{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:YJEAZ3DRWEIHOAPTKTT33RCS5T","short_pith_number":"pith:YJEAZ3DR","canonical_record":{"source":{"id":"2506.10209","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T22:03:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"85ad7841a0b7c1b1b09e56f63da6082533aa8d681d55919517c3a88c73a15314","abstract_canon_sha256":"023f1f5c7a5b381d06f1a9b10bff800923148919c08a6d52f05f21f946cdc86e"},"schema_version":"1.0"},"canonical_sha256":"c2480cec71b1107701f354e7bdc452ece0ebd6b556529231feb26e21658586df","source":{"kind":"arxiv","id":"2506.10209","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.10209","created_at":"2026-07-05T11:20:19Z"},{"alias_kind":"arxiv_version","alias_value":"2506.10209v1","created_at":"2026-07-05T11:20:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.10209","created_at":"2026-07-05T11:20:19Z"},{"alias_kind":"pith_short_12","alias_value":"YJEAZ3DRWEIH","created_at":"2026-07-05T11:20:19Z"},{"alias_kind":"pith_short_16","alias_value":"YJEAZ3DRWEIHOAPT","created_at":"2026-07-05T11:20:19Z"},{"alias_kind":"pith_short_8","alias_value":"YJEAZ3DR","created_at":"2026-07-05T11:20:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:YJEAZ3DRWEIHOAPTKTT33RCS5T","target":"record","payload":{"canonical_record":{"source":{"id":"2506.10209","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T22:03:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"85ad7841a0b7c1b1b09e56f63da6082533aa8d681d55919517c3a88c73a15314","abstract_canon_sha256":"023f1f5c7a5b381d06f1a9b10bff800923148919c08a6d52f05f21f946cdc86e"},"schema_version":"1.0"},"canonical_sha256":"c2480cec71b1107701f354e7bdc452ece0ebd6b556529231feb26e21658586df","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:20:19.286277Z","signature_b64":"KRNA7x3+wiYKR6EzBGRXW62EFm+A09kvarls1Bbjfi1LSFwyqLfOdJXWZmONi+ctwuJqjKkeoLMz9t+fByq/CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c2480cec71b1107701f354e7bdc452ece0ebd6b556529231feb26e21658586df","last_reissued_at":"2026-07-05T11:20:19.285789Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:20:19.285789Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.10209","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:20:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AI9uSCEh4sb54emUQSR12MBuwC16NXdQv6z4Fd5+/l6UzDcbB8U5GpoJeFMDiY+3KBZQBiwl6jHW08D01Sf2Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T09:59:35.626877Z"},"content_sha256":"296b2928d00d4832863b7598ba2c3e1933c6aadaf0a2a6013063523c9f65800b","schema_version":"1.0","event_id":"sha256:296b2928d00d4832863b7598ba2c3e1933c6aadaf0a2a6013063523c9f65800b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:YJEAZ3DRWEIHOAPTKTT33RCS5T","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"TTT-Bench: A Benchmark for Evaluating Reasoning Ability with Simple and Novel Tic-Tac-Toe-style Games","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Emad Barsoum, Jialian Wu, Jiang Liu, Prakamya Mishra, Xiaodong Yu, Zicheng Liu","submitted_at":"2025-06-11T22:03:19Z","abstract_excerpt":"Large reasoning models (LRMs) have demonstrated impressive reasoning capabilities across a broad range of tasks including Olympiad-level mathematical problems, indicating evidence of their complex reasoning abilities. While many reasoning benchmarks focus on the STEM domain, the ability of LRMs to reason correctly in broader task domains remains underexplored. In this work, we introduce \\textbf{TTT-Bench}, a new benchmark that is designed to evaluate basic strategic, spatial, and logical reasoning abilities in LRMs through a suite of four two-player Tic-Tac-Toe-style games that humans can effo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.10209","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/2506.10209/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:20:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8+zjy3GhngHThBPFLRBqQUz1XHn8EJRZvbI89W7lxtJsGPe5OOqYoisDM88b0g5CbReRzOhZHHPalPHIQOB3Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T09:59:35.627559Z"},"content_sha256":"50b2f120253eff40434d7011c932cff9977a682ea25c1aa14ddfe280f6f15f82","schema_version":"1.0","event_id":"sha256:50b2f120253eff40434d7011c932cff9977a682ea25c1aa14ddfe280f6f15f82"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YJEAZ3DRWEIHOAPTKTT33RCS5T/bundle.json","state_url":"https://pith.science/pith/YJEAZ3DRWEIHOAPTKTT33RCS5T/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YJEAZ3DRWEIHOAPTKTT33RCS5T/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-01T09:59:35Z","links":{"resolver":"https://pith.science/pith/YJEAZ3DRWEIHOAPTKTT33RCS5T","bundle":"https://pith.science/pith/YJEAZ3DRWEIHOAPTKTT33RCS5T/bundle.json","state":"https://pith.science/pith/YJEAZ3DRWEIHOAPTKTT33RCS5T/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YJEAZ3DRWEIHOAPTKTT33RCS5T/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:YJEAZ3DRWEIHOAPTKTT33RCS5T","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":"023f1f5c7a5b381d06f1a9b10bff800923148919c08a6d52f05f21f946cdc86e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T22:03:19Z","title_canon_sha256":"85ad7841a0b7c1b1b09e56f63da6082533aa8d681d55919517c3a88c73a15314"},"schema_version":"1.0","source":{"id":"2506.10209","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.10209","created_at":"2026-07-05T11:20:19Z"},{"alias_kind":"arxiv_version","alias_value":"2506.10209v1","created_at":"2026-07-05T11:20:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.10209","created_at":"2026-07-05T11:20:19Z"},{"alias_kind":"pith_short_12","alias_value":"YJEAZ3DRWEIH","created_at":"2026-07-05T11:20:19Z"},{"alias_kind":"pith_short_16","alias_value":"YJEAZ3DRWEIHOAPT","created_at":"2026-07-05T11:20:19Z"},{"alias_kind":"pith_short_8","alias_value":"YJEAZ3DR","created_at":"2026-07-05T11:20:19Z"}],"graph_snapshots":[{"event_id":"sha256:50b2f120253eff40434d7011c932cff9977a682ea25c1aa14ddfe280f6f15f82","target":"graph","created_at":"2026-07-05T11:20:19Z","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/2506.10209/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large reasoning models (LRMs) have demonstrated impressive reasoning capabilities across a broad range of tasks including Olympiad-level mathematical problems, indicating evidence of their complex reasoning abilities. While many reasoning benchmarks focus on the STEM domain, the ability of LRMs to reason correctly in broader task domains remains underexplored. In this work, we introduce \\textbf{TTT-Bench}, a new benchmark that is designed to evaluate basic strategic, spatial, and logical reasoning abilities in LRMs through a suite of four two-player Tic-Tac-Toe-style games that humans can effo","authors_text":"Emad Barsoum, Jialian Wu, Jiang Liu, Prakamya Mishra, Xiaodong Yu, Zicheng Liu","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T22:03:19Z","title":"TTT-Bench: A Benchmark for Evaluating Reasoning Ability with Simple and Novel Tic-Tac-Toe-style Games"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.10209","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":"sha256:296b2928d00d4832863b7598ba2c3e1933c6aadaf0a2a6013063523c9f65800b","target":"record","created_at":"2026-07-05T11:20:19Z","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":"023f1f5c7a5b381d06f1a9b10bff800923148919c08a6d52f05f21f946cdc86e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T22:03:19Z","title_canon_sha256":"85ad7841a0b7c1b1b09e56f63da6082533aa8d681d55919517c3a88c73a15314"},"schema_version":"1.0","source":{"id":"2506.10209","kind":"arxiv","version":1}},"canonical_sha256":"c2480cec71b1107701f354e7bdc452ece0ebd6b556529231feb26e21658586df","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c2480cec71b1107701f354e7bdc452ece0ebd6b556529231feb26e21658586df","first_computed_at":"2026-07-05T11:20:19.285789Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:20:19.285789Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"KRNA7x3+wiYKR6EzBGRXW62EFm+A09kvarls1Bbjfi1LSFwyqLfOdJXWZmONi+ctwuJqjKkeoLMz9t+fByq/CQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:20:19.286277Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.10209","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:296b2928d00d4832863b7598ba2c3e1933c6aadaf0a2a6013063523c9f65800b","sha256:50b2f120253eff40434d7011c932cff9977a682ea25c1aa14ddfe280f6f15f82"],"state_sha256":"146041b7faa20515c5ddcfc45322ad198c35339d2c1cf82e5c39f9e263f90416"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Kj2wNwYJzNMMOpcdGp0HLFh7HSauSz0NotwPDp18C/Qvfjh23lHLxLvTM0D5oe8S7xhzjDnihyhUIf7kD3MhAg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-01T09:59:35.632756Z","bundle_sha256":"123af4094a50818e0b30304bc747d54c60d9627f0dd7ce8698212a6bcaf594e9"}}