{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:B4WFNZQMAVQIXNEWMX6UWEJOLZ","short_pith_number":"pith:B4WFNZQM","canonical_record":{"source":{"id":"2505.02018","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-05-04T07:48:36Z","cross_cats_sorted":[],"title_canon_sha256":"d9abea7ec371e5826d56e9146b14b20b493324f6795fafaa09c1f11ec1836405","abstract_canon_sha256":"c580bcba27d56353ee610bfd4b291c9a291cc2440fe4fdb675c776f3bbbb0e7c"},"schema_version":"1.0"},"canonical_sha256":"0f2c56e60c05608bb49665fd4b112e5e645409526d64f8ae58ead9c9597b318b","source":{"kind":"arxiv","id":"2505.02018","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.02018","created_at":"2026-07-05T10:58:30Z"},{"alias_kind":"arxiv_version","alias_value":"2505.02018v1","created_at":"2026-07-05T10:58:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.02018","created_at":"2026-07-05T10:58:30Z"},{"alias_kind":"pith_short_12","alias_value":"B4WFNZQMAVQI","created_at":"2026-07-05T10:58:30Z"},{"alias_kind":"pith_short_16","alias_value":"B4WFNZQMAVQIXNEW","created_at":"2026-07-05T10:58:30Z"},{"alias_kind":"pith_short_8","alias_value":"B4WFNZQM","created_at":"2026-07-05T10:58:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:B4WFNZQMAVQIXNEWMX6UWEJOLZ","target":"record","payload":{"canonical_record":{"source":{"id":"2505.02018","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-05-04T07:48:36Z","cross_cats_sorted":[],"title_canon_sha256":"d9abea7ec371e5826d56e9146b14b20b493324f6795fafaa09c1f11ec1836405","abstract_canon_sha256":"c580bcba27d56353ee610bfd4b291c9a291cc2440fe4fdb675c776f3bbbb0e7c"},"schema_version":"1.0"},"canonical_sha256":"0f2c56e60c05608bb49665fd4b112e5e645409526d64f8ae58ead9c9597b318b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:58:30.535870Z","signature_b64":"/7XqLmgzvzEW0nFnRSXfcVxlqa9WhWQZzPwih9tFvZ6OM7nSQYMSzfbhMtW9HUpsxC0Hs6k+xlrgTi/oR/JRBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0f2c56e60c05608bb49665fd4b112e5e645409526d64f8ae58ead9c9597b318b","last_reissued_at":"2026-07-05T10:58:30.535340Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:58:30.535340Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.02018","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":"2026-07-05T10:58:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GslstJWRyPlrSCwTTLJeXvbXCFodn694HF9088heTiF1FCekyea576HNDiovgnD1mgTaA8pH77YAtxdaIkMmCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-17T00:45:44.748080Z"},"content_sha256":"8f331d04956658bb16fc2aabc5e06b3502afc64ae2d77f90830f848afe56f219","schema_version":"1.0","event_id":"sha256:8f331d04956658bb16fc2aabc5e06b3502afc64ae2d77f90830f848afe56f219"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:B4WFNZQMAVQIXNEWMX6UWEJOLZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"R-Bench: Graduate-level Multi-disciplinary Benchmarks for LLM & MLLM Complex Reasoning Evaluation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Bolin Ni, Chen Wang, Gordon Wetzstein, Guo-Wei Yang, Han Hu, Haoyang Peng, Houwen Peng, Jiajun Xu, Jiaxi Song, Kiyohiro Nakayama, Meng-Hao Guo, Shi-Min Hu, Xinzhi Dong, Yi-Xuan Deng, Yi Zhang, Yongming Rao, Zhengyang Geng","submitted_at":"2025-05-04T07:48:36Z","abstract_excerpt":"Reasoning stands as a cornerstone of intelligence, enabling the synthesis of existing knowledge to solve complex problems. Despite remarkable progress, existing reasoning benchmarks often fail to rigorously evaluate the nuanced reasoning capabilities required for complex, real-world problemsolving, particularly in multi-disciplinary and multimodal contexts. In this paper, we introduce a graduate-level, multi-disciplinary, EnglishChinese benchmark, dubbed as Reasoning Bench (R-Bench), for assessing the reasoning capability of both language and multimodal models. RBench spans 1,094 questions acr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.02018","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/2505.02018/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":"2026-07-05T10:58:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HZVagNeCCQIGkJXi1pkzh11o1wRdia95si8+kjjg06kpqF+1YLZwHpkPCEodSvmloGeY1P4w0ov8wDTidgpiDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-17T00:45:44.748474Z"},"content_sha256":"2c5da17359c468d8d99e4930e62b65bfa9361843d2a1b21bd43ad4b72f2e8979","schema_version":"1.0","event_id":"sha256:2c5da17359c468d8d99e4930e62b65bfa9361843d2a1b21bd43ad4b72f2e8979"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/B4WFNZQMAVQIXNEWMX6UWEJOLZ/bundle.json","state_url":"https://pith.science/pith/B4WFNZQMAVQIXNEWMX6UWEJOLZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/B4WFNZQMAVQIXNEWMX6UWEJOLZ/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":"2026-07-17T00:45:44Z","links":{"resolver":"https://pith.science/pith/B4WFNZQMAVQIXNEWMX6UWEJOLZ","bundle":"https://pith.science/pith/B4WFNZQMAVQIXNEWMX6UWEJOLZ/bundle.json","state":"https://pith.science/pith/B4WFNZQMAVQIXNEWMX6UWEJOLZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/B4WFNZQMAVQIXNEWMX6UWEJOLZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:B4WFNZQMAVQIXNEWMX6UWEJOLZ","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":"c580bcba27d56353ee610bfd4b291c9a291cc2440fe4fdb675c776f3bbbb0e7c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-05-04T07:48:36Z","title_canon_sha256":"d9abea7ec371e5826d56e9146b14b20b493324f6795fafaa09c1f11ec1836405"},"schema_version":"1.0","source":{"id":"2505.02018","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.02018","created_at":"2026-07-05T10:58:30Z"},{"alias_kind":"arxiv_version","alias_value":"2505.02018v1","created_at":"2026-07-05T10:58:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.02018","created_at":"2026-07-05T10:58:30Z"},{"alias_kind":"pith_short_12","alias_value":"B4WFNZQMAVQI","created_at":"2026-07-05T10:58:30Z"},{"alias_kind":"pith_short_16","alias_value":"B4WFNZQMAVQIXNEW","created_at":"2026-07-05T10:58:30Z"},{"alias_kind":"pith_short_8","alias_value":"B4WFNZQM","created_at":"2026-07-05T10:58:30Z"}],"graph_snapshots":[{"event_id":"sha256:2c5da17359c468d8d99e4930e62b65bfa9361843d2a1b21bd43ad4b72f2e8979","target":"graph","created_at":"2026-07-05T10:58:30Z","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/2505.02018/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reasoning stands as a cornerstone of intelligence, enabling the synthesis of existing knowledge to solve complex problems. Despite remarkable progress, existing reasoning benchmarks often fail to rigorously evaluate the nuanced reasoning capabilities required for complex, real-world problemsolving, particularly in multi-disciplinary and multimodal contexts. In this paper, we introduce a graduate-level, multi-disciplinary, EnglishChinese benchmark, dubbed as Reasoning Bench (R-Bench), for assessing the reasoning capability of both language and multimodal models. RBench spans 1,094 questions acr","authors_text":"Bolin Ni, Chen Wang, Gordon Wetzstein, Guo-Wei Yang, Han Hu, Haoyang Peng, Houwen Peng, Jiajun Xu, Jiaxi Song, Kiyohiro Nakayama, Meng-Hao Guo, Shi-Min Hu, Xinzhi Dong, Yi-Xuan Deng, Yi Zhang, Yongming Rao, Zhengyang Geng","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-05-04T07:48:36Z","title":"R-Bench: Graduate-level Multi-disciplinary Benchmarks for LLM & MLLM Complex Reasoning Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.02018","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":"sha256:8f331d04956658bb16fc2aabc5e06b3502afc64ae2d77f90830f848afe56f219","target":"record","created_at":"2026-07-05T10:58:30Z","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":"c580bcba27d56353ee610bfd4b291c9a291cc2440fe4fdb675c776f3bbbb0e7c","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-05-04T07:48:36Z","title_canon_sha256":"d9abea7ec371e5826d56e9146b14b20b493324f6795fafaa09c1f11ec1836405"},"schema_version":"1.0","source":{"id":"2505.02018","kind":"arxiv","version":1}},"canonical_sha256":"0f2c56e60c05608bb49665fd4b112e5e645409526d64f8ae58ead9c9597b318b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0f2c56e60c05608bb49665fd4b112e5e645409526d64f8ae58ead9c9597b318b","first_computed_at":"2026-07-05T10:58:30.535340Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:58:30.535340Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"/7XqLmgzvzEW0nFnRSXfcVxlqa9WhWQZzPwih9tFvZ6OM7nSQYMSzfbhMtW9HUpsxC0Hs6k+xlrgTi/oR/JRBg==","signature_status":"signed_v1","signed_at":"2026-07-05T10:58:30.535870Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.02018","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8f331d04956658bb16fc2aabc5e06b3502afc64ae2d77f90830f848afe56f219","sha256:2c5da17359c468d8d99e4930e62b65bfa9361843d2a1b21bd43ad4b72f2e8979"],"state_sha256":"a44f4298f0f35e35bc92b2e4bb0505ff0e89f9ec633399f4ef77355f95180a10"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OvCMi8+MhLXs7onvbg8wwtcnDnyXxePbbXZqYAV+AEQxc6d+cIpenvBClSrLi9z45GbOJlafaXbvgpGn9qExCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-17T00:45:44.750639Z","bundle_sha256":"5411550d9c8745c7e14cb1752e55cbbd93d25543bf099a045d2036b075818ab1"}}