{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:W5RZNYN2LB4FPCQ5CZWCQDGC2M","short_pith_number":"pith:W5RZNYN2","canonical_record":{"source":{"id":"2605.12741","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T20:46:05Z","cross_cats_sorted":[],"title_canon_sha256":"84b5e45e462f5ae833fc31dc3254bca3bc0371a78c94b09cb6a4ed5b273cd3ad","abstract_canon_sha256":"6f65a56df01cf0dd1fd513dcb7360816b6c921d45d91542a9c01ce95339c81d8"},"schema_version":"1.0"},"canonical_sha256":"b76396e1ba5878578a1d166c280cc2d335cf77b5ea15d9f651b4507273e81ef6","source":{"kind":"arxiv","id":"2605.12741","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.12741","created_at":"2026-05-18T03:09:49Z"},{"alias_kind":"arxiv_version","alias_value":"2605.12741v1","created_at":"2026-05-18T03:09:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12741","created_at":"2026-05-18T03:09:49Z"},{"alias_kind":"pith_short_12","alias_value":"W5RZNYN2LB4F","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"W5RZNYN2LB4FPCQ5","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"W5RZNYN2","created_at":"2026-05-18T12:33:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:W5RZNYN2LB4FPCQ5CZWCQDGC2M","target":"record","payload":{"canonical_record":{"source":{"id":"2605.12741","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T20:46:05Z","cross_cats_sorted":[],"title_canon_sha256":"84b5e45e462f5ae833fc31dc3254bca3bc0371a78c94b09cb6a4ed5b273cd3ad","abstract_canon_sha256":"6f65a56df01cf0dd1fd513dcb7360816b6c921d45d91542a9c01ce95339c81d8"},"schema_version":"1.0"},"canonical_sha256":"b76396e1ba5878578a1d166c280cc2d335cf77b5ea15d9f651b4507273e81ef6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T03:09:49.090701Z","signature_b64":"SCuf/4WrnQgRXKvDJi7t/bhZxi9QfZJBT0ghod6i61pI993pYpixhIETGha/aM/Ia/dRSPJQIrq5S0eyhKQRDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b76396e1ba5878578a1d166c280cc2d335cf77b5ea15d9f651b4507273e81ef6","last_reissued_at":"2026-05-18T03:09:49.089887Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T03:09:49.089887Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.12741","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T03:09:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hljpK019pkfPlWTmZe0RG2T2HDHJM2ZHe00jwiFgNqgcueN+pUJqRzCRCJUoPKzzgfxsyjdyiofF7Qh3CFJdCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-05T04:51:13.994787Z"},"content_sha256":"0ed73fb12e7cd0d66f5a33566940b3e9eb8b6f06d6455e3a10073de455166e9d","schema_version":"1.0","event_id":"sha256:0ed73fb12e7cd0d66f5a33566940b3e9eb8b6f06d6455e3a10073de455166e9d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:W5RZNYN2LB4FPCQ5CZWCQDGC2M","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Learning with Rare Success but Rich Feedback via Reflection-Enhanced Self-Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Reflection-Enhanced Self-Distillation lets models learn from failure feedback by creating diagnostic reflections and a reusable global playbook.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Bing Yin, Changlong Yu, Chengyu Dong, Haoran Liu, Ilgee Hong, Jingbo Shang, Qin Lu, Sha Li, Shuowei Jin, Xintong Li, Yuwei Zhang, Zhenyu Shi","submitted_at":"2026-05-12T20:46:05Z","abstract_excerpt":"Enabling Large Language Models (LLMs) to continuously improve from environmental interactions is a central challenge in post-training. While on-policy self-distillation offers a promising paradigm, existing methods predominantly treat environmental feedback as a passive conditioning signal. Consequently, they heavily rely on successful demonstrations and struggle to learn in rare-success regimes. To bridge this gap, we introduce Reflection-Enhanced Self-Distillation (RESD), a framework that transforms raw failure feedback into an active source of corrective supervision. Instead of passively ap"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"RESD substantially outperforms standard self-distillation baselines and achieves significantly faster early-stage improvement than GRPO with 8× samples using only a single rollout per prompt.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the model-generated retrospective reflections accurately diagnose local errors and that the curated global playbook preserves reusable lessons without introducing noise or compounding errors across training steps.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"RESD turns failure trajectories into token-level supervision via retrospective reflections and a persistent global playbook, enabling faster improvement than standard self-distillation or GRPO with only one rollout per prompt.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Reflection-Enhanced Self-Distillation lets models learn from failure feedback by creating diagnostic reflections and a reusable global playbook.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"8daeae4eeeb4395ca66bc16ddc1f418291aa8157d8709b88316242c5e4caca0a"},"source":{"id":"2605.12741","kind":"arxiv","version":1},"verdict":{"id":"1bc594d4-bc48-4675-a39f-5986b6c1d8ee","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T21:20:07.083787Z","strongest_claim":"RESD substantially outperforms standard self-distillation baselines and achieves significantly faster early-stage improvement than GRPO with 8× samples using only a single rollout per prompt.","one_line_summary":"RESD turns failure trajectories into token-level supervision via retrospective reflections and a persistent global playbook, enabling faster improvement than standard self-distillation or GRPO with only one rollout per prompt.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the model-generated retrospective reflections accurately diagnose local errors and that the curated global playbook preserves reusable lessons without introducing noise or compounding errors across training steps.","pith_extraction_headline":"Reflection-Enhanced Self-Distillation lets models learn from failure feedback by creating diagnostic reflections and a reusable global playbook."},"references":{"count":32,"sample":[{"doi":"","year":2023,"title":"GKD: Generalized knowledge distillation for auto-regressive se- quence models.arXiv preprint arXiv:2306.13649","work_id":"78b7a553-fd43-4995-b559-e95779797d3f","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"On-policy distillation of language models: Learning from self-generated mistakes","work_id":"3733ff2d-cd95-4776-9fa9-1b2328326749","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Retaining by doing: The role of on-policy data in mitigating forgetting, 2025","work_id":"4d844db6-2d57-4fe8-ace2-574a47636b4a","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2024,"title":"Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models","work_id":"6b7f0773-4e99-4274-9d8e-279a1f25c5e1","ref_index":4,"cited_arxiv_id":"2401.01335","is_internal_anchor":true},{"doi":"","year":2026,"title":"Deepseek-v4: Towards highly efficient million-token context intelligence","work_id":"b5a5843d-47c0-432e-88d5-c9eaac555a7f","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":32,"snapshot_sha256":"baac3e7471575d1284ac586b57d1d22460486b0c63fd6a5bc820423b0f2c6c1c","internal_anchors":13},"formal_canon":{"evidence_count":1,"snapshot_sha256":"87f5a5f78cca71786cd7005c46a6a050286f8ebb09ae3b3d620050bf441c370b"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"1bc594d4-bc48-4675-a39f-5986b6c1d8ee"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T03:09:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"O4Yv/Tg0N7TPpwqwcEOTTloYKEJXHuMIBAelo5AABUjHFi+HNUTTGpShrLe/8cXDN6XibhkwK5xD/Ta7fm41Cg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-05T04:51:13.995685Z"},"content_sha256":"f4a2a8db6022d03ecbe1aa6e11f2a60ba507ed075b493c90c647bdd7b0980ca8","schema_version":"1.0","event_id":"sha256:f4a2a8db6022d03ecbe1aa6e11f2a60ba507ed075b493c90c647bdd7b0980ca8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/W5RZNYN2LB4FPCQ5CZWCQDGC2M/bundle.json","state_url":"https://pith.science/pith/W5RZNYN2LB4FPCQ5CZWCQDGC2M/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/W5RZNYN2LB4FPCQ5CZWCQDGC2M/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":"2026-06-05T04:51:13Z","links":{"resolver":"https://pith.science/pith/W5RZNYN2LB4FPCQ5CZWCQDGC2M","bundle":"https://pith.science/pith/W5RZNYN2LB4FPCQ5CZWCQDGC2M/bundle.json","state":"https://pith.science/pith/W5RZNYN2LB4FPCQ5CZWCQDGC2M/state.json","well_known_bundle":"https://pith.science/.well-known/pith/W5RZNYN2LB4FPCQ5CZWCQDGC2M/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:W5RZNYN2LB4FPCQ5CZWCQDGC2M","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":"6f65a56df01cf0dd1fd513dcb7360816b6c921d45d91542a9c01ce95339c81d8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T20:46:05Z","title_canon_sha256":"84b5e45e462f5ae833fc31dc3254bca3bc0371a78c94b09cb6a4ed5b273cd3ad"},"schema_version":"1.0","source":{"id":"2605.12741","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.12741","created_at":"2026-05-18T03:09:49Z"},{"alias_kind":"arxiv_version","alias_value":"2605.12741v1","created_at":"2026-05-18T03:09:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12741","created_at":"2026-05-18T03:09:49Z"},{"alias_kind":"pith_short_12","alias_value":"W5RZNYN2LB4F","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"W5RZNYN2LB4FPCQ5","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"W5RZNYN2","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:f4a2a8db6022d03ecbe1aa6e11f2a60ba507ed075b493c90c647bdd7b0980ca8","target":"graph","created_at":"2026-05-18T03:09:49Z","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":"RESD substantially outperforms standard self-distillation baselines and achieves significantly faster early-stage improvement than GRPO with 8× samples using only a single rollout per prompt."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the model-generated retrospective reflections accurately diagnose local errors and that the curated global playbook preserves reusable lessons without introducing noise or compounding errors across training steps."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"RESD turns failure trajectories into token-level supervision via retrospective reflections and a persistent global playbook, enabling faster improvement than standard self-distillation or GRPO with only one rollout per prompt."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Reflection-Enhanced Self-Distillation lets models learn from failure feedback by creating diagnostic reflections and a reusable global playbook."}],"snapshot_sha256":"8daeae4eeeb4395ca66bc16ddc1f418291aa8157d8709b88316242c5e4caca0a"},"formal_canon":{"evidence_count":1,"snapshot_sha256":"87f5a5f78cca71786cd7005c46a6a050286f8ebb09ae3b3d620050bf441c370b"},"paper":{"abstract_excerpt":"Enabling Large Language Models (LLMs) to continuously improve from environmental interactions is a central challenge in post-training. While on-policy self-distillation offers a promising paradigm, existing methods predominantly treat environmental feedback as a passive conditioning signal. Consequently, they heavily rely on successful demonstrations and struggle to learn in rare-success regimes. To bridge this gap, we introduce Reflection-Enhanced Self-Distillation (RESD), a framework that transforms raw failure feedback into an active source of corrective supervision. Instead of passively ap","authors_text":"Bing Yin, Changlong Yu, Chengyu Dong, Haoran Liu, Ilgee Hong, Jingbo Shang, Qin Lu, Sha Li, Shuowei Jin, Xintong Li, Yuwei Zhang, Zhenyu Shi","cross_cats":[],"headline":"Reflection-Enhanced Self-Distillation lets models learn from failure feedback by creating diagnostic reflections and a reusable global playbook.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T20:46:05Z","title":"Learning with Rare Success but Rich Feedback via Reflection-Enhanced Self-Distillation"},"references":{"count":32,"internal_anchors":13,"resolved_work":32,"sample":[{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":1,"title":"GKD: Generalized knowledge distillation for auto-regressive se- quence models.arXiv preprint arXiv:2306.13649","work_id":"78b7a553-fd43-4995-b559-e95779797d3f","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"On-policy distillation of language models: Learning from self-generated mistakes","work_id":"3733ff2d-cd95-4776-9fa9-1b2328326749","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Retaining by doing: The role of on-policy data in mitigating forgetting, 2025","work_id":"4d844db6-2d57-4fe8-ace2-574a47636b4a","year":2025},{"cited_arxiv_id":"2401.01335","doi":"","is_internal_anchor":true,"ref_index":4,"title":"Self-Play Fine-Tuning Converts Weak Language Models to Strong Language Models","work_id":"6b7f0773-4e99-4274-9d8e-279a1f25c5e1","year":2024},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Deepseek-v4: Towards highly efficient million-token context intelligence","work_id":"b5a5843d-47c0-432e-88d5-c9eaac555a7f","year":2026}],"snapshot_sha256":"baac3e7471575d1284ac586b57d1d22460486b0c63fd6a5bc820423b0f2c6c1c"},"source":{"id":"2605.12741","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T21:20:07.083787Z","id":"1bc594d4-bc48-4675-a39f-5986b6c1d8ee","model_set":{"reader":"grok-4.3"},"one_line_summary":"RESD turns failure trajectories into token-level supervision via retrospective reflections and a persistent global playbook, enabling faster improvement than standard self-distillation or GRPO with only one rollout per prompt.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Reflection-Enhanced Self-Distillation lets models learn from failure feedback by creating diagnostic reflections and a reusable global playbook.","strongest_claim":"RESD substantially outperforms standard self-distillation baselines and achieves significantly faster early-stage improvement than GRPO with 8× samples using only a single rollout per prompt.","weakest_assumption":"That the model-generated retrospective reflections accurately diagnose local errors and that the curated global playbook preserves reusable lessons without introducing noise or compounding errors across training steps."}},"verdict_id":"1bc594d4-bc48-4675-a39f-5986b6c1d8ee"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:0ed73fb12e7cd0d66f5a33566940b3e9eb8b6f06d6455e3a10073de455166e9d","target":"record","created_at":"2026-05-18T03:09:49Z","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":"6f65a56df01cf0dd1fd513dcb7360816b6c921d45d91542a9c01ce95339c81d8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-12T20:46:05Z","title_canon_sha256":"84b5e45e462f5ae833fc31dc3254bca3bc0371a78c94b09cb6a4ed5b273cd3ad"},"schema_version":"1.0","source":{"id":"2605.12741","kind":"arxiv","version":1}},"canonical_sha256":"b76396e1ba5878578a1d166c280cc2d335cf77b5ea15d9f651b4507273e81ef6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b76396e1ba5878578a1d166c280cc2d335cf77b5ea15d9f651b4507273e81ef6","first_computed_at":"2026-05-18T03:09:49.089887Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T03:09:49.089887Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"SCuf/4WrnQgRXKvDJi7t/bhZxi9QfZJBT0ghod6i61pI993pYpixhIETGha/aM/Ia/dRSPJQIrq5S0eyhKQRDw==","signature_status":"signed_v1","signed_at":"2026-05-18T03:09:49.090701Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.12741","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0ed73fb12e7cd0d66f5a33566940b3e9eb8b6f06d6455e3a10073de455166e9d","sha256:f4a2a8db6022d03ecbe1aa6e11f2a60ba507ed075b493c90c647bdd7b0980ca8"],"state_sha256":"e7b47b4d11abcdfceec3c13b05f330fd846139d9b44a532cbc9b650aa630f0d8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+WnIUsM4Mssy4pXFAFBRy9Io9SwNzIAfB+8yqxUcoXH99ZA+TijX7bzW2LL3na0r4EJvaqELzoIfuAVm7iGQCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-05T04:51:13.999276Z","bundle_sha256":"189fffec5e395e24670b18ec092dd672a620b67509956d41e32b30ab066c854d"}}