{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:6IW4Z36OOPTX2HATIYXDKGTVRK","short_pith_number":"pith:6IW4Z36O","canonical_record":{"source":{"id":"2504.15219","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-21T16:43:50Z","cross_cats_sorted":[],"title_canon_sha256":"2a549495bfffd8bab9e1f3548da13cb2e80c34c82487520668a6de204c6b5671","abstract_canon_sha256":"bbc22af0e54967beedda2de96d1d029ac9bee744c97d75d2e4aabb0fb6bbad96"},"schema_version":"1.0"},"canonical_sha256":"f22dccefce73e77d1c13462e351a758a94ec4441179374bba62ee4ba1075d274","source":{"kind":"arxiv","id":"2504.15219","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.15219","created_at":"2026-07-05T11:55:18Z"},{"alias_kind":"arxiv_version","alias_value":"2504.15219v2","created_at":"2026-07-05T11:55:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.15219","created_at":"2026-07-05T11:55:18Z"},{"alias_kind":"pith_short_12","alias_value":"6IW4Z36OOPTX","created_at":"2026-07-05T11:55:18Z"},{"alias_kind":"pith_short_16","alias_value":"6IW4Z36OOPTX2HAT","created_at":"2026-07-05T11:55:18Z"},{"alias_kind":"pith_short_8","alias_value":"6IW4Z36O","created_at":"2026-07-05T11:55:18Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:6IW4Z36OOPTX2HATIYXDKGTVRK","target":"record","payload":{"canonical_record":{"source":{"id":"2504.15219","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-21T16:43:50Z","cross_cats_sorted":[],"title_canon_sha256":"2a549495bfffd8bab9e1f3548da13cb2e80c34c82487520668a6de204c6b5671","abstract_canon_sha256":"bbc22af0e54967beedda2de96d1d029ac9bee744c97d75d2e4aabb0fb6bbad96"},"schema_version":"1.0"},"canonical_sha256":"f22dccefce73e77d1c13462e351a758a94ec4441179374bba62ee4ba1075d274","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:55:18.815461Z","signature_b64":"zs3zoWS2NGC2DJHqdKyMhXHbTg6NVk9+QhRGdAJQppOA+egDTb0tGOvYdw/oWvPprQDiMGeFUx3uQIjJLlcyBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f22dccefce73e77d1c13462e351a758a94ec4441179374bba62ee4ba1075d274","last_reissued_at":"2026-07-05T11:55:18.815029Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:55:18.815029Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2504.15219","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":"2026-07-05T11:55:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"L/1ya/T4FcuTMEFmCl88Pp80BfVw6P0RXtOgIeWdSq3s7WECeQ/LQsJrM7ToAlGWU4wmNXZlkmXBCVdOwcmeCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T06:20:38.533462Z"},"content_sha256":"c353bd233ccabbebaf2677dffde658bd7cede8af0ac0a1469bf47d2ea0bd5b20","schema_version":"1.0","event_id":"sha256:c353bd233ccabbebaf2677dffde658bd7cede8af0ac0a1469bf47d2ea0bd5b20"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:6IW4Z36OOPTX2HATIYXDKGTVRK","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"EvalAgent: Discovering Implicit Evaluation Criteria from the Web","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Chaitanya Malaviya, Greg Durrett, Junyi Jessy Li, Manya Wadhwa, Philippe Laban, Zayne Sprague","submitted_at":"2025-04-21T16:43:50Z","abstract_excerpt":"Evaluation of language model outputs on structured writing tasks is typically conducted with a number of desirable criteria presented to human evaluators or large language models (LLMs). For instance, on a prompt like \"Help me draft an academic talk on coffee intake vs research productivity\", a model response may be evaluated for criteria like accuracy and coherence. However, high-quality responses should do more than just satisfy basic task requirements. An effective response to this query should include quintessential features of an academic talk, such as a compelling opening, clear research"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.15219","kind":"arxiv","version":2},"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/2504.15219/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":"2026-07-05T11:55:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TdBRUZDIaWJW9/jdHs+WyKJXxhcxJYQgbcr9hUD1NC9ZZcM3wIjkAFrdkMiQ2HhCDn2vgedszs8nl8f3Y31yCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-13T06:20:38.533979Z"},"content_sha256":"f72136c443543bf988a2296f808a05462b1e8440c29983cd28d5b554b064f478","schema_version":"1.0","event_id":"sha256:f72136c443543bf988a2296f808a05462b1e8440c29983cd28d5b554b064f478"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6IW4Z36OOPTX2HATIYXDKGTVRK/bundle.json","state_url":"https://pith.science/pith/6IW4Z36OOPTX2HATIYXDKGTVRK/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6IW4Z36OOPTX2HATIYXDKGTVRK/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":"2026-08-13T06:20:38Z","links":{"resolver":"https://pith.science/pith/6IW4Z36OOPTX2HATIYXDKGTVRK","bundle":"https://pith.science/pith/6IW4Z36OOPTX2HATIYXDKGTVRK/bundle.json","state":"https://pith.science/pith/6IW4Z36OOPTX2HATIYXDKGTVRK/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6IW4Z36OOPTX2HATIYXDKGTVRK/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:6IW4Z36OOPTX2HATIYXDKGTVRK","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":"bbc22af0e54967beedda2de96d1d029ac9bee744c97d75d2e4aabb0fb6bbad96","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-21T16:43:50Z","title_canon_sha256":"2a549495bfffd8bab9e1f3548da13cb2e80c34c82487520668a6de204c6b5671"},"schema_version":"1.0","source":{"id":"2504.15219","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.15219","created_at":"2026-07-05T11:55:18Z"},{"alias_kind":"arxiv_version","alias_value":"2504.15219v2","created_at":"2026-07-05T11:55:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.15219","created_at":"2026-07-05T11:55:18Z"},{"alias_kind":"pith_short_12","alias_value":"6IW4Z36OOPTX","created_at":"2026-07-05T11:55:18Z"},{"alias_kind":"pith_short_16","alias_value":"6IW4Z36OOPTX2HAT","created_at":"2026-07-05T11:55:18Z"},{"alias_kind":"pith_short_8","alias_value":"6IW4Z36O","created_at":"2026-07-05T11:55:18Z"}],"graph_snapshots":[{"event_id":"sha256:f72136c443543bf988a2296f808a05462b1e8440c29983cd28d5b554b064f478","target":"graph","created_at":"2026-07-05T11:55:18Z","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/2504.15219/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Evaluation of language model outputs on structured writing tasks is typically conducted with a number of desirable criteria presented to human evaluators or large language models (LLMs). For instance, on a prompt like \"Help me draft an academic talk on coffee intake vs research productivity\", a model response may be evaluated for criteria like accuracy and coherence. However, high-quality responses should do more than just satisfy basic task requirements. An effective response to this query should include quintessential features of an academic talk, such as a compelling opening, clear research","authors_text":"Chaitanya Malaviya, Greg Durrett, Junyi Jessy Li, Manya Wadhwa, Philippe Laban, Zayne Sprague","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-21T16:43:50Z","title":"EvalAgent: Discovering Implicit Evaluation Criteria from the Web"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.15219","kind":"arxiv","version":2},"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:c353bd233ccabbebaf2677dffde658bd7cede8af0ac0a1469bf47d2ea0bd5b20","target":"record","created_at":"2026-07-05T11:55:18Z","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":"bbc22af0e54967beedda2de96d1d029ac9bee744c97d75d2e4aabb0fb6bbad96","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-04-21T16:43:50Z","title_canon_sha256":"2a549495bfffd8bab9e1f3548da13cb2e80c34c82487520668a6de204c6b5671"},"schema_version":"1.0","source":{"id":"2504.15219","kind":"arxiv","version":2}},"canonical_sha256":"f22dccefce73e77d1c13462e351a758a94ec4441179374bba62ee4ba1075d274","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f22dccefce73e77d1c13462e351a758a94ec4441179374bba62ee4ba1075d274","first_computed_at":"2026-07-05T11:55:18.815029Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:55:18.815029Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zs3zoWS2NGC2DJHqdKyMhXHbTg6NVk9+QhRGdAJQppOA+egDTb0tGOvYdw/oWvPprQDiMGeFUx3uQIjJLlcyBA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:55:18.815461Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.15219","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c353bd233ccabbebaf2677dffde658bd7cede8af0ac0a1469bf47d2ea0bd5b20","sha256:f72136c443543bf988a2296f808a05462b1e8440c29983cd28d5b554b064f478"],"state_sha256":"77d095b3e2257e2a58449c597e0453453a0b81984db70021da9ecb50cf0ec693"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pi8IOXPUbPJVkFv8qtxg2VxQEnxXfImscdMy18gVvgr2St8bOAoaHUQoKJ4BE1S2UW5kwmaeX4mrK69+R1NuBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-13T06:20:38.540067Z","bundle_sha256":"982a10d39129a21389eefe8e0ce4c5b7223f3eb095e3fc2204aa054d953a89f4"}}