{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:WBPSXKPKHYFY4BA6SGVNF3M3UX","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":"92a5c1ce6afa21dcb4c3b59636a1c60b93d48f0394935ec5f410e314893e807d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-30T08:18:45Z","title_canon_sha256":"ee471f87183bf9548daf3b5ebe107bf9971d73ff2889a7685628f56f0962fa0f"},"schema_version":"1.0","source":{"id":"2606.31308","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.31308","created_at":"2026-07-01T01:17:58Z"},{"alias_kind":"arxiv_version","alias_value":"2606.31308v1","created_at":"2026-07-01T01:17:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.31308","created_at":"2026-07-01T01:17:58Z"},{"alias_kind":"pith_short_12","alias_value":"WBPSXKPKHYFY","created_at":"2026-07-01T01:17:58Z"},{"alias_kind":"pith_short_16","alias_value":"WBPSXKPKHYFY4BA6","created_at":"2026-07-01T01:17:58Z"},{"alias_kind":"pith_short_8","alias_value":"WBPSXKPK","created_at":"2026-07-01T01:17:58Z"}],"graph_snapshots":[{"event_id":"sha256:6bab24435382c13b00ba9cc5433160319c70e59671de8253d96bc7572115db11","target":"graph","created_at":"2026-07-01T01:17:58Z","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/2606.31308/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper investigates the capability of Large Language Models (LLMs) to detect and classify floating-point errors statically in software code. We introduce InterFLOPBench, a benchmark of 90 C kernels with 1 130 test samples designed to evaluate LLMs across six categories of floating-point error: cancellation, comparison, division by zero, overflow, underflow and NaN, compared across 14 LLMs. The evaluation framework treats floating-point error detection as a multi-label classification problem and employs the F1-score metric to measure performance. Results demonstrate that latest models (Qwen","authors_text":"David Defour, Eric Petit, Lisa Taldir, Muhammad Ahmad Saeed, Pablo de Oliveira Castro (LI-PaRAD)","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-30T08:18:45Z","title":"Benchmarking Large Language Models on Floating-Point Error Classification"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.31308","kind":"arxiv","version":1},"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":"sha256:3a51c2153507924cca8ebef5d3c0a4bb87cdb850cace3fbf66f81851ab1c68c7","target":"record","created_at":"2026-07-01T01:17:58Z","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":"92a5c1ce6afa21dcb4c3b59636a1c60b93d48f0394935ec5f410e314893e807d","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-30T08:18:45Z","title_canon_sha256":"ee471f87183bf9548daf3b5ebe107bf9971d73ff2889a7685628f56f0962fa0f"},"schema_version":"1.0","source":{"id":"2606.31308","kind":"arxiv","version":1}},"canonical_sha256":"b05f2ba9ea3e0b8e041e91aad2ed9ba5ffba6dd00258cdd42bc1ec75f37cb591","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b05f2ba9ea3e0b8e041e91aad2ed9ba5ffba6dd00258cdd42bc1ec75f37cb591","first_computed_at":"2026-07-01T01:17:58.721337Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-01T01:17:58.721337Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"y3bEGnNR7meFgp5gX4OAPRbrbkmtZdCMptySmeiebSx/Xpa/suEdr2L5rrs2msz4uoef8JaUCIFEwpQgacgRBQ==","signature_status":"signed_v1","signed_at":"2026-07-01T01:17:58.721753Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.31308","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:3a51c2153507924cca8ebef5d3c0a4bb87cdb850cace3fbf66f81851ab1c68c7","sha256:6bab24435382c13b00ba9cc5433160319c70e59671de8253d96bc7572115db11"],"state_sha256":"1829620bc04f0fc4f6f8d299370c53afe9f21f2b56b1ca09d905d1dda704938d"}