{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:2QE6M6KHAOIPRHZWANQJSVLOZR","short_pith_number":"pith:2QE6M6KH","canonical_record":{"source":{"id":"2605.09253","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-10T01:41:43Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f0fe8d86a87a7b9d7c45fa7b0042d6ff0e5c4180c3fceaf30d14b62f7deef40f","abstract_canon_sha256":"92339fc440ab4d2fbc71c519d06d3f52f4852462f92dd4ccd22dfb127e656fdb"},"schema_version":"1.0"},"canonical_sha256":"d409e679470390f89f36036099556ecc5f6b72f6c63d0ddf0907eb3bcb265a29","source":{"kind":"arxiv","id":"2605.09253","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.09253","created_at":"2026-06-02T01:03:48Z"},{"alias_kind":"arxiv_version","alias_value":"2605.09253v2","created_at":"2026-06-02T01:03:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.09253","created_at":"2026-06-02T01:03:48Z"},{"alias_kind":"pith_short_12","alias_value":"2QE6M6KHAOIP","created_at":"2026-06-02T01:03:48Z"},{"alias_kind":"pith_short_16","alias_value":"2QE6M6KHAOIPRHZW","created_at":"2026-06-02T01:03:48Z"},{"alias_kind":"pith_short_8","alias_value":"2QE6M6KH","created_at":"2026-06-02T01:03:48Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:2QE6M6KHAOIPRHZWANQJSVLOZR","target":"record","payload":{"canonical_record":{"source":{"id":"2605.09253","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-10T01:41:43Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f0fe8d86a87a7b9d7c45fa7b0042d6ff0e5c4180c3fceaf30d14b62f7deef40f","abstract_canon_sha256":"92339fc440ab4d2fbc71c519d06d3f52f4852462f92dd4ccd22dfb127e656fdb"},"schema_version":"1.0"},"canonical_sha256":"d409e679470390f89f36036099556ecc5f6b72f6c63d0ddf0907eb3bcb265a29","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:03:48.286723Z","signature_b64":"aHoR6Q0KHR9wGiy+1n3CJC8ML04Ta6Izdoqy0UYSqlnaCs87GoXWMCnwy8zJbCHH3GKt28eQ/sTr2eK863KFCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d409e679470390f89f36036099556ecc5f6b72f6c63d0ddf0907eb3bcb265a29","last_reissued_at":"2026-06-02T01:03:48.286268Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:03:48.286268Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.09253","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":"2026-06-02T01:03:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HoR87s6aF7nsYtAA/c8pQo/54q/LGDC5kx/o7EK/TW7ifE+sI8LjduRmY++Qd1eENNu7iips+izHYZhQ2G1+Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-29T13:08:53.794269Z"},"content_sha256":"97bf1f60fbf4ca2da4db916da75d7277c20d5db5443dba5c8a13ea0c92dc71d2","schema_version":"1.0","event_id":"sha256:97bf1f60fbf4ca2da4db916da75d7277c20d5db5443dba5c8a13ea0c92dc71d2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:2QE6M6KHAOIPRHZWANQJSVLOZR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Cornerstones or Stumbling Blocks? Deciphering the Rock Tokens in On-Policy Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"High-loss Rock Tokens in on-policy distillation resist training yet add almost nothing to reasoning performance.","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Dawei Li, Runchao Li, Shubhashis Roy Dipta, Yuxuan Jiang, Zhao Yang","submitted_at":"2026-05-10T01:41:43Z","abstract_excerpt":"While recent work in Reinforcement Learning with Verifiable Rewards (RLVR) has shown that a small subset of critical tokens disproportionately drives reasoning gains, an analogous token-level understanding of On-Policy Distillation (OPD) remains largely unexplored. In this work, we investigate high-loss tokens, a token type that--as the most direct signal of student-teacher mismatch under OPD's per-token KL objective--should progressively diminish as training converges according to existing studies; however, our empirical analysis shows otherwise. Even after OPD training reaches apparent satur"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"strategically bypassing these ``stumbling blocks'' can significantly streamline the alignment process, challenging the necessity of uniform token weighting and offering a more efficient paradigm for large-scale model distillation.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The causal interventions used to measure negligible functional contribution to reasoning performance are valid and complete, and that high-loss tokens identified as Rock Tokens truly have no downstream effect on model outputs.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Persistent 'Rock Tokens' in on-policy distillation resist teacher corrections, consume large gradient norms, yet add negligible value to reasoning, allowing targeted bypassing to streamline alignment.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"High-loss Rock Tokens in on-policy distillation resist training yet add almost nothing to reasoning performance.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"3cd6937433c316d9573fe2a6908f813c086f6d98025b071b8a2ffe296cb237b3"},"source":{"id":"2605.09253","kind":"arxiv","version":2},"verdict":{"id":"6fac69b6-7c6a-4e25-97bd-69f1b378a820","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T02:28:31.072089Z","strongest_claim":"strategically bypassing these ``stumbling blocks'' can significantly streamline the alignment process, challenging the necessity of uniform token weighting and offering a more efficient paradigm for large-scale model distillation.","one_line_summary":"Persistent 'Rock Tokens' in on-policy distillation resist teacher corrections, consume large gradient norms, yet add negligible value to reasoning, allowing targeted bypassing to streamline alignment.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The causal interventions used to measure negligible functional contribution to reasoning performance are valid and complete, and that high-loss tokens identified as Rock Tokens truly have no downstream effect on model outputs.","pith_extraction_headline":"High-loss Rock Tokens in on-policy distillation resist training yet add almost nothing to reasoning performance."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.09253/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T08:02:09.161304Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T20:34:42.788462Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T13:31:17.954079Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T10:22:49.473097Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"66825782fb90cc711e333e9db65763a0f205c05bf41fbf2f753261da0689952c"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":2,"snapshot_sha256":"2509bd32ac75aa3c878a3682c0a3aef683f04124243d8863a4849671392be29f"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"6fac69b6-7c6a-4e25-97bd-69f1b378a820"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-02T01:03:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OmUHfmb9lGzD3WOhplCQHQHBmjTG3aVdmz0uFg7wd9ixJmxmhswNU1BfNeKkqlDtLNbGr7DXRFMBxQ+D16WwCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-29T13:08:53.794929Z"},"content_sha256":"fe06d6aad379e4732f42cf9f666ceb97f1a5127bddb45993c9025fa43bc1a870","schema_version":"1.0","event_id":"sha256:fe06d6aad379e4732f42cf9f666ceb97f1a5127bddb45993c9025fa43bc1a870"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/2QE6M6KHAOIPRHZWANQJSVLOZR/bundle.json","state_url":"https://pith.science/pith/2QE6M6KHAOIPRHZWANQJSVLOZR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/2QE6M6KHAOIPRHZWANQJSVLOZR/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":"2026-06-29T13:08:53Z","links":{"resolver":"https://pith.science/pith/2QE6M6KHAOIPRHZWANQJSVLOZR","bundle":"https://pith.science/pith/2QE6M6KHAOIPRHZWANQJSVLOZR/bundle.json","state":"https://pith.science/pith/2QE6M6KHAOIPRHZWANQJSVLOZR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/2QE6M6KHAOIPRHZWANQJSVLOZR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:2QE6M6KHAOIPRHZWANQJSVLOZR","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":"92339fc440ab4d2fbc71c519d06d3f52f4852462f92dd4ccd22dfb127e656fdb","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-10T01:41:43Z","title_canon_sha256":"f0fe8d86a87a7b9d7c45fa7b0042d6ff0e5c4180c3fceaf30d14b62f7deef40f"},"schema_version":"1.0","source":{"id":"2605.09253","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.09253","created_at":"2026-06-02T01:03:48Z"},{"alias_kind":"arxiv_version","alias_value":"2605.09253v2","created_at":"2026-06-02T01:03:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.09253","created_at":"2026-06-02T01:03:48Z"},{"alias_kind":"pith_short_12","alias_value":"2QE6M6KHAOIP","created_at":"2026-06-02T01:03:48Z"},{"alias_kind":"pith_short_16","alias_value":"2QE6M6KHAOIPRHZW","created_at":"2026-06-02T01:03:48Z"},{"alias_kind":"pith_short_8","alias_value":"2QE6M6KH","created_at":"2026-06-02T01:03:48Z"}],"graph_snapshots":[{"event_id":"sha256:fe06d6aad379e4732f42cf9f666ceb97f1a5127bddb45993c9025fa43bc1a870","target":"graph","created_at":"2026-06-02T01:03:48Z","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":"strategically bypassing these ``stumbling blocks'' can significantly streamline the alignment process, challenging the necessity of uniform token weighting and offering a more efficient paradigm for large-scale model distillation."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The causal interventions used to measure negligible functional contribution to reasoning performance are valid and complete, and that high-loss tokens identified as Rock Tokens truly have no downstream effect on model outputs."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Persistent 'Rock Tokens' in on-policy distillation resist teacher corrections, consume large gradient norms, yet add negligible value to reasoning, allowing targeted bypassing to streamline alignment."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"High-loss Rock Tokens in on-policy distillation resist training yet add almost nothing to reasoning performance."}],"snapshot_sha256":"3cd6937433c316d9573fe2a6908f813c086f6d98025b071b8a2ffe296cb237b3"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"2509bd32ac75aa3c878a3682c0a3aef683f04124243d8863a4849671392be29f"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-20T08:02:09.161304Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T20:34:42.788462Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T13:31:17.954079Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T10:22:49.473097Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.09253/integrity.json","findings":[],"snapshot_sha256":"66825782fb90cc711e333e9db65763a0f205c05bf41fbf2f753261da0689952c","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"While recent work in Reinforcement Learning with Verifiable Rewards (RLVR) has shown that a small subset of critical tokens disproportionately drives reasoning gains, an analogous token-level understanding of On-Policy Distillation (OPD) remains largely unexplored. In this work, we investigate high-loss tokens, a token type that--as the most direct signal of student-teacher mismatch under OPD's per-token KL objective--should progressively diminish as training converges according to existing studies; however, our empirical analysis shows otherwise. Even after OPD training reaches apparent satur","authors_text":"Dawei Li, Runchao Li, Shubhashis Roy Dipta, Yuxuan Jiang, Zhao Yang","cross_cats":["cs.AI"],"headline":"High-loss Rock Tokens in on-policy distillation resist training yet add almost nothing to reasoning performance.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-10T01:41:43Z","title":"Cornerstones or Stumbling Blocks? Deciphering the Rock Tokens in On-Policy Distillation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.09253","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-12T02:28:31.072089Z","id":"6fac69b6-7c6a-4e25-97bd-69f1b378a820","model_set":{"reader":"grok-4.3"},"one_line_summary":"Persistent 'Rock Tokens' in on-policy distillation resist teacher corrections, consume large gradient norms, yet add negligible value to reasoning, allowing targeted bypassing to streamline alignment.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"High-loss Rock Tokens in on-policy distillation resist training yet add almost nothing to reasoning performance.","strongest_claim":"strategically bypassing these ``stumbling blocks'' can significantly streamline the alignment process, challenging the necessity of uniform token weighting and offering a more efficient paradigm for large-scale model distillation.","weakest_assumption":"The causal interventions used to measure negligible functional contribution to reasoning performance are valid and complete, and that high-loss tokens identified as Rock Tokens truly have no downstream effect on model outputs."}},"verdict_id":"6fac69b6-7c6a-4e25-97bd-69f1b378a820"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:97bf1f60fbf4ca2da4db916da75d7277c20d5db5443dba5c8a13ea0c92dc71d2","target":"record","created_at":"2026-06-02T01:03:48Z","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":"92339fc440ab4d2fbc71c519d06d3f52f4852462f92dd4ccd22dfb127e656fdb","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-10T01:41:43Z","title_canon_sha256":"f0fe8d86a87a7b9d7c45fa7b0042d6ff0e5c4180c3fceaf30d14b62f7deef40f"},"schema_version":"1.0","source":{"id":"2605.09253","kind":"arxiv","version":2}},"canonical_sha256":"d409e679470390f89f36036099556ecc5f6b72f6c63d0ddf0907eb3bcb265a29","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d409e679470390f89f36036099556ecc5f6b72f6c63d0ddf0907eb3bcb265a29","first_computed_at":"2026-06-02T01:03:48.286268Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-02T01:03:48.286268Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"aHoR6Q0KHR9wGiy+1n3CJC8ML04Ta6Izdoqy0UYSqlnaCs87GoXWMCnwy8zJbCHH3GKt28eQ/sTr2eK863KFCQ==","signature_status":"signed_v1","signed_at":"2026-06-02T01:03:48.286723Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.09253","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:97bf1f60fbf4ca2da4db916da75d7277c20d5db5443dba5c8a13ea0c92dc71d2","sha256:fe06d6aad379e4732f42cf9f666ceb97f1a5127bddb45993c9025fa43bc1a870"],"state_sha256":"0739d46f768622d5c240bccce6f3b73be0303fda9048aa6310542200a7cdf4c9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HaTP5dTFPL4FmuoS94R6FLbfImgJJFpyUN7hJ656ADv23FTW6ETATaTMec2NtZl4BDz2bzhISN/API+68apGAQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-29T13:08:53.797193Z","bundle_sha256":"e6c2cd913af114f37e62904a6ddb1f81ac32ceea1972df90aebdbac31dabf372"}}