{"_canonicalization":{"envelope_id":"axm_ + sha256(envelope minus {signature, axiom_id, anchors})","envelope_signature":"ed25519(envelope minus {signature, axiom_id})","json":"sort_keys=True, separators=(',',':'), ensure_ascii=False, allow_nan=False, utf-8","leaf_hash":"sha256(0x00 || canonical_json(envelope_full))","seal_signature":"ed25519(seal minus {signature, sig_algorithm})"},"axiom_id":"axm_fe942729ed803eadbe4e17bd2eba54f0c8f896444535286ba2f4d5259c47d7fd","bitcoin_anchor":{"bitcoin_attestations":[],"calendar_attestations":[],"ots_url":"","stamped_at":"","status":"pending_next_stamp"},"envelope":{"anchors":[{"chain":"crovia.axiom_graph","height":0,"merkle_proof":"spider_vendor_press_v1","root_at_anchor":"spider_vendor_press_v1"}],"axiom_id":"axm_fe942729ed803eadbe4e17bd2eba54f0c8f896444535286ba2f4d5259c47d7fd","axiom_type":"AX.OBS","body":{"axiom_subtype":"news.vendor_press.v1","category":"news","fingerprint":"c2c9ad7d7a0a7720bca932b63e7d66c426cf352f1a722d65add968a5840cf134","published":"Fri, 15 May 2026 00:00:00 -0400","receipt_hash":"c2c9ad7d7a0a7720bca932b63e7d66c426cf352f1a722d65add968a5840cf134","schema":"spider.news.vendor_press.v1","spider":"vendor_press","spider_record":{"axiom_subtype":"news.vendor_press.v1","category":"news","decision_hint":"POSITIVE","envelope_target":"AX.OBS","fingerprint":"c2c9ad7d7a0a7720bca932b63e7d66c426cf352f1a722d65add968a5840cf134","observed_at":"2026-05-15T04:43:17.611638Z","parent_run_hash":"5c64f85625fabd323e9c4a1cf068c012fb88a248deda9a9ac702fb2f9799f2e5","published":"Fri, 15 May 2026 00:00:00 -0400","runtime_version":"0.1.0","schema":"spider.news.vendor_press.v1","source_status":200,"source_url":"https://export.arxiv.org/rss/cs.AI","spider":"vendor_press","summary_excerpt":"arXiv:2605.12483v2 Announce Type: replace-cross \nAbstract: When labeled verifiable training data is scarce, each checked example should be used where it has the most value. A common approach is to train the deployment student model directly with sparse RL methods such as GRPO. We argue that this is often inefficient. Sparse sequence-level reward is most useful for strong models that can explore and discover better behavior, while dense token-level teacher supervision is better suited for compressing that behavior into a smaller student. This suggests a simple allocation rule: use scarce labeled data upstream to improve the strongest available teacher, then transfer the improved behavior downstream through dense supervision. In this view, GRPO-style sparse RL and OPD-style distillation are not competing methods, but two reward-density regimes used at different stages. We evaluate this rule on verifiable math tasks with Qwen3 and Llama models. For a fixed Qwen3-1.7B deployment student, d","title":"Beyond GRPO and On-Policy Distillation: An Empirical Sparse-to-Dense Reward Principle for Language-Model Post-Training","url":"https://arxiv.org/abs/2605.12483","vendor":"arxiv_cs_ai"},"summary":"arXiv:2605.12483v2 Announce Type: replace-cross \nAbstract: When labeled verifiable training data is scarce, each checked example should be used where it has the most value. A common approach is to train the deployment student model directly with sparse RL methods such as GRPO. We argue that this is often inefficient. Sparse sequence-level reward is most useful for strong models that can explore and discover better behavior, while dense token-level teacher supervision is better suited for compressing that behavior into a smaller student. This suggests a simple allocation rule: use scarce labeled data upstream to improve the strongest available teacher, then transfer the improved behavior downstream through dense supervision. In this view, GRPO-style sparse RL and OPD-style distillation are not competing methods, but two reward-density regimes used at different stages. We evaluate this rule on verifiable math tasks with Qwen3 and Llama models. For a fixed Qwen3-1.7B deployment student, d","title":"Beyond GRPO and On-Policy Distillation: An Empirical Sparse-to-Dense Reward Principle for Language-Model Post-Training","vendor":"arxiv_cs_ai"},"confidence":{"method":"deterministic"},"decision":"POSITIVE","issued_at":"2026-05-15T04:43:17Z","notes":"Spider vendor_press (news) news.vendor_press.v1","object":{"captured_by":"crovia.spider.vendor_press","primary_source_url":"https://arxiv.org/abs/2605.12483"},"predecessors":[],"schema":"crovia.axiom.v1","signature":"ed25519:3a8264df98075ccd8a573ed0bafb7dfc2b8556fddd6d0ca12fed1fc06e8deb67cce5feaef025f21fdf72d19818df02160c94642f5b6dc1266758f4180bd70b0c","signer":"crovia.substrate","subject":{"observed_at":"2026-05-15T04:43:17Z","source_collector":"spider:vendor_press","target_id":"https://arxiv.org/abs/2605.12483"},"tsa":{"authority":"crovia.substrate.bootstrap","rfc3161_token":"{\"kind\":\"crovia.bootstrap.tsa\",\"source_jsonl\":\"/opt/crovia/spider/data/news/vendor_press_v1.jsonl\",\"source_seal_merkle_root\":\"spider_vendor_press_v1\",\"upgrade_path\":\"Sessione H \\u2014 OpenTimestamps weekly anchor\"}"},"zk_mode":"clear","zk_proof":null},"ledger":{"leaf_hash":"d95c55ac89d65e10b5079a269a945368cb950e4cdf9f29ef84f674125ca69a10","leaf_index":134824,"ledger_path":"/opt/crovia/substrate/axiom_ledger.jsonl"},"merkle_proof":{"hash_alg":"sha256","leaf_prefix":"0x00","node_prefix":"0x01","odd_leaf_rule":"duplicate_last","path":[{"sibling":"e8e6d364d53d6059d8f1eca2e8ce9f52da498e4a250830d38d18c5caf2eb992d","side":"right"},{"sibling":"37ae0b5f404660ef98a3c218f5afbadcea9d62c938bee7f56260bbc046adc3ea","side":"right"},{"sibling":"3c232f75fed9014a0c0ef168724a72a48d746a08da01050c4656cbcbbcb5a3e5","side":"right"},{"sibling":"859c8ad567383b60cabd7653e93757b189d8a387fc1f47ec79e721b211448593","side":"left"},{"sibling":"b0a9ed28676718773b429698350751b67059b6659614997a62c21816d017dfdb","side":"right"},{"sibling":"6f9b6af9807c5aa1d9839cdea9eaebc45e0476787d000a69e7feaa627336733a","side":"left"},{"sibling":"7fb440277b43a0a448cbfa0e516ee5aa0504495adea486e423b8bbf045b1f798","side":"right"},{"sibling":"c007bf5d6c3a3a484205f30ad38ce03dff757769991af941772ae831255ede50","side":"left"},{"sibling":"a17a0eaa87574a308e2c02cf125072b9e69566b9e8d6d2e64a02880192b885d5","side":"right"},{"sibling":"6bd0475fd3a73322a4f73095c96b072de89e462ce5839161aa552e0208619bce","side":"left"},{"sibling":"36672459e5ed50c64ee1842b69cb6d2eb682c2a04844555be8d124257571994a","side":"left"},{"sibling":"727783827652adfa99c455bd80a01bfb33836228e51068b4f654ef3da468ca69","side":"left"},{"sibling":"623194cd30880ed223e306737fdb111aa0d781751bfc47553c404a6af6aad2c4","side":"right"},{"sibling":"fc8f53ed42756907fb79ee19a4ed09f72c560e5302b3d98198b96bf1da635a4a","side":"right"},{"sibling":"d6607539da7ba39ec68be2d12f27ed6768766c745e3120fd915f88c5e288e07c","side":"right"},{"sibling":"b63408a424d27cd6a75e0fb155e69a58a328f41e9cb9dba1eddef9a5289cc7fd","side":"right"},{"sibling":"356fb36a4e188f03d7a05c54cd8789bdd40eda454b9bc9560f667acc08e6c4e0","side":"right"},{"sibling":"d841ad93efda0869e5eb97678f348f03f5caab4353e05ff4bf18f47fb945b822","side":"left"}]},"schema":"crovia.axiom_proof.v1","seal":{"first_collector_run_id":"","first_receipt_hash":"","jsonl_path":"/opt/crovia/substrate/axiom_ledger.jsonl","key_id":"430895f101d38164","last_collector_run_id":"","last_receipt_hash":"","leaf_count":134886,"merkle_root":"c6c7ae28c065bced89e7f844216b073f1a7cc4b378db0d41a98bcd21b28066db","public_key_hex":"cf742e26f75669dc673cb5c0786a1ae23ae8ca19c347317192ce40c28a7ff25c","run_id":"hourly_json_retrofit_20260515T053701Z","schema":"crovia.seal.v1","seal_family_version":"crovia-seal-family/1","seal_kind":"substrate_batch","sealed_at":"2026-05-15T05:37:26Z","sig_algorithm":"ed25519","signature":"5a3978c26017daf4104adbb3e1c3099c5750157acbfc7242ece1815dc6740fe08a690291ce0afe42011e20cc565b5ebe64ec016bf658b5bab563a63337985c05","signer_version":"1.1.0"},"trust_root":{"key_id":"430895f101d38164","public_key_hex":"cf742e26f75669dc673cb5c0786a1ae23ae8ca19c347317192ce40c28a7ff25c","signature_algorithm":"ed25519","url":"/registry/canon/TRUST_ROOT.md"},"verifier":{"spec":"/registry/canon/AXIOM_RECEIPT_v1.md","url":"/v/axm_fe942729ed803eadbe4e17bd2eba54f0c8f896444535286ba2f4d5259c47d7fd"}}