{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:YXZBOYLONQ52GSO6EGMYPPYJF3","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":"9297e286be1bb99c4687d344eda848f5eea49d4c70bf405f581ad60c5bfb054c","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2026-08-10T12:36:16Z","title_canon_sha256":"c300f5a8f2564a81fd07a41ec54ca8e7c6c01d1207f7293b0658e3ed3eead7c5"},"schema_version":"1.0","source":{"id":"2608.09538","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.09538","created_at":"2026-08-11T02:24:26Z"},{"alias_kind":"arxiv_version","alias_value":"2608.09538v1","created_at":"2026-08-11T02:24:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.09538","created_at":"2026-08-11T02:24:26Z"},{"alias_kind":"pith_short_12","alias_value":"YXZBOYLONQ52","created_at":"2026-08-11T02:24:26Z"},{"alias_kind":"pith_short_16","alias_value":"YXZBOYLONQ52GSO6","created_at":"2026-08-11T02:24:26Z"},{"alias_kind":"pith_short_8","alias_value":"YXZBOYLO","created_at":"2026-08-11T02:24:26Z"}],"graph_snapshots":[{"event_id":"sha256:9305edde5a8146feb99d020958b7301a3b08ad17a8e10e01aa055b7711fc79e2","target":"graph","created_at":"2026-08-11T02:24:26Z","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/2608.09538/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce TCS-Bench, a benchmark for evaluating Large Language Models (LLMs) on research-level Theoretical Computer Science (TCS) proof generation. TCS-Bench consists of theorem-proving tasks from papers published at top theoretical computer science venues (STOC, FOCS, and SODA). Each task provides the necessary context to derive a self-contained proof for a target result. We evaluate state-of-the-art models on this benchmark. We verify the correctness of generated proofs via a verification agent, and further benchmark the verifier against human-expert proof judgements on a set of target st","authors_text":"Adarsh Kumarappan, David Woodruff, Dimitris Paparas, Ernest van Wijland, Honghao Lin, Julien Canitrot-Paradis, Lalit Jain, Max Springer, Mislav Balunovic, Ola Svensson, Rajesh Jayaram, Rudrajit Das, Silvio Lattanzi, Theophane Weber, Vahab Mirrokni, Vincent Cohen-Addad","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2026-08-10T12:36:16Z","title":"TCS-BENCH: Benchmarking State-of-the-Art Generative AI Theoretical Computer Science Research Ability"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.09538","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":"sha256:34176fa1d5f217264cdc6b5f53f24a245bde75084c46eeb14fcafb159c24e72e","target":"record","created_at":"2026-08-11T02:24:26Z","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":"9297e286be1bb99c4687d344eda848f5eea49d4c70bf405f581ad60c5bfb054c","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/publicdomain/zero/1.0/","primary_cat":"cs.CL","submitted_at":"2026-08-10T12:36:16Z","title_canon_sha256":"c300f5a8f2564a81fd07a41ec54ca8e7c6c01d1207f7293b0658e3ed3eead7c5"},"schema_version":"1.0","source":{"id":"2608.09538","kind":"arxiv","version":1}},"canonical_sha256":"c5f217616e6c3ba349de219987bf092ed63e648bc9f4bb977a4cb17d7ec4aa4a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c5f217616e6c3ba349de219987bf092ed63e648bc9f4bb977a4cb17d7ec4aa4a","first_computed_at":"2026-08-11T02:24:26.779232Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-11T02:24:26.779232Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"h+vta732Cs2uc5XcK0GLHkuT/MILzu/uwsMcAqu3ysJy4BXRtQooSK9uv1SO9ig/zEpz+S23wuJcOniTyZDOCw==","signature_status":"signed_v1","signed_at":"2026-08-11T02:24:26.780942Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.09538","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:34176fa1d5f217264cdc6b5f53f24a245bde75084c46eeb14fcafb159c24e72e","sha256:9305edde5a8146feb99d020958b7301a3b08ad17a8e10e01aa055b7711fc79e2"],"state_sha256":"0d6852a166c86b280de7c8097f667cc8db2c21844e0b65e94c3c9f9ebacf3f3b"}