{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:7HS36GOIZU26I5KRG2LITUXWU2","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":"bd8115472b4131024766a5ea6556e9f1cd47161d2634653ff8791823636d8e32","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-08-14T05:55:42Z","title_canon_sha256":"5db17abd59502e3e23b905c1522151fa806cd66738addf6be08c452f357a331d"},"schema_version":"1.0","source":{"id":"2508.10358","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2508.10358","created_at":"2026-07-05T11:53:54Z"},{"alias_kind":"arxiv_version","alias_value":"2508.10358v1","created_at":"2026-07-05T11:53:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.10358","created_at":"2026-07-05T11:53:54Z"},{"alias_kind":"pith_short_12","alias_value":"7HS36GOIZU26","created_at":"2026-07-05T11:53:54Z"},{"alias_kind":"pith_short_16","alias_value":"7HS36GOIZU26I5KR","created_at":"2026-07-05T11:53:54Z"},{"alias_kind":"pith_short_8","alias_value":"7HS36GOI","created_at":"2026-07-05T11:53:54Z"}],"graph_snapshots":[{"event_id":"sha256:6563632a31c635ef54b81cc7c68ba9c72cf0f85bf2023fb8afb923ed3e3356ab","target":"graph","created_at":"2026-07-05T11:53:54Z","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/2508.10358/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We investigate the capacity of Large Language Models (LLMs) for imaginative reasoning--the proactive construction, testing, and revision of hypotheses in information-sparse environments. Existing benchmarks, often static or focused on social deduction, fail to capture the dynamic, exploratory nature of this reasoning process. To address this gap, we introduce a comprehensive research framework based on the classic \"Turtle Soup\" game, integrating a benchmark, an agent, and an evaluation protocol. We present TurtleSoup-Bench, the first large-scale, bilingual, interactive benchmark for imaginativ","authors_text":"Bang Liu, Huan Zhang, Mengtao Zhou, Qi Sima, Sifan Wu","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-08-14T05:55:42Z","title":"What to Ask Next? Probing the Imaginative Reasoning of LLMs with TurtleSoup Puzzles"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.10358","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:e61a716c316195fb14fed51ae7d6d2cadc85049ff74705d2af5409631023e299","target":"record","created_at":"2026-07-05T11:53:54Z","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":"bd8115472b4131024766a5ea6556e9f1cd47161d2634653ff8791823636d8e32","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-08-14T05:55:42Z","title_canon_sha256":"5db17abd59502e3e23b905c1522151fa806cd66738addf6be08c452f357a331d"},"schema_version":"1.0","source":{"id":"2508.10358","kind":"arxiv","version":1}},"canonical_sha256":"f9e5bf19c8cd35e47551369689d2f6a6b2cdaa3ffcb09c9993fefa8ecd67f6c6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f9e5bf19c8cd35e47551369689d2f6a6b2cdaa3ffcb09c9993fefa8ecd67f6c6","first_computed_at":"2026-07-05T11:53:54.519346Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:53:54.519346Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"NrsGQBojDK8t0uOvXo56loSA/Mz558CoFuzmcYUaW59GcN2hJ8BmjapAetyQkuNiMA0K8GWrlzhrvuUQ8Q6XCA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:53:54.519820Z","signed_message":"canonical_sha256_bytes"},"source_id":"2508.10358","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e61a716c316195fb14fed51ae7d6d2cadc85049ff74705d2af5409631023e299","sha256:6563632a31c635ef54b81cc7c68ba9c72cf0f85bf2023fb8afb923ed3e3356ab"],"state_sha256":"f733a4e67bf4ff34c43f33cbeacf9954c03427a321b859d465d21c4b07564503"}