{"_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_54c6d5bf2f11a7cae2b7f4cddef24271d27e7eab2abf8173e2c7a7858705b556","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_54c6d5bf2f11a7cae2b7f4cddef24271d27e7eab2abf8173e2c7a7858705b556","axiom_type":"AX.OBS","body":{"axiom_subtype":"news.vendor_press.v1","category":"news","fingerprint":"901674812daf619498f6d1e25ded1a91e26c83acd62c87a137ce4c8a19c4b89c","published":"Mon, 18 May 2026 00:00:00 -0400","receipt_hash":"901674812daf619498f6d1e25ded1a91e26c83acd62c87a137ce4c8a19c4b89c","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":"901674812daf619498f6d1e25ded1a91e26c83acd62c87a137ce4c8a19c4b89c","observed_at":"2026-05-18T04:43:11.219741Z","parent_run_hash":"a8aad7414ebb6b75c726f09cd673410576a7f87e191fbdb9ddac99e9b2b95a05","published":"Mon, 18 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:2605.15417v1 Announce Type: cross \nAbstract: In GFlowNets and variational inference, it has been shown that the mean square error between target and model log probabilities is an effective, low variance, surrogate loss for training generative models.\n  This loss has the property that when evaluated \\emph{on-policy} its gradients correspond to those of the KL divergence, while \\emph{off-policy} it remains a valid loss with the same global minimizer. In this work, we demonstrate that this construction can be extended to the whole family of $f$-divergences, leading to a family of losses whose on-policy gradients are that of the corresponding $f$-divergence, but retain the same global minimizer off-policy. Specifically, we show that the on-policy gradients lead to a one to one correspondence between translation invariant loss functions on the target and model log probabilities, and $f$-divergences. This equivalence allows us to design new surrogate loss functions for tuning a wide cl","title":"$f$-Trajectory Balance: A Loss Family for Tuning GFlowNets, Generative Models, and LLMs with Off- and On-Policy Data","url":"https://arxiv.org/abs/2605.15417","vendor":"arxiv_cs_ai"},"summary":"arXiv:2605.15417v1 Announce Type: cross \nAbstract: In GFlowNets and variational inference, it has been shown that the mean square error between target and model log probabilities is an effective, low variance, surrogate loss for training generative models.\n  This loss has the property that when evaluated \\emph{on-policy} its gradients correspond to those of the KL divergence, while \\emph{off-policy} it remains a valid loss with the same global minimizer. In this work, we demonstrate that this construction can be extended to the whole family of $f$-divergences, leading to a family of losses whose on-policy gradients are that of the corresponding $f$-divergence, but retain the same global minimizer off-policy. Specifically, we show that the on-policy gradients lead to a one to one correspondence between translation invariant loss functions on the target and model log probabilities, and $f$-divergences. This equivalence allows us to design new surrogate loss functions for tuning a wide cl","title":"$f$-Trajectory Balance: A Loss Family for Tuning GFlowNets, Generative Models, and LLMs with Off- and On-Policy Data","vendor":"arxiv_cs_ai"},"confidence":{"method":"deterministic"},"decision":"POSITIVE","issued_at":"2026-05-18T04:43:11Z","notes":"Spider vendor_press (news) news.vendor_press.v1","object":{"captured_by":"crovia.spider.vendor_press","primary_source_url":"https://arxiv.org/abs/2605.15417"},"predecessors":[],"schema":"crovia.axiom.v1","signature":"ed25519:18f187e36ea655f577dc3591ea13c78cc9131720467cc18e7969bca970e09bbe6a86d824ebac6c64bb49a22a0cd8a8fb0804546d4d246e053bc4da73b3065302","signer":"crovia.substrate","subject":{"observed_at":"2026-05-18T04:43:11Z","source_collector":"spider:vendor_press","target_id":"https://arxiv.org/abs/2605.15417"},"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":"d4a47c79e74b12bdf40264e372c2e47258c50d9266fae6f51f032340088ce46b","leaf_index":140589,"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":"64acca17cc77279968d81988764b1146f7c61498138b23a0c6602dbb62f92266","side":"left"},{"sibling":"a83dae78ca13df2ce078af2f366d94062b218810924dcb8f878cd6dbbcb5bdaf","side":"right"},{"sibling":"0cf9eb9a08f43e74a4b8cafd96bc33c6906807d3f9373cd2c04a5de4379f7d4a","side":"left"},{"sibling":"151c022e01fc44e0b20e6a1ae23bca00708a6e070ebc3f12a864434d5646241a","side":"left"},{"sibling":"dead59a16a46ffb59968c18ba6918b1b54799c6d66ed856b1f99236d78ac0c57","side":"right"},{"sibling":"e927ee761b610b6646108660a1c36882541add1974b059be960ba3967a40c96b","side":"left"},{"sibling":"d10f43d638e7e4b5fe1190bff28b23af4dede5ac008b59c8c88d6e509f8c5960","side":"right"},{"sibling":"b29bcbd6adca3228d98aa21ca6aabf98e279340820cd7a960b41569223eebf29","side":"right"},{"sibling":"6bedf73520cf3dd8758d8bdedf3be245de9aea97abd42934aae25539176ae1b2","side":"left"},{"sibling":"07abc3bad689e74e6304772503dc9372a118e6f66883b8e88c43414efddac063","side":"right"},{"sibling":"28b78fb112bcf26b6801664db97eb8f52a9bccbf0a7ae6766e11845d443692df","side":"left"},{"sibling":"68d0a4634c1460a19c92edd9480df3aa733b814463e7420d1e14471bf61b2f83","side":"right"},{"sibling":"8af64f275b862349aa3bbb9d5cd7fa9a7fdd5620af3bf1b36b2a4519b0b53bdf","side":"right"},{"sibling":"8f4c0fbe56b6c010fbb8c782ebcd478079bb3f991d8704e2534209a075d9163c","side":"left"},{"sibling":"b98c2afadb358e5387e88f19588f8343a81b488d9b44a6f7e57a032db3a1b030","side":"right"},{"sibling":"11b0c1591747f09f7c8971a6caa19befcd81317ca9dfd417b143234df4e10c79","side":"right"},{"sibling":"87206f3bcc342797c990d87f7235c01f78d32ca59cfaf8ad18d71afc879ba477","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":140892,"merkle_root":"6cca56ead155990456b8a014cc50bddbe710f409b26e3d1bfa6fb12b0bfcf6bf","public_key_hex":"cf742e26f75669dc673cb5c0786a1ae23ae8ca19c347317192ce40c28a7ff25c","run_id":"hourly_json_retrofit_20260518T053701Z","schema":"crovia.seal.v1","seal_family_version":"crovia-seal-family/1","seal_kind":"substrate_batch","sealed_at":"2026-05-18T05:37:30Z","sig_algorithm":"ed25519","signature":"1e1135f7595f79b14fb11f5fa81a2e17ad31b11b44b427a5e40a7d511cd86447daf492babd368ab571cf26404c8c74c450d460130fca4b064eb2760367489a0f","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_54c6d5bf2f11a7cae2b7f4cddef24271d27e7eab2abf8173e2c7a7858705b556"}}