{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:OHT6LHTPQMD5EICVSVAMEDHFGA","short_pith_number":"pith:OHT6LHTP","canonical_record":{"source":{"id":"2608.11323","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-11T18:16:51Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7d0cb65c3d291d84155a15fc8de85311f1349199a4d931ea6d6fea178c97c54f","abstract_canon_sha256":"ca8d51413a37ad4d98762ff9e8f07be68dd5c40148aa0c5d9c4306e427f86e55"},"schema_version":"1.0"},"canonical_sha256":"71e7e59e6f8307d220559540c20ce5303a508f1978e94dee1d2887b7f1907e79","source":{"kind":"arxiv","id":"2608.11323","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.11323","created_at":"2026-08-13T00:21:35Z"},{"alias_kind":"arxiv_version","alias_value":"2608.11323v1","created_at":"2026-08-13T00:21:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.11323","created_at":"2026-08-13T00:21:35Z"},{"alias_kind":"pith_short_12","alias_value":"OHT6LHTPQMD5","created_at":"2026-08-13T00:21:35Z"},{"alias_kind":"pith_short_16","alias_value":"OHT6LHTPQMD5EICV","created_at":"2026-08-13T00:21:35Z"},{"alias_kind":"pith_short_8","alias_value":"OHT6LHTP","created_at":"2026-08-13T00:21:35Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:OHT6LHTPQMD5EICVSVAMEDHFGA","target":"record","payload":{"canonical_record":{"source":{"id":"2608.11323","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-11T18:16:51Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7d0cb65c3d291d84155a15fc8de85311f1349199a4d931ea6d6fea178c97c54f","abstract_canon_sha256":"ca8d51413a37ad4d98762ff9e8f07be68dd5c40148aa0c5d9c4306e427f86e55"},"schema_version":"1.0"},"canonical_sha256":"71e7e59e6f8307d220559540c20ce5303a508f1978e94dee1d2887b7f1907e79","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-13T00:21:35.571484Z","signature_b64":"4QJoKoeY+54BviaiVrKW2ib0R/gkk+JW/99cv9uUwLtUZdKqs5eUqlK4msXu1DUtBjINuvbxoN8LIV62y1I4AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"71e7e59e6f8307d220559540c20ce5303a508f1978e94dee1d2887b7f1907e79","last_reissued_at":"2026-08-13T00:21:35.569699Z","signature_status":"signed_v1","first_computed_at":"2026-08-13T00:21:35.569699Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.11323","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":"2026-08-13T00:21:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"BhzGw8CXacgzq0nFiEB3SAVZgq8e+YY3qam4pXxgolphtgawmHePQyUJ5wWZ1AOaprMNWPJ92lp+kxsAET9OBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T21:32:47.348589Z"},"content_sha256":"665e09482d5e0693a69d9a631065d1d8fa3ae062e1d083b9a256e2c9a19bc384","schema_version":"1.0","event_id":"sha256:665e09482d5e0693a69d9a631065d1d8fa3ae062e1d083b9a256e2c9a19bc384"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:OHT6LHTPQMD5EICVSVAMEDHFGA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Deployment Decision Reliability: A Generalizability-Theory Framework for Sizing Long-Horizon Agent Evaluations","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Vasundra Srinivasan","submitted_at":"2026-08-11T18:16:51Z","abstract_excerpt":"Enterprise practitioners read agent leaderboards as if they ranked agent capability. We show, across three open agent-trace benchmarks (TheAgentCompany, $\\tau^2$-bench, and AppWorld), that the agent main effect accounts for less than 3% of total variance in every dataset and check type, while the agent-by-task interaction accounts for 7-23%. Leaderboards rank specialization, not capability. We arrive at this through a four-facet Generalizability Theory variance decomposition, fit with three estimators (Henderson Method-I, REML via lme4, and a Bayesian binomial GLMM) that agree to three decimal"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.11323","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/2608.11323/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":"2026-08-13T00:21:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"A2OD6kQe+JdVGXMSKXnGok951mrNnuaocz0a90rVc+uZ8JSpOj1tkRPel2X5ib0vpAawnCLWbPiagmt5x2LrAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T21:32:47.349776Z"},"content_sha256":"a41a2a41d41ad321be45b11f280c82657dc7df57a522296ad7296e40fbf13540","schema_version":"1.0","event_id":"sha256:a41a2a41d41ad321be45b11f280c82657dc7df57a522296ad7296e40fbf13540"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OHT6LHTPQMD5EICVSVAMEDHFGA/bundle.json","state_url":"https://pith.science/pith/OHT6LHTPQMD5EICVSVAMEDHFGA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OHT6LHTPQMD5EICVSVAMEDHFGA/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":"2026-08-17T21:32:47Z","links":{"resolver":"https://pith.science/pith/OHT6LHTPQMD5EICVSVAMEDHFGA","bundle":"https://pith.science/pith/OHT6LHTPQMD5EICVSVAMEDHFGA/bundle.json","state":"https://pith.science/pith/OHT6LHTPQMD5EICVSVAMEDHFGA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OHT6LHTPQMD5EICVSVAMEDHFGA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:OHT6LHTPQMD5EICVSVAMEDHFGA","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":"ca8d51413a37ad4d98762ff9e8f07be68dd5c40148aa0c5d9c4306e427f86e55","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-11T18:16:51Z","title_canon_sha256":"7d0cb65c3d291d84155a15fc8de85311f1349199a4d931ea6d6fea178c97c54f"},"schema_version":"1.0","source":{"id":"2608.11323","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.11323","created_at":"2026-08-13T00:21:35Z"},{"alias_kind":"arxiv_version","alias_value":"2608.11323v1","created_at":"2026-08-13T00:21:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.11323","created_at":"2026-08-13T00:21:35Z"},{"alias_kind":"pith_short_12","alias_value":"OHT6LHTPQMD5","created_at":"2026-08-13T00:21:35Z"},{"alias_kind":"pith_short_16","alias_value":"OHT6LHTPQMD5EICV","created_at":"2026-08-13T00:21:35Z"},{"alias_kind":"pith_short_8","alias_value":"OHT6LHTP","created_at":"2026-08-13T00:21:35Z"}],"graph_snapshots":[{"event_id":"sha256:a41a2a41d41ad321be45b11f280c82657dc7df57a522296ad7296e40fbf13540","target":"graph","created_at":"2026-08-13T00:21:35Z","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/2608.11323/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Enterprise practitioners read agent leaderboards as if they ranked agent capability. We show, across three open agent-trace benchmarks (TheAgentCompany, $\\tau^2$-bench, and AppWorld), that the agent main effect accounts for less than 3% of total variance in every dataset and check type, while the agent-by-task interaction accounts for 7-23%. Leaderboards rank specialization, not capability. We arrive at this through a four-facet Generalizability Theory variance decomposition, fit with three estimators (Henderson Method-I, REML via lme4, and a Bayesian binomial GLMM) that agree to three decimal","authors_text":"Vasundra Srinivasan","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-11T18:16:51Z","title":"Deployment Decision Reliability: A Generalizability-Theory Framework for Sizing Long-Horizon Agent Evaluations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.11323","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":"sha256:665e09482d5e0693a69d9a631065d1d8fa3ae062e1d083b9a256e2c9a19bc384","target":"record","created_at":"2026-08-13T00:21:35Z","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":"ca8d51413a37ad4d98762ff9e8f07be68dd5c40148aa0c5d9c4306e427f86e55","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-11T18:16:51Z","title_canon_sha256":"7d0cb65c3d291d84155a15fc8de85311f1349199a4d931ea6d6fea178c97c54f"},"schema_version":"1.0","source":{"id":"2608.11323","kind":"arxiv","version":1}},"canonical_sha256":"71e7e59e6f8307d220559540c20ce5303a508f1978e94dee1d2887b7f1907e79","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"71e7e59e6f8307d220559540c20ce5303a508f1978e94dee1d2887b7f1907e79","first_computed_at":"2026-08-13T00:21:35.569699Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-13T00:21:35.569699Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"4QJoKoeY+54BviaiVrKW2ib0R/gkk+JW/99cv9uUwLtUZdKqs5eUqlK4msXu1DUtBjINuvbxoN8LIV62y1I4AQ==","signature_status":"signed_v1","signed_at":"2026-08-13T00:21:35.571484Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.11323","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:665e09482d5e0693a69d9a631065d1d8fa3ae062e1d083b9a256e2c9a19bc384","sha256:a41a2a41d41ad321be45b11f280c82657dc7df57a522296ad7296e40fbf13540"],"state_sha256":"88238c1ffd29796162ac749b5ba748d60400177cc877663ca38d1a85524f0a94"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aY2Q9L7aCqYrVwBrPTVjqJ1wQTEEHrrfLXxXsLXQXq/qV8udd0v7axLWQwPm7IIKmYmsaUOVtCoxdXT1VDpRCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T21:32:47.360228Z","bundle_sha256":"9b9eef178d26155f09f45fa265e4f0190ac7a5cf161d5585e6ab0d524d69a392"}}