{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:C736ZUBGBV2W54DOKSEWM4KLHN","short_pith_number":"pith:C736ZUBG","canonical_record":{"source":{"id":"2607.24780","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-19T06:20:58Z","cross_cats_sorted":[],"title_canon_sha256":"6fb738a19669c2eb75950f556897b305e7e2100f875f66a287a3d33dd4bf2270","abstract_canon_sha256":"7a580da1ac40ef4b312039fa49019b2d34c864bcad25f063745db3fc98a05264"},"schema_version":"1.0"},"canonical_sha256":"17f7ecd0260d756ef06e548966714b3b54d8c4d0bde236d69e67792d48e4570f","source":{"kind":"arxiv","id":"2607.24780","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.24780","created_at":"2026-07-29T00:24:43Z"},{"alias_kind":"arxiv_version","alias_value":"2607.24780v1","created_at":"2026-07-29T00:24:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.24780","created_at":"2026-07-29T00:24:43Z"},{"alias_kind":"pith_short_12","alias_value":"C736ZUBGBV2W","created_at":"2026-07-29T00:24:43Z"},{"alias_kind":"pith_short_16","alias_value":"C736ZUBGBV2W54DO","created_at":"2026-07-29T00:24:43Z"},{"alias_kind":"pith_short_8","alias_value":"C736ZUBG","created_at":"2026-07-29T00:24:43Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:C736ZUBGBV2W54DOKSEWM4KLHN","target":"record","payload":{"canonical_record":{"source":{"id":"2607.24780","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-19T06:20:58Z","cross_cats_sorted":[],"title_canon_sha256":"6fb738a19669c2eb75950f556897b305e7e2100f875f66a287a3d33dd4bf2270","abstract_canon_sha256":"7a580da1ac40ef4b312039fa49019b2d34c864bcad25f063745db3fc98a05264"},"schema_version":"1.0"},"canonical_sha256":"17f7ecd0260d756ef06e548966714b3b54d8c4d0bde236d69e67792d48e4570f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-29T00:24:43.661522Z","signature_b64":"65Unytn22nNbWG/4SgNmOZ41J67r1Wz7EYWYGKod1oGSxQAtai9esrMH1O0SUN4qMsC9ELTlU6gLBMIJ3Lr2Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"17f7ecd0260d756ef06e548966714b3b54d8c4d0bde236d69e67792d48e4570f","last_reissued_at":"2026-07-29T00:24:43.660659Z","signature_status":"signed_v1","first_computed_at":"2026-07-29T00:24:43.660659Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2607.24780","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":"2026-07-29T00:24:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"55bPDhTrozCKam9ZmKBcWqWApwYiIaytuquTnANGnrH7n0iNlGMPOiMHOVk3AGc6JughmUcaaWvwezvw1c5PDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T11:49:40.337614Z"},"content_sha256":"1ad1413f342e382e54b2c5bc3bc066b5241d8e6940aa0cd5ffbc6f2b65a104c4","schema_version":"1.0","event_id":"sha256:1ad1413f342e382e54b2c5bc3bc066b5241d8e6940aa0cd5ffbc6f2b65a104c4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:C736ZUBGBV2W54DOKSEWM4KLHN","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"LivingArena: Do LLMs Know What Other LLMs Don't? Peer-Probing as Scalable Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Liefeng Bo, Rui Wang, Xingyu Chen, Zhaopeng Tu","submitted_at":"2026-06-19T06:20:58Z","abstract_excerpt":"Evaluating frontier LLMs is challenging: static benchmarks suffer from contamination and saturation -- leaving users unable to distinguish top models and developers blind to specific failure modes -- while human preference is subjective. In this paper, our question is: \\emph{Do LLMs know what other LLMs don't? And can we leverage this dynamic for evaluation?} We present \\textbf{LivingArena}, an automated, contamination-resistant evaluation framework. In this framework, models take turns proposing questions, aiming to pose items that opponents cannot answer correctly. Questioners are encouraged"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.24780","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/2607.24780/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":"2026-07-29T00:24:43Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rvHiNoXp+oPgBpnaNmpfaRG3eZ5J3xGBnPEI+I+AP/RY68dmV6E4MXkEUpr7+fu/Lo1veX0PQVZuCJzhQBwkDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T11:49:40.337993Z"},"content_sha256":"93a6f759145dd2eb58b377035a6b6cc97a4558d10883a9eb638e108083fb1935","schema_version":"1.0","event_id":"sha256:93a6f759145dd2eb58b377035a6b6cc97a4558d10883a9eb638e108083fb1935"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/C736ZUBGBV2W54DOKSEWM4KLHN/bundle.json","state_url":"https://pith.science/pith/C736ZUBGBV2W54DOKSEWM4KLHN/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/C736ZUBGBV2W54DOKSEWM4KLHN/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":"2026-08-06T11:49:40Z","links":{"resolver":"https://pith.science/pith/C736ZUBGBV2W54DOKSEWM4KLHN","bundle":"https://pith.science/pith/C736ZUBGBV2W54DOKSEWM4KLHN/bundle.json","state":"https://pith.science/pith/C736ZUBGBV2W54DOKSEWM4KLHN/state.json","well_known_bundle":"https://pith.science/.well-known/pith/C736ZUBGBV2W54DOKSEWM4KLHN/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:C736ZUBGBV2W54DOKSEWM4KLHN","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":"7a580da1ac40ef4b312039fa49019b2d34c864bcad25f063745db3fc98a05264","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-19T06:20:58Z","title_canon_sha256":"6fb738a19669c2eb75950f556897b305e7e2100f875f66a287a3d33dd4bf2270"},"schema_version":"1.0","source":{"id":"2607.24780","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2607.24780","created_at":"2026-07-29T00:24:43Z"},{"alias_kind":"arxiv_version","alias_value":"2607.24780v1","created_at":"2026-07-29T00:24:43Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.24780","created_at":"2026-07-29T00:24:43Z"},{"alias_kind":"pith_short_12","alias_value":"C736ZUBGBV2W","created_at":"2026-07-29T00:24:43Z"},{"alias_kind":"pith_short_16","alias_value":"C736ZUBGBV2W54DO","created_at":"2026-07-29T00:24:43Z"},{"alias_kind":"pith_short_8","alias_value":"C736ZUBG","created_at":"2026-07-29T00:24:43Z"}],"graph_snapshots":[{"event_id":"sha256:93a6f759145dd2eb58b377035a6b6cc97a4558d10883a9eb638e108083fb1935","target":"graph","created_at":"2026-07-29T00:24:43Z","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/2607.24780/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Evaluating frontier LLMs is challenging: static benchmarks suffer from contamination and saturation -- leaving users unable to distinguish top models and developers blind to specific failure modes -- while human preference is subjective. In this paper, our question is: \\emph{Do LLMs know what other LLMs don't? And can we leverage this dynamic for evaluation?} We present \\textbf{LivingArena}, an automated, contamination-resistant evaluation framework. In this framework, models take turns proposing questions, aiming to pose items that opponents cannot answer correctly. Questioners are encouraged","authors_text":"Liefeng Bo, Rui Wang, Xingyu Chen, Zhaopeng Tu","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-19T06:20:58Z","title":"LivingArena: Do LLMs Know What Other LLMs Don't? Peer-Probing as Scalable Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.24780","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":"sha256:1ad1413f342e382e54b2c5bc3bc066b5241d8e6940aa0cd5ffbc6f2b65a104c4","target":"record","created_at":"2026-07-29T00:24:43Z","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":"7a580da1ac40ef4b312039fa49019b2d34c864bcad25f063745db3fc98a05264","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-06-19T06:20:58Z","title_canon_sha256":"6fb738a19669c2eb75950f556897b305e7e2100f875f66a287a3d33dd4bf2270"},"schema_version":"1.0","source":{"id":"2607.24780","kind":"arxiv","version":1}},"canonical_sha256":"17f7ecd0260d756ef06e548966714b3b54d8c4d0bde236d69e67792d48e4570f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"17f7ecd0260d756ef06e548966714b3b54d8c4d0bde236d69e67792d48e4570f","first_computed_at":"2026-07-29T00:24:43.660659Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-29T00:24:43.660659Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"65Unytn22nNbWG/4SgNmOZ41J67r1Wz7EYWYGKod1oGSxQAtai9esrMH1O0SUN4qMsC9ELTlU6gLBMIJ3Lr2Aw==","signature_status":"signed_v1","signed_at":"2026-07-29T00:24:43.661522Z","signed_message":"canonical_sha256_bytes"},"source_id":"2607.24780","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1ad1413f342e382e54b2c5bc3bc066b5241d8e6940aa0cd5ffbc6f2b65a104c4","sha256:93a6f759145dd2eb58b377035a6b6cc97a4558d10883a9eb638e108083fb1935"],"state_sha256":"d84c823a62a3cdd258e9e90973ad8e4a719728936c11d71bdfa67be12ff24611"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Q9dexTLdqDewFwSOr4HZIcmy7s57+JXrXUvkWHS+vQ8YEF7472eNneCVjKGNg8JzfesDxa1sg70GhlVWOiYjCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T11:49:40.341687Z","bundle_sha256":"8e31d5ac92bbf631f63d12e1a19b6b9548c4c0403702cf212972e1490508fb6f"}}