{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:MULZLS7HDBBOPG427ZX6RDYG3O","short_pith_number":"pith:MULZLS7H","canonical_record":{"source":{"id":"2607.26873","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-29T13:03:07Z","cross_cats_sorted":[],"title_canon_sha256":"c04eab36a0bd3750bf7c81d4d163f8bf5f8cbbd6e9b68cda19d16a8801543202","abstract_canon_sha256":"3040e19887fabb366990990589c2ca8271a3a20c1764901eeb3fded84b189eb0"},"schema_version":"1.0"},"canonical_sha256":"651795cbe71842e79b9afe6fe88f06db8dddf7744da7622826199f11d317a364","source":{"kind":"arxiv","id":"2607.26873","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.26873","created_at":"2026-07-30T01:22:50Z"},{"alias_kind":"arxiv_version","alias_value":"2607.26873v1","created_at":"2026-07-30T01:22:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.26873","created_at":"2026-07-30T01:22:50Z"},{"alias_kind":"pith_short_12","alias_value":"MULZLS7HDBBO","created_at":"2026-07-30T01:22:50Z"},{"alias_kind":"pith_short_16","alias_value":"MULZLS7HDBBOPG42","created_at":"2026-07-30T01:22:50Z"},{"alias_kind":"pith_short_8","alias_value":"MULZLS7H","created_at":"2026-07-30T01:22:50Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:MULZLS7HDBBOPG427ZX6RDYG3O","target":"record","payload":{"canonical_record":{"source":{"id":"2607.26873","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-29T13:03:07Z","cross_cats_sorted":[],"title_canon_sha256":"c04eab36a0bd3750bf7c81d4d163f8bf5f8cbbd6e9b68cda19d16a8801543202","abstract_canon_sha256":"3040e19887fabb366990990589c2ca8271a3a20c1764901eeb3fded84b189eb0"},"schema_version":"1.0"},"canonical_sha256":"651795cbe71842e79b9afe6fe88f06db8dddf7744da7622826199f11d317a364","receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"651795cbe71842e79b9afe6fe88f06db8dddf7744da7622826199f11d317a364","last_reissued_at":"2026-07-30T01:22:50.354964Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-30T01:22:50.354964Z"},"source_kind":"arxiv","source_id":"2607.26873","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":"2026-07-30T01:22:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"too9vxrAgt0jQom7V/qFSM8sGmgVctcI5CM5wu1b39GCUQHrHVGkQfPk6OIIlAbeuqqj08hKL2ljiy4qbykCAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T21:26:23.403300Z"},"content_sha256":"973bf945989322c4aa7bd18e03f7a0cb5d1523fb8e14867751f0d69355d8023b","schema_version":"1.0","event_id":"sha256:973bf945989322c4aa7bd18e03f7a0cb5d1523fb8e14867751f0d69355d8023b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:MULZLS7HDBBOPG427ZX6RDYG3O","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"SERPO: Self-Evolving Rubric Policy Optimization for Open-Ended Test-Time Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Hua Yang, Jianze Wang, Jinlong Chen, Kunwang Zheng, Qianglong Chen, Qilong Zhang, Ying Liu, Yu Cao","submitted_at":"2026-07-29T13:03:07Z","abstract_excerpt":"Test-time reinforcement learning (TTRL) enables language models to self-evolve at inference time without labeled feedback. Existing methods rely on answer voting and therefore do not extend naturally to open-ended generation, where valid responses cannot be mapped to a shared canonical answer. Without external reward models or stronger judges, adaptation must instead construct reliable rewards from the model's own outputs. We introduce SERPO (Self-Evolving Rubric Policy Optimization), which replaces answer voting with a closed loop that co-evolves response evidence, query-specific rubrics, and"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.26873","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/2607.26873/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":"2026-07-30T01:22:50Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tfRHLSBMyrMkq/r9JCpYaRPb9vPYMWMJObtTXnSjirdDpZneDKEW+K++IhyGcpBxMSk/M71nIj30RSllMgEYAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-31T21:26:23.403607Z"},"content_sha256":"33353a048fb6d0fd3f1326203b43362377e6593694c4f61c01b3f29f08529980","schema_version":"1.0","event_id":"sha256:33353a048fb6d0fd3f1326203b43362377e6593694c4f61c01b3f29f08529980"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/MULZLS7HDBBOPG427ZX6RDYG3O/bundle.json","state_url":"https://pith.science/pith/MULZLS7HDBBOPG427ZX6RDYG3O/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/MULZLS7HDBBOPG427ZX6RDYG3O/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":"2026-07-31T21:26:23Z","links":{"resolver":"https://pith.science/pith/MULZLS7HDBBOPG427ZX6RDYG3O","bundle":"https://pith.science/pith/MULZLS7HDBBOPG427ZX6RDYG3O/bundle.json","state":"https://pith.science/pith/MULZLS7HDBBOPG427ZX6RDYG3O/state.json","well_known_bundle":"https://pith.science/.well-known/pith/MULZLS7HDBBOPG427ZX6RDYG3O/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:MULZLS7HDBBOPG427ZX6RDYG3O","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":"3040e19887fabb366990990589c2ca8271a3a20c1764901eeb3fded84b189eb0","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-29T13:03:07Z","title_canon_sha256":"c04eab36a0bd3750bf7c81d4d163f8bf5f8cbbd6e9b68cda19d16a8801543202"},"schema_version":"1.0","source":{"id":"2607.26873","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.26873","created_at":"2026-07-30T01:22:50Z"},{"alias_kind":"arxiv_version","alias_value":"2607.26873v1","created_at":"2026-07-30T01:22:50Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.26873","created_at":"2026-07-30T01:22:50Z"},{"alias_kind":"pith_short_12","alias_value":"MULZLS7HDBBO","created_at":"2026-07-30T01:22:50Z"},{"alias_kind":"pith_short_16","alias_value":"MULZLS7HDBBOPG42","created_at":"2026-07-30T01:22:50Z"},{"alias_kind":"pith_short_8","alias_value":"MULZLS7H","created_at":"2026-07-30T01:22:50Z"}],"graph_snapshots":[{"event_id":"sha256:33353a048fb6d0fd3f1326203b43362377e6593694c4f61c01b3f29f08529980","target":"graph","created_at":"2026-07-30T01:22:50Z","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/2607.26873/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Test-time reinforcement learning (TTRL) enables language models to self-evolve at inference time without labeled feedback. Existing methods rely on answer voting and therefore do not extend naturally to open-ended generation, where valid responses cannot be mapped to a shared canonical answer. Without external reward models or stronger judges, adaptation must instead construct reliable rewards from the model's own outputs. We introduce SERPO (Self-Evolving Rubric Policy Optimization), which replaces answer voting with a closed loop that co-evolves response evidence, query-specific rubrics, and","authors_text":"Hua Yang, Jianze Wang, Jinlong Chen, Kunwang Zheng, Qianglong Chen, Qilong Zhang, Ying Liu, Yu Cao","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-29T13:03:07Z","title":"SERPO: Self-Evolving Rubric Policy Optimization for Open-Ended Test-Time Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.26873","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":"sha256:973bf945989322c4aa7bd18e03f7a0cb5d1523fb8e14867751f0d69355d8023b","target":"record","created_at":"2026-07-30T01:22:50Z","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":"3040e19887fabb366990990589c2ca8271a3a20c1764901eeb3fded84b189eb0","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-07-29T13:03:07Z","title_canon_sha256":"c04eab36a0bd3750bf7c81d4d163f8bf5f8cbbd6e9b68cda19d16a8801543202"},"schema_version":"1.0","source":{"id":"2607.26873","kind":"arxiv","version":1}},"canonical_sha256":"651795cbe71842e79b9afe6fe88f06db8dddf7744da7622826199f11d317a364","receipt":{"builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"651795cbe71842e79b9afe6fe88f06db8dddf7744da7622826199f11d317a364","first_computed_at":"2026-07-30T01:22:50.354964Z","kind":"pith_receipt","last_reissued_at":"2026-07-30T01:22:50.354964Z","receipt_version":"0.3","signature_status":"unsigned_v0"},"source_id":"2607.26873","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:973bf945989322c4aa7bd18e03f7a0cb5d1523fb8e14867751f0d69355d8023b","sha256:33353a048fb6d0fd3f1326203b43362377e6593694c4f61c01b3f29f08529980"],"state_sha256":"5394650dc5a93b6e1b61bce706c56cbfa0ce3aa3f31c50a4297b48b29b60f5cf"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"w91XQ1Yh73oNqENcoXqg4p8LM9kj/2T9ZjAVE21k9bpcxX004mhoDovkimvl/P8AemFHdDxp0yR9u8fRDa1aAA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-31T21:26:23.406516Z","bundle_sha256":"42a261308891498d81dcf453a0a08c7a7f57b8eaf33d336d248e48f8d4014927"}}