{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:NGKYKIAK2YNQHDSTDVY4WVD2XY","short_pith_number":"pith:NGKYKIAK","canonical_record":{"source":{"id":"2605.11458","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-12T03:15:58Z","cross_cats_sorted":["cs.CL","cs.LO"],"title_canon_sha256":"c6d1d4231c0b2deef372a95b664109c6d744977b41c04ca0d03397984278f7c7","abstract_canon_sha256":"904a1023c98c8cc7d93e5da3ce5481f1f93879c80f6c2466377bc5f15cb4ccb9"},"schema_version":"1.0"},"canonical_sha256":"699585200ad61b038e531d71cb547abe2edccb32c3e145ebfe106541a40a44fd","source":{"kind":"arxiv","id":"2605.11458","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.11458","created_at":"2026-05-28T01:04:42Z"},{"alias_kind":"arxiv_version","alias_value":"2605.11458v2","created_at":"2026-05-28T01:04:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.11458","created_at":"2026-05-28T01:04:42Z"},{"alias_kind":"pith_short_12","alias_value":"NGKYKIAK2YNQ","created_at":"2026-05-28T01:04:42Z"},{"alias_kind":"pith_short_16","alias_value":"NGKYKIAK2YNQHDST","created_at":"2026-05-28T01:04:42Z"},{"alias_kind":"pith_short_8","alias_value":"NGKYKIAK","created_at":"2026-05-28T01:04:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:NGKYKIAK2YNQHDSTDVY4WVD2XY","target":"record","payload":{"canonical_record":{"source":{"id":"2605.11458","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-12T03:15:58Z","cross_cats_sorted":["cs.CL","cs.LO"],"title_canon_sha256":"c6d1d4231c0b2deef372a95b664109c6d744977b41c04ca0d03397984278f7c7","abstract_canon_sha256":"904a1023c98c8cc7d93e5da3ce5481f1f93879c80f6c2466377bc5f15cb4ccb9"},"schema_version":"1.0"},"canonical_sha256":"699585200ad61b038e531d71cb547abe2edccb32c3e145ebfe106541a40a44fd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-28T01:04:42.194831Z","signature_b64":"viS9WilWCU6GnQsrvK5PHm0wC86EvWPsFMj/vlyJqEYUHqMNZbZzCrE2Yt0oQkJhig/g6ag1l0Z541bpfeFCAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"699585200ad61b038e531d71cb547abe2edccb32c3e145ebfe106541a40a44fd","last_reissued_at":"2026-05-28T01:04:42.194431Z","signature_status":"signed_v1","first_computed_at":"2026-05-28T01:04:42.194431Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.11458","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":"2026-05-28T01:04:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zA4br1nRGgwcTBd+MhpBEsLGJfX22dQyD7g3ttiq4Vdlkz2lKwFC1B9Yc0gYjmU1cIt+/ar+CGM0DrmksuLZBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T03:52:52.319262Z"},"content_sha256":"4bc8ed7715949896419a8f5f7b81ed9199639e336f38d1775280e4f53cf7142c","schema_version":"1.0","event_id":"sha256:4bc8ed7715949896419a8f5f7b81ed9199639e336f38d1775280e4f53cf7142c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:NGKYKIAK2YNQHDSTDVY4WVD2XY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Adaptive Teacher Exposure for Self-Distillation in LLM Reasoning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Adaptive control of how much reference reasoning the teacher sees during self-distillation improves LLM performance on math tasks.","cross_cats":["cs.CL","cs.LO"],"primary_cat":"cs.AI","authors_text":"Huaibin Wang, Tiangang Zhang, Yilun Sun, Zihao Han","submitted_at":"2026-05-12T03:15:58Z","abstract_excerpt":"On-policy self-distillation has become a strong recipe for LLM reasoning, where a privileged teacher supervises the student's own rollouts while conditioning on the reference solution. A design choice shared by nearly all such methods, however, has gone unquestioned: the teacher always sees the full reference reasoning. We argue that this default itself is part of the problem and identify a teacher-side exposure mismatch: when the teacher conditions on reasoning far beyond the student's current competence, the resulting token targets become too strong to absorb. A controlled fixed-exposure swe"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Experiments on AIME 24, AIME 25, and HMMT 25 across Qwen3-{1.7B, 4B, 8B} show that ATESD consistently outperforms competitive self-distillation and RL baselines, improving over OPSD by +0.95, +2.05, and +2.33 Average@12 points respectively.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That a lightweight Beta-policy controller, optimized via a discounted learning-progress reward on compact training-state statistics, will reliably produce exposure decisions that improve long-term student performance without introducing training instability or benchmark-specific overfitting.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"ATESD makes teacher exposure to reference reasoning a learnable control variable via a Beta-policy optimized on future student improvement, yielding gains of up to +2.33 points over fixed-exposure self-distillation on AIME and HMMT math benchmarks.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Adaptive control of how much reference reasoning the teacher sees during self-distillation improves LLM performance on math tasks.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"247392f2a7a4b3ce7677691686c4452dd9cf055493a5c128ece74144e3dda313"},"source":{"id":"2605.11458","kind":"arxiv","version":2},"verdict":{"id":"280b6ff4-8e7c-40fd-a0ac-4edfcda05f18","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-13T01:58:33.413358Z","strongest_claim":"Experiments on AIME 24, AIME 25, and HMMT 25 across Qwen3-{1.7B, 4B, 8B} show that ATESD consistently outperforms competitive self-distillation and RL baselines, improving over OPSD by +0.95, +2.05, and +2.33 Average@12 points respectively.","one_line_summary":"ATESD makes teacher exposure to reference reasoning a learnable control variable via a Beta-policy optimized on future student improvement, yielding gains of up to +2.33 points over fixed-exposure self-distillation on AIME and HMMT math benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That a lightweight Beta-policy controller, optimized via a discounted learning-progress reward on compact training-state statistics, will reliably produce exposure decisions that improve long-term student performance without introducing training instability or benchmark-specific overfitting.","pith_extraction_headline":"Adaptive control of how much reference reasoning the teacher sees during self-distillation improves LLM performance on math tasks."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.11458/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T04:22:00.356800Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-19T12:35:50.141606Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T10:01:16.403452Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T08:27:17.389902Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"38adbe07a56e833b42a95bc4bdf342f3b25d1dbdf58aeae49e5fe179f68572f6"},"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":"280b6ff4-8e7c-40fd-a0ac-4edfcda05f18"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-28T01:04:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"J6O4O3ISQvVQ8wAR64f936DbxzP1TW6rbiRquQZzH6qz2hEf2AyC+GIJtNyBc5uoQKgMefLkijRZ4lTxRswvBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T03:52:52.319806Z"},"content_sha256":"4f937306cb9a4d474d9e7764fa30bab5f00173f5b2b059ebb1f28b7cb512aaa4","schema_version":"1.0","event_id":"sha256:4f937306cb9a4d474d9e7764fa30bab5f00173f5b2b059ebb1f28b7cb512aaa4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NGKYKIAK2YNQHDSTDVY4WVD2XY/bundle.json","state_url":"https://pith.science/pith/NGKYKIAK2YNQHDSTDVY4WVD2XY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NGKYKIAK2YNQHDSTDVY4WVD2XY/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":"2026-05-30T03:52:52Z","links":{"resolver":"https://pith.science/pith/NGKYKIAK2YNQHDSTDVY4WVD2XY","bundle":"https://pith.science/pith/NGKYKIAK2YNQHDSTDVY4WVD2XY/bundle.json","state":"https://pith.science/pith/NGKYKIAK2YNQHDSTDVY4WVD2XY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NGKYKIAK2YNQHDSTDVY4WVD2XY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:NGKYKIAK2YNQHDSTDVY4WVD2XY","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":"904a1023c98c8cc7d93e5da3ce5481f1f93879c80f6c2466377bc5f15cb4ccb9","cross_cats_sorted":["cs.CL","cs.LO"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-12T03:15:58Z","title_canon_sha256":"c6d1d4231c0b2deef372a95b664109c6d744977b41c04ca0d03397984278f7c7"},"schema_version":"1.0","source":{"id":"2605.11458","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.11458","created_at":"2026-05-28T01:04:42Z"},{"alias_kind":"arxiv_version","alias_value":"2605.11458v2","created_at":"2026-05-28T01:04:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.11458","created_at":"2026-05-28T01:04:42Z"},{"alias_kind":"pith_short_12","alias_value":"NGKYKIAK2YNQ","created_at":"2026-05-28T01:04:42Z"},{"alias_kind":"pith_short_16","alias_value":"NGKYKIAK2YNQHDST","created_at":"2026-05-28T01:04:42Z"},{"alias_kind":"pith_short_8","alias_value":"NGKYKIAK","created_at":"2026-05-28T01:04:42Z"}],"graph_snapshots":[{"event_id":"sha256:4f937306cb9a4d474d9e7764fa30bab5f00173f5b2b059ebb1f28b7cb512aaa4","target":"graph","created_at":"2026-05-28T01:04:42Z","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":"Experiments on AIME 24, AIME 25, and HMMT 25 across Qwen3-{1.7B, 4B, 8B} show that ATESD consistently outperforms competitive self-distillation and RL baselines, improving over OPSD by +0.95, +2.05, and +2.33 Average@12 points respectively."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That a lightweight Beta-policy controller, optimized via a discounted learning-progress reward on compact training-state statistics, will reliably produce exposure decisions that improve long-term student performance without introducing training instability or benchmark-specific overfitting."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"ATESD makes teacher exposure to reference reasoning a learnable control variable via a Beta-policy optimized on future student improvement, yielding gains of up to +2.33 points over fixed-exposure self-distillation on AIME and HMMT math benchmarks."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Adaptive control of how much reference reasoning the teacher sees during self-distillation improves LLM performance on math tasks."}],"snapshot_sha256":"247392f2a7a4b3ce7677691686c4452dd9cf055493a5c128ece74144e3dda313"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-20T04:22:00.356800Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-19T12:35:50.141606Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T10:01:16.403452Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T08:27:17.389902Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.11458/integrity.json","findings":[],"snapshot_sha256":"38adbe07a56e833b42a95bc4bdf342f3b25d1dbdf58aeae49e5fe179f68572f6","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"On-policy self-distillation has become a strong recipe for LLM reasoning, where a privileged teacher supervises the student's own rollouts while conditioning on the reference solution. A design choice shared by nearly all such methods, however, has gone unquestioned: the teacher always sees the full reference reasoning. We argue that this default itself is part of the problem and identify a teacher-side exposure mismatch: when the teacher conditions on reasoning far beyond the student's current competence, the resulting token targets become too strong to absorb. A controlled fixed-exposure swe","authors_text":"Huaibin Wang, Tiangang Zhang, Yilun Sun, Zihao Han","cross_cats":["cs.CL","cs.LO"],"headline":"Adaptive control of how much reference reasoning the teacher sees during self-distillation improves LLM performance on math tasks.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-12T03:15:58Z","title":"Adaptive Teacher Exposure for Self-Distillation in LLM Reasoning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.11458","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-13T01:58:33.413358Z","id":"280b6ff4-8e7c-40fd-a0ac-4edfcda05f18","model_set":{"reader":"grok-4.3"},"one_line_summary":"ATESD makes teacher exposure to reference reasoning a learnable control variable via a Beta-policy optimized on future student improvement, yielding gains of up to +2.33 points over fixed-exposure self-distillation on AIME and HMMT math benchmarks.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Adaptive control of how much reference reasoning the teacher sees during self-distillation improves LLM performance on math tasks.","strongest_claim":"Experiments on AIME 24, AIME 25, and HMMT 25 across Qwen3-{1.7B, 4B, 8B} show that ATESD consistently outperforms competitive self-distillation and RL baselines, improving over OPSD by +0.95, +2.05, and +2.33 Average@12 points respectively.","weakest_assumption":"That a lightweight Beta-policy controller, optimized via a discounted learning-progress reward on compact training-state statistics, will reliably produce exposure decisions that improve long-term student performance without introducing training instability or benchmark-specific overfitting."}},"verdict_id":"280b6ff4-8e7c-40fd-a0ac-4edfcda05f18"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:4bc8ed7715949896419a8f5f7b81ed9199639e336f38d1775280e4f53cf7142c","target":"record","created_at":"2026-05-28T01:04:42Z","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":"904a1023c98c8cc7d93e5da3ce5481f1f93879c80f6c2466377bc5f15cb4ccb9","cross_cats_sorted":["cs.CL","cs.LO"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-12T03:15:58Z","title_canon_sha256":"c6d1d4231c0b2deef372a95b664109c6d744977b41c04ca0d03397984278f7c7"},"schema_version":"1.0","source":{"id":"2605.11458","kind":"arxiv","version":2}},"canonical_sha256":"699585200ad61b038e531d71cb547abe2edccb32c3e145ebfe106541a40a44fd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"699585200ad61b038e531d71cb547abe2edccb32c3e145ebfe106541a40a44fd","first_computed_at":"2026-05-28T01:04:42.194431Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-28T01:04:42.194431Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"viS9WilWCU6GnQsrvK5PHm0wC86EvWPsFMj/vlyJqEYUHqMNZbZzCrE2Yt0oQkJhig/g6ag1l0Z541bpfeFCAA==","signature_status":"signed_v1","signed_at":"2026-05-28T01:04:42.194831Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.11458","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:4bc8ed7715949896419a8f5f7b81ed9199639e336f38d1775280e4f53cf7142c","sha256:4f937306cb9a4d474d9e7764fa30bab5f00173f5b2b059ebb1f28b7cb512aaa4"],"state_sha256":"360665985fda41b4c1c02550667d4a6e374ba6985716003067e3a89d9d224324"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"34aY5WgL+EH/w/pyHzXpZDO6fXA37PX9Uz8zYY0cOB/1tyhHXoTdCh/j8lj8gUdQGUo3tDJ8f40LmdZfKBjtBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-30T03:52:52.322276Z","bundle_sha256":"62ffa848488d0f48c8c5965a93fd8f1496d9ab0881f22d9c139fddc0335ad109"}}