{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:WSTJLLNXSCEO2JHMTZAFX7GB6N","short_pith_number":"pith:WSTJLLNX","schema_version":"1.0","canonical_sha256":"b4a695adb79088ed24ec9e405bfcc1f34ef52cee56b3eecdefbcfc913096eca2","source":{"kind":"arxiv","id":"2605.27935","version":1},"attestation_state":"computed","paper":{"title":"Do Agents Think Deeper? A Mechanistic Investigation of Layer-Wise Dynamics in Sequential Planning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Xiangzhong Luo, Zhenyu Cui","submitted_at":"2026-05-27T04:08:08Z","abstract_excerpt":"Recent mechanistic studies suggest that large language models (LLMs) may utilize their depth inefficiently in standard single-turn tasks. Whether this still holds in autonomous agent settings, where models must perform multi-turn planning, tool use, and iterative state updates, remains unclear. We study this question through a systematic layer-wise analysis of complete user-agent trajectories spanning three domains: Deep Research, Code Generation, and Tabular Processing. Using residual stream probes, causal layer-skipping interventions, and effective-depth measurements, we show that agentic re"},"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":"2605.27935","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-05-27T04:08:08Z","cross_cats_sorted":[],"title_canon_sha256":"98b1085a3e7b9e4c5c4b62dc36cb4abb9dc171283a2ae914e4847e738feba547","abstract_canon_sha256":"d1d195c05ebaf34f61326e5b9144b670fefb650b84eb2b873ec2ce9f80bff4f0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:04:53.001796Z","signature_b64":"MtXzcQ76cBstYVPQbWtYWMDgQzFXzoV60GV8ob2ubaGnjZoq+GysRyDtmpZf2CdfQoZndS8R0cf1yGvKpV77DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b4a695adb79088ed24ec9e405bfcc1f34ef52cee56b3eecdefbcfc913096eca2","last_reissued_at":"2026-05-28T01:04:52.998323Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:04:52.998323Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Do Agents Think Deeper? A Mechanistic Investigation of Layer-Wise Dynamics in Sequential Planning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Xiangzhong Luo, Zhenyu Cui","submitted_at":"2026-05-27T04:08:08Z","abstract_excerpt":"Recent mechanistic studies suggest that large language models (LLMs) may utilize their depth inefficiently in standard single-turn tasks. Whether this still holds in autonomous agent settings, where models must perform multi-turn planning, tool use, and iterative state updates, remains unclear. We study this question through a systematic layer-wise analysis of complete user-agent trajectories spanning three domains: Deep Research, Code Generation, and Tabular Processing. Using residual stream probes, causal layer-skipping interventions, and effective-depth measurements, we show that agentic re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.27935","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/2605.27935/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":"2605.27935","created_at":"2026-05-28T01:04:53.001334+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.27935v1","created_at":"2026-05-28T01:04:53.001334+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.27935","created_at":"2026-05-28T01:04:53.001334+00:00"},{"alias_kind":"pith_short_12","alias_value":"WSTJLLNXSCEO","created_at":"2026-05-28T01:04:53.001334+00:00"},{"alias_kind":"pith_short_16","alias_value":"WSTJLLNXSCEO2JHM","created_at":"2026-05-28T01:04:53.001334+00:00"},{"alias_kind":"pith_short_8","alias_value":"WSTJLLNX","created_at":"2026-05-28T01:04:53.001334+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/WSTJLLNXSCEO2JHMTZAFX7GB6N","json":"https://pith.science/pith/WSTJLLNXSCEO2JHMTZAFX7GB6N.json","graph_json":"https://pith.science/api/pith-number/WSTJLLNXSCEO2JHMTZAFX7GB6N/graph.json","events_json":"https://pith.science/api/pith-number/WSTJLLNXSCEO2JHMTZAFX7GB6N/events.json","paper":"https://pith.science/paper/WSTJLLNX"},"agent_actions":{"view_html":"https://pith.science/pith/WSTJLLNXSCEO2JHMTZAFX7GB6N","download_json":"https://pith.science/pith/WSTJLLNXSCEO2JHMTZAFX7GB6N.json","view_paper":"https://pith.science/paper/WSTJLLNX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.27935&json=true","fetch_graph":"https://pith.science/api/pith-number/WSTJLLNXSCEO2JHMTZAFX7GB6N/graph.json","fetch_events":"https://pith.science/api/pith-number/WSTJLLNXSCEO2JHMTZAFX7GB6N/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WSTJLLNXSCEO2JHMTZAFX7GB6N/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WSTJLLNXSCEO2JHMTZAFX7GB6N/action/storage_attestation","attest_author":"https://pith.science/pith/WSTJLLNXSCEO2JHMTZAFX7GB6N/action/author_attestation","sign_citation":"https://pith.science/pith/WSTJLLNXSCEO2JHMTZAFX7GB6N/action/citation_signature","submit_replication":"https://pith.science/pith/WSTJLLNXSCEO2JHMTZAFX7GB6N/action/replication_record"}},"created_at":"2026-05-28T01:04:53.001334+00:00","updated_at":"2026-05-28T01:04:53.001334+00:00"}