{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:IBOON75NXU35CHZHNVCJSICENA","short_pith_number":"pith:IBOON75N","canonical_record":{"source":{"id":"2604.28020","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-04-30T15:39:09Z","cross_cats_sorted":[],"title_canon_sha256":"ceb970895d645166143f06e611672c4dc852112665822d98373e8c5c76a46a20","abstract_canon_sha256":"e203bea884ca6872e40e862b8e31299cd16b3707afb0f879b35534da2de819c6"},"schema_version":"1.0"},"canonical_sha256":"405ce6ffadbd37d11f276d449920446828560d01b951f46c8598b9f01d0b33aa","source":{"kind":"arxiv","id":"2604.28020","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.28020","created_at":"2026-06-01T02:03:42Z"},{"alias_kind":"arxiv_version","alias_value":"2604.28020v2","created_at":"2026-06-01T02:03:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.28020","created_at":"2026-06-01T02:03:42Z"},{"alias_kind":"pith_short_12","alias_value":"IBOON75NXU35","created_at":"2026-06-01T02:03:42Z"},{"alias_kind":"pith_short_16","alias_value":"IBOON75NXU35CHZH","created_at":"2026-06-01T02:03:42Z"},{"alias_kind":"pith_short_8","alias_value":"IBOON75N","created_at":"2026-06-01T02:03:42Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:IBOON75NXU35CHZHNVCJSICENA","target":"record","payload":{"canonical_record":{"source":{"id":"2604.28020","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-04-30T15:39:09Z","cross_cats_sorted":[],"title_canon_sha256":"ceb970895d645166143f06e611672c4dc852112665822d98373e8c5c76a46a20","abstract_canon_sha256":"e203bea884ca6872e40e862b8e31299cd16b3707afb0f879b35534da2de819c6"},"schema_version":"1.0"},"canonical_sha256":"405ce6ffadbd37d11f276d449920446828560d01b951f46c8598b9f01d0b33aa","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T02:03:42.200805Z","signature_b64":"fWix5Lo5oUgZTQHYnEk1YYcNCAWvJ7n2AjBR2CFcxMouwtA2eZTzE+OMz/ySeF3GOA6SE3x5NBNhBM7AQ+1ODg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"405ce6ffadbd37d11f276d449920446828560d01b951f46c8598b9f01d0b33aa","last_reissued_at":"2026-06-01T02:03:42.199752Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T02:03:42.199752Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.28020","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-01T02:03:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JYtptH8WByQNqzLdmPZRf4jl55RhhpnySBIfTA1K+6TeopdbWoq6QpLrrQYZ4UMAqBekq/H/QLI2dhkF7pRRAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T09:01:44.083850Z"},"content_sha256":"49eb019337d309dda3f6403cf882858899e9f38baf22dfedd0b8753dc3bbcac2","schema_version":"1.0","event_id":"sha256:49eb019337d309dda3f6403cf882858899e9f38baf22dfedd0b8753dc3bbcac2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:IBOON75NXU35CHZHNVCJSICENA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Cost-Aware Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"By accounting for different sampling costs, cost-aware stochastic gradient descent reaches target accuracy at lower total cost and reduces token usage by up to 30 percent in LLM policy optimization.","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Amir Globerson, Clara Mohri, Haim Kaplan, Tomer Koren, Yishay Mansour","submitted_at":"2026-04-30T15:39:09Z","abstract_excerpt":"We consider the problem of Cost-Aware Learning, where sampling different components of a finite-sum objective incurs different costs. The objective is to reach a target error while minimizing the total cost. We propose Cost-Aware SGD, which uses a distribution based on gradient norms and costs to sample components. We provide a thorough analysis of this algorithm, including cost-improvement bounds over baselines, a characterization of distribution proxy sub-optimality, and a lower bound. We apply our theoretical insights to reinforcement learning with language models, where the computational c"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Empirical results on 1.5B and 8B LLMs demonstrate that our approach reduces the tokens used in policy optimization by up to about 30% while matching or exceeding baseline accuracy.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The per-component sampling costs are known in advance and can be used to set sampling probabilities without introducing bias that harms convergence; this is stated implicitly in the cost-aware SGD derivation and the GRPO adaptation.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Cost-aware SGD achieves target error with lower total sampling cost than standard methods, and Cost-Aware GRPO reduces token usage by up to 30% in LLM reinforcement learning while matching baseline performance.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"By accounting for different sampling costs, cost-aware stochastic gradient descent reaches target accuracy at lower total cost and reduces token usage by up to 30 percent in LLM policy optimization.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"9490acccddeb05ab3d7e0e2d4accb8b6f60bc06438e4cfa77fbd02f8b118e23b"},"source":{"id":"2604.28020","kind":"arxiv","version":2},"verdict":{"id":"b34bcc42-13f1-456f-8a51-02a3738b6194","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-07T05:01:06.202809Z","strongest_claim":"Empirical results on 1.5B and 8B LLMs demonstrate that our approach reduces the tokens used in policy optimization by up to about 30% while matching or exceeding baseline accuracy.","one_line_summary":"Cost-aware SGD achieves target error with lower total sampling cost than standard methods, and Cost-Aware GRPO reduces token usage by up to 30% in LLM reinforcement learning while matching baseline performance.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The per-component sampling costs are known in advance and can be used to set sampling probabilities without introducing bias that harms convergence; this is stated implicitly in the cost-aware SGD derivation and the GRPO adaptation.","pith_extraction_headline":"By accounting for different sampling costs, cost-aware stochastic gradient descent reaches target accuracy at lower total cost and reduces token usage by up to 30 percent in LLM policy optimization."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.28020/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T20:42:12.032805Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T18:41:30.656157Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"2edd81f7537ba101f25a5110eb2e237f033fd9e99a20aa2e221961cb7c58f48f"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"b34bcc42-13f1-456f-8a51-02a3738b6194"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-01T02:03:42Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Bmf4nNtK9jrTn1wNYgGvfsB8AQFEQiWbpUDFZWN6d+uc4GahMcd5W1aXSxu1i9+9FLLbcg6vC1OyKW97ShujBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-07T09:01:44.084830Z"},"content_sha256":"1452302bb39780f969bd31f15b53ae507778b0935ff6c02d0058cea2a6436a1b","schema_version":"1.0","event_id":"sha256:1452302bb39780f969bd31f15b53ae507778b0935ff6c02d0058cea2a6436a1b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/IBOON75NXU35CHZHNVCJSICENA/bundle.json","state_url":"https://pith.science/pith/IBOON75NXU35CHZHNVCJSICENA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/IBOON75NXU35CHZHNVCJSICENA/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":"2026-06-07T09:01:44Z","links":{"resolver":"https://pith.science/pith/IBOON75NXU35CHZHNVCJSICENA","bundle":"https://pith.science/pith/IBOON75NXU35CHZHNVCJSICENA/bundle.json","state":"https://pith.science/pith/IBOON75NXU35CHZHNVCJSICENA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/IBOON75NXU35CHZHNVCJSICENA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:IBOON75NXU35CHZHNVCJSICENA","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":"e203bea884ca6872e40e862b8e31299cd16b3707afb0f879b35534da2de819c6","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-04-30T15:39:09Z","title_canon_sha256":"ceb970895d645166143f06e611672c4dc852112665822d98373e8c5c76a46a20"},"schema_version":"1.0","source":{"id":"2604.28020","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.28020","created_at":"2026-06-01T02:03:42Z"},{"alias_kind":"arxiv_version","alias_value":"2604.28020v2","created_at":"2026-06-01T02:03:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.28020","created_at":"2026-06-01T02:03:42Z"},{"alias_kind":"pith_short_12","alias_value":"IBOON75NXU35","created_at":"2026-06-01T02:03:42Z"},{"alias_kind":"pith_short_16","alias_value":"IBOON75NXU35CHZH","created_at":"2026-06-01T02:03:42Z"},{"alias_kind":"pith_short_8","alias_value":"IBOON75N","created_at":"2026-06-01T02:03:42Z"}],"graph_snapshots":[{"event_id":"sha256:1452302bb39780f969bd31f15b53ae507778b0935ff6c02d0058cea2a6436a1b","target":"graph","created_at":"2026-06-01T02:03:42Z","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":"Empirical results on 1.5B and 8B LLMs demonstrate that our approach reduces the tokens used in policy optimization by up to about 30% while matching or exceeding baseline accuracy."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The per-component sampling costs are known in advance and can be used to set sampling probabilities without introducing bias that harms convergence; this is stated implicitly in the cost-aware SGD derivation and the GRPO adaptation."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Cost-aware SGD achieves target error with lower total sampling cost than standard methods, and Cost-Aware GRPO reduces token usage by up to 30% in LLM reinforcement learning while matching baseline performance."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"By accounting for different sampling costs, cost-aware stochastic gradient descent reaches target accuracy at lower total cost and reduces token usage by up to 30 percent in LLM policy optimization."}],"snapshot_sha256":"9490acccddeb05ab3d7e0e2d4accb8b6f60bc06438e4cfa77fbd02f8b118e23b"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-20T20:42:12.032805Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T18:41:30.656157Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2604.28020/integrity.json","findings":[],"snapshot_sha256":"2edd81f7537ba101f25a5110eb2e237f033fd9e99a20aa2e221961cb7c58f48f","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We consider the problem of Cost-Aware Learning, where sampling different components of a finite-sum objective incurs different costs. The objective is to reach a target error while minimizing the total cost. We propose Cost-Aware SGD, which uses a distribution based on gradient norms and costs to sample components. We provide a thorough analysis of this algorithm, including cost-improvement bounds over baselines, a characterization of distribution proxy sub-optimality, and a lower bound. We apply our theoretical insights to reinforcement learning with language models, where the computational c","authors_text":"Amir Globerson, Clara Mohri, Haim Kaplan, Tomer Koren, Yishay Mansour","cross_cats":[],"headline":"By accounting for different sampling costs, cost-aware stochastic gradient descent reaches target accuracy at lower total cost and reduces token usage by up to 30 percent in LLM policy optimization.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-04-30T15:39:09Z","title":"Cost-Aware Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.28020","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-07T05:01:06.202809Z","id":"b34bcc42-13f1-456f-8a51-02a3738b6194","model_set":{"reader":"grok-4.3"},"one_line_summary":"Cost-aware SGD achieves target error with lower total sampling cost than standard methods, and Cost-Aware GRPO reduces token usage by up to 30% in LLM reinforcement learning while matching baseline performance.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"By accounting for different sampling costs, cost-aware stochastic gradient descent reaches target accuracy at lower total cost and reduces token usage by up to 30 percent in LLM policy optimization.","strongest_claim":"Empirical results on 1.5B and 8B LLMs demonstrate that our approach reduces the tokens used in policy optimization by up to about 30% while matching or exceeding baseline accuracy.","weakest_assumption":"The per-component sampling costs are known in advance and can be used to set sampling probabilities without introducing bias that harms convergence; this is stated implicitly in the cost-aware SGD derivation and the GRPO adaptation."}},"verdict_id":"b34bcc42-13f1-456f-8a51-02a3738b6194"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:49eb019337d309dda3f6403cf882858899e9f38baf22dfedd0b8753dc3bbcac2","target":"record","created_at":"2026-06-01T02:03:42Z","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":"e203bea884ca6872e40e862b8e31299cd16b3707afb0f879b35534da2de819c6","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-04-30T15:39:09Z","title_canon_sha256":"ceb970895d645166143f06e611672c4dc852112665822d98373e8c5c76a46a20"},"schema_version":"1.0","source":{"id":"2604.28020","kind":"arxiv","version":2}},"canonical_sha256":"405ce6ffadbd37d11f276d449920446828560d01b951f46c8598b9f01d0b33aa","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"405ce6ffadbd37d11f276d449920446828560d01b951f46c8598b9f01d0b33aa","first_computed_at":"2026-06-01T02:03:42.199752Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-01T02:03:42.199752Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fWix5Lo5oUgZTQHYnEk1YYcNCAWvJ7n2AjBR2CFcxMouwtA2eZTzE+OMz/ySeF3GOA6SE3x5NBNhBM7AQ+1ODg==","signature_status":"signed_v1","signed_at":"2026-06-01T02:03:42.200805Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.28020","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:49eb019337d309dda3f6403cf882858899e9f38baf22dfedd0b8753dc3bbcac2","sha256:1452302bb39780f969bd31f15b53ae507778b0935ff6c02d0058cea2a6436a1b"],"state_sha256":"de4d3b8bed05147a8056237b9991e687abeb540ad4eebd24a523c318a0f3f62a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vSIFjbu0iw4ajKfA+gQMiTmljMXgfBD+hhrhZ6xZD23TXGX7QUtnlLErzOI2K7i8Q6sSpwio9Q6Va7QhH9ByBA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-07T09:01:44.089630Z","bundle_sha256":"7166fa3f362d69d8895c955da50d283dddd5beb0bf32ef1475ed6809d5796685"}}