{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:TVH354FXPS5X7FIH3XVUIXLHZ5","short_pith_number":"pith:TVH354FX","canonical_record":{"source":{"id":"2605.26684","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-26T08:23:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"79130d0b80f897eef16a488951d2d0179123bee083e23a8ebd9f9060086fcae3","abstract_canon_sha256":"dc022c4bcfeffb8bfd3d5a868a98876720bb3facd04dc5711a8644a968e5d151"},"schema_version":"1.0"},"canonical_sha256":"9d4fbef0b77cbb7f9507ddeb445d67cf664a2030291340552fad284568a41007","source":{"kind":"arxiv","id":"2605.26684","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.26684","created_at":"2026-05-27T01:06:05Z"},{"alias_kind":"arxiv_version","alias_value":"2605.26684v1","created_at":"2026-05-27T01:06:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.26684","created_at":"2026-05-27T01:06:05Z"},{"alias_kind":"pith_short_12","alias_value":"TVH354FXPS5X","created_at":"2026-05-27T01:06:05Z"},{"alias_kind":"pith_short_16","alias_value":"TVH354FXPS5X7FIH","created_at":"2026-05-27T01:06:05Z"},{"alias_kind":"pith_short_8","alias_value":"TVH354FX","created_at":"2026-05-27T01:06:05Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:TVH354FXPS5X7FIH3XVUIXLHZ5","target":"record","payload":{"canonical_record":{"source":{"id":"2605.26684","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-26T08:23:34Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"79130d0b80f897eef16a488951d2d0179123bee083e23a8ebd9f9060086fcae3","abstract_canon_sha256":"dc022c4bcfeffb8bfd3d5a868a98876720bb3facd04dc5711a8644a968e5d151"},"schema_version":"1.0"},"canonical_sha256":"9d4fbef0b77cbb7f9507ddeb445d67cf664a2030291340552fad284568a41007","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-27T01:06:05.985107Z","signature_b64":"L66pNuRslWqzWLb0xO1cXwtNVpPGTTps0uqcYIvADxx8Uqqcqe9+4j3Q+9nknZR0Ke0FsGUbAMxWEMaV49QEDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9d4fbef0b77cbb7f9507ddeb445d67cf664a2030291340552fad284568a41007","last_reissued_at":"2026-05-27T01:06:05.984212Z","signature_status":"signed_v1","first_computed_at":"2026-05-27T01:06:05.984212Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.26684","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":"2026-05-27T01:06:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nYLPBB75N5Km6+poiaGgOmPB1kmivCwqRDpoIYfCfURqZrX9YjUuYiIPLCBZ7xiecjCeXEm2HOinrL36OqjlAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T12:14:00.412832Z"},"content_sha256":"427f18e00d8bd37df8beb8dfda001fe75f770d2a371718769d191e0fae7c0a85","schema_version":"1.0","event_id":"sha256:427f18e00d8bd37df8beb8dfda001fe75f770d2a371718769d191e0fae7c0a85"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:TVH354FXPS5X7FIH3XVUIXLHZ5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Beyond Trajectory-Level Attribution: Graph-Based Credit Assignment for Agentic Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Bo An, Haiyang Xu, Lang Feng, Lei Feng, Ming Yan, Shuo He, Xin Cheng","submitted_at":"2026-05-26T08:23:34Z","abstract_excerpt":"Group-based reinforcement learning (RL) methods have achieved remarkable success in improving the performance of large language models (LLMs) and have been rapidly extended to agentic tasks. However, their credit assignment relies heavily on coarse-grained trajectory-level attribution according to final outcomes, making it difficult to capture the contribution of individual steps, such as valuable steps obscured within failed trajectories. To uncover latent information and enable more faithful step-level credit assignment, we propose Graph-based Group Policy Optimization (GraphGPO), which firs"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.26684","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/2605.26684/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":"2026-05-27T01:06:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TPQE6+uT8ituJ/4kF0bu9yos5ceINZ7CLzJTlo/YemqLm9gEDdTqkf6CTGLoDI2BR2W9QeqK+u3ueoK1+gSHCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-01T12:14:00.413210Z"},"content_sha256":"cf0a87ad300c67ac425c3d5252797f414a8eb77380a2e5ecd0a3d8c98b5bd762","schema_version":"1.0","event_id":"sha256:cf0a87ad300c67ac425c3d5252797f414a8eb77380a2e5ecd0a3d8c98b5bd762"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/TVH354FXPS5X7FIH3XVUIXLHZ5/bundle.json","state_url":"https://pith.science/pith/TVH354FXPS5X7FIH3XVUIXLHZ5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/TVH354FXPS5X7FIH3XVUIXLHZ5/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":"2026-06-01T12:14:00Z","links":{"resolver":"https://pith.science/pith/TVH354FXPS5X7FIH3XVUIXLHZ5","bundle":"https://pith.science/pith/TVH354FXPS5X7FIH3XVUIXLHZ5/bundle.json","state":"https://pith.science/pith/TVH354FXPS5X7FIH3XVUIXLHZ5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/TVH354FXPS5X7FIH3XVUIXLHZ5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:TVH354FXPS5X7FIH3XVUIXLHZ5","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":"dc022c4bcfeffb8bfd3d5a868a98876720bb3facd04dc5711a8644a968e5d151","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-26T08:23:34Z","title_canon_sha256":"79130d0b80f897eef16a488951d2d0179123bee083e23a8ebd9f9060086fcae3"},"schema_version":"1.0","source":{"id":"2605.26684","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.26684","created_at":"2026-05-27T01:06:05Z"},{"alias_kind":"arxiv_version","alias_value":"2605.26684v1","created_at":"2026-05-27T01:06:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.26684","created_at":"2026-05-27T01:06:05Z"},{"alias_kind":"pith_short_12","alias_value":"TVH354FXPS5X","created_at":"2026-05-27T01:06:05Z"},{"alias_kind":"pith_short_16","alias_value":"TVH354FXPS5X7FIH","created_at":"2026-05-27T01:06:05Z"},{"alias_kind":"pith_short_8","alias_value":"TVH354FX","created_at":"2026-05-27T01:06:05Z"}],"graph_snapshots":[{"event_id":"sha256:cf0a87ad300c67ac425c3d5252797f414a8eb77380a2e5ecd0a3d8c98b5bd762","target":"graph","created_at":"2026-05-27T01:06:05Z","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/2605.26684/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Group-based reinforcement learning (RL) methods have achieved remarkable success in improving the performance of large language models (LLMs) and have been rapidly extended to agentic tasks. However, their credit assignment relies heavily on coarse-grained trajectory-level attribution according to final outcomes, making it difficult to capture the contribution of individual steps, such as valuable steps obscured within failed trajectories. To uncover latent information and enable more faithful step-level credit assignment, we propose Graph-based Group Policy Optimization (GraphGPO), which firs","authors_text":"Bo An, Haiyang Xu, Lang Feng, Lei Feng, Ming Yan, Shuo He, Xin Cheng","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-26T08:23:34Z","title":"Beyond Trajectory-Level Attribution: Graph-Based Credit Assignment for Agentic Reinforcement Learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.26684","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":"sha256:427f18e00d8bd37df8beb8dfda001fe75f770d2a371718769d191e0fae7c0a85","target":"record","created_at":"2026-05-27T01:06:05Z","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":"dc022c4bcfeffb8bfd3d5a868a98876720bb3facd04dc5711a8644a968e5d151","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2026-05-26T08:23:34Z","title_canon_sha256":"79130d0b80f897eef16a488951d2d0179123bee083e23a8ebd9f9060086fcae3"},"schema_version":"1.0","source":{"id":"2605.26684","kind":"arxiv","version":1}},"canonical_sha256":"9d4fbef0b77cbb7f9507ddeb445d67cf664a2030291340552fad284568a41007","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9d4fbef0b77cbb7f9507ddeb445d67cf664a2030291340552fad284568a41007","first_computed_at":"2026-05-27T01:06:05.984212Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-27T01:06:05.984212Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"L66pNuRslWqzWLb0xO1cXwtNVpPGTTps0uqcYIvADxx8Uqqcqe9+4j3Q+9nknZR0Ke0FsGUbAMxWEMaV49QEDQ==","signature_status":"signed_v1","signed_at":"2026-05-27T01:06:05.985107Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.26684","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:427f18e00d8bd37df8beb8dfda001fe75f770d2a371718769d191e0fae7c0a85","sha256:cf0a87ad300c67ac425c3d5252797f414a8eb77380a2e5ecd0a3d8c98b5bd762"],"state_sha256":"a5075230d239c142c76adb395598fdb4f405d26d45fb4276087fd6cf2ac8b587"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"NGU17gesv6HUCp7SmlDNf5FD1C+PW2uIFQVIo3u1dyzVW4cib96+QLw9ZHpwWCHvFsn+H5aw55VMhaCNXMWkBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-01T12:14:00.415827Z","bundle_sha256":"7d8372a2fa19ec8eae5ff3e31b8e2b81a6fa8ab1d23be4a681c336d05fc362d3"}}