{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:XKB45ZKZUTCRDF57FQAKOY2L2X","short_pith_number":"pith:XKB45ZKZ","canonical_record":{"source":{"id":"2502.17848","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-25T04:51:17Z","cross_cats_sorted":[],"title_canon_sha256":"f17b0d6e42c04ec44fd31394832cea322cc0b32ad9b77a58b7938f10bf1ec13c","abstract_canon_sha256":"ac9e9c3705655cb80e6bded8a7dcb99321724bc4215f515dab7b3e8e74bb5f51"},"schema_version":"1.0"},"canonical_sha256":"ba83cee559a4c51197bf2c00a7634bd5e8cd3c836d933a0b89966103e026a9fa","source":{"kind":"arxiv","id":"2502.17848","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.17848","created_at":"2026-07-05T11:26:48Z"},{"alias_kind":"arxiv_version","alias_value":"2502.17848v4","created_at":"2026-07-05T11:26:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.17848","created_at":"2026-07-05T11:26:48Z"},{"alias_kind":"pith_short_12","alias_value":"XKB45ZKZUTCR","created_at":"2026-07-05T11:26:48Z"},{"alias_kind":"pith_short_16","alias_value":"XKB45ZKZUTCRDF57","created_at":"2026-07-05T11:26:48Z"},{"alias_kind":"pith_short_8","alias_value":"XKB45ZKZ","created_at":"2026-07-05T11:26:48Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:XKB45ZKZUTCRDF57FQAKOY2L2X","target":"record","payload":{"canonical_record":{"source":{"id":"2502.17848","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-25T04:51:17Z","cross_cats_sorted":[],"title_canon_sha256":"f17b0d6e42c04ec44fd31394832cea322cc0b32ad9b77a58b7938f10bf1ec13c","abstract_canon_sha256":"ac9e9c3705655cb80e6bded8a7dcb99321724bc4215f515dab7b3e8e74bb5f51"},"schema_version":"1.0"},"canonical_sha256":"ba83cee559a4c51197bf2c00a7634bd5e8cd3c836d933a0b89966103e026a9fa","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:26:48.184917Z","signature_b64":"GRaPEFGlvBLsSKBAZwnCaVnDFiFf++jmvokPO4eHvYbLHjH93RvOsiJkylj+eoe05kzdE7R6rMs8BJ6a7DzDDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ba83cee559a4c51197bf2c00a7634bd5e8cd3c836d933a0b89966103e026a9fa","last_reissued_at":"2026-07-05T11:26:48.184450Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:26:48.184450Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2502.17848","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:26:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"g0C2QMn19flNI+IXwwp7j13IYpscv//2fUEHvYwr8Cv+gI9qUjJF/9lqkfh8xPCibkp1dQL7f4o/4uqehrnnCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T16:23:34.610318Z"},"content_sha256":"a37e38c593e02044c3fed019d667bf5d9106a2398ea5b841a2ae0ee36ba044e7","schema_version":"1.0","event_id":"sha256:a37e38c593e02044c3fed019d667bf5d9106a2398ea5b841a2ae0ee36ba044e7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:XKB45ZKZUTCRDF57FQAKOY2L2X","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LR^2Bench: Evaluating Long-chain Reflective Reasoning Capabilities of Large Language Models via Constraint Satisfaction Problems","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Jiajun Zhang, Jianghao Chen, Zhenjiang Ren, Zhenlin Wei, Ziyong Li","submitted_at":"2025-02-25T04:51:17Z","abstract_excerpt":"Recent progress in Large Reasoning Models (LRMs) has significantly enhanced the reasoning abilities of Large Language Models (LLMs), empowering them to tackle increasingly complex tasks through reflection capabilities, such as making assumptions, backtracking, and self-refinement. However, effectively evaluating such reflection capabilities remains challenging due to the lack of appropriate benchmarks. To bridge this gap, we introduce LR$^2$Bench, a novel benchmark designed to evaluate the Long-chain Reflective Reasoning capabilities of LLMs. LR$^2$Bench comprises 850 samples across six Constr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.17848","kind":"arxiv","version":4},"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/2502.17848/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":"2026-07-05T11:26:48Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"nC9DXuFRfQVRG7idykW/5XU6i9E+M3SADPWPgQ3KXQ0wHbfKf4dEiQKu452Pz0oCuzP/fbwA/PDR4Opy856MCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T16:23:34.611035Z"},"content_sha256":"9f2581fcef11026c3240c1d655a8f291756dd99e20c7bd3d2b5d1361cd959de6","schema_version":"1.0","event_id":"sha256:9f2581fcef11026c3240c1d655a8f291756dd99e20c7bd3d2b5d1361cd959de6"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XKB45ZKZUTCRDF57FQAKOY2L2X/bundle.json","state_url":"https://pith.science/pith/XKB45ZKZUTCRDF57FQAKOY2L2X/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XKB45ZKZUTCRDF57FQAKOY2L2X/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":"2026-08-06T16:23:34Z","links":{"resolver":"https://pith.science/pith/XKB45ZKZUTCRDF57FQAKOY2L2X","bundle":"https://pith.science/pith/XKB45ZKZUTCRDF57FQAKOY2L2X/bundle.json","state":"https://pith.science/pith/XKB45ZKZUTCRDF57FQAKOY2L2X/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XKB45ZKZUTCRDF57FQAKOY2L2X/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:XKB45ZKZUTCRDF57FQAKOY2L2X","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":"ac9e9c3705655cb80e6bded8a7dcb99321724bc4215f515dab7b3e8e74bb5f51","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-25T04:51:17Z","title_canon_sha256":"f17b0d6e42c04ec44fd31394832cea322cc0b32ad9b77a58b7938f10bf1ec13c"},"schema_version":"1.0","source":{"id":"2502.17848","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2502.17848","created_at":"2026-07-05T11:26:48Z"},{"alias_kind":"arxiv_version","alias_value":"2502.17848v4","created_at":"2026-07-05T11:26:48Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.17848","created_at":"2026-07-05T11:26:48Z"},{"alias_kind":"pith_short_12","alias_value":"XKB45ZKZUTCR","created_at":"2026-07-05T11:26:48Z"},{"alias_kind":"pith_short_16","alias_value":"XKB45ZKZUTCRDF57","created_at":"2026-07-05T11:26:48Z"},{"alias_kind":"pith_short_8","alias_value":"XKB45ZKZ","created_at":"2026-07-05T11:26:48Z"}],"graph_snapshots":[{"event_id":"sha256:9f2581fcef11026c3240c1d655a8f291756dd99e20c7bd3d2b5d1361cd959de6","target":"graph","created_at":"2026-07-05T11:26:48Z","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/2502.17848/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent progress in Large Reasoning Models (LRMs) has significantly enhanced the reasoning abilities of Large Language Models (LLMs), empowering them to tackle increasingly complex tasks through reflection capabilities, such as making assumptions, backtracking, and self-refinement. However, effectively evaluating such reflection capabilities remains challenging due to the lack of appropriate benchmarks. To bridge this gap, we introduce LR$^2$Bench, a novel benchmark designed to evaluate the Long-chain Reflective Reasoning capabilities of LLMs. LR$^2$Bench comprises 850 samples across six Constr","authors_text":"Jiajun Zhang, Jianghao Chen, Zhenjiang Ren, Zhenlin Wei, Ziyong Li","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-25T04:51:17Z","title":"LR^2Bench: Evaluating Long-chain Reflective Reasoning Capabilities of Large Language Models via Constraint Satisfaction Problems"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.17848","kind":"arxiv","version":4},"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:a37e38c593e02044c3fed019d667bf5d9106a2398ea5b841a2ae0ee36ba044e7","target":"record","created_at":"2026-07-05T11:26:48Z","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":"ac9e9c3705655cb80e6bded8a7dcb99321724bc4215f515dab7b3e8e74bb5f51","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-02-25T04:51:17Z","title_canon_sha256":"f17b0d6e42c04ec44fd31394832cea322cc0b32ad9b77a58b7938f10bf1ec13c"},"schema_version":"1.0","source":{"id":"2502.17848","kind":"arxiv","version":4}},"canonical_sha256":"ba83cee559a4c51197bf2c00a7634bd5e8cd3c836d933a0b89966103e026a9fa","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ba83cee559a4c51197bf2c00a7634bd5e8cd3c836d933a0b89966103e026a9fa","first_computed_at":"2026-07-05T11:26:48.184450Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:26:48.184450Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"GRaPEFGlvBLsSKBAZwnCaVnDFiFf++jmvokPO4eHvYbLHjH93RvOsiJkylj+eoe05kzdE7R6rMs8BJ6a7DzDDA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:26:48.184917Z","signed_message":"canonical_sha256_bytes"},"source_id":"2502.17848","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a37e38c593e02044c3fed019d667bf5d9106a2398ea5b841a2ae0ee36ba044e7","sha256:9f2581fcef11026c3240c1d655a8f291756dd99e20c7bd3d2b5d1361cd959de6"],"state_sha256":"6f61479b3cf2643fd15794593f9631bb596fff78d4e7dad0dcbc765d7988b6a1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+WRvd7szPcApruz+0CIeQLhDMFjEopDhcrLO8qNjA1tNKz1bAjv8u8qmalPtlRQ+kssXNiDJetE6vh/nkQPoDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T16:23:34.620569Z","bundle_sha256":"c989e51c98cbf7c9757ceef064ff1b368c5595be1c736e19d8537fa9f4f9cbe2"}}