{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:DO5NSUKNYRK6PWEBFGOHD5IJJX","short_pith_number":"pith:DO5NSUKN","canonical_record":{"source":{"id":"2608.07077","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-07T10:29:19Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"07c5b4ea62a3fe0461f50b07e25ddff057d1396732846202599eee1f55d9bb12","abstract_canon_sha256":"7ceeb488b8e2acb324ae5bb141e16b4bdea459ed936a598c704c198ec5d396fd"},"schema_version":"1.0"},"canonical_sha256":"1bbad9514dc455e7d881299c71f5094def27e6cd4e18564d6f793f62f2a79e79","source":{"kind":"arxiv","id":"2608.07077","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.07077","created_at":"2026-08-10T01:13:08Z"},{"alias_kind":"arxiv_version","alias_value":"2608.07077v1","created_at":"2026-08-10T01:13:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.07077","created_at":"2026-08-10T01:13:08Z"},{"alias_kind":"pith_short_12","alias_value":"DO5NSUKNYRK6","created_at":"2026-08-10T01:13:08Z"},{"alias_kind":"pith_short_16","alias_value":"DO5NSUKNYRK6PWEB","created_at":"2026-08-10T01:13:08Z"},{"alias_kind":"pith_short_8","alias_value":"DO5NSUKN","created_at":"2026-08-10T01:13:08Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:DO5NSUKNYRK6PWEBFGOHD5IJJX","target":"record","payload":{"canonical_record":{"source":{"id":"2608.07077","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-07T10:29:19Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"07c5b4ea62a3fe0461f50b07e25ddff057d1396732846202599eee1f55d9bb12","abstract_canon_sha256":"7ceeb488b8e2acb324ae5bb141e16b4bdea459ed936a598c704c198ec5d396fd"},"schema_version":"1.0"},"canonical_sha256":"1bbad9514dc455e7d881299c71f5094def27e6cd4e18564d6f793f62f2a79e79","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-10T01:13:08.630814Z","signature_b64":"coTtxFyv8YLI1OEKrzBq3byETCZFZWtZEXyGacxZYNfQN0pkrsERbwwntOsvqBujdDDEkG+qysNZR6yaKDxbBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1bbad9514dc455e7d881299c71f5094def27e6cd4e18564d6f793f62f2a79e79","last_reissued_at":"2026-08-10T01:13:08.626643Z","signature_status":"signed_v1","first_computed_at":"2026-08-10T01:13:08.626643Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.07077","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":"2026-08-10T01:13:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gXa0V2RXGZlufqpVAtPVZ37aCZxpXSO8ar5P0PTsyS1Uo1kg+GD/nLFW0kJ+cRUX/+O+klJosEJP+aCyXpbKDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T22:31:43.909490Z"},"content_sha256":"4ca249312774995c92ebe7db4f67c391300f723350177f89bbe53c0372bddde5","schema_version":"1.0","event_id":"sha256:4ca249312774995c92ebe7db4f67c391300f723350177f89bbe53c0372bddde5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:DO5NSUKNYRK6PWEBFGOHD5IJJX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Transformers Struggle to Use Their Emergent World Models: Revisiting the Tower of Hanoi, and the Illusion of Thinking","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Devin Pereira, Willem Zuidema","submitted_at":"2026-08-07T10:29:19Z","abstract_excerpt":"The Tower of Hanoi is a simple planning puzzle that in prior work has proven challenging for large reasoning models (LRMs). Current models solve the standard formulation of the puzzle, but still struggle with the flat-to-flat variant (where initial and goal states are not restricted to have all rings on a single peg). This paper presents an in-depth study of how both small, in-house Transformers and large, third-party LRMs solve this task. To understand the failures mechanistically, we first train small Transformers from scratch on precomputed solution traces. Using a variety of interpretabili"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.07077","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/2608.07077/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":"2026-08-10T01:13:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4ZVYqDiZ8JgBAjd7YleYr1YcYWsfyW9nOoGtDx/lYP0R5S8DjS5/nLVbNPZ42TVj0vMucIz/pUZWEC6Msi9SAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-21T22:31:43.909987Z"},"content_sha256":"8d628e21a5364d6e5db55354ace8b82b0f3a8990a4e26968a307f7c403673203","schema_version":"1.0","event_id":"sha256:8d628e21a5364d6e5db55354ace8b82b0f3a8990a4e26968a307f7c403673203"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DO5NSUKNYRK6PWEBFGOHD5IJJX/bundle.json","state_url":"https://pith.science/pith/DO5NSUKNYRK6PWEBFGOHD5IJJX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DO5NSUKNYRK6PWEBFGOHD5IJJX/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":"2026-08-21T22:31:43Z","links":{"resolver":"https://pith.science/pith/DO5NSUKNYRK6PWEBFGOHD5IJJX","bundle":"https://pith.science/pith/DO5NSUKNYRK6PWEBFGOHD5IJJX/bundle.json","state":"https://pith.science/pith/DO5NSUKNYRK6PWEBFGOHD5IJJX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DO5NSUKNYRK6PWEBFGOHD5IJJX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:DO5NSUKNYRK6PWEBFGOHD5IJJX","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":"7ceeb488b8e2acb324ae5bb141e16b4bdea459ed936a598c704c198ec5d396fd","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-07T10:29:19Z","title_canon_sha256":"07c5b4ea62a3fe0461f50b07e25ddff057d1396732846202599eee1f55d9bb12"},"schema_version":"1.0","source":{"id":"2608.07077","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.07077","created_at":"2026-08-10T01:13:08Z"},{"alias_kind":"arxiv_version","alias_value":"2608.07077v1","created_at":"2026-08-10T01:13:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.07077","created_at":"2026-08-10T01:13:08Z"},{"alias_kind":"pith_short_12","alias_value":"DO5NSUKNYRK6","created_at":"2026-08-10T01:13:08Z"},{"alias_kind":"pith_short_16","alias_value":"DO5NSUKNYRK6PWEB","created_at":"2026-08-10T01:13:08Z"},{"alias_kind":"pith_short_8","alias_value":"DO5NSUKN","created_at":"2026-08-10T01:13:08Z"}],"graph_snapshots":[{"event_id":"sha256:8d628e21a5364d6e5db55354ace8b82b0f3a8990a4e26968a307f7c403673203","target":"graph","created_at":"2026-08-10T01:13:08Z","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/2608.07077/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The Tower of Hanoi is a simple planning puzzle that in prior work has proven challenging for large reasoning models (LRMs). Current models solve the standard formulation of the puzzle, but still struggle with the flat-to-flat variant (where initial and goal states are not restricted to have all rings on a single peg). This paper presents an in-depth study of how both small, in-house Transformers and large, third-party LRMs solve this task. To understand the failures mechanistically, we first train small Transformers from scratch on precomputed solution traces. Using a variety of interpretabili","authors_text":"Devin Pereira, Willem Zuidema","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-07T10:29:19Z","title":"Transformers Struggle to Use Their Emergent World Models: Revisiting the Tower of Hanoi, and the Illusion of Thinking"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.07077","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":"sha256:4ca249312774995c92ebe7db4f67c391300f723350177f89bbe53c0372bddde5","target":"record","created_at":"2026-08-10T01:13:08Z","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":"7ceeb488b8e2acb324ae5bb141e16b4bdea459ed936a598c704c198ec5d396fd","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-07T10:29:19Z","title_canon_sha256":"07c5b4ea62a3fe0461f50b07e25ddff057d1396732846202599eee1f55d9bb12"},"schema_version":"1.0","source":{"id":"2608.07077","kind":"arxiv","version":1}},"canonical_sha256":"1bbad9514dc455e7d881299c71f5094def27e6cd4e18564d6f793f62f2a79e79","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1bbad9514dc455e7d881299c71f5094def27e6cd4e18564d6f793f62f2a79e79","first_computed_at":"2026-08-10T01:13:08.626643Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-10T01:13:08.626643Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"coTtxFyv8YLI1OEKrzBq3byETCZFZWtZEXyGacxZYNfQN0pkrsERbwwntOsvqBujdDDEkG+qysNZR6yaKDxbBA==","signature_status":"signed_v1","signed_at":"2026-08-10T01:13:08.630814Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.07077","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4ca249312774995c92ebe7db4f67c391300f723350177f89bbe53c0372bddde5","sha256:8d628e21a5364d6e5db55354ace8b82b0f3a8990a4e26968a307f7c403673203"],"state_sha256":"7e82e74423253d3350864263cf1bdb4a339dc1a7ef145759e42537bb38ce9d8e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eIREenl0RZEZPzZDFQ4HyZrOvn3FOdNkZjCsqKVeCrlnTOEo6zLLkDsd6x8JWqjCpfGuggwG0RJ799dJ/g5xDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-21T22:31:43.914539Z","bundle_sha256":"9dde836da343fa5c2ff3cc70f0987c8731e116a09101e6043ca2b591f0adbd62"}}