{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:NF7B3YFJXZR6ZE7XO2MFWQ2I7J","short_pith_number":"pith:NF7B3YFJ","canonical_record":{"source":{"id":"2606.22737","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-22T00:41:16Z","cross_cats_sorted":["cs.CL","cs.SE"],"title_canon_sha256":"e0bddb12ee29d64d94b31556b609d257b16456865f8880fa5458d78c8c2e1d3d","abstract_canon_sha256":"37badc9e1609124510ea289a73a4dc35e0a05f632029f85f1e6e8809747d5010"},"schema_version":"1.0"},"canonical_sha256":"697e1de0a9be63ec93f776985b4348fa490872e8bcaa5a8a887f7d5cf3de3bc8","source":{"kind":"arxiv","id":"2606.22737","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.22737","created_at":"2026-06-23T02:13:46Z"},{"alias_kind":"arxiv_version","alias_value":"2606.22737v1","created_at":"2026-06-23T02:13:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.22737","created_at":"2026-06-23T02:13:46Z"},{"alias_kind":"pith_short_12","alias_value":"NF7B3YFJXZR6","created_at":"2026-06-23T02:13:46Z"},{"alias_kind":"pith_short_16","alias_value":"NF7B3YFJXZR6ZE7X","created_at":"2026-06-23T02:13:46Z"},{"alias_kind":"pith_short_8","alias_value":"NF7B3YFJ","created_at":"2026-06-23T02:13:46Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:NF7B3YFJXZR6ZE7XO2MFWQ2I7J","target":"record","payload":{"canonical_record":{"source":{"id":"2606.22737","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-22T00:41:16Z","cross_cats_sorted":["cs.CL","cs.SE"],"title_canon_sha256":"e0bddb12ee29d64d94b31556b609d257b16456865f8880fa5458d78c8c2e1d3d","abstract_canon_sha256":"37badc9e1609124510ea289a73a4dc35e0a05f632029f85f1e6e8809747d5010"},"schema_version":"1.0"},"canonical_sha256":"697e1de0a9be63ec93f776985b4348fa490872e8bcaa5a8a887f7d5cf3de3bc8","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T02:13:46.001621Z","signature_b64":"bqbPh/jqN+80FQp+l+ml+VSOQCWAGS4/34/IKo04yS5IjSlSJ+NlmQ+HWmjAljRHBw2n9Q3XxNKxCvAmZusGAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"697e1de0a9be63ec93f776985b4348fa490872e8bcaa5a8a887f7d5cf3de3bc8","last_reissued_at":"2026-06-23T02:13:46.001229Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T02:13:46.001229Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.22737","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":"2026-06-23T02:13:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"s/zn3Leo0dc9MM7UnEsGzJ/uoAMb/S3qwfbpbOcIi309w9GC3MTM9kOQ2tmEIvtdx6LjegyvYeO6TAkgEhTjAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-28T18:43:41.220364Z"},"content_sha256":"05d0bb7f94ec77a8f8c442a46d8edd4f8329389b838e7cf5ac4546efff1acf8b","schema_version":"1.0","event_id":"sha256:05d0bb7f94ec77a8f8c442a46d8edd4f8329389b838e7cf5ac4546efff1acf8b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:NF7B3YFJXZR6ZE7XO2MFWQ2I7J","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"GroundEval: A Deterministic Replacement for LLM-as-Judge in Stateful Agent Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.SE"],"primary_cat":"cs.AI","authors_text":"Jeffrey Flynt","submitted_at":"2026-06-22T00:41:16Z","abstract_excerpt":"Before letting an agent operate over real context, can you prove it used the right evidence? GroundEval turns that question into a deterministic test of what the agent searched, fetched, cited, and was permitted to access. In one case study, two frontier LLM judges scored a plausible agent response above 0.85. But the trace told a different story: the agent had never retrieved the artifact its answer depended on, yielding a GroundEval score of 0.000.\n  We introduce GroundEval, a judge-free framework for evaluating agents against grounded, time-bounded, and access-controlled evidence. GroundEva"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.22737","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/2606.22737/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":"2026-06-23T02:13:46Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uq8ovviIX4cdlWfphKz8BEsNWAZ5/CDhjUry9m+ZtYGejS+2fhhhZtO86EZEuP5L+8H/D8oRGTxVyVoExtQ6DA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-28T18:43:41.220742Z"},"content_sha256":"754429b3a1f76bcd202b3f8b5f1283d7b1b8406d3cdeff15db49b1659a28f50f","schema_version":"1.0","event_id":"sha256:754429b3a1f76bcd202b3f8b5f1283d7b1b8406d3cdeff15db49b1659a28f50f"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NF7B3YFJXZR6ZE7XO2MFWQ2I7J/bundle.json","state_url":"https://pith.science/pith/NF7B3YFJXZR6ZE7XO2MFWQ2I7J/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NF7B3YFJXZR6ZE7XO2MFWQ2I7J/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":"2026-06-28T18:43:41Z","links":{"resolver":"https://pith.science/pith/NF7B3YFJXZR6ZE7XO2MFWQ2I7J","bundle":"https://pith.science/pith/NF7B3YFJXZR6ZE7XO2MFWQ2I7J/bundle.json","state":"https://pith.science/pith/NF7B3YFJXZR6ZE7XO2MFWQ2I7J/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NF7B3YFJXZR6ZE7XO2MFWQ2I7J/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NF7B3YFJXZR6ZE7XO2MFWQ2I7J","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":"37badc9e1609124510ea289a73a4dc35e0a05f632029f85f1e6e8809747d5010","cross_cats_sorted":["cs.CL","cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-22T00:41:16Z","title_canon_sha256":"e0bddb12ee29d64d94b31556b609d257b16456865f8880fa5458d78c8c2e1d3d"},"schema_version":"1.0","source":{"id":"2606.22737","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.22737","created_at":"2026-06-23T02:13:46Z"},{"alias_kind":"arxiv_version","alias_value":"2606.22737v1","created_at":"2026-06-23T02:13:46Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.22737","created_at":"2026-06-23T02:13:46Z"},{"alias_kind":"pith_short_12","alias_value":"NF7B3YFJXZR6","created_at":"2026-06-23T02:13:46Z"},{"alias_kind":"pith_short_16","alias_value":"NF7B3YFJXZR6ZE7X","created_at":"2026-06-23T02:13:46Z"},{"alias_kind":"pith_short_8","alias_value":"NF7B3YFJ","created_at":"2026-06-23T02:13:46Z"}],"graph_snapshots":[{"event_id":"sha256:754429b3a1f76bcd202b3f8b5f1283d7b1b8406d3cdeff15db49b1659a28f50f","target":"graph","created_at":"2026-06-23T02:13:46Z","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/2606.22737/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Before letting an agent operate over real context, can you prove it used the right evidence? GroundEval turns that question into a deterministic test of what the agent searched, fetched, cited, and was permitted to access. In one case study, two frontier LLM judges scored a plausible agent response above 0.85. But the trace told a different story: the agent had never retrieved the artifact its answer depended on, yielding a GroundEval score of 0.000.\n  We introduce GroundEval, a judge-free framework for evaluating agents against grounded, time-bounded, and access-controlled evidence. GroundEva","authors_text":"Jeffrey Flynt","cross_cats":["cs.CL","cs.SE"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-22T00:41:16Z","title":"GroundEval: A Deterministic Replacement for LLM-as-Judge in Stateful Agent Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.22737","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":"sha256:05d0bb7f94ec77a8f8c442a46d8edd4f8329389b838e7cf5ac4546efff1acf8b","target":"record","created_at":"2026-06-23T02:13:46Z","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":"37badc9e1609124510ea289a73a4dc35e0a05f632029f85f1e6e8809747d5010","cross_cats_sorted":["cs.CL","cs.SE"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-22T00:41:16Z","title_canon_sha256":"e0bddb12ee29d64d94b31556b609d257b16456865f8880fa5458d78c8c2e1d3d"},"schema_version":"1.0","source":{"id":"2606.22737","kind":"arxiv","version":1}},"canonical_sha256":"697e1de0a9be63ec93f776985b4348fa490872e8bcaa5a8a887f7d5cf3de3bc8","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"697e1de0a9be63ec93f776985b4348fa490872e8bcaa5a8a887f7d5cf3de3bc8","first_computed_at":"2026-06-23T02:13:46.001229Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-23T02:13:46.001229Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bqbPh/jqN+80FQp+l+ml+VSOQCWAGS4/34/IKo04yS5IjSlSJ+NlmQ+HWmjAljRHBw2n9Q3XxNKxCvAmZusGAQ==","signature_status":"signed_v1","signed_at":"2026-06-23T02:13:46.001621Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.22737","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:05d0bb7f94ec77a8f8c442a46d8edd4f8329389b838e7cf5ac4546efff1acf8b","sha256:754429b3a1f76bcd202b3f8b5f1283d7b1b8406d3cdeff15db49b1659a28f50f"],"state_sha256":"3197af41cf20fefb87c69bafd506a8ed27aa6908b3d7b8297cf03deaf409ee75"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"88jJxlyfhEozTEjFyyvWfFwLmz2QFRkslv+F/fO5RAhLf+ocKzlU4JS5Zxgj3rXfiIM01gju7FhXIfzNFhdVDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-28T18:43:41.222815Z","bundle_sha256":"2271dd5d9c2b118efbc6da859a2c0710541fe803627e179cea7348f5fce2b575"}}