{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:OGPC33FFYMVU7AOHCOLYSGKUGA","short_pith_number":"pith:OGPC33FF","canonical_record":{"source":{"id":"2504.05520","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-07T21:31:31Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"5958956eaae622b8906b00b75373a53063848a853c258c32efa331fe963e4b0d","abstract_canon_sha256":"fb4234b8ba5bbdc0e2ae165d35b08e8026ec7e99ade82f6fb677ca346e74aef8"},"schema_version":"1.0"},"canonical_sha256":"719e2deca5c32b4f81c71397891954300188b9461e776f22764b6f8f8675396f","source":{"kind":"arxiv","id":"2504.05520","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.05520","created_at":"2026-06-23T02:13:14Z"},{"alias_kind":"arxiv_version","alias_value":"2504.05520v4","created_at":"2026-06-23T02:13:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.05520","created_at":"2026-06-23T02:13:14Z"},{"alias_kind":"pith_short_12","alias_value":"OGPC33FFYMVU","created_at":"2026-06-23T02:13:14Z"},{"alias_kind":"pith_short_16","alias_value":"OGPC33FFYMVU7AOH","created_at":"2026-06-23T02:13:14Z"},{"alias_kind":"pith_short_8","alias_value":"OGPC33FF","created_at":"2026-06-23T02:13:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:OGPC33FFYMVU7AOHCOLYSGKUGA","target":"record","payload":{"canonical_record":{"source":{"id":"2504.05520","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-07T21:31:31Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"5958956eaae622b8906b00b75373a53063848a853c258c32efa331fe963e4b0d","abstract_canon_sha256":"fb4234b8ba5bbdc0e2ae165d35b08e8026ec7e99ade82f6fb677ca346e74aef8"},"schema_version":"1.0"},"canonical_sha256":"719e2deca5c32b4f81c71397891954300188b9461e776f22764b6f8f8675396f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-23T02:13:14.853152Z","signature_b64":"tABYj7qkGKwu7joYUDB6KeYF9TAvNOYvDsA1izabZxcyYe2yir2RAaNYDEwtXwHihUH+Uv9Cjfss45KMj1T5Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"719e2deca5c32b4f81c71397891954300188b9461e776f22764b6f8f8675396f","last_reissued_at":"2026-06-23T02:13:14.852673Z","signature_status":"signed_v1","first_computed_at":"2026-06-23T02:13:14.852673Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2504.05520","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-23T02:13:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NMtSQHwEKuXmclftgqoLMeabrwn43EOpwDI93ikkousXOo6dA1YXqeyLv+unuXpkA8b9sN6+subsVeYT3EnnAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-28T08:37:27.830382Z"},"content_sha256":"46aa3a53b241f23370fa920e4d616d9cca2e974177f896901800c5e24f636473","schema_version":"1.0","event_id":"sha256:46aa3a53b241f23370fa920e4d616d9cca2e974177f896901800c5e24f636473"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:OGPC33FFYMVU7AOHCOLYSGKUGA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Efficient Reinforcement Finetuning via Adaptive Curriculum Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.LG","authors_text":"Jieyu Zhao, Linxin Song, Taiwei Shi, Tianyi Zhou, Yiyang Wu","submitted_at":"2025-04-07T21:31:31Z","abstract_excerpt":"Reinforcement finetuning (RFT) has shown great potential for enhancing the mathematical reasoning capabilities of large language models (LLMs), but it is often sample- and compute-inefficient, requiring extensive training. In this work, we introduce AdaRFT (Adaptive Curriculum Reinforcement Finetuning), a method that significantly improves the efficiency of RFT through adaptive curriculum learning. AdaRFT dynamically adjusts the difficulty of training problems based on the model's recent reward signals, ensuring that the model consistently trains on tasks that are challenging but solvable. Thi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.05520","kind":"arxiv","version":4},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2504.05520/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"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":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-23T02:13:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Vo+7g+7KRAD1+SKxC8li9YdZ0x6mQFmVBnuTDaRe3Za5k8HSy+DqiGfBAyrNqVJimkvh2j3Nm+auwNYpA77HAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-28T08:37:27.830763Z"},"content_sha256":"c87b5564c364d2f1d9a72ce09390f77becb357ee360b0316a168f3cbcbf69615","schema_version":"1.0","event_id":"sha256:c87b5564c364d2f1d9a72ce09390f77becb357ee360b0316a168f3cbcbf69615"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OGPC33FFYMVU7AOHCOLYSGKUGA/bundle.json","state_url":"https://pith.science/pith/OGPC33FFYMVU7AOHCOLYSGKUGA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OGPC33FFYMVU7AOHCOLYSGKUGA/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":"2026-06-28T08:37:27Z","links":{"resolver":"https://pith.science/pith/OGPC33FFYMVU7AOHCOLYSGKUGA","bundle":"https://pith.science/pith/OGPC33FFYMVU7AOHCOLYSGKUGA/bundle.json","state":"https://pith.science/pith/OGPC33FFYMVU7AOHCOLYSGKUGA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OGPC33FFYMVU7AOHCOLYSGKUGA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:OGPC33FFYMVU7AOHCOLYSGKUGA","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":"fb4234b8ba5bbdc0e2ae165d35b08e8026ec7e99ade82f6fb677ca346e74aef8","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-07T21:31:31Z","title_canon_sha256":"5958956eaae622b8906b00b75373a53063848a853c258c32efa331fe963e4b0d"},"schema_version":"1.0","source":{"id":"2504.05520","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2504.05520","created_at":"2026-06-23T02:13:14Z"},{"alias_kind":"arxiv_version","alias_value":"2504.05520v4","created_at":"2026-06-23T02:13:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.05520","created_at":"2026-06-23T02:13:14Z"},{"alias_kind":"pith_short_12","alias_value":"OGPC33FFYMVU","created_at":"2026-06-23T02:13:14Z"},{"alias_kind":"pith_short_16","alias_value":"OGPC33FFYMVU7AOH","created_at":"2026-06-23T02:13:14Z"},{"alias_kind":"pith_short_8","alias_value":"OGPC33FF","created_at":"2026-06-23T02:13:14Z"}],"graph_snapshots":[{"event_id":"sha256:c87b5564c364d2f1d9a72ce09390f77becb357ee360b0316a168f3cbcbf69615","target":"graph","created_at":"2026-06-23T02:13:14Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2504.05520/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement finetuning (RFT) has shown great potential for enhancing the mathematical reasoning capabilities of large language models (LLMs), but it is often sample- and compute-inefficient, requiring extensive training. In this work, we introduce AdaRFT (Adaptive Curriculum Reinforcement Finetuning), a method that significantly improves the efficiency of RFT through adaptive curriculum learning. AdaRFT dynamically adjusts the difficulty of training problems based on the model's recent reward signals, ensuring that the model consistently trains on tasks that are challenging but solvable. Thi","authors_text":"Jieyu Zhao, Linxin Song, Taiwei Shi, Tianyi Zhou, Yiyang Wu","cross_cats":["cs.CL"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-07T21:31:31Z","title":"Efficient Reinforcement Finetuning via Adaptive Curriculum Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.05520","kind":"arxiv","version":4},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:46aa3a53b241f23370fa920e4d616d9cca2e974177f896901800c5e24f636473","target":"record","created_at":"2026-06-23T02:13:14Z","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":"fb4234b8ba5bbdc0e2ae165d35b08e8026ec7e99ade82f6fb677ca346e74aef8","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-04-07T21:31:31Z","title_canon_sha256":"5958956eaae622b8906b00b75373a53063848a853c258c32efa331fe963e4b0d"},"schema_version":"1.0","source":{"id":"2504.05520","kind":"arxiv","version":4}},"canonical_sha256":"719e2deca5c32b4f81c71397891954300188b9461e776f22764b6f8f8675396f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"719e2deca5c32b4f81c71397891954300188b9461e776f22764b6f8f8675396f","first_computed_at":"2026-06-23T02:13:14.852673Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-23T02:13:14.852673Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"tABYj7qkGKwu7joYUDB6KeYF9TAvNOYvDsA1izabZxcyYe2yir2RAaNYDEwtXwHihUH+Uv9Cjfss45KMj1T5Ag==","signature_status":"signed_v1","signed_at":"2026-06-23T02:13:14.853152Z","signed_message":"canonical_sha256_bytes"},"source_id":"2504.05520","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:46aa3a53b241f23370fa920e4d616d9cca2e974177f896901800c5e24f636473","sha256:c87b5564c364d2f1d9a72ce09390f77becb357ee360b0316a168f3cbcbf69615"],"state_sha256":"886ec4c811d83132b58b9136ca8f1b6c08c17a3b36d244f5f3f90129d4a6c4c5"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"0W4AyKgtYy2a6IyoMzPPLCBJe6kIsHzxGnzlHFlLY4x3RFaL2UIpM3MkXp8KxDWtJ8j1tc+6l9N4UJUnqzZsCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-28T08:37:27.832716Z","bundle_sha256":"876f6b3e8aedc90f305228d84f52952c37ead58af554bc9da73ca210bf40ca04"}}