{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:DYTUVJFTFI3PGBXGV4DDR4VNYR","short_pith_number":"pith:DYTUVJFT","canonical_record":{"source":{"id":"2604.24697","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-27T16:58:04Z","cross_cats_sorted":[],"title_canon_sha256":"9785c223a98d408354a325a84a0468629332f3f6de1add786d9f7ba0129efa02","abstract_canon_sha256":"e0bde6e485ee69e4f4a0f7e4b788628dff1d2bb51e0a855ca56496f3889945bd"},"schema_version":"1.0"},"canonical_sha256":"1e274aa4b32a36f306e6af0638f2adc467c8f64e7ba775e55deba28c2325496c","source":{"kind":"arxiv","id":"2604.24697","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.24697","created_at":"2026-05-21T01:04:26Z"},{"alias_kind":"arxiv_version","alias_value":"2604.24697v2","created_at":"2026-05-21T01:04:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.24697","created_at":"2026-05-21T01:04:26Z"},{"alias_kind":"pith_short_12","alias_value":"DYTUVJFTFI3P","created_at":"2026-05-21T01:04:26Z"},{"alias_kind":"pith_short_16","alias_value":"DYTUVJFTFI3PGBXG","created_at":"2026-05-21T01:04:26Z"},{"alias_kind":"pith_short_8","alias_value":"DYTUVJFT","created_at":"2026-05-21T01:04:26Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:DYTUVJFTFI3PGBXGV4DDR4VNYR","target":"record","payload":{"canonical_record":{"source":{"id":"2604.24697","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-27T16:58:04Z","cross_cats_sorted":[],"title_canon_sha256":"9785c223a98d408354a325a84a0468629332f3f6de1add786d9f7ba0129efa02","abstract_canon_sha256":"e0bde6e485ee69e4f4a0f7e4b788628dff1d2bb51e0a855ca56496f3889945bd"},"schema_version":"1.0"},"canonical_sha256":"1e274aa4b32a36f306e6af0638f2adc467c8f64e7ba775e55deba28c2325496c","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:04:26.244780Z","signature_b64":"sP2tq/ohcLIsibvkNM3ukLcqU7MppADPe4yRUAAF6ObLGT/B+Gk4Q9SWzT51X1wHe/KQDkpgfBi61QFL6AgtDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1e274aa4b32a36f306e6af0638f2adc467c8f64e7ba775e55deba28c2325496c","last_reissued_at":"2026-05-21T01:04:26.244096Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:04:26.244096Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.24697","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":"2026-05-21T01:04:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"F/LtJLf61Cq0fXPrlw3tshqbR/HPzueKjOXJLuSZIr9TyC8/kL9mSK0KKWRnwd13nIbUcjclunXOnetmSZczCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T23:31:18.279995Z"},"content_sha256":"1ac48fa6877acde6354fe61c1edfaf59021053de0fb866fc1e31f07acdf98fb2","schema_version":"1.0","event_id":"sha256:1ac48fa6877acde6354fe61c1edfaf59021053de0fb866fc1e31f07acdf98fb2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:DYTUVJFTFI3PGBXGV4DDR4VNYR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Can Current Agents Close the Discovery-to-Application Gap? A Case Study in Minecraft","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Frontier AI agents plateau at 26% success when required to discover causal rules in Minecraft and apply them to build working systems.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bangcheng Yang, Demetri Terzopoulos, Fang Sun, Haowei Lin, Huacong Tang, Jinyuan Zhang, Qian Long, Xiaofeng Gao, Ying Nian Wu, Yitao Liang, Yizhou Sun, Zhou Ziheng","submitted_at":"2026-04-27T16:58:04Z","abstract_excerpt":"Discovering causal regularities and applying them to build functional systems--the discovery-to-application loop--is a hallmark of general intelligence, yet evaluating this capacity has been hindered by the vast complexity gap between scientific discovery and real-world engineering. We introduce SciCrafter, a Minecraft-based benchmark that operationalizes this loop through parameterized redstone circuit tasks. Agents must ignite lamps in specified patterns (e.g., simultaneously or in timed sequences); scaling target parameters substantially increases construction complexity and required knowle"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Evaluating frontier models including GPT-5.2, Gemini-3-Pro, and Claude-Opus-4.5 under a general-purpose code agent scaffold, we find that all plateau at approximately 26% success rate. [...] although the general knowledge application capability still remains as the biggest gap across all models, for frontier models the knowledge gap identification starts to become a major hurdle--indicating the bottleneck is shifting from solving problems right to raising the right problems for current AI.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"scaling target parameters substantially increases construction complexity and required knowledge, forcing genuine discovery rather than reliance on memorized solutions - this assumption underpins that the benchmark measures discovery rather than recall.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Current AI agents achieve only 26% success on SciCrafter's redstone tasks requiring causal discovery and application, indicating the discovery-to-application loop remains challenging with shifting bottlenecks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Frontier AI agents plateau at 26% success when required to discover causal rules in Minecraft and apply them to build working systems.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"7d8f99277e8a291f32a4eacb7a37ce1f426dd5c41f82adef39c64575d1bf94e4"},"source":{"id":"2604.24697","kind":"arxiv","version":2},"verdict":{"id":"6b53cffa-2fb4-4a48-8f3d-d7b8d22966fa","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T03:30:46.844360Z","strongest_claim":"Evaluating frontier models including GPT-5.2, Gemini-3-Pro, and Claude-Opus-4.5 under a general-purpose code agent scaffold, we find that all plateau at approximately 26% success rate. [...] although the general knowledge application capability still remains as the biggest gap across all models, for frontier models the knowledge gap identification starts to become a major hurdle--indicating the bottleneck is shifting from solving problems right to raising the right problems for current AI.","one_line_summary":"Current AI agents achieve only 26% success on SciCrafter's redstone tasks requiring causal discovery and application, indicating the discovery-to-application loop remains challenging with shifting bottlenecks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"scaling target parameters substantially increases construction complexity and required knowledge, forcing genuine discovery rather than reliance on memorized solutions - this assumption underpins that the benchmark measures discovery rather than recall.","pith_extraction_headline":"Frontier AI agents plateau at 26% success when required to discover causal rules in Minecraft and apply them to build working systems."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.24697/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"doi_compliance","ran_at":"2026-05-19T21:46:24.216163Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"fc920622b393ef29f096aa56c4c1be79e8aa3258e4e163eae33d8231bb37b8c0"},"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":"6b53cffa-2fb4-4a48-8f3d-d7b8d22966fa"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-21T01:04:26Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ekTZ1gFL8JTaygakLpSktNli/gxEV1eOVq7wRbMNmjv3FvnwZnQuFd/TgaP5VweP/0IGvaUeOzcLZ1DSqLHJCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T23:31:18.280895Z"},"content_sha256":"ef5b1a767f19f0a7f1ee1a372f3b4252b1b34785430758ec4f43afc1db2fe648","schema_version":"1.0","event_id":"sha256:ef5b1a767f19f0a7f1ee1a372f3b4252b1b34785430758ec4f43afc1db2fe648"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DYTUVJFTFI3PGBXGV4DDR4VNYR/bundle.json","state_url":"https://pith.science/pith/DYTUVJFTFI3PGBXGV4DDR4VNYR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DYTUVJFTFI3PGBXGV4DDR4VNYR/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":"2026-05-26T23:31:18Z","links":{"resolver":"https://pith.science/pith/DYTUVJFTFI3PGBXGV4DDR4VNYR","bundle":"https://pith.science/pith/DYTUVJFTFI3PGBXGV4DDR4VNYR/bundle.json","state":"https://pith.science/pith/DYTUVJFTFI3PGBXGV4DDR4VNYR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DYTUVJFTFI3PGBXGV4DDR4VNYR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:DYTUVJFTFI3PGBXGV4DDR4VNYR","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":"e0bde6e485ee69e4f4a0f7e4b788628dff1d2bb51e0a855ca56496f3889945bd","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-27T16:58:04Z","title_canon_sha256":"9785c223a98d408354a325a84a0468629332f3f6de1add786d9f7ba0129efa02"},"schema_version":"1.0","source":{"id":"2604.24697","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.24697","created_at":"2026-05-21T01:04:26Z"},{"alias_kind":"arxiv_version","alias_value":"2604.24697v2","created_at":"2026-05-21T01:04:26Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.24697","created_at":"2026-05-21T01:04:26Z"},{"alias_kind":"pith_short_12","alias_value":"DYTUVJFTFI3P","created_at":"2026-05-21T01:04:26Z"},{"alias_kind":"pith_short_16","alias_value":"DYTUVJFTFI3PGBXG","created_at":"2026-05-21T01:04:26Z"},{"alias_kind":"pith_short_8","alias_value":"DYTUVJFT","created_at":"2026-05-21T01:04:26Z"}],"graph_snapshots":[{"event_id":"sha256:ef5b1a767f19f0a7f1ee1a372f3b4252b1b34785430758ec4f43afc1db2fe648","target":"graph","created_at":"2026-05-21T01:04:26Z","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":"Evaluating frontier models including GPT-5.2, Gemini-3-Pro, and Claude-Opus-4.5 under a general-purpose code agent scaffold, we find that all plateau at approximately 26% success rate. [...] although the general knowledge application capability still remains as the biggest gap across all models, for frontier models the knowledge gap identification starts to become a major hurdle--indicating the bottleneck is shifting from solving problems right to raising the right problems for current AI."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"scaling target parameters substantially increases construction complexity and required knowledge, forcing genuine discovery rather than reliance on memorized solutions - this assumption underpins that the benchmark measures discovery rather than recall."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Current AI agents achieve only 26% success on SciCrafter's redstone tasks requiring causal discovery and application, indicating the discovery-to-application loop remains challenging with shifting bottlenecks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Frontier AI agents plateau at 26% success when required to discover causal rules in Minecraft and apply them to build working systems."}],"snapshot_sha256":"7d8f99277e8a291f32a4eacb7a37ce1f426dd5c41f82adef39c64575d1bf94e4"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T21:46:24.216163Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2604.24697/integrity.json","findings":[],"snapshot_sha256":"fc920622b393ef29f096aa56c4c1be79e8aa3258e4e163eae33d8231bb37b8c0","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Discovering causal regularities and applying them to build functional systems--the discovery-to-application loop--is a hallmark of general intelligence, yet evaluating this capacity has been hindered by the vast complexity gap between scientific discovery and real-world engineering. We introduce SciCrafter, a Minecraft-based benchmark that operationalizes this loop through parameterized redstone circuit tasks. Agents must ignite lamps in specified patterns (e.g., simultaneously or in timed sequences); scaling target parameters substantially increases construction complexity and required knowle","authors_text":"Bangcheng Yang, Demetri Terzopoulos, Fang Sun, Haowei Lin, Huacong Tang, Jinyuan Zhang, Qian Long, Xiaofeng Gao, Ying Nian Wu, Yitao Liang, Yizhou Sun, Zhou Ziheng","cross_cats":[],"headline":"Frontier AI agents plateau at 26% success when required to discover causal rules in Minecraft and apply them to build working systems.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-27T16:58:04Z","title":"Can Current Agents Close the Discovery-to-Application Gap? A Case Study in Minecraft"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.24697","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-08T03:30:46.844360Z","id":"6b53cffa-2fb4-4a48-8f3d-d7b8d22966fa","model_set":{"reader":"grok-4.3"},"one_line_summary":"Current AI agents achieve only 26% success on SciCrafter's redstone tasks requiring causal discovery and application, indicating the discovery-to-application loop remains challenging with shifting bottlenecks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Frontier AI agents plateau at 26% success when required to discover causal rules in Minecraft and apply them to build working systems.","strongest_claim":"Evaluating frontier models including GPT-5.2, Gemini-3-Pro, and Claude-Opus-4.5 under a general-purpose code agent scaffold, we find that all plateau at approximately 26% success rate. [...] although the general knowledge application capability still remains as the biggest gap across all models, for frontier models the knowledge gap identification starts to become a major hurdle--indicating the bottleneck is shifting from solving problems right to raising the right problems for current AI.","weakest_assumption":"scaling target parameters substantially increases construction complexity and required knowledge, forcing genuine discovery rather than reliance on memorized solutions - this assumption underpins that the benchmark measures discovery rather than recall."}},"verdict_id":"6b53cffa-2fb4-4a48-8f3d-d7b8d22966fa"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:1ac48fa6877acde6354fe61c1edfaf59021053de0fb866fc1e31f07acdf98fb2","target":"record","created_at":"2026-05-21T01:04:26Z","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":"e0bde6e485ee69e4f4a0f7e4b788628dff1d2bb51e0a855ca56496f3889945bd","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-04-27T16:58:04Z","title_canon_sha256":"9785c223a98d408354a325a84a0468629332f3f6de1add786d9f7ba0129efa02"},"schema_version":"1.0","source":{"id":"2604.24697","kind":"arxiv","version":2}},"canonical_sha256":"1e274aa4b32a36f306e6af0638f2adc467c8f64e7ba775e55deba28c2325496c","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"1e274aa4b32a36f306e6af0638f2adc467c8f64e7ba775e55deba28c2325496c","first_computed_at":"2026-05-21T01:04:26.244096Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-21T01:04:26.244096Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"sP2tq/ohcLIsibvkNM3ukLcqU7MppADPe4yRUAAF6ObLGT/B+Gk4Q9SWzT51X1wHe/KQDkpgfBi61QFL6AgtDg==","signature_status":"signed_v1","signed_at":"2026-05-21T01:04:26.244780Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.24697","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1ac48fa6877acde6354fe61c1edfaf59021053de0fb866fc1e31f07acdf98fb2","sha256:ef5b1a767f19f0a7f1ee1a372f3b4252b1b34785430758ec4f43afc1db2fe648"],"state_sha256":"b867e9ba057a2bcbc8086f586bb79209cf2233277486d7992a732cf5988fb4fc"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ktikOgSbiFXxr21aihl9dBQPL9JkJKYNR6NqFp99RJO0OYFfhxwsbDX9C5Cj7jvk+AHVgzDRE2jdVQch1pMcBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T23:31:18.284718Z","bundle_sha256":"59f147a5341c45ceeb135f0052a0e12bd7a56a3df50f26b1dd9818a7d6cc27e1"}}