{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:SOXVMNK6MJYBMJSKSNQ3LEUWWM","short_pith_number":"pith:SOXVMNK6","schema_version":"1.0","canonical_sha256":"93af56355e627016264a9361b59296b32880a3944fd04d178e43b1fbd0540c82","source":{"kind":"arxiv","id":"2606.22164","version":1},"attestation_state":"computed","paper":{"title":"Drowning in Routine: Signal Dilution in Multi-Turn Agent Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"2), (2) Polytechnique Montr\\'eal), Vi Retault (2) ((1) Mila - Qu\\'ebec AI Institute, Yann Pernot (1","submitted_at":"2026-06-20T17:55:10Z","abstract_excerpt":"Multi-turn agents interleave consequential decisions with routine execution: some actions change the downstream return distribution, while others are necessary but reward-equivalent. The cost of trajectory-level credit assignment, often attributed to long horizons, is in fact governed by decision density $\\rho$: the fraction of turns whose actions affect the return. When decision density is low, routine turns create signal dilution: they add gradient variance to trajectory-level estimators such as GRPO without adding expected signal. Under explicit assumptions, the resulting turn-level to traj"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.22164","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-06-20T17:55:10Z","cross_cats_sorted":[],"title_canon_sha256":"d928c1d670d588e3b4690aac62a7922e611a21f50c6ad5ad959cc483a30c4ba2","abstract_canon_sha256":"1c89e0a8b98213fcd5a8988015bcab9a30cdf3c80366d644604607d574dc6565"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T02:13:30.787033Z","signature_b64":"bAJW7VsFMXae7YAX/RBVnRP5/uVaJTna0eaCeNu0gzMKuURuHNlOWnFscY4tmEcRGBVNoWT269b1k7lNDIKKDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"93af56355e627016264a9361b59296b32880a3944fd04d178e43b1fbd0540c82","last_reissued_at":"2026-06-23T02:13:30.786632Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T02:13:30.786632Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Drowning in Routine: Signal Dilution in Multi-Turn Agent Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"2), (2) Polytechnique Montr\\'eal), Vi Retault (2) ((1) Mila - Qu\\'ebec AI Institute, Yann Pernot (1","submitted_at":"2026-06-20T17:55:10Z","abstract_excerpt":"Multi-turn agents interleave consequential decisions with routine execution: some actions change the downstream return distribution, while others are necessary but reward-equivalent. The cost of trajectory-level credit assignment, often attributed to long horizons, is in fact governed by decision density $\\rho$: the fraction of turns whose actions affect the return. When decision density is low, routine turns create signal dilution: they add gradient variance to trajectory-level estimators such as GRPO without adding expected signal. Under explicit assumptions, the resulting turn-level to traj"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.22164","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/2606.22164/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.22164","created_at":"2026-06-23T02:13:30.786702+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.22164v1","created_at":"2026-06-23T02:13:30.786702+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.22164","created_at":"2026-06-23T02:13:30.786702+00:00"},{"alias_kind":"pith_short_12","alias_value":"SOXVMNK6MJYB","created_at":"2026-06-23T02:13:30.786702+00:00"},{"alias_kind":"pith_short_16","alias_value":"SOXVMNK6MJYBMJSK","created_at":"2026-06-23T02:13:30.786702+00:00"},{"alias_kind":"pith_short_8","alias_value":"SOXVMNK6","created_at":"2026-06-23T02:13:30.786702+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SOXVMNK6MJYBMJSKSNQ3LEUWWM","json":"https://pith.science/pith/SOXVMNK6MJYBMJSKSNQ3LEUWWM.json","graph_json":"https://pith.science/api/pith-number/SOXVMNK6MJYBMJSKSNQ3LEUWWM/graph.json","events_json":"https://pith.science/api/pith-number/SOXVMNK6MJYBMJSKSNQ3LEUWWM/events.json","paper":"https://pith.science/paper/SOXVMNK6"},"agent_actions":{"view_html":"https://pith.science/pith/SOXVMNK6MJYBMJSKSNQ3LEUWWM","download_json":"https://pith.science/pith/SOXVMNK6MJYBMJSKSNQ3LEUWWM.json","view_paper":"https://pith.science/paper/SOXVMNK6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.22164&json=true","fetch_graph":"https://pith.science/api/pith-number/SOXVMNK6MJYBMJSKSNQ3LEUWWM/graph.json","fetch_events":"https://pith.science/api/pith-number/SOXVMNK6MJYBMJSKSNQ3LEUWWM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SOXVMNK6MJYBMJSKSNQ3LEUWWM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SOXVMNK6MJYBMJSKSNQ3LEUWWM/action/storage_attestation","attest_author":"https://pith.science/pith/SOXVMNK6MJYBMJSKSNQ3LEUWWM/action/author_attestation","sign_citation":"https://pith.science/pith/SOXVMNK6MJYBMJSKSNQ3LEUWWM/action/citation_signature","submit_replication":"https://pith.science/pith/SOXVMNK6MJYBMJSKSNQ3LEUWWM/action/replication_record"}},"created_at":"2026-06-23T02:13:30.786702+00:00","updated_at":"2026-06-23T02:13:30.786702+00:00"}