{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:TGG24TRQL6WWQHAH2ZYNOOFN7S","short_pith_number":"pith:TGG24TRQ","canonical_record":{"source":{"id":"2412.12173","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-12T16:04:31Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3612261b681adaf2559d3a6597a3a34da8646c2be36f4cf1b04dd9cc26a405e2","abstract_canon_sha256":"89f79ba3fda4e951ec3b1e0b088b9ce333047071b8169f99d4b63d70d88a07f7"},"schema_version":"1.0"},"canonical_sha256":"998dae4e305fad681c07d670d738adfc92c1b6ca37c64c9f48efe1fd80376fdc","source":{"kind":"arxiv","id":"2412.12173","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.12173","created_at":"2026-07-05T09:50:11Z"},{"alias_kind":"arxiv_version","alias_value":"2412.12173v1","created_at":"2026-07-05T09:50:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.12173","created_at":"2026-07-05T09:50:11Z"},{"alias_kind":"pith_short_12","alias_value":"TGG24TRQL6WW","created_at":"2026-07-05T09:50:11Z"},{"alias_kind":"pith_short_16","alias_value":"TGG24TRQL6WWQHAH","created_at":"2026-07-05T09:50:11Z"},{"alias_kind":"pith_short_8","alias_value":"TGG24TRQ","created_at":"2026-07-05T09:50:11Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:TGG24TRQL6WWQHAH2ZYNOOFN7S","target":"record","payload":{"canonical_record":{"source":{"id":"2412.12173","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-12T16:04:31Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"3612261b681adaf2559d3a6597a3a34da8646c2be36f4cf1b04dd9cc26a405e2","abstract_canon_sha256":"89f79ba3fda4e951ec3b1e0b088b9ce333047071b8169f99d4b63d70d88a07f7"},"schema_version":"1.0"},"canonical_sha256":"998dae4e305fad681c07d670d738adfc92c1b6ca37c64c9f48efe1fd80376fdc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:50:11.018723Z","signature_b64":"Lw3gZy043n57EHuhGwmB50LZ9d64M8k7hhsyMrGCdTAocU0Tx6olf3Nwrx5UdQG+i1pXCxIqG2vr1jFB9muyAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"998dae4e305fad681c07d670d738adfc92c1b6ca37c64c9f48efe1fd80376fdc","last_reissued_at":"2026-07-05T09:50:11.018301Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:50:11.018301Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2412.12173","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":"2026-07-05T09:50:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"eWGQ4wXA07BPnmRmJV3PsQmmWkjHxSCiUb8udZaiE1lHaLHNGuAeHedB7cYzNsHgxHGM+Yc9rCOqfOxX83R0Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T15:09:12.669571Z"},"content_sha256":"5a69fea382362811e10c2cc877060fcf10155d72d7891847cc2f4629d3ba22c1","schema_version":"1.0","event_id":"sha256:5a69fea382362811e10c2cc877060fcf10155d72d7891847cc2f4629d3ba22c1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:TGG24TRQL6WWQHAH2ZYNOOFN7S","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"A NotSo Simple Way to Beat Simple Bench","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Angus McLean, Soham Sane","submitted_at":"2024-12-12T16:04:31Z","abstract_excerpt":"This paper presents a novel framework for enhancing reasoning capabilities in large language models (LLMs) by leveraging iterative reasoning and feedback-driven methodologies. Building on the limitations identified in the SimpleBench benchmark, a dataset designed to evaluate logical coherence and real-world reasoning, we propose a multi-step prompting strategy coupled with global consistency checks to improve model accuracy and robustness. Through comparative analysis of state-of-the-art models, including Claude 3 Opus, Claude 3.5, GPT- 4o, and o1-preview, we demonstrate that iterative reasoni"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.12173","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/2412.12173/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":"2026-07-05T09:50:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4Nq22lM7LzPrQPQgzHz1jIJkOtwgYRC1GTwEbxUWki35BvSw5ipsR03xf5GExWIf7pcVWKx+u13LcJ46+/8VDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T15:09:12.670113Z"},"content_sha256":"dccfe97f976e991a3667316c574f26854ffc35bbe62dbf12a43b974234bc8913","schema_version":"1.0","event_id":"sha256:dccfe97f976e991a3667316c574f26854ffc35bbe62dbf12a43b974234bc8913"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TGG24TRQL6WWQHAH2ZYNOOFN7S/bundle.json","state_url":"https://pith.science/pith/TGG24TRQL6WWQHAH2ZYNOOFN7S/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TGG24TRQL6WWQHAH2ZYNOOFN7S/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":"2026-08-12T15:09:12Z","links":{"resolver":"https://pith.science/pith/TGG24TRQL6WWQHAH2ZYNOOFN7S","bundle":"https://pith.science/pith/TGG24TRQL6WWQHAH2ZYNOOFN7S/bundle.json","state":"https://pith.science/pith/TGG24TRQL6WWQHAH2ZYNOOFN7S/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TGG24TRQL6WWQHAH2ZYNOOFN7S/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:TGG24TRQL6WWQHAH2ZYNOOFN7S","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":"89f79ba3fda4e951ec3b1e0b088b9ce333047071b8169f99d4b63d70d88a07f7","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-12T16:04:31Z","title_canon_sha256":"3612261b681adaf2559d3a6597a3a34da8646c2be36f4cf1b04dd9cc26a405e2"},"schema_version":"1.0","source":{"id":"2412.12173","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2412.12173","created_at":"2026-07-05T09:50:11Z"},{"alias_kind":"arxiv_version","alias_value":"2412.12173v1","created_at":"2026-07-05T09:50:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.12173","created_at":"2026-07-05T09:50:11Z"},{"alias_kind":"pith_short_12","alias_value":"TGG24TRQL6WW","created_at":"2026-07-05T09:50:11Z"},{"alias_kind":"pith_short_16","alias_value":"TGG24TRQL6WWQHAH","created_at":"2026-07-05T09:50:11Z"},{"alias_kind":"pith_short_8","alias_value":"TGG24TRQ","created_at":"2026-07-05T09:50:11Z"}],"graph_snapshots":[{"event_id":"sha256:dccfe97f976e991a3667316c574f26854ffc35bbe62dbf12a43b974234bc8913","target":"graph","created_at":"2026-07-05T09:50:11Z","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/2412.12173/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper presents a novel framework for enhancing reasoning capabilities in large language models (LLMs) by leveraging iterative reasoning and feedback-driven methodologies. Building on the limitations identified in the SimpleBench benchmark, a dataset designed to evaluate logical coherence and real-world reasoning, we propose a multi-step prompting strategy coupled with global consistency checks to improve model accuracy and robustness. Through comparative analysis of state-of-the-art models, including Claude 3 Opus, Claude 3.5, GPT- 4o, and o1-preview, we demonstrate that iterative reasoni","authors_text":"Angus McLean, Soham Sane","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-12T16:04:31Z","title":"A NotSo Simple Way to Beat Simple Bench"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.12173","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":"sha256:5a69fea382362811e10c2cc877060fcf10155d72d7891847cc2f4629d3ba22c1","target":"record","created_at":"2026-07-05T09:50:11Z","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":"89f79ba3fda4e951ec3b1e0b088b9ce333047071b8169f99d4b63d70d88a07f7","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-12-12T16:04:31Z","title_canon_sha256":"3612261b681adaf2559d3a6597a3a34da8646c2be36f4cf1b04dd9cc26a405e2"},"schema_version":"1.0","source":{"id":"2412.12173","kind":"arxiv","version":1}},"canonical_sha256":"998dae4e305fad681c07d670d738adfc92c1b6ca37c64c9f48efe1fd80376fdc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"998dae4e305fad681c07d670d738adfc92c1b6ca37c64c9f48efe1fd80376fdc","first_computed_at":"2026-07-05T09:50:11.018301Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:50:11.018301Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Lw3gZy043n57EHuhGwmB50LZ9d64M8k7hhsyMrGCdTAocU0Tx6olf3Nwrx5UdQG+i1pXCxIqG2vr1jFB9muyAA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:50:11.018723Z","signed_message":"canonical_sha256_bytes"},"source_id":"2412.12173","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5a69fea382362811e10c2cc877060fcf10155d72d7891847cc2f4629d3ba22c1","sha256:dccfe97f976e991a3667316c574f26854ffc35bbe62dbf12a43b974234bc8913"],"state_sha256":"613622c11a54d3b8940f8c3c7171e3fa69f5d883ecf3332cb7dbe72bcc783c6a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+KnHzR+k+JvXzvlVSWaXbvTSpF/TkhpQ5+OtCFF2vaFimrtLea+fi9Cuye8PWA4Z7UtxPXC8YtbNmxlLfK38Cw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-12T15:09:12.674093Z","bundle_sha256":"10aafee07f09c8775faceb0a0d2baf9d09d191267178dd60a6e82f9a2a13d342"}}