{"_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_b92f428601cf79c22b451b3f2f8d147deb3d6c7e4d1dbd8ffd061cc6bcaa4e63","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_b92f428601cf79c22b451b3f2f8d147deb3d6c7e4d1dbd8ffd061cc6bcaa4e63","axiom_type":"AX.OBS","body":{"axiom_subtype":"news.vendor_press.v1","category":"news","fingerprint":"f88eadd39681ded57783d2c08169a54999f299c0d726f3b29f705cdb5e46d665","published":"Wed, 15 Jul 2026 00:00:00 -0400","receipt_hash":"f88eadd39681ded57783d2c08169a54999f299c0d726f3b29f705cdb5e46d665","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":"f88eadd39681ded57783d2c08169a54999f299c0d726f3b29f705cdb5e46d665","observed_at":"2026-07-15T04:44:03.592429Z","parent_run_hash":"d49a6cf532e74153266f377b7760fc948d950d80ed41fc3e3eb82b58f5597ead","published":"Wed, 15 Jul 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:2607.12640v1 Announce Type: new \nAbstract: Reinforcement learning with verifiable rewards, and Group Relative Policy Optimization (GRPO) in particular, is now run routinely on a supervised checkpoint in the hope of producing a stronger agent. We ask whether it adds skill to a small language and vision-language model web agent at the 4B to 8B scale, or whether it mostly reshapes behavior the supervised model already has. Across a control grid of 18 runs that varies learning rate, KL weight, seed, initialization, and clipping, no configuration credibly improves the success rate of a strong supervised baseline on tasks the agent has largely mastered. On the text track, moderate to high learning rates make it credibly worse. The null holds under paired testing, 25 evaluation seeds, 6 training seeds, changes to the recipe, both text and Set-of-Marks screenshot observations, and scaling the backbone to 8B; the credible harm is a text-track finding and is only nominal under Set-of-Marks","title":"A Learning-Rate-Gated Failure of GRPO in a Small Language and Vision-Language Model Web Agent: A Controlled Null and Its Mechanism","url":"https://arxiv.org/abs/2607.12640","vendor":"arxiv_cs_ai"},"summary":"arXiv:2607.12640v1 Announce Type: new \nAbstract: Reinforcement learning with verifiable rewards, and Group Relative Policy Optimization (GRPO) in particular, is now run routinely on a supervised checkpoint in the hope of producing a stronger agent. We ask whether it adds skill to a small language and vision-language model web agent at the 4B to 8B scale, or whether it mostly reshapes behavior the supervised model already has. Across a control grid of 18 runs that varies learning rate, KL weight, seed, initialization, and clipping, no configuration credibly improves the success rate of a strong supervised baseline on tasks the agent has largely mastered. On the text track, moderate to high learning rates make it credibly worse. The null holds under paired testing, 25 evaluation seeds, 6 training seeds, changes to the recipe, both text and Set-of-Marks screenshot observations, and scaling the backbone to 8B; the credible harm is a text-track finding and is only nominal under Set-of-Marks","title":"A Learning-Rate-Gated Failure of GRPO in a Small Language and Vision-Language Model Web Agent: A Controlled Null and Its Mechanism","vendor":"arxiv_cs_ai"},"confidence":{"method":"deterministic"},"decision":"POSITIVE","issued_at":"2026-07-15T04:44:03Z","notes":"Spider vendor_press (news) news.vendor_press.v1","object":{"captured_by":"crovia.spider.vendor_press","primary_source_url":"https://arxiv.org/abs/2607.12640"},"predecessors":[],"schema":"crovia.axiom.v1","signature":"ed25519:607068e7d5b59a2141102bc872cb97b4aca9b937497f0b01d1a9315ee97a80f4cfd84f56a419d1d65e792a2a669dae288e05784d96179f445f11c5d670a88302","signer":"crovia.substrate","subject":{"observed_at":"2026-07-15T04:44:03Z","source_collector":"spider:vendor_press","target_id":"https://arxiv.org/abs/2607.12640"},"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":"8227b52b019d50277c43316d90aa41e55786ec87a93574ead34b1376da9691fa","leaf_index":316389,"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":"b60353f6db5d1b4ff3c94708eec6030cfc0d606435fdaae90f2f6fc314eb680d","side":"left"},{"sibling":"55846b22489fc503c1932771eec75d247fa7e9dcd7957c07d49619aea6839855","side":"right"},{"sibling":"bc0bae33ac57b915d4b41ed4305fc2a9f5b2c80a7b42393e22df62b9b1b49ab8","side":"left"},{"sibling":"2621a73eec78a5ccc5b25509399cc6966b41111562e556ba2502a2a4f22c798e","side":"right"},{"sibling":"cdc0db623f671a908513680b838b37849335a442f43950d1c6337bf1400035dd","side":"right"},{"sibling":"b922a2728fd9d538aa8b17f643ec62d6ceefaa53e320aeeedde1104caba3409d","side":"left"},{"sibling":"8dc6b151fc86f1b43f490c2263dd78189a9778cebe1755f2bcd01eb2c6239831","side":"left"},{"sibling":"9a178a31a5d97bdb1bf054b31ec2fcc6cf9b633c6d3f460e934451a35bac9c25","side":"left"},{"sibling":"31b0ff6eae164fcff8882af2b581471a08fffa2373505e1f33f01e361c94a42f","side":"left"},{"sibling":"e1ebad13fafe4a9a53c35e1f19fa67a19c940f70a4211d1fab42cf49be486a3c","side":"left"},{"sibling":"c265283bb70fb86740bdfa059cfe36cfea4a5f903b841822b2954814237de972","side":"right"},{"sibling":"0cb62c0ada57a2406a6bcb100889d3e8b29a15efeed07adaff5bb90a5e80612a","side":"right"},{"sibling":"0b69289b25462ddd6166f6f49004cfc8ada0ab4f10adafe188347817bdd46e37","side":"left"},{"sibling":"1418b281cd985b5ed411ef25f2017a1826cc14919b6fad3934e6ceeec693699b","side":"right"},{"sibling":"f302542c38ba7c3aab7c9280dd60259ecec777dca6e6f71b6f0729b0b8791b72","side":"left"},{"sibling":"d8b9143917b539c543cf4448cec00131f8b807bd8004979c54ebe09798748c66","side":"left"},{"sibling":"abe4a8c706e530484d1e96a8988cb09eab85928b2050985500ab289753fe3eec","side":"right"},{"sibling":"f436dccf82aa2c1eb7bfa3eb84316e116aaf64dc55cd9592597118f6cb0648f6","side":"right"},{"sibling":"1cecb7f447febd025aac272837c80de218aecc6485d2395a509b2a1f1b9c746e","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":316730,"merkle_root":"a8e6e5be81ea6f5b5f2227422459bf39455fe9f0b6602b4d1ce6977dbfd78bc7","public_key_hex":"cf742e26f75669dc673cb5c0786a1ae23ae8ca19c347317192ce40c28a7ff25c","run_id":"hourly_json_retrofit_20260715T053701Z","schema":"crovia.seal.v1","seal_family_version":"crovia-seal-family/1","seal_kind":"substrate_batch","sealed_at":"2026-07-15T05:38:24Z","sig_algorithm":"ed25519","signature":"df1678d268b5a07413e2ca6e748c3f40b6cfedea930a18d843489a4ab513da791bf0a886caab1b918d0989f8ebaaf3d0035ca2aa777913b1ad29979f1deb9a0c","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_b92f428601cf79c22b451b3f2f8d147deb3d6c7e4d1dbd8ffd061cc6bcaa4e63"}}