{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:BJSTLSDCFP6MAL2GUGYCHP5WWS","short_pith_number":"pith:BJSTLSDC","canonical_record":{"source":{"id":"2606.23937","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-22T20:57:11Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"3707f47895b30d801d4bca6a513fb37d6a7ecc8bafdc6f19594eca3924bd22db","abstract_canon_sha256":"62bcbab2d54d327b1da4d57804ecc3cd0c988e8740dea1849bbc0098d44e97d3"},"schema_version":"1.0"},"canonical_sha256":"0a6535c8622bfcc02f46a1b023bfb6b4802a261fd13a285d807b0cd1340cb1f5","source":{"kind":"arxiv","id":"2606.23937","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.23937","created_at":"2026-06-24T00:14:30Z"},{"alias_kind":"arxiv_version","alias_value":"2606.23937v1","created_at":"2026-06-24T00:14:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.23937","created_at":"2026-06-24T00:14:30Z"},{"alias_kind":"pith_short_12","alias_value":"BJSTLSDCFP6M","created_at":"2026-06-24T00:14:30Z"},{"alias_kind":"pith_short_16","alias_value":"BJSTLSDCFP6MAL2G","created_at":"2026-06-24T00:14:30Z"},{"alias_kind":"pith_short_8","alias_value":"BJSTLSDC","created_at":"2026-06-24T00:14:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:BJSTLSDCFP6MAL2GUGYCHP5WWS","target":"record","payload":{"canonical_record":{"source":{"id":"2606.23937","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-22T20:57:11Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"3707f47895b30d801d4bca6a513fb37d6a7ecc8bafdc6f19594eca3924bd22db","abstract_canon_sha256":"62bcbab2d54d327b1da4d57804ecc3cd0c988e8740dea1849bbc0098d44e97d3"},"schema_version":"1.0"},"canonical_sha256":"0a6535c8622bfcc02f46a1b023bfb6b4802a261fd13a285d807b0cd1340cb1f5","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-24T00:14:30.800205Z","signature_b64":"os4DhyHqRjsY/ANVVFKYK4u9mlpxxfDtTU4iiYZ3GtfOOwAM1FgWoRDsdgo5yJFyct5jENLbUDg5f9TjpLCzBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0a6535c8622bfcc02f46a1b023bfb6b4802a261fd13a285d807b0cd1340cb1f5","last_reissued_at":"2026-06-24T00:14:30.799837Z","signature_status":"signed_v1","first_computed_at":"2026-06-24T00:14:30.799837Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.23937","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":"2026-06-24T00:14:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VX1kqDJ1+fNIlqlm2D9Prg0b46YHStuf7Lq0NJz3cL9Zg6QiQN96wSCYEydFGYVCAYkgfqajkCFNnHYHd6yGCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T21:23:06.806349Z"},"content_sha256":"ecfb063836f743c4a7ffcff53c620588401efda96a59d7d01ad1fa19a65c24fe","schema_version":"1.0","event_id":"sha256:ecfb063836f743c4a7ffcff53c620588401efda96a59d7d01ad1fa19a65c24fe"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:BJSTLSDCFP6MAL2GUGYCHP5WWS","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"When Retrieval Metrics Mislead: Measuring Policy Signal in Long-Horizon Tool-Use Agents","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Juan Pablo De la Cruz Weinstein, Tianyu Ding","submitted_at":"2026-06-22T20:57:11Z","abstract_excerpt":"Exact-match retrieval recall is often used as a proxy for whether a retriever supplies useful policy context to a downstream decision model. We test this proxy for pre-action policy classification in tau-bench using Qwen2.5-3B/7B classifiers. Under gold-policy conditioning, a compact structured state improves macro-F1 over raw trajectories by 0.13-0.17 after tuning. We then replace the benchmark-designated policy clause with the top-ranked clause retrieved from decision-time context. Although the exact governing clause is retrieved at rank 1 for only 7% of airline states, the primary 3B classi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.23937","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/2606.23937/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":"2026-06-24T00:14:30Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nQIb1mbL8r1p4QTqEtn3fvN+KQ7yirmh7Vg4spKuuBqB4rA3TFmHWpBphMcemzHji32nSFdHN3KRqnZ+E+W/CA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T21:23:06.806737Z"},"content_sha256":"29ee798331b8da7be61a10b2fe8af23c8c7d65b255013625897a3219278ed7f8","schema_version":"1.0","event_id":"sha256:29ee798331b8da7be61a10b2fe8af23c8c7d65b255013625897a3219278ed7f8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BJSTLSDCFP6MAL2GUGYCHP5WWS/bundle.json","state_url":"https://pith.science/pith/BJSTLSDCFP6MAL2GUGYCHP5WWS/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BJSTLSDCFP6MAL2GUGYCHP5WWS/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":"2026-06-30T21:23:06Z","links":{"resolver":"https://pith.science/pith/BJSTLSDCFP6MAL2GUGYCHP5WWS","bundle":"https://pith.science/pith/BJSTLSDCFP6MAL2GUGYCHP5WWS/bundle.json","state":"https://pith.science/pith/BJSTLSDCFP6MAL2GUGYCHP5WWS/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BJSTLSDCFP6MAL2GUGYCHP5WWS/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:BJSTLSDCFP6MAL2GUGYCHP5WWS","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":"62bcbab2d54d327b1da4d57804ecc3cd0c988e8740dea1849bbc0098d44e97d3","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-22T20:57:11Z","title_canon_sha256":"3707f47895b30d801d4bca6a513fb37d6a7ecc8bafdc6f19594eca3924bd22db"},"schema_version":"1.0","source":{"id":"2606.23937","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.23937","created_at":"2026-06-24T00:14:30Z"},{"alias_kind":"arxiv_version","alias_value":"2606.23937v1","created_at":"2026-06-24T00:14:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.23937","created_at":"2026-06-24T00:14:30Z"},{"alias_kind":"pith_short_12","alias_value":"BJSTLSDCFP6M","created_at":"2026-06-24T00:14:30Z"},{"alias_kind":"pith_short_16","alias_value":"BJSTLSDCFP6MAL2G","created_at":"2026-06-24T00:14:30Z"},{"alias_kind":"pith_short_8","alias_value":"BJSTLSDC","created_at":"2026-06-24T00:14:30Z"}],"graph_snapshots":[{"event_id":"sha256:29ee798331b8da7be61a10b2fe8af23c8c7d65b255013625897a3219278ed7f8","target":"graph","created_at":"2026-06-24T00:14:30Z","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/2606.23937/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Exact-match retrieval recall is often used as a proxy for whether a retriever supplies useful policy context to a downstream decision model. We test this proxy for pre-action policy classification in tau-bench using Qwen2.5-3B/7B classifiers. Under gold-policy conditioning, a compact structured state improves macro-F1 over raw trajectories by 0.13-0.17 after tuning. We then replace the benchmark-designated policy clause with the top-ranked clause retrieved from decision-time context. Although the exact governing clause is retrieved at rank 1 for only 7% of airline states, the primary 3B classi","authors_text":"Juan Pablo De la Cruz Weinstein, Tianyu Ding","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-22T20:57:11Z","title":"When Retrieval Metrics Mislead: Measuring Policy Signal in Long-Horizon Tool-Use Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.23937","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:ecfb063836f743c4a7ffcff53c620588401efda96a59d7d01ad1fa19a65c24fe","target":"record","created_at":"2026-06-24T00:14:30Z","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":"62bcbab2d54d327b1da4d57804ecc3cd0c988e8740dea1849bbc0098d44e97d3","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-06-22T20:57:11Z","title_canon_sha256":"3707f47895b30d801d4bca6a513fb37d6a7ecc8bafdc6f19594eca3924bd22db"},"schema_version":"1.0","source":{"id":"2606.23937","kind":"arxiv","version":1}},"canonical_sha256":"0a6535c8622bfcc02f46a1b023bfb6b4802a261fd13a285d807b0cd1340cb1f5","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0a6535c8622bfcc02f46a1b023bfb6b4802a261fd13a285d807b0cd1340cb1f5","first_computed_at":"2026-06-24T00:14:30.799837Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-24T00:14:30.799837Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"os4DhyHqRjsY/ANVVFKYK4u9mlpxxfDtTU4iiYZ3GtfOOwAM1FgWoRDsdgo5yJFyct5jENLbUDg5f9TjpLCzBg==","signature_status":"signed_v1","signed_at":"2026-06-24T00:14:30.800205Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.23937","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:ecfb063836f743c4a7ffcff53c620588401efda96a59d7d01ad1fa19a65c24fe","sha256:29ee798331b8da7be61a10b2fe8af23c8c7d65b255013625897a3219278ed7f8"],"state_sha256":"d2c53ec7963e34e20f90b0c3d4a0847aec54474c7218738b0740de8f63e12aa7"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xwWxxU9uk6xnW2RXZ0WuJUq+u57qY41AP1//FVCO51VCHKvyv+9kMCijDSD/XZKfvu+wFTIPUEOj1N6hRttzDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-30T21:23:06.809775Z","bundle_sha256":"686ecde9daaa215b23b3ae0d898708ae39b7097e5b16215173753e1a1c291b8c"}}