{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:OFXK4YJZV636PJQVHLWDIXGNRP","short_pith_number":"pith:OFXK4YJZ","canonical_record":{"source":{"id":"2607.07508","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-08T15:02:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8f2a275f3755ef0674d8e6d9770677b0961bb54c288140853ab58e8d5f3da9cb","abstract_canon_sha256":"d2eff236798f1e53ae28065c651f03618bda8753cb7286a98ed109cfaffc295e"},"schema_version":"1.0"},"canonical_sha256":"716eae6139afb7e7a6153aec345ccd8be021244bafcde8a8c024486d496242bb","source":{"kind":"arxiv","id":"2607.07508","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.07508","created_at":"2026-07-09T01:20:27Z"},{"alias_kind":"arxiv_version","alias_value":"2607.07508v1","created_at":"2026-07-09T01:20:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.07508","created_at":"2026-07-09T01:20:27Z"},{"alias_kind":"pith_short_12","alias_value":"OFXK4YJZV636","created_at":"2026-07-09T01:20:27Z"},{"alias_kind":"pith_short_16","alias_value":"OFXK4YJZV636PJQV","created_at":"2026-07-09T01:20:27Z"},{"alias_kind":"pith_short_8","alias_value":"OFXK4YJZ","created_at":"2026-07-09T01:20:27Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:OFXK4YJZV636PJQVHLWDIXGNRP","target":"record","payload":{"canonical_record":{"source":{"id":"2607.07508","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-08T15:02:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8f2a275f3755ef0674d8e6d9770677b0961bb54c288140853ab58e8d5f3da9cb","abstract_canon_sha256":"d2eff236798f1e53ae28065c651f03618bda8753cb7286a98ed109cfaffc295e"},"schema_version":"1.0"},"canonical_sha256":"716eae6139afb7e7a6153aec345ccd8be021244bafcde8a8c024486d496242bb","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-09T01:20:27.611757Z","signature_b64":"wONyD9SSlEZoqjSrRcqGH0H3mx44vhbj7jyZggpMgJOv1yi58+kpdeWRawW3PXoCzqg4z5bb7OuZQ3vIx/FjBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"716eae6139afb7e7a6153aec345ccd8be021244bafcde8a8c024486d496242bb","last_reissued_at":"2026-07-09T01:20:27.611346Z","signature_status":"signed_v1","first_computed_at":"2026-07-09T01:20:27.611346Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.07508","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-09T01:20:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Rn6aex7J4pchTWh2YNudjzm4p380d3rPLmG0CsRh+9Rv7CyCEbReim2q0aCMxlQ23JDHfDHiudh+WMCb6W+FCA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:15:06.138659Z"},"content_sha256":"f9168bf77340b67a282299dbbd70f980293224db7a15122274497f8b4fd42236","schema_version":"1.0","event_id":"sha256:f9168bf77340b67a282299dbbd70f980293224db7a15122274497f8b4fd42236"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:OFXK4YJZV636PJQVHLWDIXGNRP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Single-Rollout Asynchronous Optimization for Agentic Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jie Tang, Yujiang Li, Yuxiao Dong, Zhenyu Hou","submitted_at":"2026-07-08T15:02:19Z","abstract_excerpt":"Reinforcement learning (RL) is becoming increasingly important for post-training large language models (LLMs). Previous RL pipelines for LLMs were mostly synchronous and batch-interleaved, which is inefficient for long-horizon agentic tasks. Recently, asynchronous RL has emerged as a more efficient alternative by updating the model as rollouts arrive. However, existing asynchronous RL systems often emphasize throughput, while leaving training stability and task effectiveness largely underexplored. For example, a key challenge is that group-wise sampling in the widely-used GRPO framework does n"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.07508","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/2607.07508/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":"2026-07-09T01:20:27Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"MFfnZgN7u4YWWatcT4D/0cZGvu46JxGViJulCCYHC48LDZ9IuRbAAc2FPMiKV86MHazUIvFdxp9IOnYORFBLDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-03T16:15:06.139387Z"},"content_sha256":"2260d8f793a5c367e185549ea610727e8409fe8efc63d5c4675663552f43c022","schema_version":"1.0","event_id":"sha256:2260d8f793a5c367e185549ea610727e8409fe8efc63d5c4675663552f43c022"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/OFXK4YJZV636PJQVHLWDIXGNRP/bundle.json","state_url":"https://pith.science/pith/OFXK4YJZV636PJQVHLWDIXGNRP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/OFXK4YJZV636PJQVHLWDIXGNRP/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":"2026-08-03T16:15:06Z","links":{"resolver":"https://pith.science/pith/OFXK4YJZV636PJQVHLWDIXGNRP","bundle":"https://pith.science/pith/OFXK4YJZV636PJQVHLWDIXGNRP/bundle.json","state":"https://pith.science/pith/OFXK4YJZV636PJQVHLWDIXGNRP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/OFXK4YJZV636PJQVHLWDIXGNRP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:OFXK4YJZV636PJQVHLWDIXGNRP","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":"d2eff236798f1e53ae28065c651f03618bda8753cb7286a98ed109cfaffc295e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-08T15:02:19Z","title_canon_sha256":"8f2a275f3755ef0674d8e6d9770677b0961bb54c288140853ab58e8d5f3da9cb"},"schema_version":"1.0","source":{"id":"2607.07508","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.07508","created_at":"2026-07-09T01:20:27Z"},{"alias_kind":"arxiv_version","alias_value":"2607.07508v1","created_at":"2026-07-09T01:20:27Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.07508","created_at":"2026-07-09T01:20:27Z"},{"alias_kind":"pith_short_12","alias_value":"OFXK4YJZV636","created_at":"2026-07-09T01:20:27Z"},{"alias_kind":"pith_short_16","alias_value":"OFXK4YJZV636PJQV","created_at":"2026-07-09T01:20:27Z"},{"alias_kind":"pith_short_8","alias_value":"OFXK4YJZ","created_at":"2026-07-09T01:20:27Z"}],"graph_snapshots":[{"event_id":"sha256:2260d8f793a5c367e185549ea610727e8409fe8efc63d5c4675663552f43c022","target":"graph","created_at":"2026-07-09T01:20:27Z","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/2607.07508/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning (RL) is becoming increasingly important for post-training large language models (LLMs). Previous RL pipelines for LLMs were mostly synchronous and batch-interleaved, which is inefficient for long-horizon agentic tasks. Recently, asynchronous RL has emerged as a more efficient alternative by updating the model as rollouts arrive. However, existing asynchronous RL systems often emphasize throughput, while leaving training stability and task effectiveness largely underexplored. For example, a key challenge is that group-wise sampling in the widely-used GRPO framework does n","authors_text":"Jie Tang, Yujiang Li, Yuxiao Dong, Zhenyu Hou","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-08T15:02:19Z","title":"Single-Rollout Asynchronous Optimization for Agentic Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.07508","kind":"arxiv","version":1},"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:f9168bf77340b67a282299dbbd70f980293224db7a15122274497f8b4fd42236","target":"record","created_at":"2026-07-09T01:20:27Z","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":"d2eff236798f1e53ae28065c651f03618bda8753cb7286a98ed109cfaffc295e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-07-08T15:02:19Z","title_canon_sha256":"8f2a275f3755ef0674d8e6d9770677b0961bb54c288140853ab58e8d5f3da9cb"},"schema_version":"1.0","source":{"id":"2607.07508","kind":"arxiv","version":1}},"canonical_sha256":"716eae6139afb7e7a6153aec345ccd8be021244bafcde8a8c024486d496242bb","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"716eae6139afb7e7a6153aec345ccd8be021244bafcde8a8c024486d496242bb","first_computed_at":"2026-07-09T01:20:27.611346Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-09T01:20:27.611346Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"wONyD9SSlEZoqjSrRcqGH0H3mx44vhbj7jyZggpMgJOv1yi58+kpdeWRawW3PXoCzqg4z5bb7OuZQ3vIx/FjBw==","signature_status":"signed_v1","signed_at":"2026-07-09T01:20:27.611757Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.07508","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:f9168bf77340b67a282299dbbd70f980293224db7a15122274497f8b4fd42236","sha256:2260d8f793a5c367e185549ea610727e8409fe8efc63d5c4675663552f43c022"],"state_sha256":"43497e20986fb9bc6e312a89c1ef56ee92ff294ccfeb21e0c5057f0a4cf0b860"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"cBQCyY9LFCc5d4pRYbjOhAl/fD6PINUCLD5ooelJQP+ACsGjv8LmxkoUoQmP8SsHP6CgPme6NqK1PXhmsNJkCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-03T16:15:06.147331Z","bundle_sha256":"46061ad5e7924a31e6d819004859eecb468a03aba27c3bad5fddd5b121124aac"}}