{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:M6BJ42EQAUBHRF663DUU5WRTBQ","short_pith_number":"pith:M6BJ42EQ","canonical_record":{"source":{"id":"2506.19095","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-23T20:16:28Z","cross_cats_sorted":[],"title_canon_sha256":"578c76bf90dbe3757c767f14d5db37833e942021b9688fd0e32225825a1ba60e","abstract_canon_sha256":"b22aecbea470e9024cdf77ba81e44edc42ce4bc301f498d16de143b09da95430"},"schema_version":"1.0"},"canonical_sha256":"67829e689005027897ded8e94eda330c316a2dba869220e9fa6515040e844158","source":{"kind":"arxiv","id":"2506.19095","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.19095","created_at":"2026-07-05T11:26:09Z"},{"alias_kind":"arxiv_version","alias_value":"2506.19095v1","created_at":"2026-07-05T11:26:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.19095","created_at":"2026-07-05T11:26:09Z"},{"alias_kind":"pith_short_12","alias_value":"M6BJ42EQAUBH","created_at":"2026-07-05T11:26:09Z"},{"alias_kind":"pith_short_16","alias_value":"M6BJ42EQAUBHRF66","created_at":"2026-07-05T11:26:09Z"},{"alias_kind":"pith_short_8","alias_value":"M6BJ42EQ","created_at":"2026-07-05T11:26:09Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:M6BJ42EQAUBHRF663DUU5WRTBQ","target":"record","payload":{"canonical_record":{"source":{"id":"2506.19095","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-23T20:16:28Z","cross_cats_sorted":[],"title_canon_sha256":"578c76bf90dbe3757c767f14d5db37833e942021b9688fd0e32225825a1ba60e","abstract_canon_sha256":"b22aecbea470e9024cdf77ba81e44edc42ce4bc301f498d16de143b09da95430"},"schema_version":"1.0"},"canonical_sha256":"67829e689005027897ded8e94eda330c316a2dba869220e9fa6515040e844158","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:26:09.354412Z","signature_b64":"Sp8BqP5GGEwbxXBD3pB6KyVxfU7RzlqNxnM5zHGTreTChC5CiRNH+y7XFlDihtvcTm4Ou9SVrZSJzP1KvJzUBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"67829e689005027897ded8e94eda330c316a2dba869220e9fa6515040e844158","last_reissued_at":"2026-07-05T11:26:09.353896Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:26:09.353896Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.19095","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":"2026-07-05T11:26:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"X9SxCs20jq39fLuPYeM+dF76ajXJDwZ/AFo6pHKQAnvzI8jtMzZ1BpgfGzzp9FJqRSdHVNcN10imNu6d8WkRAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T22:11:33.378905Z"},"content_sha256":"85826f976c4b1fbf5f77788f3e6ec717df4572ce329e28c3c9974e9152aaf3f1","schema_version":"1.0","event_id":"sha256:85826f976c4b1fbf5f77788f3e6ec717df4572ce329e28c3c9974e9152aaf3f1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:M6BJ42EQAUBHRF663DUU5WRTBQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Baba is LLM: Reasoning in a Game with Dynamic Rules","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Aske Plaat, Fien van Wetten, Max van Duijn","submitted_at":"2025-06-23T20:16:28Z","abstract_excerpt":"Large language models (LLMs) are known to perform well on language tasks, but struggle with reasoning tasks. This paper explores the ability of LLMs to play the 2D puzzle game Baba is You, in which players manipulate rules by rearranging text blocks that define object properties. Given that this rule-manipulation relies on language abilities and reasoning, it is a compelling challenge for LLMs. Six LLMs are evaluated using different prompt types, including (1) simple, (2) rule-extended and (3) action-extended prompts. In addition, two models (Mistral, OLMo) are finetuned using textual and stru"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.19095","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/2506.19095/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":"2026-07-05T11:26:09Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wyHWKSoLeZVWq1pFe496V49Jme6iaYdfpqIDlBSqgTo9a1m6xpya7IZQCPkuZ2nEZn1xJmvOydejxS/H886uAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T22:11:33.379564Z"},"content_sha256":"f76dc217b44f78aab3e50cafdc36dbc9cbf8658a6806191aa6caf6a987914551","schema_version":"1.0","event_id":"sha256:f76dc217b44f78aab3e50cafdc36dbc9cbf8658a6806191aa6caf6a987914551"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/M6BJ42EQAUBHRF663DUU5WRTBQ/bundle.json","state_url":"https://pith.science/pith/M6BJ42EQAUBHRF663DUU5WRTBQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/M6BJ42EQAUBHRF663DUU5WRTBQ/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":"2026-08-04T22:11:33Z","links":{"resolver":"https://pith.science/pith/M6BJ42EQAUBHRF663DUU5WRTBQ","bundle":"https://pith.science/pith/M6BJ42EQAUBHRF663DUU5WRTBQ/bundle.json","state":"https://pith.science/pith/M6BJ42EQAUBHRF663DUU5WRTBQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/M6BJ42EQAUBHRF663DUU5WRTBQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:M6BJ42EQAUBHRF663DUU5WRTBQ","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":"b22aecbea470e9024cdf77ba81e44edc42ce4bc301f498d16de143b09da95430","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-23T20:16:28Z","title_canon_sha256":"578c76bf90dbe3757c767f14d5db37833e942021b9688fd0e32225825a1ba60e"},"schema_version":"1.0","source":{"id":"2506.19095","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.19095","created_at":"2026-07-05T11:26:09Z"},{"alias_kind":"arxiv_version","alias_value":"2506.19095v1","created_at":"2026-07-05T11:26:09Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.19095","created_at":"2026-07-05T11:26:09Z"},{"alias_kind":"pith_short_12","alias_value":"M6BJ42EQAUBH","created_at":"2026-07-05T11:26:09Z"},{"alias_kind":"pith_short_16","alias_value":"M6BJ42EQAUBHRF66","created_at":"2026-07-05T11:26:09Z"},{"alias_kind":"pith_short_8","alias_value":"M6BJ42EQ","created_at":"2026-07-05T11:26:09Z"}],"graph_snapshots":[{"event_id":"sha256:f76dc217b44f78aab3e50cafdc36dbc9cbf8658a6806191aa6caf6a987914551","target":"graph","created_at":"2026-07-05T11:26:09Z","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/2506.19095/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models (LLMs) are known to perform well on language tasks, but struggle with reasoning tasks. This paper explores the ability of LLMs to play the 2D puzzle game Baba is You, in which players manipulate rules by rearranging text blocks that define object properties. Given that this rule-manipulation relies on language abilities and reasoning, it is a compelling challenge for LLMs. Six LLMs are evaluated using different prompt types, including (1) simple, (2) rule-extended and (3) action-extended prompts. In addition, two models (Mistral, OLMo) are finetuned using textual and stru","authors_text":"Aske Plaat, Fien van Wetten, Max van Duijn","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-23T20:16:28Z","title":"Baba is LLM: Reasoning in a Game with Dynamic Rules"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.19095","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":"sha256:85826f976c4b1fbf5f77788f3e6ec717df4572ce329e28c3c9974e9152aaf3f1","target":"record","created_at":"2026-07-05T11:26:09Z","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":"b22aecbea470e9024cdf77ba81e44edc42ce4bc301f498d16de143b09da95430","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-06-23T20:16:28Z","title_canon_sha256":"578c76bf90dbe3757c767f14d5db37833e942021b9688fd0e32225825a1ba60e"},"schema_version":"1.0","source":{"id":"2506.19095","kind":"arxiv","version":1}},"canonical_sha256":"67829e689005027897ded8e94eda330c316a2dba869220e9fa6515040e844158","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"67829e689005027897ded8e94eda330c316a2dba869220e9fa6515040e844158","first_computed_at":"2026-07-05T11:26:09.353896Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:26:09.353896Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Sp8BqP5GGEwbxXBD3pB6KyVxfU7RzlqNxnM5zHGTreTChC5CiRNH+y7XFlDihtvcTm4Ou9SVrZSJzP1KvJzUBg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:26:09.354412Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.19095","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:85826f976c4b1fbf5f77788f3e6ec717df4572ce329e28c3c9974e9152aaf3f1","sha256:f76dc217b44f78aab3e50cafdc36dbc9cbf8658a6806191aa6caf6a987914551"],"state_sha256":"28dfbc0fc7f7bedb3fb035325d0e10babb284e9b51bd4cdb61c7106ea9931813"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nO1+otNeK8yzgoV7WU2Fg07O8h8ZZGBHhlyCzyi9UMn2woavwPlPj23xNDvl9Xksujbj957FudUsLx9Bcv6oCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T22:11:33.385389Z","bundle_sha256":"3425f4c50d369dd9f2744eac1708fb510c2f40e793fc8b70b2a8cd37794d9a84"}}