{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:VCFARHJ7UWQ63TKD25KJWVFRS5","short_pith_number":"pith:VCFARHJ7","canonical_record":{"source":{"id":"2506.05062","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T14:06:51Z","cross_cats_sorted":[],"title_canon_sha256":"c5f7391fca8aa1f7eb7c917914ea8942193a6ea35808f736a1b58c9ce2817ca3","abstract_canon_sha256":"ecd20c5e227e5e182cdfe9a9f3f2ff34b395fd81e6b738978cfa8b3f27e9bfd8"},"schema_version":"1.0"},"canonical_sha256":"a88a089d3fa5a1edcd43d7549b54b197532b72700a750ddb2e8658c0fff785cf","source":{"kind":"arxiv","id":"2506.05062","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.05062","created_at":"2026-07-05T12:07:08Z"},{"alias_kind":"arxiv_version","alias_value":"2506.05062v2","created_at":"2026-07-05T12:07:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.05062","created_at":"2026-07-05T12:07:08Z"},{"alias_kind":"pith_short_12","alias_value":"VCFARHJ7UWQ6","created_at":"2026-07-05T12:07:08Z"},{"alias_kind":"pith_short_16","alias_value":"VCFARHJ7UWQ63TKD","created_at":"2026-07-05T12:07:08Z"},{"alias_kind":"pith_short_8","alias_value":"VCFARHJ7","created_at":"2026-07-05T12:07:08Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:VCFARHJ7UWQ63TKD25KJWVFRS5","target":"record","payload":{"canonical_record":{"source":{"id":"2506.05062","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T14:06:51Z","cross_cats_sorted":[],"title_canon_sha256":"c5f7391fca8aa1f7eb7c917914ea8942193a6ea35808f736a1b58c9ce2817ca3","abstract_canon_sha256":"ecd20c5e227e5e182cdfe9a9f3f2ff34b395fd81e6b738978cfa8b3f27e9bfd8"},"schema_version":"1.0"},"canonical_sha256":"a88a089d3fa5a1edcd43d7549b54b197532b72700a750ddb2e8658c0fff785cf","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:07:08.288728Z","signature_b64":"kZyiX3pkj699jL+jZKbkgqY7s7l9lvqbK4A+MSaaqZUTjIa2kLZUiiz2/q8HTwRtek6OGCuc65+9ij14bloACA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a88a089d3fa5a1edcd43d7549b54b197532b72700a750ddb2e8658c0fff785cf","last_reissued_at":"2026-07-05T12:07:08.288144Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:07:08.288144Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.05062","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T12:07:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Dbjwanp9yOBB2+cW2mciSAzhnLs2mUlY7nIapORSW82NFQOUFEtlt+ODDnJnRz4OwdkY/+26cfarUT1jfP2+Aw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T17:41:25.991995Z"},"content_sha256":"647aa7875a03d914b212dd9c3a800be4406db4b760c895906ab1a816bd7a81f2","schema_version":"1.0","event_id":"sha256:647aa7875a03d914b212dd9c3a800be4406db4b760c895906ab1a816bd7a81f2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:VCFARHJ7UWQ63TKD25KJWVFRS5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Debatable Intelligence: Benchmarking LLM Judges via Debate Speech Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ariel Gera, Noam Slonim, Noy Sternlicht, Roy Bar-Haim, Tom Hope","submitted_at":"2025-06-05T14:06:51Z","abstract_excerpt":"We introduce Debate Speech Evaluation as a novel and challenging benchmark for assessing LLM judges. Evaluating debate speeches requires a deep understanding of the speech at multiple levels, including argument strength and relevance, the coherence and organization of the speech, the appropriateness of its style and tone, and so on. This task involves a unique set of cognitive abilities that previously received limited attention in systematic LLM benchmarking. To explore such skills, we leverage a dataset of over 600 meticulously annotated debate speeches and present the first in-depth analysi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.05062","kind":"arxiv","version":2},"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/2506.05062/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":"2026-07-05T12:07:08Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SLXv8BIk0NJZQIeKx6XreyF3Wrbje3gAskewVTPjAIxgurrPMVXZ8jFTKalS0AOAB92xEByawdlG4ORlmIKhDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T17:41:25.993058Z"},"content_sha256":"3dc074e5aa2dc8d8ff9b3d3981b7d33de47c714f5beae686de4a9c9e16f5652c","schema_version":"1.0","event_id":"sha256:3dc074e5aa2dc8d8ff9b3d3981b7d33de47c714f5beae686de4a9c9e16f5652c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VCFARHJ7UWQ63TKD25KJWVFRS5/bundle.json","state_url":"https://pith.science/pith/VCFARHJ7UWQ63TKD25KJWVFRS5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VCFARHJ7UWQ63TKD25KJWVFRS5/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":"2026-08-08T17:41:25Z","links":{"resolver":"https://pith.science/pith/VCFARHJ7UWQ63TKD25KJWVFRS5","bundle":"https://pith.science/pith/VCFARHJ7UWQ63TKD25KJWVFRS5/bundle.json","state":"https://pith.science/pith/VCFARHJ7UWQ63TKD25KJWVFRS5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VCFARHJ7UWQ63TKD25KJWVFRS5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:VCFARHJ7UWQ63TKD25KJWVFRS5","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":"ecd20c5e227e5e182cdfe9a9f3f2ff34b395fd81e6b738978cfa8b3f27e9bfd8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T14:06:51Z","title_canon_sha256":"c5f7391fca8aa1f7eb7c917914ea8942193a6ea35808f736a1b58c9ce2817ca3"},"schema_version":"1.0","source":{"id":"2506.05062","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.05062","created_at":"2026-07-05T12:07:08Z"},{"alias_kind":"arxiv_version","alias_value":"2506.05062v2","created_at":"2026-07-05T12:07:08Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.05062","created_at":"2026-07-05T12:07:08Z"},{"alias_kind":"pith_short_12","alias_value":"VCFARHJ7UWQ6","created_at":"2026-07-05T12:07:08Z"},{"alias_kind":"pith_short_16","alias_value":"VCFARHJ7UWQ63TKD","created_at":"2026-07-05T12:07:08Z"},{"alias_kind":"pith_short_8","alias_value":"VCFARHJ7","created_at":"2026-07-05T12:07:08Z"}],"graph_snapshots":[{"event_id":"sha256:3dc074e5aa2dc8d8ff9b3d3981b7d33de47c714f5beae686de4a9c9e16f5652c","target":"graph","created_at":"2026-07-05T12:07:08Z","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/2506.05062/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"We introduce Debate Speech Evaluation as a novel and challenging benchmark for assessing LLM judges. Evaluating debate speeches requires a deep understanding of the speech at multiple levels, including argument strength and relevance, the coherence and organization of the speech, the appropriateness of its style and tone, and so on. This task involves a unique set of cognitive abilities that previously received limited attention in systematic LLM benchmarking. To explore such skills, we leverage a dataset of over 600 meticulously annotated debate speeches and present the first in-depth analysi","authors_text":"Ariel Gera, Noam Slonim, Noy Sternlicht, Roy Bar-Haim, Tom Hope","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T14:06:51Z","title":"Debatable Intelligence: Benchmarking LLM Judges via Debate Speech Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.05062","kind":"arxiv","version":2},"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":"sha256:647aa7875a03d914b212dd9c3a800be4406db4b760c895906ab1a816bd7a81f2","target":"record","created_at":"2026-07-05T12:07:08Z","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":"ecd20c5e227e5e182cdfe9a9f3f2ff34b395fd81e6b738978cfa8b3f27e9bfd8","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-05T14:06:51Z","title_canon_sha256":"c5f7391fca8aa1f7eb7c917914ea8942193a6ea35808f736a1b58c9ce2817ca3"},"schema_version":"1.0","source":{"id":"2506.05062","kind":"arxiv","version":2}},"canonical_sha256":"a88a089d3fa5a1edcd43d7549b54b197532b72700a750ddb2e8658c0fff785cf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a88a089d3fa5a1edcd43d7549b54b197532b72700a750ddb2e8658c0fff785cf","first_computed_at":"2026-07-05T12:07:08.288144Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T12:07:08.288144Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"kZyiX3pkj699jL+jZKbkgqY7s7l9lvqbK4A+MSaaqZUTjIa2kLZUiiz2/q8HTwRtek6OGCuc65+9ij14bloACA==","signature_status":"signed_v1","signed_at":"2026-07-05T12:07:08.288728Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.05062","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:647aa7875a03d914b212dd9c3a800be4406db4b760c895906ab1a816bd7a81f2","sha256:3dc074e5aa2dc8d8ff9b3d3981b7d33de47c714f5beae686de4a9c9e16f5652c"],"state_sha256":"f2d81e92b2883fe08c5264b4aa7770a2643574cb7699b239c8a2839e91180e99"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"8+3p8cBEq+x41aoIYW02mbgZfEWcOdOV0u9+1LRaqkZQYYcz4wOTxlUVBS26ympXDh5+zQKmornvMwh5+jN1DQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T17:41:26.001518Z","bundle_sha256":"59a801445f8fd6177e473289db7e446011aa7c2f9f2e1fcc4b97650ee16e6b54"}}