{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:V2XX74F4UQTPVFWN67XAORFV5V","short_pith_number":"pith:V2XX74F4","schema_version":"1.0","canonical_sha256":"aeaf7ff0bca426fa96cdf7ee0744b5ed4d9d53d2f6a673138ff35145847662c6","source":{"kind":"arxiv","id":"2606.24369","version":1},"attestation_state":"computed","paper":{"title":"Accelerating Disaggregated RL for Visual Generative LLMs with Diffusion-Based Parallelism and Trainer-Assisted Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.NI","cs.PF"],"primary_cat":"cs.AI","authors_text":"Qiang Wang, Shaohuai Shi, Sijie Wang, Xiaowen Chu, Yaoyuan Wang, Yeqing Zhang, Yiming Yin, Zhengyu Qing, Zhiqiang Tan","submitted_at":"2026-06-23T09:59:35Z","abstract_excerpt":"Reinforcement learning (RL) has become a dominant post-training paradigm, driving the emergence of high-performance RL systems such as veRL for autoregressive large language models (LLMs). In parallel, diffusion-oriented RL algorithms, e.g., DanceGRPO and FlowGRPO, have rapidly expanded the scope of RL from language reasoning to diffusion-based visual and flow-based generation. However, efficient RL systems for diffusion generative LLMs remain underexplored. Existing implementations, e.g., veRL-Omni, still rely on colocated execution, which simplifies synchronization but couples rollout and tr"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2606.24369","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-23T09:59:35Z","cross_cats_sorted":["cs.DC","cs.NI","cs.PF"],"title_canon_sha256":"230b6efa6f343c7bb37b27786a94cd088ea77829dca862b2c8bbd0b4a2f1760a","abstract_canon_sha256":"a9276a4adece2bdf801abdd6a5009c0177be3ee76da92b8c5926ead1d5c42dd6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-24T01:15:28.745726Z","signature_b64":"yHP+3Kv9FIWAJFiWrMqx3g64pjoIaPS5e6T2lV58KqVvULH32VGfc1lNUrYh9Wd4NZO+9n4owS/Tw97NnevkAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aeaf7ff0bca426fa96cdf7ee0744b5ed4d9d53d2f6a673138ff35145847662c6","last_reissued_at":"2026-06-24T01:15:28.745371Z","signature_status":"signed_v1","first_computed_at":"2026-06-24T01:15:28.745371Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Accelerating Disaggregated RL for Visual Generative LLMs with Diffusion-Based Parallelism and Trainer-Assisted Generation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.DC","cs.NI","cs.PF"],"primary_cat":"cs.AI","authors_text":"Qiang Wang, Shaohuai Shi, Sijie Wang, Xiaowen Chu, Yaoyuan Wang, Yeqing Zhang, Yiming Yin, Zhengyu Qing, Zhiqiang Tan","submitted_at":"2026-06-23T09:59:35Z","abstract_excerpt":"Reinforcement learning (RL) has become a dominant post-training paradigm, driving the emergence of high-performance RL systems such as veRL for autoregressive large language models (LLMs). In parallel, diffusion-oriented RL algorithms, e.g., DanceGRPO and FlowGRPO, have rapidly expanded the scope of RL from language reasoning to diffusion-based visual and flow-based generation. However, efficient RL systems for diffusion generative LLMs remain underexplored. Existing implementations, e.g., veRL-Omni, still rely on colocated execution, which simplifies synchronization but couples rollout and tr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.24369","kind":"arxiv","version":1},"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/2606.24369/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2606.24369","created_at":"2026-06-24T01:15:28.745437+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.24369v1","created_at":"2026-06-24T01:15:28.745437+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.24369","created_at":"2026-06-24T01:15:28.745437+00:00"},{"alias_kind":"pith_short_12","alias_value":"V2XX74F4UQTP","created_at":"2026-06-24T01:15:28.745437+00:00"},{"alias_kind":"pith_short_16","alias_value":"V2XX74F4UQTPVFWN","created_at":"2026-06-24T01:15:28.745437+00:00"},{"alias_kind":"pith_short_8","alias_value":"V2XX74F4","created_at":"2026-06-24T01:15:28.745437+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/V2XX74F4UQTPVFWN67XAORFV5V","json":"https://pith.science/pith/V2XX74F4UQTPVFWN67XAORFV5V.json","graph_json":"https://pith.science/api/pith-number/V2XX74F4UQTPVFWN67XAORFV5V/graph.json","events_json":"https://pith.science/api/pith-number/V2XX74F4UQTPVFWN67XAORFV5V/events.json","paper":"https://pith.science/paper/V2XX74F4"},"agent_actions":{"view_html":"https://pith.science/pith/V2XX74F4UQTPVFWN67XAORFV5V","download_json":"https://pith.science/pith/V2XX74F4UQTPVFWN67XAORFV5V.json","view_paper":"https://pith.science/paper/V2XX74F4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.24369&json=true","fetch_graph":"https://pith.science/api/pith-number/V2XX74F4UQTPVFWN67XAORFV5V/graph.json","fetch_events":"https://pith.science/api/pith-number/V2XX74F4UQTPVFWN67XAORFV5V/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V2XX74F4UQTPVFWN67XAORFV5V/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V2XX74F4UQTPVFWN67XAORFV5V/action/storage_attestation","attest_author":"https://pith.science/pith/V2XX74F4UQTPVFWN67XAORFV5V/action/author_attestation","sign_citation":"https://pith.science/pith/V2XX74F4UQTPVFWN67XAORFV5V/action/citation_signature","submit_replication":"https://pith.science/pith/V2XX74F4UQTPVFWN67XAORFV5V/action/replication_record"}},"created_at":"2026-06-24T01:15:28.745437+00:00","updated_at":"2026-06-24T01:15:28.745437+00:00"}