{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:XGE6GUEMKDUZEL3N6ETNMDGFXY","short_pith_number":"pith:XGE6GUEM","canonical_record":{"source":{"id":"2606.12117","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-10T14:12:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"97d717a8b90ca94a1fc83151bdeef9d885d405516821a651607cb98dfee4c89a","abstract_canon_sha256":"00464b526e06fe7a5deb6c685847acc2e7c66430950971527412607535ba97ec"},"schema_version":"1.0"},"canonical_sha256":"b989e3508c50e9922f6df126d60cc5be3d9454948f81088b34cacffecb5f0a7f","source":{"kind":"arxiv","id":"2606.12117","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.12117","created_at":"2026-06-11T01:10:49Z"},{"alias_kind":"arxiv_version","alias_value":"2606.12117v1","created_at":"2026-06-11T01:10:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.12117","created_at":"2026-06-11T01:10:49Z"},{"alias_kind":"pith_short_12","alias_value":"XGE6GUEMKDUZ","created_at":"2026-06-11T01:10:49Z"},{"alias_kind":"pith_short_16","alias_value":"XGE6GUEMKDUZEL3N","created_at":"2026-06-11T01:10:49Z"},{"alias_kind":"pith_short_8","alias_value":"XGE6GUEM","created_at":"2026-06-11T01:10:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:XGE6GUEMKDUZEL3N6ETNMDGFXY","target":"record","payload":{"canonical_record":{"source":{"id":"2606.12117","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-10T14:12:19Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"97d717a8b90ca94a1fc83151bdeef9d885d405516821a651607cb98dfee4c89a","abstract_canon_sha256":"00464b526e06fe7a5deb6c685847acc2e7c66430950971527412607535ba97ec"},"schema_version":"1.0"},"canonical_sha256":"b989e3508c50e9922f6df126d60cc5be3d9454948f81088b34cacffecb5f0a7f","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-11T01:10:49.314542Z","signature_b64":"1lj85gHQ29I7XYIp1ZvUmdVMN0yp70zWPCK0MlQarFemsb1uGpcXOyRsSi69l4pIo4fC+bAI7SUfijAuB+rhCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b989e3508c50e9922f6df126d60cc5be3d9454948f81088b34cacffecb5f0a7f","last_reissued_at":"2026-06-11T01:10:49.313649Z","signature_status":"signed_v1","first_computed_at":"2026-06-11T01:10:49.313649Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.12117","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":"2026-06-11T01:10:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"5BgR8hhpr8djnS26rOhxXgvo2h91DvEddZiqMXaRtV6RKqo9tnPiuYyULOiUw+74vUA/z+E+Y6y1fGx4iRBHAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-28T10:38:10.207506Z"},"content_sha256":"568c68287263e3d8ea835925be26c9a526c95677895cb5add58bc49f5620da74","schema_version":"1.0","event_id":"sha256:568c68287263e3d8ea835925be26c9a526c95677895cb5add58bc49f5620da74"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:XGE6GUEMKDUZEL3N6ETNMDGFXY","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Soft-Prompt Tuning for Fair and Efficient LLM Benchmark Evaluation","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Bastian Boll, Bj\\\"orn Deiseroth, Kristian Kersting, Letitia Parcalabescu, Selen Erkan","submitted_at":"2026-06-10T14:12:19Z","abstract_excerpt":"Benchmark scores often misrepresent a large language model's (LLM's) knowledge, because they rely, e.g., on the model's ability to follow specific formatting requirements. This especially penalizes base models that may know the correct answers but lack the ability -- typically introduced in post-training -- to structure them as instructed. To overcome this, we propose soft-prompt tuning, an efficient, fair, and architecture-agnostic model evaluation. By optimizing only 10 soft-prompt vectors (roughly 0.0006% parameters for a 7B model) over a short tuning period, we adapt models to specific ben"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.12117","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/2606.12117/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":"2026-06-11T01:10:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"hsDpT6c0SaOICI43HEta3Edq1PKS5CPRwnvui46AcysshP8mL/XHXerNI5CkSBFywPfX8c6JA723lGYhd8nhBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-28T10:38:10.207864Z"},"content_sha256":"c72e61141d126122274102985ec4239b36d2311544d40c1f062e45a14d7af29c","schema_version":"1.0","event_id":"sha256:c72e61141d126122274102985ec4239b36d2311544d40c1f062e45a14d7af29c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XGE6GUEMKDUZEL3N6ETNMDGFXY/bundle.json","state_url":"https://pith.science/pith/XGE6GUEMKDUZEL3N6ETNMDGFXY/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XGE6GUEMKDUZEL3N6ETNMDGFXY/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":"2026-06-28T10:38:10Z","links":{"resolver":"https://pith.science/pith/XGE6GUEMKDUZEL3N6ETNMDGFXY","bundle":"https://pith.science/pith/XGE6GUEMKDUZEL3N6ETNMDGFXY/bundle.json","state":"https://pith.science/pith/XGE6GUEMKDUZEL3N6ETNMDGFXY/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XGE6GUEMKDUZEL3N6ETNMDGFXY/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:XGE6GUEMKDUZEL3N6ETNMDGFXY","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":"00464b526e06fe7a5deb6c685847acc2e7c66430950971527412607535ba97ec","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-10T14:12:19Z","title_canon_sha256":"97d717a8b90ca94a1fc83151bdeef9d885d405516821a651607cb98dfee4c89a"},"schema_version":"1.0","source":{"id":"2606.12117","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.12117","created_at":"2026-06-11T01:10:49Z"},{"alias_kind":"arxiv_version","alias_value":"2606.12117v1","created_at":"2026-06-11T01:10:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.12117","created_at":"2026-06-11T01:10:49Z"},{"alias_kind":"pith_short_12","alias_value":"XGE6GUEMKDUZ","created_at":"2026-06-11T01:10:49Z"},{"alias_kind":"pith_short_16","alias_value":"XGE6GUEMKDUZEL3N","created_at":"2026-06-11T01:10:49Z"},{"alias_kind":"pith_short_8","alias_value":"XGE6GUEM","created_at":"2026-06-11T01:10:49Z"}],"graph_snapshots":[{"event_id":"sha256:c72e61141d126122274102985ec4239b36d2311544d40c1f062e45a14d7af29c","target":"graph","created_at":"2026-06-11T01:10:49Z","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/2606.12117/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Benchmark scores often misrepresent a large language model's (LLM's) knowledge, because they rely, e.g., on the model's ability to follow specific formatting requirements. This especially penalizes base models that may know the correct answers but lack the ability -- typically introduced in post-training -- to structure them as instructed. To overcome this, we propose soft-prompt tuning, an efficient, fair, and architecture-agnostic model evaluation. By optimizing only 10 soft-prompt vectors (roughly 0.0006% parameters for a 7B model) over a short tuning period, we adapt models to specific ben","authors_text":"Bastian Boll, Bj\\\"orn Deiseroth, Kristian Kersting, Letitia Parcalabescu, Selen Erkan","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-10T14:12:19Z","title":"Soft-Prompt Tuning for Fair and Efficient LLM Benchmark Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.12117","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":"sha256:568c68287263e3d8ea835925be26c9a526c95677895cb5add58bc49f5620da74","target":"record","created_at":"2026-06-11T01:10:49Z","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":"00464b526e06fe7a5deb6c685847acc2e7c66430950971527412607535ba97ec","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-10T14:12:19Z","title_canon_sha256":"97d717a8b90ca94a1fc83151bdeef9d885d405516821a651607cb98dfee4c89a"},"schema_version":"1.0","source":{"id":"2606.12117","kind":"arxiv","version":1}},"canonical_sha256":"b989e3508c50e9922f6df126d60cc5be3d9454948f81088b34cacffecb5f0a7f","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b989e3508c50e9922f6df126d60cc5be3d9454948f81088b34cacffecb5f0a7f","first_computed_at":"2026-06-11T01:10:49.313649Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-11T01:10:49.313649Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"1lj85gHQ29I7XYIp1ZvUmdVMN0yp70zWPCK0MlQarFemsb1uGpcXOyRsSi69l4pIo4fC+bAI7SUfijAuB+rhCw==","signature_status":"signed_v1","signed_at":"2026-06-11T01:10:49.314542Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.12117","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:568c68287263e3d8ea835925be26c9a526c95677895cb5add58bc49f5620da74","sha256:c72e61141d126122274102985ec4239b36d2311544d40c1f062e45a14d7af29c"],"state_sha256":"3da1c24d6213d09eaf0cfa4104f6a0a5fcf7de27bbaca9f0bdc0fbeec7eabfa8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6fdMfnKdCkh5i51kqdtQbGPZFVTFWpt6H2yGa3S/e4c7ZxKkWDJBF1KhgPGo7I5wK6/5a8QbaY9smKX61dHJBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-28T10:38:10.209747Z","bundle_sha256":"0467e225cd32bd117b159db22c13e12509567c9222710c3ed703dca1d1077250"}}