{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:Y7HJXN23UUAPLZ4HJL3Y32LJQP","short_pith_number":"pith:Y7HJXN23","canonical_record":{"source":{"id":"2308.02151","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-08-04T06:14:23Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f4f107b6e22e521642f3b190cd039b994d0a3ee3421b508b9ed77d3c87ced750","abstract_canon_sha256":"3b83c42db64b140fc516df950e2f969c0c406b75625dcb86e49ad75b48bc9826"},"schema_version":"1.0"},"canonical_sha256":"c7ce9bb75ba500f5e7874af78de96983f8cb09aba7bad6eda1ed3de67a36c9ed","source":{"kind":"arxiv","id":"2308.02151","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2308.02151","created_at":"2026-07-05T08:15:38Z"},{"alias_kind":"arxiv_version","alias_value":"2308.02151v3","created_at":"2026-07-05T08:15:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.02151","created_at":"2026-07-05T08:15:38Z"},{"alias_kind":"pith_short_12","alias_value":"Y7HJXN23UUAP","created_at":"2026-07-05T08:15:38Z"},{"alias_kind":"pith_short_16","alias_value":"Y7HJXN23UUAPLZ4H","created_at":"2026-07-05T08:15:38Z"},{"alias_kind":"pith_short_8","alias_value":"Y7HJXN23","created_at":"2026-07-05T08:15:38Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:Y7HJXN23UUAPLZ4HJL3Y32LJQP","target":"record","payload":{"canonical_record":{"source":{"id":"2308.02151","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-08-04T06:14:23Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"f4f107b6e22e521642f3b190cd039b994d0a3ee3421b508b9ed77d3c87ced750","abstract_canon_sha256":"3b83c42db64b140fc516df950e2f969c0c406b75625dcb86e49ad75b48bc9826"},"schema_version":"1.0"},"canonical_sha256":"c7ce9bb75ba500f5e7874af78de96983f8cb09aba7bad6eda1ed3de67a36c9ed","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:15:38.685901Z","signature_b64":"bSgrRtQIlX5UV/Fu8Hj+bLU2H0HxtD5BjiRbSkQLygAs11wpa7DTG0BhgUMJ7hjnR5MUgUHGSup1hPbb1s8vAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c7ce9bb75ba500f5e7874af78de96983f8cb09aba7bad6eda1ed3de67a36c9ed","last_reissued_at":"2026-07-05T08:15:38.685298Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:15:38.685298Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2308.02151","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:15:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"4Y/pHvzq5Kf87k/H76Nfb9pkz441wEzkZd5eNw45tqcH1/LVFA37uf4rS1/ujdJpWNlj48aLvDG+/M81bI5pAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T18:28:22.273029Z"},"content_sha256":"2be8e85e466513d03a5075a68c4f33ac3e400db3b61b374d61ba6781bbd50388","schema_version":"1.0","event_id":"sha256:2be8e85e466513d03a5075a68c4f33ac3e400db3b61b374d61ba6781bbd50388"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:Y7HJXN23UUAPLZ4HJL3Y32LJQP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Retroformer: Retrospective Large Language Agents with Policy Gradient Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Caiming Xiong, Devansh Arpit, Huan Wang, Jianguo Zhang, Juan Carlos Niebles, Le Xue, Phil Mui, Ran Xu, Rithesh Murthy, Shelby Heinecke, Silvio Savarese, Weiran Yao, Yihao Feng, Zeyuan Chen, Zhiwei Liu","submitted_at":"2023-08-04T06:14:23Z","abstract_excerpt":"Recent months have seen the emergence of a powerful new trend in which large language models (LLMs) are augmented to become autonomous language agents capable of performing objective oriented multi-step tasks on their own, rather than merely responding to queries from human users. Most existing language agents, however, are not optimized using environment-specific rewards. Although some agents enable iterative refinement through verbal feedback, they do not reason and plan in ways that are compatible with gradient-based learning from rewards. This paper introduces a principled framework for re"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.02151","kind":"arxiv","version":3},"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/2308.02151/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":"2026-07-05T08:15:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"J41DPsJkWanqpcpJjypJ1ewuP+1RWe0g+7XQHcKnewwDlCJCyChJtULHGq3bsV1An0Zt6F4aq87p6+wGv8EEDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-14T18:28:22.273914Z"},"content_sha256":"a3df8e1605ae13bea3fe6b397d470bdd9060671fbd481c9dc91ca942e12da4bc","schema_version":"1.0","event_id":"sha256:a3df8e1605ae13bea3fe6b397d470bdd9060671fbd481c9dc91ca942e12da4bc"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Y7HJXN23UUAPLZ4HJL3Y32LJQP/bundle.json","state_url":"https://pith.science/pith/Y7HJXN23UUAPLZ4HJL3Y32LJQP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Y7HJXN23UUAPLZ4HJL3Y32LJQP/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":"2026-08-14T18:28:22Z","links":{"resolver":"https://pith.science/pith/Y7HJXN23UUAPLZ4HJL3Y32LJQP","bundle":"https://pith.science/pith/Y7HJXN23UUAPLZ4HJL3Y32LJQP/bundle.json","state":"https://pith.science/pith/Y7HJXN23UUAPLZ4HJL3Y32LJQP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Y7HJXN23UUAPLZ4HJL3Y32LJQP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:Y7HJXN23UUAPLZ4HJL3Y32LJQP","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":"3b83c42db64b140fc516df950e2f969c0c406b75625dcb86e49ad75b48bc9826","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-08-04T06:14:23Z","title_canon_sha256":"f4f107b6e22e521642f3b190cd039b994d0a3ee3421b508b9ed77d3c87ced750"},"schema_version":"1.0","source":{"id":"2308.02151","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2308.02151","created_at":"2026-07-05T08:15:38Z"},{"alias_kind":"arxiv_version","alias_value":"2308.02151v3","created_at":"2026-07-05T08:15:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.02151","created_at":"2026-07-05T08:15:38Z"},{"alias_kind":"pith_short_12","alias_value":"Y7HJXN23UUAP","created_at":"2026-07-05T08:15:38Z"},{"alias_kind":"pith_short_16","alias_value":"Y7HJXN23UUAPLZ4H","created_at":"2026-07-05T08:15:38Z"},{"alias_kind":"pith_short_8","alias_value":"Y7HJXN23","created_at":"2026-07-05T08:15:38Z"}],"graph_snapshots":[{"event_id":"sha256:a3df8e1605ae13bea3fe6b397d470bdd9060671fbd481c9dc91ca942e12da4bc","target":"graph","created_at":"2026-07-05T08:15:38Z","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/2308.02151/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent months have seen the emergence of a powerful new trend in which large language models (LLMs) are augmented to become autonomous language agents capable of performing objective oriented multi-step tasks on their own, rather than merely responding to queries from human users. Most existing language agents, however, are not optimized using environment-specific rewards. Although some agents enable iterative refinement through verbal feedback, they do not reason and plan in ways that are compatible with gradient-based learning from rewards. This paper introduces a principled framework for re","authors_text":"Caiming Xiong, Devansh Arpit, Huan Wang, Jianguo Zhang, Juan Carlos Niebles, Le Xue, Phil Mui, Ran Xu, Rithesh Murthy, Shelby Heinecke, Silvio Savarese, Weiran Yao, Yihao Feng, Zeyuan Chen, Zhiwei Liu","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-08-04T06:14:23Z","title":"Retroformer: Retrospective Large Language Agents with Policy Gradient Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.02151","kind":"arxiv","version":3},"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":"sha256:2be8e85e466513d03a5075a68c4f33ac3e400db3b61b374d61ba6781bbd50388","target":"record","created_at":"2026-07-05T08:15:38Z","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":"3b83c42db64b140fc516df950e2f969c0c406b75625dcb86e49ad75b48bc9826","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-08-04T06:14:23Z","title_canon_sha256":"f4f107b6e22e521642f3b190cd039b994d0a3ee3421b508b9ed77d3c87ced750"},"schema_version":"1.0","source":{"id":"2308.02151","kind":"arxiv","version":3}},"canonical_sha256":"c7ce9bb75ba500f5e7874af78de96983f8cb09aba7bad6eda1ed3de67a36c9ed","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c7ce9bb75ba500f5e7874af78de96983f8cb09aba7bad6eda1ed3de67a36c9ed","first_computed_at":"2026-07-05T08:15:38.685298Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:15:38.685298Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bSgrRtQIlX5UV/Fu8Hj+bLU2H0HxtD5BjiRbSkQLygAs11wpa7DTG0BhgUMJ7hjnR5MUgUHGSup1hPbb1s8vAw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:15:38.685901Z","signed_message":"canonical_sha256_bytes"},"source_id":"2308.02151","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:2be8e85e466513d03a5075a68c4f33ac3e400db3b61b374d61ba6781bbd50388","sha256:a3df8e1605ae13bea3fe6b397d470bdd9060671fbd481c9dc91ca942e12da4bc"],"state_sha256":"e6676e53685280e2f514407f85c848d8a5e75c50f101b6cec530f8882454d38f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"uPTXFJYXVw4SqihkY7+pPZAgoJibWmbH2OnC9EXF1W1hF2Sqmzlzv7bLhtFv3IUqjmIHN1wrMkpS8VrspWGSDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-14T18:28:22.283833Z","bundle_sha256":"b3c351a0956dcdd5a3d3eb387b89805145119f8610d5d9ebf3003ae8cc318c6d"}}