{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:NHH3DAQX7PZZSXV3UHHI7VEOFW","short_pith_number":"pith:NHH3DAQX","schema_version":"1.0","canonical_sha256":"69cfb18217fbf3995ebba1ce8fd48e2d84a8602c6fd2a80512dbb23304548dcb","source":{"kind":"arxiv","id":"2607.17558","version":1},"attestation_state":"computed","paper":{"title":"Why Does Feedback-Augmented Self-Distillation Fail to Improve Retrieval-Interleaved Search Agents?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Fan Yang, Rui Meng, Yuxin Wen","submitted_at":"2026-07-20T05:07:22Z","abstract_excerpt":"On-policy self-distillation (OPSD) offers a promising approach for training large language models without relying on a separate teacher model. However, its effectiveness on complex agentic tasks remains largely unexplored. In this work, we instantiate Feedback-Augmented Self-Distillation (FA-SD), a self-distillation algorithm for agentic search that leverages successful demonstrations as privileged information. We identify that models can rely on recurring reasoning-and-search output templates, producing trajectories that appear diverse but are largely agnostic to the input question, making th"},"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":"2607.17558","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-07-20T05:07:22Z","cross_cats_sorted":[],"title_canon_sha256":"4cf5133d7735c76a0224467a2f91d55f5a8945b19217586e0651cec84577679d","abstract_canon_sha256":"d820c77d397dd5c5c7069f3b01d195d4e927db018d57613ad24e78a7685f5a97"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T01:21:52.245269Z","signature_b64":"X2OxuZxY+PJIhaGJUwZhqUQuGBqAoZeo772twsAnMUvUPJnLzmkeB/hOiQEj3DkRy9J9N3Hu3ytZ3JkJpH62Dg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"69cfb18217fbf3995ebba1ce8fd48e2d84a8602c6fd2a80512dbb23304548dcb","last_reissued_at":"2026-07-21T01:21:52.244428Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T01:21:52.244428Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Why Does Feedback-Augmented Self-Distillation Fail to Improve Retrieval-Interleaved Search Agents?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Fan Yang, Rui Meng, Yuxin Wen","submitted_at":"2026-07-20T05:07:22Z","abstract_excerpt":"On-policy self-distillation (OPSD) offers a promising approach for training large language models without relying on a separate teacher model. However, its effectiveness on complex agentic tasks remains largely unexplored. In this work, we instantiate Feedback-Augmented Self-Distillation (FA-SD), a self-distillation algorithm for agentic search that leverages successful demonstrations as privileged information. We identify that models can rely on recurring reasoning-and-search output templates, producing trajectories that appear diverse but are largely agnostic to the input question, making th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.17558","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/2607.17558/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":"2607.17558","created_at":"2026-07-21T01:21:52.244852+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.17558v1","created_at":"2026-07-21T01:21:52.244852+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.17558","created_at":"2026-07-21T01:21:52.244852+00:00"},{"alias_kind":"pith_short_12","alias_value":"NHH3DAQX7PZZ","created_at":"2026-07-21T01:21:52.244852+00:00"},{"alias_kind":"pith_short_16","alias_value":"NHH3DAQX7PZZSXV3","created_at":"2026-07-21T01:21:52.244852+00:00"},{"alias_kind":"pith_short_8","alias_value":"NHH3DAQX","created_at":"2026-07-21T01:21:52.244852+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/NHH3DAQX7PZZSXV3UHHI7VEOFW","json":"https://pith.science/pith/NHH3DAQX7PZZSXV3UHHI7VEOFW.json","graph_json":"https://pith.science/api/pith-number/NHH3DAQX7PZZSXV3UHHI7VEOFW/graph.json","events_json":"https://pith.science/api/pith-number/NHH3DAQX7PZZSXV3UHHI7VEOFW/events.json","paper":"https://pith.science/paper/NHH3DAQX"},"agent_actions":{"view_html":"https://pith.science/pith/NHH3DAQX7PZZSXV3UHHI7VEOFW","download_json":"https://pith.science/pith/NHH3DAQX7PZZSXV3UHHI7VEOFW.json","view_paper":"https://pith.science/paper/NHH3DAQX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.17558&json=true","fetch_graph":"https://pith.science/api/pith-number/NHH3DAQX7PZZSXV3UHHI7VEOFW/graph.json","fetch_events":"https://pith.science/api/pith-number/NHH3DAQX7PZZSXV3UHHI7VEOFW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/NHH3DAQX7PZZSXV3UHHI7VEOFW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/NHH3DAQX7PZZSXV3UHHI7VEOFW/action/storage_attestation","attest_author":"https://pith.science/pith/NHH3DAQX7PZZSXV3UHHI7VEOFW/action/author_attestation","sign_citation":"https://pith.science/pith/NHH3DAQX7PZZSXV3UHHI7VEOFW/action/citation_signature","submit_replication":"https://pith.science/pith/NHH3DAQX7PZZSXV3UHHI7VEOFW/action/replication_record"}},"created_at":"2026-07-21T01:21:52.244852+00:00","updated_at":"2026-07-21T01:21:52.244852+00:00"}