{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:BCAIFHOP6NLL7UEUZTO5NQ5J2X","short_pith_number":"pith:BCAIFHOP","canonical_record":{"source":{"id":"2605.06139","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-07T12:38:17Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2a779b4493234406e077ac567af69c5e1cdcfa6ec05c6092ab9b0e7fd7c50c8f","abstract_canon_sha256":"3cd68bc954a2a0014707915b20910a779644b64c252b024558bfbda80aa3dd1f"},"schema_version":"1.0"},"canonical_sha256":"0880829dcff356bfd094ccddd6c3a9d5e6d42ed0aff31d36e228015d8e78af88","source":{"kind":"arxiv","id":"2605.06139","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.06139","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"arxiv_version","alias_value":"2605.06139v2","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.06139","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"pith_short_12","alias_value":"BCAIFHOP6NLL","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"pith_short_16","alias_value":"BCAIFHOP6NLL7UEU","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"pith_short_8","alias_value":"BCAIFHOP","created_at":"2026-05-21T01:05:20Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:BCAIFHOP6NLL7UEUZTO5NQ5J2X","target":"record","payload":{"canonical_record":{"source":{"id":"2605.06139","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-07T12:38:17Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2a779b4493234406e077ac567af69c5e1cdcfa6ec05c6092ab9b0e7fd7c50c8f","abstract_canon_sha256":"3cd68bc954a2a0014707915b20910a779644b64c252b024558bfbda80aa3dd1f"},"schema_version":"1.0"},"canonical_sha256":"0880829dcff356bfd094ccddd6c3a9d5e6d42ed0aff31d36e228015d8e78af88","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T01:05:20.390715Z","signature_b64":"x+sbUNzZHcINjHYoIH3PZPGboKhMqKgC0MLSwboqfhK4RW7fciZQIl1EFC+WdyYA0C8M+QT+dvoqgZ3JMLfxAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0880829dcff356bfd094ccddd6c3a9d5e6d42ed0aff31d36e228015d8e78af88","last_reissued_at":"2026-05-21T01:05:20.390028Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T01:05:20.390028Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.06139","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":"2026-05-21T01:05:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OAPO2iBZNcbkGNV5CGerpRAHKFAfK1ciuCtkiHglOlLt30MrWqA0P4F+SyS1H1hqd/K6qqGhAs9QtAcM/sZkAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T03:17:53.740894Z"},"content_sha256":"110790a997e035c730339acdd02187265e583f3cb62a0c1ccfea1c598dc62a9e","schema_version":"1.0","event_id":"sha256:110790a997e035c730339acdd02187265e583f3cb62a0c1ccfea1c598dc62a9e"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:BCAIFHOP6NLL7UEUZTO5NQ5J2X","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Listwise Policy Optimization: Group-based RLVR as Target-Projection on the LLM Response Simplex","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Group-based RL for LLMs implicitly projects policies toward targets on the response simplex; LPO makes the projection explicit and exact.","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Clive Bai, Heming Zou, Kai Yang, Lizhou Cai, Qi Wang, Saiyong Yang, Weijie Liu, Wutong Xu, Xiangyang Ji, Yangkun Chen, Yingyue Li, Yixiu Mao, Yuhang Jiang, Yun Qu","submitted_at":"2026-05-07T12:38:17Z","abstract_excerpt":"Reinforcement learning with verifiable rewards (RLVR) has become a standard approach for large language models (LLMs) post-training to incentivize reasoning capacity. Among existing recipes, group-based policy gradient is prevalent, which samples a group of responses per prompt and updates the policy via group-relative advantage signals. This work reveals that these optimization strategies share a common geometric structure: each implicitly defines a target distribution on the response simplex and projects toward it via first-order approximation. Building on this insight, we propose Listwise P"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"LPO provides (i) monotonic improvement on the listwise objective with bounded, zero-sum, and self-correcting projection gradients, and (ii) flexibility in divergence selection with distinct structural properties through the decoupled projection step.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the implicit targets defined by existing group-based methods can be exactly recovered or improved upon by restricting the proximal RL objective to the response simplex and performing exact divergence minimization.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"LPO reframes group-based RLVR as explicit target-projection on the LLM response simplex and performs exact divergence minimization to achieve monotonic listwise improvement with bounded gradients.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Group-based RL for LLMs implicitly projects policies toward targets on the response simplex; LPO makes the projection explicit and exact.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"2a3211ca39bba360b8e843db855779fe770eacd7c73506edb778bd9392ae86b5"},"source":{"id":"2605.06139","kind":"arxiv","version":2},"verdict":{"id":"6cc46c2c-bc25-4323-9757-eb4547d0eca6","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-08T13:51:51.705037Z","strongest_claim":"LPO provides (i) monotonic improvement on the listwise objective with bounded, zero-sum, and self-correcting projection gradients, and (ii) flexibility in divergence selection with distinct structural properties through the decoupled projection step.","one_line_summary":"LPO reframes group-based RLVR as explicit target-projection on the LLM response simplex and performs exact divergence minimization to achieve monotonic listwise improvement with bounded gradients.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the implicit targets defined by existing group-based methods can be exactly recovered or improved upon by restricting the proximal RL objective to the response simplex and performing exact divergence minimization.","pith_extraction_headline":"Group-based RL for LLMs implicitly projects policies toward targets on the response simplex; LPO makes the projection explicit and exact."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2605.06139/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"claim_evidence","ran_at":"2026-05-20T13:02:04.296347Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"ai_meta_artifact","ran_at":"2026-05-20T08:36:39.607964Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-19T19:01:19.376208Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T12:55:45.727798Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"aa11b66d4e677d0cc7dcc831cb371755b7e60c53f031d482c4143a2cf307e86c"},"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":"6cc46c2c-bc25-4323-9757-eb4547d0eca6"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-21T01:05:20Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"E492IuCj4jYQy5j8/Y1lfjMNFzkNhsp6Qirmh+Z49YxYqZN9v/O3jcav7J5GXGWNdIR4AlLNpJ2ht3F/qHFSAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T03:17:53.741422Z"},"content_sha256":"54d377c230d729755a568675be4a1e2927c6271b0905e4573b522ccc2048f6f4","schema_version":"1.0","event_id":"sha256:54d377c230d729755a568675be4a1e2927c6271b0905e4573b522ccc2048f6f4"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/BCAIFHOP6NLL7UEUZTO5NQ5J2X/bundle.json","state_url":"https://pith.science/pith/BCAIFHOP6NLL7UEUZTO5NQ5J2X/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/BCAIFHOP6NLL7UEUZTO5NQ5J2X/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":"2026-05-31T03:17:53Z","links":{"resolver":"https://pith.science/pith/BCAIFHOP6NLL7UEUZTO5NQ5J2X","bundle":"https://pith.science/pith/BCAIFHOP6NLL7UEUZTO5NQ5J2X/bundle.json","state":"https://pith.science/pith/BCAIFHOP6NLL7UEUZTO5NQ5J2X/state.json","well_known_bundle":"https://pith.science/.well-known/pith/BCAIFHOP6NLL7UEUZTO5NQ5J2X/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:BCAIFHOP6NLL7UEUZTO5NQ5J2X","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":"3cd68bc954a2a0014707915b20910a779644b64c252b024558bfbda80aa3dd1f","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-07T12:38:17Z","title_canon_sha256":"2a779b4493234406e077ac567af69c5e1cdcfa6ec05c6092ab9b0e7fd7c50c8f"},"schema_version":"1.0","source":{"id":"2605.06139","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.06139","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"arxiv_version","alias_value":"2605.06139v2","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.06139","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"pith_short_12","alias_value":"BCAIFHOP6NLL","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"pith_short_16","alias_value":"BCAIFHOP6NLL7UEU","created_at":"2026-05-21T01:05:20Z"},{"alias_kind":"pith_short_8","alias_value":"BCAIFHOP","created_at":"2026-05-21T01:05:20Z"}],"graph_snapshots":[{"event_id":"sha256:54d377c230d729755a568675be4a1e2927c6271b0905e4573b522ccc2048f6f4","target":"graph","created_at":"2026-05-21T01:05:20Z","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":"LPO provides (i) monotonic improvement on the listwise objective with bounded, zero-sum, and self-correcting projection gradients, and (ii) flexibility in divergence selection with distinct structural properties through the decoupled projection step."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the implicit targets defined by existing group-based methods can be exactly recovered or improved upon by restricting the proximal RL objective to the response simplex and performing exact divergence minimization."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"LPO reframes group-based RLVR as explicit target-projection on the LLM response simplex and performs exact divergence minimization to achieve monotonic listwise improvement with bounded gradients."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Group-based RL for LLMs implicitly projects policies toward targets on the response simplex; LPO makes the projection explicit and exact."}],"snapshot_sha256":"2a3211ca39bba360b8e843db855779fe770eacd7c73506edb778bd9392ae86b5"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-20T13:02:04.296347Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-20T08:36:39.607964Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-19T19:01:19.376208Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T12:55:45.727798Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.06139/integrity.json","findings":[],"snapshot_sha256":"aa11b66d4e677d0cc7dcc831cb371755b7e60c53f031d482c4143a2cf307e86c","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reinforcement learning with verifiable rewards (RLVR) has become a standard approach for large language models (LLMs) post-training to incentivize reasoning capacity. Among existing recipes, group-based policy gradient is prevalent, which samples a group of responses per prompt and updates the policy via group-relative advantage signals. This work reveals that these optimization strategies share a common geometric structure: each implicitly defines a target distribution on the response simplex and projects toward it via first-order approximation. Building on this insight, we propose Listwise P","authors_text":"Clive Bai, Heming Zou, Kai Yang, Lizhou Cai, Qi Wang, Saiyong Yang, Weijie Liu, Wutong Xu, Xiangyang Ji, Yangkun Chen, Yingyue Li, Yixiu Mao, Yuhang Jiang, Yun Qu","cross_cats":["cs.AI"],"headline":"Group-based RL for LLMs implicitly projects policies toward targets on the response simplex; LPO makes the projection explicit and exact.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-07T12:38:17Z","title":"Listwise Policy Optimization: Group-based RLVR as Target-Projection on the LLM Response Simplex"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.06139","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-08T13:51:51.705037Z","id":"6cc46c2c-bc25-4323-9757-eb4547d0eca6","model_set":{"reader":"grok-4.3"},"one_line_summary":"LPO reframes group-based RLVR as explicit target-projection on the LLM response simplex and performs exact divergence minimization to achieve monotonic listwise improvement with bounded gradients.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Group-based RL for LLMs implicitly projects policies toward targets on the response simplex; LPO makes the projection explicit and exact.","strongest_claim":"LPO provides (i) monotonic improvement on the listwise objective with bounded, zero-sum, and self-correcting projection gradients, and (ii) flexibility in divergence selection with distinct structural properties through the decoupled projection step.","weakest_assumption":"That the implicit targets defined by existing group-based methods can be exactly recovered or improved upon by restricting the proximal RL objective to the response simplex and performing exact divergence minimization."}},"verdict_id":"6cc46c2c-bc25-4323-9757-eb4547d0eca6"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:110790a997e035c730339acdd02187265e583f3cb62a0c1ccfea1c598dc62a9e","target":"record","created_at":"2026-05-21T01:05:20Z","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":"3cd68bc954a2a0014707915b20910a779644b64c252b024558bfbda80aa3dd1f","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-07T12:38:17Z","title_canon_sha256":"2a779b4493234406e077ac567af69c5e1cdcfa6ec05c6092ab9b0e7fd7c50c8f"},"schema_version":"1.0","source":{"id":"2605.06139","kind":"arxiv","version":2}},"canonical_sha256":"0880829dcff356bfd094ccddd6c3a9d5e6d42ed0aff31d36e228015d8e78af88","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"0880829dcff356bfd094ccddd6c3a9d5e6d42ed0aff31d36e228015d8e78af88","first_computed_at":"2026-05-21T01:05:20.390028Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-21T01:05:20.390028Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"x+sbUNzZHcINjHYoIH3PZPGboKhMqKgC0MLSwboqfhK4RW7fciZQIl1EFC+WdyYA0C8M+QT+dvoqgZ3JMLfxAA==","signature_status":"signed_v1","signed_at":"2026-05-21T01:05:20.390715Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.06139","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:110790a997e035c730339acdd02187265e583f3cb62a0c1ccfea1c598dc62a9e","sha256:54d377c230d729755a568675be4a1e2927c6271b0905e4573b522ccc2048f6f4"],"state_sha256":"55f31b5ee709899464940eb61b180ff3ab44c86ae05860b275eebad99496445f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"23a1Jiv34wsUlO51mNvHkiQDZDBs/6Y8tQxBRWcTT/rTNcedyW7ndFP1hTrctHGLdqMQnKMFSFRdq1LbOftSDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T03:17:53.744049Z","bundle_sha256":"140d06b2c862e1e38aee4556e9bef5bb4a413cff705271e4bc8af9239235dc45"}}