{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:VFLFREZPN6POFUG3NPVHGIOBY4","short_pith_number":"pith:VFLFREZP","canonical_record":{"source":{"id":"2605.11182","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-11T19:44:59Z","cross_cats_sorted":[],"title_canon_sha256":"b248cb51e667023b57534f1c43b260f0bb7b7d2661aab3993870fca78c420d80","abstract_canon_sha256":"de0bcebca2a87255dcfef7f5533ad79685999585cf31a57ff359212454eb5959"},"schema_version":"1.0"},"canonical_sha256":"a95658932f6f9ee2d0db6bea7321c1c71826e3bb411c55a1db19227701c45158","source":{"kind":"arxiv","id":"2605.11182","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.11182","created_at":"2026-05-26T01:03:33Z"},{"alias_kind":"arxiv_version","alias_value":"2605.11182v2","created_at":"2026-05-26T01:03:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.11182","created_at":"2026-05-26T01:03:33Z"},{"alias_kind":"pith_short_12","alias_value":"VFLFREZPN6PO","created_at":"2026-05-26T01:03:33Z"},{"alias_kind":"pith_short_16","alias_value":"VFLFREZPN6POFUG3","created_at":"2026-05-26T01:03:33Z"},{"alias_kind":"pith_short_8","alias_value":"VFLFREZP","created_at":"2026-05-26T01:03:33Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:VFLFREZPN6POFUG3NPVHGIOBY4","target":"record","payload":{"canonical_record":{"source":{"id":"2605.11182","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-11T19:44:59Z","cross_cats_sorted":[],"title_canon_sha256":"b248cb51e667023b57534f1c43b260f0bb7b7d2661aab3993870fca78c420d80","abstract_canon_sha256":"de0bcebca2a87255dcfef7f5533ad79685999585cf31a57ff359212454eb5959"},"schema_version":"1.0"},"canonical_sha256":"a95658932f6f9ee2d0db6bea7321c1c71826e3bb411c55a1db19227701c45158","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:03:33.409959Z","signature_b64":"7trjWTxlCnNDCdEdnrPq1XMKZyi9NBY/HqfKqfQxdFIPIu/zbAEYKbjkqJtSf44W119BDXpUGBSB3BSLG5XqCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a95658932f6f9ee2d0db6bea7321c1c71826e3bb411c55a1db19227701c45158","last_reissued_at":"2026-05-26T01:03:33.409214Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:03:33.409214Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.11182","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-26T01:03:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"bmUbiWMWORym1AYcXOhpSWNeRrCiMTMLmKI6PEw9CUXgm1vkDJnG2l/wBUucXI5wpr49R8vaW2ObYXLIKH6fCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-27T23:43:49.740194Z"},"content_sha256":"8c24232f1936cd0a61a688ce594a1b557c6279421730d638eb6f913249bcb7a5","schema_version":"1.0","event_id":"sha256:8c24232f1936cd0a61a688ce594a1b557c6279421730d638eb6f913249bcb7a5"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:VFLFREZPN6POFUG3NPVHGIOBY4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"The Many Faces of On-Policy Distillation: Pitfalls, Mechanisms, and Fixes","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"On-policy distillation fails in LLMs due to distribution mismatch, biased gradients, and privileged information aggregation but targeted fixes restore effectiveness.","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Ge Liu, Hongyu Lu, Siqi Zhu, Weiye Shi, Xuyan Ye","submitted_at":"2026-05-11T19:44:59Z","abstract_excerpt":"On-policy distillation (OPD) and on-policy self-distillation (OPSD) have emerged as promising post-training methods for large language models, offering dense token-level supervision on trajectories sampled from the model's own policy. However, existing results on their effectiveness remain mixed: while OP(S)D has shown promise in system prompt and knowledge internalization, recent studies also report instability and degradation. In this work, we present a comprehensive empirical study of when OPD and OPSD work, when they fail, and why. We find that OPD on mathematical reasoning is highly sensi"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We identify three failure mechanisms: (1) distribution mismatch between teacher and student caused by conditioning on student-generated prefixes, (2) optimization instability from biased TopK reverse-KL gradients, and (3) an OPSD-specific limitation where the student learns a PI-free policy that aggregates PI-conditioned teachers, which is insufficient when PI is instance-specific. We further show that stop-gradient TopK objectives, RLVR-adapted teachers, and SFT-stabilized students mitigate these failures.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The tested settings (mathematical reasoning trajectories and system-prompt/alignment PI) are representative enough that the three failure mechanisms and proposed fixes will apply to other LLM tasks, model scales, and data distributions without additional confounding factors.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"On-policy distillation for LLMs is sensitive to teacher choice and loss design, while self-distillation fails on instance-specific information but succeeds on shared rules, with stop-gradient TopK, adapted teachers, and SFT stabilization as mitigations.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"On-policy distillation fails in LLMs due to distribution mismatch, biased gradients, and privileged information aggregation but targeted fixes restore effectiveness.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"832de19b3cd4f2b63119cd12226e252ddd28bde55f16c215f062eb805aecbf04"},"source":{"id":"2605.11182","kind":"arxiv","version":2},"verdict":{"id":"25146744-6cee-46de-be8c-ce7581fd0189","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T02:13:55.930380Z","strongest_claim":"We identify three failure mechanisms: (1) distribution mismatch between teacher and student caused by conditioning on student-generated prefixes, (2) optimization instability from biased TopK reverse-KL gradients, and (3) an OPSD-specific limitation where the student learns a PI-free policy that aggregates PI-conditioned teachers, which is insufficient when PI is instance-specific. We further show that stop-gradient TopK objectives, RLVR-adapted teachers, and SFT-stabilized students mitigate these failures.","one_line_summary":"On-policy distillation for LLMs is sensitive to teacher choice and loss design, while self-distillation fails on instance-specific information but succeeds on shared rules, with stop-gradient TopK, adapted teachers, and SFT stabilization as mitigations.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The tested settings (mathematical reasoning trajectories and system-prompt/alignment PI) are representative enough that the three failure mechanisms and proposed fixes will apply to other LLM tasks, model scales, and data distributions without additional confounding factors.","pith_extraction_headline":"On-policy distillation fails in LLMs due to distribution mismatch, biased gradients, and privileged information aggregation but targeted fixes restore effectiveness."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.11182/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T04:42:00.971795Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:52.618077Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T10:01:17.612652Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T08:42:08.036066Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"e3f86565937dc8f02f8385eeaa5b3f6abf8d1dad995e7e7b0150e5577541d36e"},"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":"25146744-6cee-46de-be8c-ce7581fd0189"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-26T01:03:33Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Y9ospIFd1Kg1D5FNRl/6C0KYOgF7JSjzr1XC+t/WfHeH72cpS/yAJ+FSaqrqmBTI7C/jzNGYqdilDcj8gj2RBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-27T23:43:49.740700Z"},"content_sha256":"2ffdac2896e3e9d62710e13d0604c789fb6330719f9667fa9f4087ebc23b6da1","schema_version":"1.0","event_id":"sha256:2ffdac2896e3e9d62710e13d0604c789fb6330719f9667fa9f4087ebc23b6da1"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VFLFREZPN6POFUG3NPVHGIOBY4/bundle.json","state_url":"https://pith.science/pith/VFLFREZPN6POFUG3NPVHGIOBY4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VFLFREZPN6POFUG3NPVHGIOBY4/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":"2026-06-27T23:43:49Z","links":{"resolver":"https://pith.science/pith/VFLFREZPN6POFUG3NPVHGIOBY4","bundle":"https://pith.science/pith/VFLFREZPN6POFUG3NPVHGIOBY4/bundle.json","state":"https://pith.science/pith/VFLFREZPN6POFUG3NPVHGIOBY4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VFLFREZPN6POFUG3NPVHGIOBY4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:VFLFREZPN6POFUG3NPVHGIOBY4","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":"de0bcebca2a87255dcfef7f5533ad79685999585cf31a57ff359212454eb5959","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-11T19:44:59Z","title_canon_sha256":"b248cb51e667023b57534f1c43b260f0bb7b7d2661aab3993870fca78c420d80"},"schema_version":"1.0","source":{"id":"2605.11182","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.11182","created_at":"2026-05-26T01:03:33Z"},{"alias_kind":"arxiv_version","alias_value":"2605.11182v2","created_at":"2026-05-26T01:03:33Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.11182","created_at":"2026-05-26T01:03:33Z"},{"alias_kind":"pith_short_12","alias_value":"VFLFREZPN6PO","created_at":"2026-05-26T01:03:33Z"},{"alias_kind":"pith_short_16","alias_value":"VFLFREZPN6POFUG3","created_at":"2026-05-26T01:03:33Z"},{"alias_kind":"pith_short_8","alias_value":"VFLFREZP","created_at":"2026-05-26T01:03:33Z"}],"graph_snapshots":[{"event_id":"sha256:2ffdac2896e3e9d62710e13d0604c789fb6330719f9667fa9f4087ebc23b6da1","target":"graph","created_at":"2026-05-26T01:03:33Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"We identify three failure mechanisms: (1) distribution mismatch between teacher and student caused by conditioning on student-generated prefixes, (2) optimization instability from biased TopK reverse-KL gradients, and (3) an OPSD-specific limitation where the student learns a PI-free policy that aggregates PI-conditioned teachers, which is insufficient when PI is instance-specific. We further show that stop-gradient TopK objectives, RLVR-adapted teachers, and SFT-stabilized students mitigate these failures."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The tested settings (mathematical reasoning trajectories and system-prompt/alignment PI) are representative enough that the three failure mechanisms and proposed fixes will apply to other LLM tasks, model scales, and data distributions without additional confounding factors."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"On-policy distillation for LLMs is sensitive to teacher choice and loss design, while self-distillation fails on instance-specific information but succeeds on shared rules, with stop-gradient TopK, adapted teachers, and SFT stabilization as mitigations."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"On-policy distillation fails in LLMs due to distribution mismatch, biased gradients, and privileged information aggregation but targeted fixes restore effectiveness."}],"snapshot_sha256":"832de19b3cd4f2b63119cd12226e252ddd28bde55f16c215f062eb805aecbf04"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-20T04:42:00.971795Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T13:33:52.618077Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T10:01:17.612652Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T08:42:08.036066Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.11182/integrity.json","findings":[],"snapshot_sha256":"e3f86565937dc8f02f8385eeaa5b3f6abf8d1dad995e7e7b0150e5577541d36e","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"On-policy distillation (OPD) and on-policy self-distillation (OPSD) have emerged as promising post-training methods for large language models, offering dense token-level supervision on trajectories sampled from the model's own policy. However, existing results on their effectiveness remain mixed: while OP(S)D has shown promise in system prompt and knowledge internalization, recent studies also report instability and degradation. In this work, we present a comprehensive empirical study of when OPD and OPSD work, when they fail, and why. We find that OPD on mathematical reasoning is highly sensi","authors_text":"Ge Liu, Hongyu Lu, Siqi Zhu, Weiye Shi, Xuyan Ye","cross_cats":[],"headline":"On-policy distillation fails in LLMs due to distribution mismatch, biased gradients, and privileged information aggregation but targeted fixes restore effectiveness.","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-11T19:44:59Z","title":"The Many Faces of On-Policy Distillation: Pitfalls, Mechanisms, and Fixes"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.11182","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-13T02:13:55.930380Z","id":"25146744-6cee-46de-be8c-ce7581fd0189","model_set":{"reader":"grok-4.3"},"one_line_summary":"On-policy distillation for LLMs is sensitive to teacher choice and loss design, while self-distillation fails on instance-specific information but succeeds on shared rules, with stop-gradient TopK, adapted teachers, and SFT stabilization as mitigations.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"On-policy distillation fails in LLMs due to distribution mismatch, biased gradients, and privileged information aggregation but targeted fixes restore effectiveness.","strongest_claim":"We identify three failure mechanisms: (1) distribution mismatch between teacher and student caused by conditioning on student-generated prefixes, (2) optimization instability from biased TopK reverse-KL gradients, and (3) an OPSD-specific limitation where the student learns a PI-free policy that aggregates PI-conditioned teachers, which is insufficient when PI is instance-specific. We further show that stop-gradient TopK objectives, RLVR-adapted teachers, and SFT-stabilized students mitigate these failures.","weakest_assumption":"The tested settings (mathematical reasoning trajectories and system-prompt/alignment PI) are representative enough that the three failure mechanisms and proposed fixes will apply to other LLM tasks, model scales, and data distributions without additional confounding factors."}},"verdict_id":"25146744-6cee-46de-be8c-ce7581fd0189"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:8c24232f1936cd0a61a688ce594a1b557c6279421730d638eb6f913249bcb7a5","target":"record","created_at":"2026-05-26T01:03:33Z","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":"de0bcebca2a87255dcfef7f5533ad79685999585cf31a57ff359212454eb5959","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-11T19:44:59Z","title_canon_sha256":"b248cb51e667023b57534f1c43b260f0bb7b7d2661aab3993870fca78c420d80"},"schema_version":"1.0","source":{"id":"2605.11182","kind":"arxiv","version":2}},"canonical_sha256":"a95658932f6f9ee2d0db6bea7321c1c71826e3bb411c55a1db19227701c45158","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a95658932f6f9ee2d0db6bea7321c1c71826e3bb411c55a1db19227701c45158","first_computed_at":"2026-05-26T01:03:33.409214Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-26T01:03:33.409214Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7trjWTxlCnNDCdEdnrPq1XMKZyi9NBY/HqfKqfQxdFIPIu/zbAEYKbjkqJtSf44W119BDXpUGBSB3BSLG5XqCg==","signature_status":"signed_v1","signed_at":"2026-05-26T01:03:33.409959Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.11182","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8c24232f1936cd0a61a688ce594a1b557c6279421730d638eb6f913249bcb7a5","sha256:2ffdac2896e3e9d62710e13d0604c789fb6330719f9667fa9f4087ebc23b6da1"],"state_sha256":"24caa778e825bddd13579e9f4e73058ed521f74c4666bd5c6103c25ce8d2d6ed"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mrUuGAE4cmLXbffhyDq61m1s2wzeACKoylA4Fs6tSf7UPiLEplOXJz7awJ+bGWbEt9NWFUAo569YQz25ZYN9Aw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-27T23:43:49.743022Z","bundle_sha256":"9ed5a1cc932d5ba7a31c8968c6379bf08eb2845de651c715e654baf217e578d0"}}