{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:QUIUDW2BI3PKQC2MXKYGZVF76L","short_pith_number":"pith:QUIUDW2B","canonical_record":{"source":{"id":"2608.00014","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-24T16:14:12Z","cross_cats_sorted":[],"title_canon_sha256":"d9dd1e1e0aa91740be86aad2bbe11994325d0da9db439be8bfa8ddf0fce3d573","abstract_canon_sha256":"3c24d92754436f9be873de3448537c0dcbceaf557c92e8a019f6c3d71f314db0"},"schema_version":"1.0"},"canonical_sha256":"851141db4146dea80b4cbab06cd4bff2e1d151e8b399dbbf990179eb29542320","source":{"kind":"arxiv","id":"2608.00014","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.00014","created_at":"2026-08-04T00:31:38Z"},{"alias_kind":"arxiv_version","alias_value":"2608.00014v1","created_at":"2026-08-04T00:31:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.00014","created_at":"2026-08-04T00:31:38Z"},{"alias_kind":"pith_short_12","alias_value":"QUIUDW2BI3PK","created_at":"2026-08-04T00:31:38Z"},{"alias_kind":"pith_short_16","alias_value":"QUIUDW2BI3PKQC2M","created_at":"2026-08-04T00:31:38Z"},{"alias_kind":"pith_short_8","alias_value":"QUIUDW2B","created_at":"2026-08-04T00:31:38Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:QUIUDW2BI3PKQC2MXKYGZVF76L","target":"record","payload":{"canonical_record":{"source":{"id":"2608.00014","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-24T16:14:12Z","cross_cats_sorted":[],"title_canon_sha256":"d9dd1e1e0aa91740be86aad2bbe11994325d0da9db439be8bfa8ddf0fce3d573","abstract_canon_sha256":"3c24d92754436f9be873de3448537c0dcbceaf557c92e8a019f6c3d71f314db0"},"schema_version":"1.0"},"canonical_sha256":"851141db4146dea80b4cbab06cd4bff2e1d151e8b399dbbf990179eb29542320","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T00:31:38.644226Z","signature_b64":"MpcGgQkAAcr0w61zmpkzft+ppIzjZGuTe5w5bzj7rj5vF477l7MsROnbdEizoE16kBOPGoPdRdwfolMYVSpnAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"851141db4146dea80b4cbab06cd4bff2e1d151e8b399dbbf990179eb29542320","last_reissued_at":"2026-08-04T00:31:38.642525Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T00:31:38.642525Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2608.00014","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":"2026-08-04T00:31:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/qniRVy0HhCaEa3H7hFX94KNdQuCyCkM49CIczuXcFidoyQu/ydxcNWYId7o2wiyupzrdlzWyENmZKJ5Px6dAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T08:15:41.429931Z"},"content_sha256":"54284e6453d091d8d64bb5bc2460a613beaa8fe62a6fbf4ed75ca52320ed298d","schema_version":"1.0","event_id":"sha256:54284e6453d091d8d64bb5bc2460a613beaa8fe62a6fbf4ed75ca52320ed298d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:QUIUDW2BI3PKQC2MXKYGZVF76L","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"CoT-Core: Accelerating LLM Evaluation via CoT-Aware Coreset Selection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Bo Li, Huacan Wang, Peijie Dong, Qihua Pan, Xiang Liu, Xiaowen Chu, Zhenheng Tang","submitted_at":"2026-06-24T16:14:12Z","abstract_excerpt":"Evaluating Large Language Models (LLMs) incurs prohibitive computational overhead during continuous development processes. While coreset selection accelerates evaluation, existing methods either suffer from a severe ``cold start'' bottleneck requiring massive historical logs (e.g., Item Response Theory) or exhibit a surface lexical bias that misses the underlying reasoning manifold of tasks. We propose CoT-Core, a novel training-free core question selection framework. Recognizing that lexically disparate questions can share equivalent underlying logic, CoT-Core prompts LLMs to unroll zero-shot"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.00014","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/2608.00014/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":"2026-08-04T00:31:38Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"TlPVlL/wDr8onHbS/2emKsw1g8Xcj4Vj+8I6662JkGRiAZc3ZFDarr+J2tHD0RKKeC1uQcCfiqC24cdIKfybDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T08:15:41.430858Z"},"content_sha256":"e53f4b678a387a54f1ff0f51b2c94645bffdf3f7e3f89342b60202ceacac057d","schema_version":"1.0","event_id":"sha256:e53f4b678a387a54f1ff0f51b2c94645bffdf3f7e3f89342b60202ceacac057d"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/QUIUDW2BI3PKQC2MXKYGZVF76L/bundle.json","state_url":"https://pith.science/pith/QUIUDW2BI3PKQC2MXKYGZVF76L/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/QUIUDW2BI3PKQC2MXKYGZVF76L/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":"2026-08-05T08:15:41Z","links":{"resolver":"https://pith.science/pith/QUIUDW2BI3PKQC2MXKYGZVF76L","bundle":"https://pith.science/pith/QUIUDW2BI3PKQC2MXKYGZVF76L/bundle.json","state":"https://pith.science/pith/QUIUDW2BI3PKQC2MXKYGZVF76L/state.json","well_known_bundle":"https://pith.science/.well-known/pith/QUIUDW2BI3PKQC2MXKYGZVF76L/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:QUIUDW2BI3PKQC2MXKYGZVF76L","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":"3c24d92754436f9be873de3448537c0dcbceaf557c92e8a019f6c3d71f314db0","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-24T16:14:12Z","title_canon_sha256":"d9dd1e1e0aa91740be86aad2bbe11994325d0da9db439be8bfa8ddf0fce3d573"},"schema_version":"1.0","source":{"id":"2608.00014","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.00014","created_at":"2026-08-04T00:31:38Z"},{"alias_kind":"arxiv_version","alias_value":"2608.00014v1","created_at":"2026-08-04T00:31:38Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.00014","created_at":"2026-08-04T00:31:38Z"},{"alias_kind":"pith_short_12","alias_value":"QUIUDW2BI3PK","created_at":"2026-08-04T00:31:38Z"},{"alias_kind":"pith_short_16","alias_value":"QUIUDW2BI3PKQC2M","created_at":"2026-08-04T00:31:38Z"},{"alias_kind":"pith_short_8","alias_value":"QUIUDW2B","created_at":"2026-08-04T00:31:38Z"}],"graph_snapshots":[{"event_id":"sha256:e53f4b678a387a54f1ff0f51b2c94645bffdf3f7e3f89342b60202ceacac057d","target":"graph","created_at":"2026-08-04T00:31:38Z","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/2608.00014/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Evaluating Large Language Models (LLMs) incurs prohibitive computational overhead during continuous development processes. While coreset selection accelerates evaluation, existing methods either suffer from a severe ``cold start'' bottleneck requiring massive historical logs (e.g., Item Response Theory) or exhibit a surface lexical bias that misses the underlying reasoning manifold of tasks. We propose CoT-Core, a novel training-free core question selection framework. Recognizing that lexically disparate questions can share equivalent underlying logic, CoT-Core prompts LLMs to unroll zero-shot","authors_text":"Bo Li, Huacan Wang, Peijie Dong, Qihua Pan, Xiang Liu, Xiaowen Chu, Zhenheng Tang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-24T16:14:12Z","title":"CoT-Core: Accelerating LLM Evaluation via CoT-Aware Coreset Selection"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.00014","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":"sha256:54284e6453d091d8d64bb5bc2460a613beaa8fe62a6fbf4ed75ca52320ed298d","target":"record","created_at":"2026-08-04T00:31:38Z","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":"3c24d92754436f9be873de3448537c0dcbceaf557c92e8a019f6c3d71f314db0","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-06-24T16:14:12Z","title_canon_sha256":"d9dd1e1e0aa91740be86aad2bbe11994325d0da9db439be8bfa8ddf0fce3d573"},"schema_version":"1.0","source":{"id":"2608.00014","kind":"arxiv","version":1}},"canonical_sha256":"851141db4146dea80b4cbab06cd4bff2e1d151e8b399dbbf990179eb29542320","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"851141db4146dea80b4cbab06cd4bff2e1d151e8b399dbbf990179eb29542320","first_computed_at":"2026-08-04T00:31:38.642525Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-04T00:31:38.642525Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"MpcGgQkAAcr0w61zmpkzft+ppIzjZGuTe5w5bzj7rj5vF477l7MsROnbdEizoE16kBOPGoPdRdwfolMYVSpnAw==","signature_status":"signed_v1","signed_at":"2026-08-04T00:31:38.644226Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.00014","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:54284e6453d091d8d64bb5bc2460a613beaa8fe62a6fbf4ed75ca52320ed298d","sha256:e53f4b678a387a54f1ff0f51b2c94645bffdf3f7e3f89342b60202ceacac057d"],"state_sha256":"42ddabe3b609fbba82af9dfda9fce46867a451c4e4b8a2cae8279ac02f698225"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"t7vMpueMOD6J/+seYM2pvxUjqTbvyoWlO7x2Na6D7Ht3Zh7dJvjFKYW7/KLiFr7AUmp5GBLZ+z2iVFrQGc57DQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T08:15:41.438246Z","bundle_sha256":"d03d08a07884884a3b377470f497c2c422445f65f33cd09d1837abd96c35e417"}}