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":"411b92e258a18b8e257e89beabb62836ef53a4cf6035a98c034d709dd0eefeee","cross_cats_sorted":["cs.CR","cs.SY","eess.SY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-14T14:24:25Z","title_canon_sha256":"b98f3ff9120e1de72e61e6eeb162f4c274a45e33ecc386e3894173cb46c58d7a"},"schema_version":"1.0","source":{"id":"2606.15834","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.15834","created_at":"2026-06-25T01:18:38Z"},{"alias_kind":"arxiv_version","alias_value":"2606.15834v2","created_at":"2026-06-25T01:18:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.15834","created_at":"2026-06-25T01:18:38Z"},{"alias_kind":"pith_short_12","alias_value":"6FVGTDQS54ZA","created_at":"2026-06-25T01:18:38Z"},{"alias_kind":"pith_short_16","alias_value":"6FVGTDQS54ZABDYI","created_at":"2026-06-25T01:18:38Z"},{"alias_kind":"pith_short_8","alias_value":"6FVGTDQS","created_at":"2026-06-25T01:18:38Z"}],"graph_snapshots":[{"event_id":"sha256:296f36780ff2680b1aa8230748d3f906c46d4b086a5375336196a4a0be592ac9","target":"graph","created_at":"2026-06-25T01:18:38Z","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/2606.15834/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The computer systems community has recently seen growing interest in AI-driven system evolution, where AI agents iteratively rewrite systems. Frameworks such as AdaEvolve and Engram report 12-60% score improvements over human-designed algorithms. While these results are promising, there are practical concerns if these AI-evolved programs can perform worse on unseen workloads and exhibit scalability regressions. Given the speed and scale of AI-generated code, we need automated mechanisms to uncover such identify hidden weaknesses in AI-evolved systems programs. To this end, we develop AIChilles","authors_text":"Ao Li, Ashwin Silla, Vyas Sekar, Yajie Zhou, Zaoxing Liu","cross_cats":["cs.CR","cs.SY","eess.SY"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-14T14:24:25Z","title":"AIChilles: Automatically Uncovering Hidden Weaknesses in AI-Evolved Systems"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.15834","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:fc4ba6dd3ab4f4b11679035cc2814196cdf2084e6782d8ab0636a9d0546d7791","target":"record","created_at":"2026-06-25T01:18:38Z","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":"411b92e258a18b8e257e89beabb62836ef53a4cf6035a98c034d709dd0eefeee","cross_cats_sorted":["cs.CR","cs.SY","eess.SY"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-14T14:24:25Z","title_canon_sha256":"b98f3ff9120e1de72e61e6eeb162f4c274a45e33ecc386e3894173cb46c58d7a"},"schema_version":"1.0","source":{"id":"2606.15834","kind":"arxiv","version":2}},"canonical_sha256":"f16a698e12ef32008f08a3e51ea9c5265f959aa6bacc4e4270fd1b75b6417537","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f16a698e12ef32008f08a3e51ea9c5265f959aa6bacc4e4270fd1b75b6417537","first_computed_at":"2026-06-25T01:18:38.921922Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-25T01:18:38.921922Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"GyyvCDcoxZibBhKvqtiHxQ/hXmCxQBZF/7WCVrQQ6+WD6nVXC+2ZevUaUdVwsrODO670K0YmnuSypU7OoLc9DA==","signature_status":"signed_v1","signed_at":"2026-06-25T01:18:38.922413Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.15834","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fc4ba6dd3ab4f4b11679035cc2814196cdf2084e6782d8ab0636a9d0546d7791","sha256:296f36780ff2680b1aa8230748d3f906c46d4b086a5375336196a4a0be592ac9"],"state_sha256":"32139dad7bda4d12bdce20f90fc0129c52fa9876256c0aae3b04df16d707c74d"}