{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:76GERYO5NN6HEVSYZEDF47PZP6","short_pith_number":"pith:76GERYO5","canonical_record":{"source":{"id":"2506.01793","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T15:33:29Z","cross_cats_sorted":[],"title_canon_sha256":"3da39a2c377fc87a3e0ce53fd1ae6e8c0226cc8bbc5d64a66b6b577baf398d82","abstract_canon_sha256":"b774717fe995c06038947cca31536cf7c055113f077e4b64b98ddf740f543860"},"schema_version":"1.0"},"canonical_sha256":"ff8c48e1dd6b7c725658c9065e7df97f981c6cd5e97b9efed1398db8ed98ef7a","source":{"kind":"arxiv","id":"2506.01793","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.01793","created_at":"2026-07-05T11:14:19Z"},{"alias_kind":"arxiv_version","alias_value":"2506.01793v1","created_at":"2026-07-05T11:14:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.01793","created_at":"2026-07-05T11:14:19Z"},{"alias_kind":"pith_short_12","alias_value":"76GERYO5NN6H","created_at":"2026-07-05T11:14:19Z"},{"alias_kind":"pith_short_16","alias_value":"76GERYO5NN6HEVSY","created_at":"2026-07-05T11:14:19Z"},{"alias_kind":"pith_short_8","alias_value":"76GERYO5","created_at":"2026-07-05T11:14:19Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:76GERYO5NN6HEVSYZEDF47PZP6","target":"record","payload":{"canonical_record":{"source":{"id":"2506.01793","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T15:33:29Z","cross_cats_sorted":[],"title_canon_sha256":"3da39a2c377fc87a3e0ce53fd1ae6e8c0226cc8bbc5d64a66b6b577baf398d82","abstract_canon_sha256":"b774717fe995c06038947cca31536cf7c055113f077e4b64b98ddf740f543860"},"schema_version":"1.0"},"canonical_sha256":"ff8c48e1dd6b7c725658c9065e7df97f981c6cd5e97b9efed1398db8ed98ef7a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:19.412752Z","signature_b64":"35eqLCYQm/9so/ri6P9nGWt7F8XbyslTWj09z8RTUHfSGuOU4iK5c2aJBt43zN+mAeS9Whbjbqe6rFSyHn0FBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ff8c48e1dd6b7c725658c9065e7df97f981c6cd5e97b9efed1398db8ed98ef7a","last_reissued_at":"2026-07-05T11:14:19.412270Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:19.412270Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.01793","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":"2026-07-05T11:14:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"gBJ0AKuKWqNKUWapy/dcJ368EG0R6EQFBX+fz+cUnDWLfTg9odP3V4AVPTATMe/QVdgiFbXgoVQrx0cW5eImCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T22:43:53.114984Z"},"content_sha256":"5dba5cf62a2e2599de0dd73771b01f303dbbd70295c44e5075b7a6cb85a76ac1","schema_version":"1.0","event_id":"sha256:5dba5cf62a2e2599de0dd73771b01f303dbbd70295c44e5075b7a6cb85a76ac1"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:76GERYO5NN6HEVSYZEDF47PZP6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Human-Centric Evaluation for Foundation Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chunyi Li, Farong Wen, Guangtao Zhai, Junying Wang, Kaiyuan Ji, Xiaorong Zhu, Yijin Guo, Zicheng Zhang","submitted_at":"2025-06-02T15:33:29Z","abstract_excerpt":"Currently, nearly all evaluations of foundation models focus on objective metrics, emphasizing quiz performance to define model capabilities. While this model-centric approach enables rapid performance assessment, it fails to reflect authentic human experiences. To address this gap, we propose a Human-Centric subjective Evaluation (HCE) framework, focusing on three core dimensions: problem-solving ability, information quality, and interaction experience. Through experiments involving Deepseek R1, OpenAI o3 mini, Grok 3, and Gemini 2.5, we conduct over 540 participant-driven evaluations, where "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.01793","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/2506.01793/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":"2026-07-05T11:14:19Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nWupkp3Z5w2bH/br1NHfsJIJ76mOxbk+g1MUt02JKaEIARBTHy4VTZV+IQxfoT1OEwlNb6aY9nSojZ+nGDtBCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-07T22:43:53.115651Z"},"content_sha256":"8b68a6dccef77cf3ad94fdca949e41dda61e50fc3f5586d453790dfd4219748a","schema_version":"1.0","event_id":"sha256:8b68a6dccef77cf3ad94fdca949e41dda61e50fc3f5586d453790dfd4219748a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/76GERYO5NN6HEVSYZEDF47PZP6/bundle.json","state_url":"https://pith.science/pith/76GERYO5NN6HEVSYZEDF47PZP6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/76GERYO5NN6HEVSYZEDF47PZP6/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":"2026-08-07T22:43:53Z","links":{"resolver":"https://pith.science/pith/76GERYO5NN6HEVSYZEDF47PZP6","bundle":"https://pith.science/pith/76GERYO5NN6HEVSYZEDF47PZP6/bundle.json","state":"https://pith.science/pith/76GERYO5NN6HEVSYZEDF47PZP6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/76GERYO5NN6HEVSYZEDF47PZP6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:76GERYO5NN6HEVSYZEDF47PZP6","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":"b774717fe995c06038947cca31536cf7c055113f077e4b64b98ddf740f543860","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T15:33:29Z","title_canon_sha256":"3da39a2c377fc87a3e0ce53fd1ae6e8c0226cc8bbc5d64a66b6b577baf398d82"},"schema_version":"1.0","source":{"id":"2506.01793","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.01793","created_at":"2026-07-05T11:14:19Z"},{"alias_kind":"arxiv_version","alias_value":"2506.01793v1","created_at":"2026-07-05T11:14:19Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.01793","created_at":"2026-07-05T11:14:19Z"},{"alias_kind":"pith_short_12","alias_value":"76GERYO5NN6H","created_at":"2026-07-05T11:14:19Z"},{"alias_kind":"pith_short_16","alias_value":"76GERYO5NN6HEVSY","created_at":"2026-07-05T11:14:19Z"},{"alias_kind":"pith_short_8","alias_value":"76GERYO5","created_at":"2026-07-05T11:14:19Z"}],"graph_snapshots":[{"event_id":"sha256:8b68a6dccef77cf3ad94fdca949e41dda61e50fc3f5586d453790dfd4219748a","target":"graph","created_at":"2026-07-05T11:14:19Z","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/2506.01793/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Currently, nearly all evaluations of foundation models focus on objective metrics, emphasizing quiz performance to define model capabilities. While this model-centric approach enables rapid performance assessment, it fails to reflect authentic human experiences. To address this gap, we propose a Human-Centric subjective Evaluation (HCE) framework, focusing on three core dimensions: problem-solving ability, information quality, and interaction experience. Through experiments involving Deepseek R1, OpenAI o3 mini, Grok 3, and Gemini 2.5, we conduct over 540 participant-driven evaluations, where ","authors_text":"Chunyi Li, Farong Wen, Guangtao Zhai, Junying Wang, Kaiyuan Ji, Xiaorong Zhu, Yijin Guo, Zicheng Zhang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T15:33:29Z","title":"Human-Centric Evaluation for Foundation Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.01793","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":"sha256:5dba5cf62a2e2599de0dd73771b01f303dbbd70295c44e5075b7a6cb85a76ac1","target":"record","created_at":"2026-07-05T11:14:19Z","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":"b774717fe995c06038947cca31536cf7c055113f077e4b64b98ddf740f543860","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-02T15:33:29Z","title_canon_sha256":"3da39a2c377fc87a3e0ce53fd1ae6e8c0226cc8bbc5d64a66b6b577baf398d82"},"schema_version":"1.0","source":{"id":"2506.01793","kind":"arxiv","version":1}},"canonical_sha256":"ff8c48e1dd6b7c725658c9065e7df97f981c6cd5e97b9efed1398db8ed98ef7a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ff8c48e1dd6b7c725658c9065e7df97f981c6cd5e97b9efed1398db8ed98ef7a","first_computed_at":"2026-07-05T11:14:19.412270Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:14:19.412270Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"35eqLCYQm/9so/ri6P9nGWt7F8XbyslTWj09z8RTUHfSGuOU4iK5c2aJBt43zN+mAeS9Whbjbqe6rFSyHn0FBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T11:14:19.412752Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.01793","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5dba5cf62a2e2599de0dd73771b01f303dbbd70295c44e5075b7a6cb85a76ac1","sha256:8b68a6dccef77cf3ad94fdca949e41dda61e50fc3f5586d453790dfd4219748a"],"state_sha256":"e0ccc8229f0edd5b4dae0d148607fb6484c3dd72fadbe80df7d207f8c7fe53d4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"94rhxJDq91VIncv9E233lnszFZr51PwZvp+6yuwFTzcyE/FXiggWBekvxDKq2rVGLmz6qHQdgmCfehPKQyMOCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-07T22:43:53.121648Z","bundle_sha256":"22249ef1474687d1174c1749c2a436eba5bbe11ca070065297fbc57c6d1e3a86"}}