{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:KA2Y4KY4PZSVYZEQN3EH7U3DP6","short_pith_number":"pith:KA2Y4KY4","canonical_record":{"source":{"id":"2502.04686","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-07T06:19:55Z","cross_cats_sorted":[],"title_canon_sha256":"7507b9a21d0a19ffe1b6a0f212c766d2e0d9df35423eb417fed5ed43725695ee","abstract_canon_sha256":"7a6ec4ce6b7b11bc168985a98444db399e510288ee3e9a00b4509d20725e649b"},"schema_version":"1.0"},"canonical_sha256":"50358e2b1c7e655c64906ec87fd3637f8a77c8e7a375f1d2251e70f29b77a199","source":{"kind":"arxiv","id":"2502.04686","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.04686","created_at":"2026-07-05T11:23:23Z"},{"alias_kind":"arxiv_version","alias_value":"2502.04686v3","created_at":"2026-07-05T11:23:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.04686","created_at":"2026-07-05T11:23:23Z"},{"alias_kind":"pith_short_12","alias_value":"KA2Y4KY4PZSV","created_at":"2026-07-05T11:23:23Z"},{"alias_kind":"pith_short_16","alias_value":"KA2Y4KY4PZSVYZEQ","created_at":"2026-07-05T11:23:23Z"},{"alias_kind":"pith_short_8","alias_value":"KA2Y4KY4","created_at":"2026-07-05T11:23:23Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:KA2Y4KY4PZSVYZEQN3EH7U3DP6","target":"record","payload":{"canonical_record":{"source":{"id":"2502.04686","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-07T06:19:55Z","cross_cats_sorted":[],"title_canon_sha256":"7507b9a21d0a19ffe1b6a0f212c766d2e0d9df35423eb417fed5ed43725695ee","abstract_canon_sha256":"7a6ec4ce6b7b11bc168985a98444db399e510288ee3e9a00b4509d20725e649b"},"schema_version":"1.0"},"canonical_sha256":"50358e2b1c7e655c64906ec87fd3637f8a77c8e7a375f1d2251e70f29b77a199","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:23:23.188549Z","signature_b64":"2q4fW1+SnbE7400vuDjGgftUFiqRffeTfD5AQUwBJh0wa2tHaWk+y5hJSps2OQwHJuhliXQgCA7MGX6bi9XSDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"50358e2b1c7e655c64906ec87fd3637f8a77c8e7a375f1d2251e70f29b77a199","last_reissued_at":"2026-07-05T11:23:23.187999Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:23:23.187999Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.04686","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":"2026-07-05T11:23:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OKw7jeQVi8seKpvbUGHYJgBbMjWrtKeAICJlL82QIAvqzYNETMdg7aLoIujDBhQPDDgF3AzYkQu+JzGRBMi9AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T08:04:59.928575Z"},"content_sha256":"19c46dcf28da3e904acd9cb851dd14ba807f3f68d022e2c21db031fa63e60fe8","schema_version":"1.0","event_id":"sha256:19c46dcf28da3e904acd9cb851dd14ba807f3f68d022e2c21db031fa63e60fe8"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:KA2Y4KY4PZSVYZEQN3EH7U3DP6","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Learning Strategic Language Agents in the Werewolf Game with Iterative Latent Space Policy Optimization","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Chao Yu, Wanjun Gu, Yi Wu, Yu Wang, Zelai Xu","submitted_at":"2025-02-07T06:19:55Z","abstract_excerpt":"Large language model (LLM) agents have recently demonstrated impressive capabilities in various domains like open-ended conversation and multi-step decision-making. However, it remains challenging for these agents to solve strategic language games, such as Werewolf, which demand both strategic decision-making and free-form language interactions. Existing LLM agents often suffer from intrinsic bias in their action distributions and limited exploration of the unbounded text action space, resulting in suboptimal performance. To address these challenges, we propose Latent Space Policy Optimization"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.04686","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/2502.04686/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":"2026-07-05T11:23:23Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hfYcn1KySrULedxXZD8ZJxIwx7+nfQImyS1jyQmlMI0W1qO4dHfXBz6eh0RdMJEToO80dUxbJH8UY1Wy1+ZxCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T08:04:59.929088Z"},"content_sha256":"433265e69904ef9f247daab3b3411416634bf50c9ff8a1aa11d350cd71106247","schema_version":"1.0","event_id":"sha256:433265e69904ef9f247daab3b3411416634bf50c9ff8a1aa11d350cd71106247"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/KA2Y4KY4PZSVYZEQN3EH7U3DP6/bundle.json","state_url":"https://pith.science/pith/KA2Y4KY4PZSVYZEQN3EH7U3DP6/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/KA2Y4KY4PZSVYZEQN3EH7U3DP6/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":"2026-08-04T08:04:59Z","links":{"resolver":"https://pith.science/pith/KA2Y4KY4PZSVYZEQN3EH7U3DP6","bundle":"https://pith.science/pith/KA2Y4KY4PZSVYZEQN3EH7U3DP6/bundle.json","state":"https://pith.science/pith/KA2Y4KY4PZSVYZEQN3EH7U3DP6/state.json","well_known_bundle":"https://pith.science/.well-known/pith/KA2Y4KY4PZSVYZEQN3EH7U3DP6/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:KA2Y4KY4PZSVYZEQN3EH7U3DP6","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":"7a6ec4ce6b7b11bc168985a98444db399e510288ee3e9a00b4509d20725e649b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-07T06:19:55Z","title_canon_sha256":"7507b9a21d0a19ffe1b6a0f212c766d2e0d9df35423eb417fed5ed43725695ee"},"schema_version":"1.0","source":{"id":"2502.04686","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.04686","created_at":"2026-07-05T11:23:23Z"},{"alias_kind":"arxiv_version","alias_value":"2502.04686v3","created_at":"2026-07-05T11:23:23Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.04686","created_at":"2026-07-05T11:23:23Z"},{"alias_kind":"pith_short_12","alias_value":"KA2Y4KY4PZSV","created_at":"2026-07-05T11:23:23Z"},{"alias_kind":"pith_short_16","alias_value":"KA2Y4KY4PZSVYZEQ","created_at":"2026-07-05T11:23:23Z"},{"alias_kind":"pith_short_8","alias_value":"KA2Y4KY4","created_at":"2026-07-05T11:23:23Z"}],"graph_snapshots":[{"event_id":"sha256:433265e69904ef9f247daab3b3411416634bf50c9ff8a1aa11d350cd71106247","target":"graph","created_at":"2026-07-05T11:23:23Z","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/2502.04686/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language model (LLM) agents have recently demonstrated impressive capabilities in various domains like open-ended conversation and multi-step decision-making. However, it remains challenging for these agents to solve strategic language games, such as Werewolf, which demand both strategic decision-making and free-form language interactions. Existing LLM agents often suffer from intrinsic bias in their action distributions and limited exploration of the unbounded text action space, resulting in suboptimal performance. To address these challenges, we propose Latent Space Policy Optimization","authors_text":"Chao Yu, Wanjun Gu, Yi Wu, Yu Wang, Zelai Xu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-07T06:19:55Z","title":"Learning Strategic Language Agents in the Werewolf Game with Iterative Latent Space Policy Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.04686","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":"sha256:19c46dcf28da3e904acd9cb851dd14ba807f3f68d022e2c21db031fa63e60fe8","target":"record","created_at":"2026-07-05T11:23:23Z","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":"7a6ec4ce6b7b11bc168985a98444db399e510288ee3e9a00b4509d20725e649b","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2025-02-07T06:19:55Z","title_canon_sha256":"7507b9a21d0a19ffe1b6a0f212c766d2e0d9df35423eb417fed5ed43725695ee"},"schema_version":"1.0","source":{"id":"2502.04686","kind":"arxiv","version":3}},"canonical_sha256":"50358e2b1c7e655c64906ec87fd3637f8a77c8e7a375f1d2251e70f29b77a199","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"50358e2b1c7e655c64906ec87fd3637f8a77c8e7a375f1d2251e70f29b77a199","first_computed_at":"2026-07-05T11:23:23.187999Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:23:23.187999Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2q4fW1+SnbE7400vuDjGgftUFiqRffeTfD5AQUwBJh0wa2tHaWk+y5hJSps2OQwHJuhliXQgCA7MGX6bi9XSDg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:23:23.188549Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.04686","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:19c46dcf28da3e904acd9cb851dd14ba807f3f68d022e2c21db031fa63e60fe8","sha256:433265e69904ef9f247daab3b3411416634bf50c9ff8a1aa11d350cd71106247"],"state_sha256":"cd48bf6ba4ec1c29141d6cf1ce3e654723adf79210415a30110cac883ef0e850"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"akl3o2dJFmfF0QPtca7fhImiqLdrOI3mbPZPbJNKqsU2l5BTCXcRvLJlorRElBwitVQ7fLsd/jEBoY3oEanvBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T08:04:59.933089Z","bundle_sha256":"b0b42ce31eab0425d95a8f30edc4a5ad77c20a0cfa2ce0561201e0eea029a051"}}