{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:UTEDE6PPJKBAFFTP5O5I2GI5Q5","short_pith_number":"pith:UTEDE6PP","canonical_record":{"source":{"id":"2506.09250","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-10T21:16:53Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"da24e437e967bc44e557da492c13e1a54dd8fb1eac990529e0d974278c68ca25","abstract_canon_sha256":"37621a46aef3abd8429db8be2782d841a4cb87c87e8779160fa2b1e931a5887f"},"schema_version":"1.0"},"canonical_sha256":"a4c83279ef4a8202966febba8d191d8755fb38ad4cd475ed45cc32f9824d68c8","source":{"kind":"arxiv","id":"2506.09250","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.09250","created_at":"2026-07-05T11:22:41Z"},{"alias_kind":"arxiv_version","alias_value":"2506.09250v2","created_at":"2026-07-05T11:22:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.09250","created_at":"2026-07-05T11:22:41Z"},{"alias_kind":"pith_short_12","alias_value":"UTEDE6PPJKBA","created_at":"2026-07-05T11:22:41Z"},{"alias_kind":"pith_short_16","alias_value":"UTEDE6PPJKBAFFTP","created_at":"2026-07-05T11:22:41Z"},{"alias_kind":"pith_short_8","alias_value":"UTEDE6PP","created_at":"2026-07-05T11:22:41Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:UTEDE6PPJKBAFFTP5O5I2GI5Q5","target":"record","payload":{"canonical_record":{"source":{"id":"2506.09250","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-10T21:16:53Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"da24e437e967bc44e557da492c13e1a54dd8fb1eac990529e0d974278c68ca25","abstract_canon_sha256":"37621a46aef3abd8429db8be2782d841a4cb87c87e8779160fa2b1e931a5887f"},"schema_version":"1.0"},"canonical_sha256":"a4c83279ef4a8202966febba8d191d8755fb38ad4cd475ed45cc32f9824d68c8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:22:41.731973Z","signature_b64":"ELs7Z9VYjgfDyaO+sbxqR7sZ9dpWPk0A9nOxEaudEyyld2xq5LAHHjLciKUmZjeZjQK5lzOON5A9uxk9AFbRCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a4c83279ef4a8202966febba8d191d8755fb38ad4cd475ed45cc32f9824d68c8","last_reissued_at":"2026-07-05T11:22:41.731354Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:22:41.731354Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.09250","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:22:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oJJo8rE4nukv+dviEY1ks6LpQgm3NrPBj7NRzVDS0y6VLicHZIfUrGXfOobOjEorZvruSEcVpxiZwGAhGqk5BA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T19:55:17.144611Z"},"content_sha256":"350583bc22f7fac5bb8190cb5ac70f93e9fb08619080743ccc7f194e400cb897","schema_version":"1.0","event_id":"sha256:350583bc22f7fac5bb8190cb5ac70f93e9fb08619080743ccc7f194e400cb897"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:UTEDE6PPJKBAFFTP5O5I2GI5Q5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Comment on The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"A. Lawsen","submitted_at":"2025-06-10T21:16:53Z","abstract_excerpt":"Shojaee et al. (2025) report that Large Reasoning Models (LRMs) exhibit \"accuracy collapse\" on planning puzzles beyond certain complexity thresholds. We demonstrate that their findings primarily reflect experimental design limitations rather than fundamental reasoning failures. Our analysis reveals three critical issues: (1) Tower of Hanoi experiments risk exceeding model output token limits, with models explicitly acknowledging these constraints in their outputs; (2) The authors' automated evaluation framework fails to distinguish between reasoning failures and practical constraints, leading "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.09250","kind":"arxiv","version":2},"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/2506.09250/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":"2026-07-05T11:22:41Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"R0hZssu/U1jLDckB/xHRapwAomnfhPi9y29+s2XpPrQz/GDGGA/lUsUpeXbPMppq5d+VJzbel3pyYZzBsp2PAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T19:55:17.145150Z"},"content_sha256":"274af61a741796dd6ecc6795c887123a3401bfab44a67aa4a0d7ddc76a7af4d3","schema_version":"1.0","event_id":"sha256:274af61a741796dd6ecc6795c887123a3401bfab44a67aa4a0d7ddc76a7af4d3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/UTEDE6PPJKBAFFTP5O5I2GI5Q5/bundle.json","state_url":"https://pith.science/pith/UTEDE6PPJKBAFFTP5O5I2GI5Q5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/UTEDE6PPJKBAFFTP5O5I2GI5Q5/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":"2026-08-07T19:55:17Z","links":{"resolver":"https://pith.science/pith/UTEDE6PPJKBAFFTP5O5I2GI5Q5","bundle":"https://pith.science/pith/UTEDE6PPJKBAFFTP5O5I2GI5Q5/bundle.json","state":"https://pith.science/pith/UTEDE6PPJKBAFFTP5O5I2GI5Q5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/UTEDE6PPJKBAFFTP5O5I2GI5Q5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:UTEDE6PPJKBAFFTP5O5I2GI5Q5","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":"37621a46aef3abd8429db8be2782d841a4cb87c87e8779160fa2b1e931a5887f","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-10T21:16:53Z","title_canon_sha256":"da24e437e967bc44e557da492c13e1a54dd8fb1eac990529e0d974278c68ca25"},"schema_version":"1.0","source":{"id":"2506.09250","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.09250","created_at":"2026-07-05T11:22:41Z"},{"alias_kind":"arxiv_version","alias_value":"2506.09250v2","created_at":"2026-07-05T11:22:41Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.09250","created_at":"2026-07-05T11:22:41Z"},{"alias_kind":"pith_short_12","alias_value":"UTEDE6PPJKBA","created_at":"2026-07-05T11:22:41Z"},{"alias_kind":"pith_short_16","alias_value":"UTEDE6PPJKBAFFTP","created_at":"2026-07-05T11:22:41Z"},{"alias_kind":"pith_short_8","alias_value":"UTEDE6PP","created_at":"2026-07-05T11:22:41Z"}],"graph_snapshots":[{"event_id":"sha256:274af61a741796dd6ecc6795c887123a3401bfab44a67aa4a0d7ddc76a7af4d3","target":"graph","created_at":"2026-07-05T11:22:41Z","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/2506.09250/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Shojaee et al. (2025) report that Large Reasoning Models (LRMs) exhibit \"accuracy collapse\" on planning puzzles beyond certain complexity thresholds. We demonstrate that their findings primarily reflect experimental design limitations rather than fundamental reasoning failures. Our analysis reveals three critical issues: (1) Tower of Hanoi experiments risk exceeding model output token limits, with models explicitly acknowledging these constraints in their outputs; (2) The authors' automated evaluation framework fails to distinguish between reasoning failures and practical constraints, leading ","authors_text":"A. Lawsen","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-10T21:16:53Z","title":"Comment on The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.09250","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":"sha256:350583bc22f7fac5bb8190cb5ac70f93e9fb08619080743ccc7f194e400cb897","target":"record","created_at":"2026-07-05T11:22:41Z","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":"37621a46aef3abd8429db8be2782d841a4cb87c87e8779160fa2b1e931a5887f","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-10T21:16:53Z","title_canon_sha256":"da24e437e967bc44e557da492c13e1a54dd8fb1eac990529e0d974278c68ca25"},"schema_version":"1.0","source":{"id":"2506.09250","kind":"arxiv","version":2}},"canonical_sha256":"a4c83279ef4a8202966febba8d191d8755fb38ad4cd475ed45cc32f9824d68c8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a4c83279ef4a8202966febba8d191d8755fb38ad4cd475ed45cc32f9824d68c8","first_computed_at":"2026-07-05T11:22:41.731354Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:22:41.731354Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ELs7Z9VYjgfDyaO+sbxqR7sZ9dpWPk0A9nOxEaudEyyld2xq5LAHHjLciKUmZjeZjQK5lzOON5A9uxk9AFbRCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:22:41.731973Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.09250","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:350583bc22f7fac5bb8190cb5ac70f93e9fb08619080743ccc7f194e400cb897","sha256:274af61a741796dd6ecc6795c887123a3401bfab44a67aa4a0d7ddc76a7af4d3"],"state_sha256":"2a6b061ce1cddc39aa0f42b412d053819c2dbf995ab9d15ba28b9c9cd0612cd3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Ru5oh0A6g64TkpUoYqNvlM8fJJRH6Dp4fbueBI8CldiF7F3jA01GE/fonSsdalevqCdfYaWwcORU/WOtd6KSAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T19:55:17.150360Z","bundle_sha256":"3c4dc9650b2b53d85a129893e8b0d1337b5b6dc48cda89c1f2aa0facb0407fd1"}}