{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:CGHLCSAZJIGR3PRSX24QUY47IR","short_pith_number":"pith:CGHLCSAZ","canonical_record":{"source":{"id":"2610.18216","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-10-13T11:47:32Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"31e09bb674285057cdabf9484479abd79485d3a44069242957a2c298f549faa1","abstract_canon_sha256":"713dd12d5218ef2b47e7fc7fc85b0f7de5b94681370e00518c6906a9c0955427"},"schema_version":"1.0"},"canonical_sha256":"118eb148194a0d1dbe32beb90a639f44518615a9c8fb8e1a72d257591b378a6d","source":{"kind":"arxiv","id":"2610.18216","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2610.18216","created_at":"2026-10-14T04:30:35Z"},{"alias_kind":"arxiv_version","alias_value":"2610.18216v1","created_at":"2026-10-14T04:30:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2610.18216","created_at":"2026-10-14T04:30:35Z"},{"alias_kind":"pith_short_12","alias_value":"CGHLCSAZJIGR","created_at":"2026-10-14T04:30:35Z"},{"alias_kind":"pith_short_16","alias_value":"CGHLCSAZJIGR3PRS","created_at":"2026-10-14T04:30:35Z"},{"alias_kind":"pith_short_8","alias_value":"CGHLCSAZ","created_at":"2026-10-14T04:30:35Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:CGHLCSAZJIGR3PRSX24QUY47IR","target":"record","payload":{"canonical_record":{"source":{"id":"2610.18216","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-10-13T11:47:32Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"31e09bb674285057cdabf9484479abd79485d3a44069242957a2c298f549faa1","abstract_canon_sha256":"713dd12d5218ef2b47e7fc7fc85b0f7de5b94681370e00518c6906a9c0955427"},"schema_version":"1.0"},"canonical_sha256":"118eb148194a0d1dbe32beb90a639f44518615a9c8fb8e1a72d257591b378a6d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-10-14T04:30:35.304717Z","signature_b64":"mGvO1g1kHLpV0C8oEKGCEWkiqfxRsNYU+BrQXXnURVMa1cMg3pp9LIzyTd5HyYuBdFaa1SiQFPws6cvFdzTkAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"118eb148194a0d1dbe32beb90a639f44518615a9c8fb8e1a72d257591b378a6d","last_reissued_at":"2026-10-14T04:30:35.302643Z","signature_status":"signed_v1","first_computed_at":"2026-10-14T04:30:35.302643Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2610.18216","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":"2026-10-14T04:30:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"fqbJTCHlGUvzjI2G0AHDFG3Elyx24hx4pW3TMF5o1ZJr4whfLIj4umpkUzWICCM6XACkwzg0wxx1i6s/7BmpBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-11-22T01:32:10.623587Z"},"content_sha256":"04006c455cfb9f766145493b1405f53486f4d9452a168a5b91a92e94fa7471e8","schema_version":"1.0","event_id":"sha256:04006c455cfb9f766145493b1405f53486f4d9452a168a5b91a92e94fa7471e8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:CGHLCSAZJIGR3PRSX24QUY47IR","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"TextualRL: Textual Reinforcement Learning for Context Management","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bo Mao, Jie Zhou, Liang He, Qin Chen, Tianyu Huai, Xipeng Qiu","submitted_at":"2026-10-13T11:47:32Z","abstract_excerpt":"LLM agents can adapt to task feedback by revising the textual context that guides their behavior, without changing their model parameters. However, turning execution rewards into reusable context updates is challenging: individual trajectories offer limited comparative evidence about what to change, while task-specific fixes may overgeneralize when incorporated into a shared context. We introduce TextualRL, a gradient-free reinforcement learning framework that treats a group of rollouts from the same task under the same context, rather than an isolated trajectory, as the unit of policy critiqu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2610.18216","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/2610.18216/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"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":"2026-10-14T04:30:35Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Nak0KNn0+FBBrs5hLu0s8Zr5gIRaFPDzUwuLojsktyX0SbrkHlB8HerunfMNjNI76I7fK4NwGnyqlvvj6F2ZAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-11-22T01:32:10.624339Z"},"content_sha256":"b230f4fa21b61dceb620c9df63f8ee1f392f771f3ba82530d92d3dc6ec78daf0","schema_version":"1.0","event_id":"sha256:b230f4fa21b61dceb620c9df63f8ee1f392f771f3ba82530d92d3dc6ec78daf0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CGHLCSAZJIGR3PRSX24QUY47IR/bundle.json","state_url":"https://pith.science/pith/CGHLCSAZJIGR3PRSX24QUY47IR/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CGHLCSAZJIGR3PRSX24QUY47IR/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":"2026-11-22T01:32:10Z","links":{"resolver":"https://pith.science/pith/CGHLCSAZJIGR3PRSX24QUY47IR","bundle":"https://pith.science/pith/CGHLCSAZJIGR3PRSX24QUY47IR/bundle.json","state":"https://pith.science/pith/CGHLCSAZJIGR3PRSX24QUY47IR/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CGHLCSAZJIGR3PRSX24QUY47IR/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:CGHLCSAZJIGR3PRSX24QUY47IR","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":"713dd12d5218ef2b47e7fc7fc85b0f7de5b94681370e00518c6906a9c0955427","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-10-13T11:47:32Z","title_canon_sha256":"31e09bb674285057cdabf9484479abd79485d3a44069242957a2c298f549faa1"},"schema_version":"1.0","source":{"id":"2610.18216","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2610.18216","created_at":"2026-10-14T04:30:35Z"},{"alias_kind":"arxiv_version","alias_value":"2610.18216v1","created_at":"2026-10-14T04:30:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2610.18216","created_at":"2026-10-14T04:30:35Z"},{"alias_kind":"pith_short_12","alias_value":"CGHLCSAZJIGR","created_at":"2026-10-14T04:30:35Z"},{"alias_kind":"pith_short_16","alias_value":"CGHLCSAZJIGR3PRS","created_at":"2026-10-14T04:30:35Z"},{"alias_kind":"pith_short_8","alias_value":"CGHLCSAZ","created_at":"2026-10-14T04:30:35Z"}],"graph_snapshots":[{"event_id":"sha256:b230f4fa21b61dceb620c9df63f8ee1f392f771f3ba82530d92d3dc6ec78daf0","target":"graph","created_at":"2026-10-14T04:30:35Z","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"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2610.18216/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"LLM agents can adapt to task feedback by revising the textual context that guides their behavior, without changing their model parameters. However, turning execution rewards into reusable context updates is challenging: individual trajectories offer limited comparative evidence about what to change, while task-specific fixes may overgeneralize when incorporated into a shared context. We introduce TextualRL, a gradient-free reinforcement learning framework that treats a group of rollouts from the same task under the same context, rather than an isolated trajectory, as the unit of policy critiqu","authors_text":"Bo Mao, Jie Zhou, Liang He, Qin Chen, Tianyu Huai, Xipeng Qiu","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-10-13T11:47:32Z","title":"TextualRL: Textual Reinforcement Learning for Context Management"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2610.18216","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":"sha256:04006c455cfb9f766145493b1405f53486f4d9452a168a5b91a92e94fa7471e8","target":"record","created_at":"2026-10-14T04:30:35Z","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":"713dd12d5218ef2b47e7fc7fc85b0f7de5b94681370e00518c6906a9c0955427","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-10-13T11:47:32Z","title_canon_sha256":"31e09bb674285057cdabf9484479abd79485d3a44069242957a2c298f549faa1"},"schema_version":"1.0","source":{"id":"2610.18216","kind":"arxiv","version":1}},"canonical_sha256":"118eb148194a0d1dbe32beb90a639f44518615a9c8fb8e1a72d257591b378a6d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"118eb148194a0d1dbe32beb90a639f44518615a9c8fb8e1a72d257591b378a6d","first_computed_at":"2026-10-14T04:30:35.302643Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-10-14T04:30:35.302643Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mGvO1g1kHLpV0C8oEKGCEWkiqfxRsNYU+BrQXXnURVMa1cMg3pp9LIzyTd5HyYuBdFaa1SiQFPws6cvFdzTkAQ==","signature_status":"signed_v1","signed_at":"2026-10-14T04:30:35.304717Z","signed_message":"canonical_sha256_bytes"},"source_id":"2610.18216","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:04006c455cfb9f766145493b1405f53486f4d9452a168a5b91a92e94fa7471e8","sha256:b230f4fa21b61dceb620c9df63f8ee1f392f771f3ba82530d92d3dc6ec78daf0"],"state_sha256":"d00c3646c76b983be6938a2d8fc560d1ffe7557b44e32766d5dd2238d1143187"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HljWTmzRsOXklvr6C0KbMW5jtoa3ty79H6gM3vaixzhXiwxkLZi1PuLQpCb3IynI5cuQQCo4rlyAVEg+0e10DQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-11-22T01:32:10.638797Z","bundle_sha256":"0c56e8e514cc800a7a5446836d189d4bea44eaffd9d08729eacacda8578b362d"}}