{"_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_1a6e88b22f4c182ff0b32cb4f37d66be12a19dd784adc49bd29ed3ee83a90a5c","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_1a6e88b22f4c182ff0b32cb4f37d66be12a19dd784adc49bd29ed3ee83a90a5c","axiom_type":"AX.OBS","body":{"axiom_subtype":"news.vendor_press.v1","category":"news","fingerprint":"96a9ccbf57e78c46c20e1505b6295c62506ec52c4a8dd77d07db7823480af3d3","published":"Fri, 15 May 2026 00:00:00 -0400","receipt_hash":"96a9ccbf57e78c46c20e1505b6295c62506ec52c4a8dd77d07db7823480af3d3","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":"96a9ccbf57e78c46c20e1505b6295c62506ec52c4a8dd77d07db7823480af3d3","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:2605.14558v1 Announce Type: cross \nAbstract: Agentic reinforcement learning trains large language models using multi-turn trajectories that interleave long reasoning traces with short environment-facing actions. Common policy-gradient methods, such as PPO and GRPO, treat each token in a trajectory equally, leading to uniform credit assignment. In this paper, we critically demonstrate that such uniform credit assignment largely misallocates token-level training signals. From an energy-based modeling perspective, we show that token-level training signals, quantified by their correlations with reward variance of different rollouts sampled from a given prompt, concentrate sharply on action tokens rather than reasoning tokens, even though action tokens account for only a small fraction of the trajectory. We refer to this phenomenon as the Action Bottleneck. Motivated by this observation, we propose an embarrassingly simple token reweighting approach, ActFocus, that downweights gradien","title":"Resolving Action Bottleneck: Agentic Reinforcement Learning Informed by Token-Level Energy","url":"https://arxiv.org/abs/2605.14558","vendor":"arxiv_cs_ai"},"summary":"arXiv:2605.14558v1 Announce Type: cross \nAbstract: Agentic reinforcement learning trains large language models using multi-turn trajectories that interleave long reasoning traces with short environment-facing actions. Common policy-gradient methods, such as PPO and GRPO, treat each token in a trajectory equally, leading to uniform credit assignment. In this paper, we critically demonstrate that such uniform credit assignment largely misallocates token-level training signals. From an energy-based modeling perspective, we show that token-level training signals, quantified by their correlations with reward variance of different rollouts sampled from a given prompt, concentrate sharply on action tokens rather than reasoning tokens, even though action tokens account for only a small fraction of the trajectory. We refer to this phenomenon as the Action Bottleneck. Motivated by this observation, we propose an embarrassingly simple token reweighting approach, ActFocus, that downweights gradien","title":"Resolving Action Bottleneck: Agentic Reinforcement Learning Informed by Token-Level Energy","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/2605.14558"},"predecessors":[],"schema":"crovia.axiom.v1","signature":"ed25519:3a27cdbd09d2d182183a20d3d1132b9634b07af972fd6b7055315475a37b2252e4e1da5454f809e696ad98100fb76519a69a14c36faf5f5868b95cd2b0f8d50e","signer":"crovia.substrate","subject":{"observed_at":"2026-05-15T04:43:17Z","source_collector":"spider:vendor_press","target_id":"https://arxiv.org/abs/2605.14558"},"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":"6f170e754df88d4a2a38a9be99ae65388ee84fecc276b5d09058ed1f87624027","leaf_index":134597,"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":"52445c0d0b5821894a9539584efd101c2a5f835a177277b74e9edffc091e2b83","side":"left"},{"sibling":"2f1ac6213b1b2c82fe056301e2b83491b2d4d393fcc6f7b1034f13f13fefa3f4","side":"right"},{"sibling":"40d21b4f3a66d62ac0e35a61f35bd055801b0df8a5022c9c43134eb1eecedb0b","side":"left"},{"sibling":"88159c1386fa806c179e984e0b1591a6c7591284a5780339097111c69dcfbbee","side":"right"},{"sibling":"c25fca3fdc7b43cc1b9b68e9cc533eacfcee1be7ce7d41bbf4f3ba8f780f260a","side":"right"},{"sibling":"636fee4d0f144f323e8d0d1097a7ed5fb4d85b2abd937d083f00f3c129363b39","side":"right"},{"sibling":"46f96cb7883dcb2712dad253af0583c5b367609f9717fd33426973df576e51f6","side":"left"},{"sibling":"cbbed0225164fe227925f0fc2929e9ecbc520f5a86b19035c412206d51935e10","side":"left"},{"sibling":"4711a7f4329f1874c3aa1ae93c336e1d4fa402bdd4b3d766c2a95304ea226882","side":"left"},{"sibling":"8ce2a4687a8ceb409df2e1cb10e44a610b21dc294e518a8550b6d1122335ca2b","side":"right"},{"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_1a6e88b22f4c182ff0b32cb4f37d66be12a19dd784adc49bd29ed3ee83a90a5c"}}