{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:4TZ3RWFY4YUNRXZDRHBU4QWBXN","short_pith_number":"pith:4TZ3RWFY","canonical_record":{"source":{"id":"2605.14386","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.NE","submitted_at":"2026-05-14T05:09:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c461aa8820472e61db747a2d0a7182cb1675e71409a9b4300174836fc1f5f49f","abstract_canon_sha256":"b2edeca0f92cfa052adfac5ff244477531547f1ea36f88c4a7da21f1c172fd40"},"schema_version":"1.0"},"canonical_sha256":"e4f3b8d8b8e628d8df2389c34e42c1bb6858967182d0c392ae44719addf25ed4","source":{"kind":"arxiv","id":"2605.14386","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14386","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14386v1","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14386","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"pith_short_12","alias_value":"4TZ3RWFY4YUN","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"4TZ3RWFY4YUNRXZD","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"4TZ3RWFY","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:4TZ3RWFY4YUNRXZDRHBU4QWBXN","target":"record","payload":{"canonical_record":{"source":{"id":"2605.14386","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.NE","submitted_at":"2026-05-14T05:09:12Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c461aa8820472e61db747a2d0a7182cb1675e71409a9b4300174836fc1f5f49f","abstract_canon_sha256":"b2edeca0f92cfa052adfac5ff244477531547f1ea36f88c4a7da21f1c172fd40"},"schema_version":"1.0"},"canonical_sha256":"e4f3b8d8b8e628d8df2389c34e42c1bb6858967182d0c392ae44719addf25ed4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:39:07.668642Z","signature_b64":"HEdNLvwcdclQU/WGQp1f2qmGYFY9UX5py1oE+tqPj+4ySEDZ5gewBw7I46N8fqnL4VsVIYg7TjIrRc1S/Ua5BQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e4f3b8d8b8e628d8df2389c34e42c1bb6858967182d0c392ae44719addf25ed4","last_reissued_at":"2026-05-17T23:39:07.667925Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:39:07.667925Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.14386","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":"2026-05-17T23:39:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xltYz/EIYC/hIFv+Yncz3GomLCpMsRY4XCxk+xDo8a4c/GJDpWzeuy3nUd+UHmVbxBaSOmpCLLGUvGVR5OFOBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T04:29:05.847204Z"},"content_sha256":"95f76e72a6b3b89b492d229d6110896ebc19af63524de5c39df877bb3b21d08e","schema_version":"1.0","event_id":"sha256:95f76e72a6b3b89b492d229d6110896ebc19af63524de5c39df877bb3b21d08e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:4TZ3RWFY4YUNRXZDRHBU4QWBXN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Darwin Family: MRI-Trust-Weighted Evolutionary Merging for Training-Free Scaling of Language-Model Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Evolutionary merging of existing language model checkpoints produces superior reasoning performance without any training.","cross_cats":["cs.AI"],"primary_cat":"cs.NE","authors_text":"Jaewon Jang, Junghoon Shin, Minseo Kim, Minsik Kim, Sunyoung Choi, Taebong Kim, Youngsik Hong","submitted_at":"2026-05-14T05:09:12Z","abstract_excerpt":"We present Darwin Family, a framework for training-free evolutionary merging of large language models via gradient-free weight-space recombination. We ask whether frontier-level reasoning performance can be improved without additional training, by reorganizing latent capabilities already encoded in existing checkpoints. Darwin introduces three key ideas: (i) a 14-dimensional adaptive merge genome enabling fine-grained component- and block-level recombination; (ii) MRI-Trust Fusion, which adaptively balances diagnostic layer-importance signals with evolutionary search through a learnable trust "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"The flagship Darwin-27B-Opus achieves 86.9% on GPQA Diamond, ranking #6 among 1,252 evaluated models, and outperforming its fully trained foundation model without any gradient-based training.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the evolutionary recombination guided by the 14-dimensional merge genome and MRI-Trust Fusion can reliably reorganize latent reasoning capabilities already present in existing checkpoints without loss of coherence or introduction of new failure modes.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Evolutionary merging with a 14-dimensional genome and MRI-Trust Fusion produces models that outperform their trained parents on reasoning benchmarks without any gradient updates.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Evolutionary merging of existing language model checkpoints produces superior reasoning performance without any training.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"1af84823694f7a281e48c186708ad9561a9f66045af0732bb7e0fd0dc86cd152"},"source":{"id":"2605.14386","kind":"arxiv","version":1},"verdict":{"id":"8c66b9a3-a779-4915-a734-ece30b7ba831","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-15T02:09:01.351243Z","strongest_claim":"The flagship Darwin-27B-Opus achieves 86.9% on GPQA Diamond, ranking #6 among 1,252 evaluated models, and outperforming its fully trained foundation model without any gradient-based training.","one_line_summary":"Evolutionary merging with a 14-dimensional genome and MRI-Trust Fusion produces models that outperform their trained parents on reasoning benchmarks without any gradient updates.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the evolutionary recombination guided by the 14-dimensional merge genome and MRI-Trust Fusion can reliably reorganize latent reasoning capabilities already present in existing checkpoints without loss of coherence or introduction of new failure modes.","pith_extraction_headline":"Evolutionary merging of existing language model checkpoints produces superior reasoning performance without any training."},"references":{"count":31,"sample":[{"doi":"","year":2022,"title":"Chain-of-thought prompting elicits reason- ing in large language models","work_id":"59e16d6c-e56a-4398-a4a3-f1d086439c8d","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Takeshi Kojima, Shixiang Gu, M. Reid, et al. Large language models are zero-shot reasoners. InNeural Information Processing Systems, 2022","work_id":"5c4ef5d3-b983-49db-86d1-2ef46e73e9d8","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Self-consistency improves chain-of-thought reasoning in language models","work_id":"f088816e-1c6c-4ac4-8a4e-0b209da8ab9a","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"Least-to-most prompting enables complex rea- soning in large language models","work_id":"87a0882f-6eb8-47cf-aa5e-41759cd0d615","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2019,"title":"Bert rediscovers the classical nlp pipeline","work_id":"c9de2a65-5b4c-48bc-98de-2b89d1846e02","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":31,"snapshot_sha256":"26e9066885c73f9ad8cefc80e611f0979829f320fdeeb33805dd4f4b2726617a","internal_anchors":3},"formal_canon":{"evidence_count":2,"snapshot_sha256":"d02bb810824e18a774590e3eb05bec50099351a9e375f14abcd36dd734e52090"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"8c66b9a3-a779-4915-a734-ece30b7ba831"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:39:07Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pHFhofz8Ewp/tENwqKNje9O2rLA3BLa2HrXOoFwwXDuAMu3CMxVN3RPxi/oXmoXiPbGIgkhHf68WbMFXaNHVDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-27T04:29:05.848143Z"},"content_sha256":"a86aaef5320a5b573589fd99416be994edad8cca3062b6e2758fa86cf212400e","schema_version":"1.0","event_id":"sha256:a86aaef5320a5b573589fd99416be994edad8cca3062b6e2758fa86cf212400e"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4TZ3RWFY4YUNRXZDRHBU4QWBXN/bundle.json","state_url":"https://pith.science/pith/4TZ3RWFY4YUNRXZDRHBU4QWBXN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4TZ3RWFY4YUNRXZDRHBU4QWBXN/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":"2026-05-27T04:29:05Z","links":{"resolver":"https://pith.science/pith/4TZ3RWFY4YUNRXZDRHBU4QWBXN","bundle":"https://pith.science/pith/4TZ3RWFY4YUNRXZDRHBU4QWBXN/bundle.json","state":"https://pith.science/pith/4TZ3RWFY4YUNRXZDRHBU4QWBXN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4TZ3RWFY4YUNRXZDRHBU4QWBXN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:4TZ3RWFY4YUNRXZDRHBU4QWBXN","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":"b2edeca0f92cfa052adfac5ff244477531547f1ea36f88c4a7da21f1c172fd40","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.NE","submitted_at":"2026-05-14T05:09:12Z","title_canon_sha256":"c461aa8820472e61db747a2d0a7182cb1675e71409a9b4300174836fc1f5f49f"},"schema_version":"1.0","source":{"id":"2605.14386","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.14386","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"arxiv_version","alias_value":"2605.14386v1","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.14386","created_at":"2026-05-17T23:39:07Z"},{"alias_kind":"pith_short_12","alias_value":"4TZ3RWFY4YUN","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"4TZ3RWFY4YUNRXZD","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"4TZ3RWFY","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:a86aaef5320a5b573589fd99416be994edad8cca3062b6e2758fa86cf212400e","target":"graph","created_at":"2026-05-17T23:39:07Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"The flagship Darwin-27B-Opus achieves 86.9% on GPQA Diamond, ranking #6 among 1,252 evaluated models, and outperforming its fully trained foundation model without any gradient-based training."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the evolutionary recombination guided by the 14-dimensional merge genome and MRI-Trust Fusion can reliably reorganize latent reasoning capabilities already present in existing checkpoints without loss of coherence or introduction of new failure modes."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Evolutionary merging with a 14-dimensional genome and MRI-Trust Fusion produces models that outperform their trained parents on reasoning benchmarks without any gradient updates."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Evolutionary merging of existing language model checkpoints produces superior reasoning performance without any training."}],"snapshot_sha256":"1af84823694f7a281e48c186708ad9561a9f66045af0732bb7e0fd0dc86cd152"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"d02bb810824e18a774590e3eb05bec50099351a9e375f14abcd36dd734e52090"},"paper":{"abstract_excerpt":"We present Darwin Family, a framework for training-free evolutionary merging of large language models via gradient-free weight-space recombination. We ask whether frontier-level reasoning performance can be improved without additional training, by reorganizing latent capabilities already encoded in existing checkpoints. Darwin introduces three key ideas: (i) a 14-dimensional adaptive merge genome enabling fine-grained component- and block-level recombination; (ii) MRI-Trust Fusion, which adaptively balances diagnostic layer-importance signals with evolutionary search through a learnable trust ","authors_text":"Jaewon Jang, Junghoon Shin, Minseo Kim, Minsik Kim, Sunyoung Choi, Taebong Kim, Youngsik Hong","cross_cats":["cs.AI"],"headline":"Evolutionary merging of existing language model checkpoints produces superior reasoning performance without any training.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.NE","submitted_at":"2026-05-14T05:09:12Z","title":"Darwin Family: MRI-Trust-Weighted Evolutionary Merging for Training-Free Scaling of Language-Model Reasoning"},"references":{"count":31,"internal_anchors":3,"resolved_work":31,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"Chain-of-thought prompting elicits reason- ing in large language models","work_id":"59e16d6c-e56a-4398-a4a3-f1d086439c8d","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Takeshi Kojima, Shixiang Gu, M. Reid, et al. Large language models are zero-shot reasoners. InNeural Information Processing Systems, 2022","work_id":"5c4ef5d3-b983-49db-86d1-2ef46e73e9d8","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Self-consistency improves chain-of-thought reasoning in language models","work_id":"f088816e-1c6c-4ac4-8a4e-0b209da8ab9a","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Least-to-most prompting enables complex rea- soning in large language models","work_id":"87a0882f-6eb8-47cf-aa5e-41759cd0d615","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Bert rediscovers the classical nlp pipeline","work_id":"c9de2a65-5b4c-48bc-98de-2b89d1846e02","year":2019}],"snapshot_sha256":"26e9066885c73f9ad8cefc80e611f0979829f320fdeeb33805dd4f4b2726617a"},"source":{"id":"2605.14386","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-15T02:09:01.351243Z","id":"8c66b9a3-a779-4915-a734-ece30b7ba831","model_set":{"reader":"grok-4.3"},"one_line_summary":"Evolutionary merging with a 14-dimensional genome and MRI-Trust Fusion produces models that outperform their trained parents on reasoning benchmarks without any gradient updates.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Evolutionary merging of existing language model checkpoints produces superior reasoning performance without any training.","strongest_claim":"The flagship Darwin-27B-Opus achieves 86.9% on GPQA Diamond, ranking #6 among 1,252 evaluated models, and outperforming its fully trained foundation model without any gradient-based training.","weakest_assumption":"That the evolutionary recombination guided by the 14-dimensional merge genome and MRI-Trust Fusion can reliably reorganize latent reasoning capabilities already present in existing checkpoints without loss of coherence or introduction of new failure modes."}},"verdict_id":"8c66b9a3-a779-4915-a734-ece30b7ba831"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:95f76e72a6b3b89b492d229d6110896ebc19af63524de5c39df877bb3b21d08e","target":"record","created_at":"2026-05-17T23:39:07Z","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":"b2edeca0f92cfa052adfac5ff244477531547f1ea36f88c4a7da21f1c172fd40","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.NE","submitted_at":"2026-05-14T05:09:12Z","title_canon_sha256":"c461aa8820472e61db747a2d0a7182cb1675e71409a9b4300174836fc1f5f49f"},"schema_version":"1.0","source":{"id":"2605.14386","kind":"arxiv","version":1}},"canonical_sha256":"e4f3b8d8b8e628d8df2389c34e42c1bb6858967182d0c392ae44719addf25ed4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e4f3b8d8b8e628d8df2389c34e42c1bb6858967182d0c392ae44719addf25ed4","first_computed_at":"2026-05-17T23:39:07.667925Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:39:07.667925Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"HEdNLvwcdclQU/WGQp1f2qmGYFY9UX5py1oE+tqPj+4ySEDZ5gewBw7I46N8fqnL4VsVIYg7TjIrRc1S/Ua5BQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:39:07.668642Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.14386","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:95f76e72a6b3b89b492d229d6110896ebc19af63524de5c39df877bb3b21d08e","sha256:a86aaef5320a5b573589fd99416be994edad8cca3062b6e2758fa86cf212400e"],"state_sha256":"28971534320d4778423d2ca3cc25b739f7c0df50b98843136f138822cc271854"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hkHWS8eJ4aUx8nCR3LljoqZJI8SITcEBfOKec7A5B6nj6OoLljds7s/7gr76M2iJOs1lCP9N7Gm80x3nxc09Ag==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-27T04:29:05.852304Z","bundle_sha256":"795049e61915d70a0ab4dc469a302ba63cf46539d00ced3aa965ebe7b445d11c"}}