{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:4PFNWQX3ERDAI3KV7SQF5345WF","short_pith_number":"pith:4PFNWQX3","canonical_record":{"source":{"id":"2506.04142","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-04T16:33:44Z","cross_cats_sorted":[],"title_canon_sha256":"ac3a6afab0449f6b01c4b9802bb791f8233d36f1f9df161e2d64de9f7c699653","abstract_canon_sha256":"11e23ae88e7a86eb5b08f9ef26428682f8e0b8685c6f7dbdb0089b3a6878b671"},"schema_version":"1.0"},"canonical_sha256":"e3cadb42fb2446046d55fca05eef9db161edbf2484ea6583875a201a96008d33","source":{"kind":"arxiv","id":"2506.04142","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.04142","created_at":"2026-07-05T11:15:57Z"},{"alias_kind":"arxiv_version","alias_value":"2506.04142v1","created_at":"2026-07-05T11:15:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.04142","created_at":"2026-07-05T11:15:57Z"},{"alias_kind":"pith_short_12","alias_value":"4PFNWQX3ERDA","created_at":"2026-07-05T11:15:57Z"},{"alias_kind":"pith_short_16","alias_value":"4PFNWQX3ERDAI3KV","created_at":"2026-07-05T11:15:57Z"},{"alias_kind":"pith_short_8","alias_value":"4PFNWQX3","created_at":"2026-07-05T11:15:57Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:4PFNWQX3ERDAI3KV7SQF5345WF","target":"record","payload":{"canonical_record":{"source":{"id":"2506.04142","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-04T16:33:44Z","cross_cats_sorted":[],"title_canon_sha256":"ac3a6afab0449f6b01c4b9802bb791f8233d36f1f9df161e2d64de9f7c699653","abstract_canon_sha256":"11e23ae88e7a86eb5b08f9ef26428682f8e0b8685c6f7dbdb0089b3a6878b671"},"schema_version":"1.0"},"canonical_sha256":"e3cadb42fb2446046d55fca05eef9db161edbf2484ea6583875a201a96008d33","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:15:57.915847Z","signature_b64":"udsPsEroiqos9Npj7yJo+O7IUNvryyz6OrLbYQBiixc+/NfdvTcJAh1UxXvPvaE5qnAlqLjQhpR6vf85fB/HDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e3cadb42fb2446046d55fca05eef9db161edbf2484ea6583875a201a96008d33","last_reissued_at":"2026-07-05T11:15:57.915129Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:15:57.915129Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2506.04142","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":"2026-07-05T11:15:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"kGUqQy81lnzjmMYQgFFTEXms+Pff7WKnb3vJWBAfjDDFU81kOxts4Y4WN0onwwi6hRIfKOGmKhSdFnvXtKVBDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T06:59:21.895470Z"},"content_sha256":"e1fd9f38b69aec52b943213fe154b7e56db48edde67c5702af2b2a4eee04e3fe","schema_version":"1.0","event_id":"sha256:e1fd9f38b69aec52b943213fe154b7e56db48edde67c5702af2b2a4eee04e3fe"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:4PFNWQX3ERDAI3KV7SQF5345WF","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Establishing Trustworthy LLM Evaluation via Shortcut Neuron Analysis","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Juanzi Li, Jun Zhao, Kejian Zhu, Lei Hou, Shangqing Tu, Zhuoran Jin","submitted_at":"2025-06-04T16:33:44Z","abstract_excerpt":"The development of large language models (LLMs) depends on trustworthy evaluation. However, most current evaluations rely on public benchmarks, which are prone to data contamination issues that significantly compromise fairness. Previous researches have focused on constructing dynamic benchmarks to address contamination. However, continuously building new benchmarks is costly and cyclical. In this work, we aim to tackle contamination by analyzing the mechanisms of contaminated models themselves. Through our experiments, we discover that the overestimation of contaminated models is likely due t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.04142","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/2506.04142/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":"2026-07-05T11:15:57Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Sj4I8xT3MqcA3kPbo4qaVIUiFNtpIsY+AGmNQUF5RXIhS9aGPtOPCDfrbfKXZGAfE1Eki5H0GW9LhZfJx4ZUCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T06:59:21.895858Z"},"content_sha256":"aa270b2f61ec467a32055b06e3674cd1356ea3ecf1df75e58ca85bf9b76e64a0","schema_version":"1.0","event_id":"sha256:aa270b2f61ec467a32055b06e3674cd1356ea3ecf1df75e58ca85bf9b76e64a0"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/4PFNWQX3ERDAI3KV7SQF5345WF/bundle.json","state_url":"https://pith.science/pith/4PFNWQX3ERDAI3KV7SQF5345WF/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/4PFNWQX3ERDAI3KV7SQF5345WF/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":"2026-07-06T06:59:21Z","links":{"resolver":"https://pith.science/pith/4PFNWQX3ERDAI3KV7SQF5345WF","bundle":"https://pith.science/pith/4PFNWQX3ERDAI3KV7SQF5345WF/bundle.json","state":"https://pith.science/pith/4PFNWQX3ERDAI3KV7SQF5345WF/state.json","well_known_bundle":"https://pith.science/.well-known/pith/4PFNWQX3ERDAI3KV7SQF5345WF/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:4PFNWQX3ERDAI3KV7SQF5345WF","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":"11e23ae88e7a86eb5b08f9ef26428682f8e0b8685c6f7dbdb0089b3a6878b671","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-04T16:33:44Z","title_canon_sha256":"ac3a6afab0449f6b01c4b9802bb791f8233d36f1f9df161e2d64de9f7c699653"},"schema_version":"1.0","source":{"id":"2506.04142","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.04142","created_at":"2026-07-05T11:15:57Z"},{"alias_kind":"arxiv_version","alias_value":"2506.04142v1","created_at":"2026-07-05T11:15:57Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.04142","created_at":"2026-07-05T11:15:57Z"},{"alias_kind":"pith_short_12","alias_value":"4PFNWQX3ERDA","created_at":"2026-07-05T11:15:57Z"},{"alias_kind":"pith_short_16","alias_value":"4PFNWQX3ERDAI3KV","created_at":"2026-07-05T11:15:57Z"},{"alias_kind":"pith_short_8","alias_value":"4PFNWQX3","created_at":"2026-07-05T11:15:57Z"}],"graph_snapshots":[{"event_id":"sha256:aa270b2f61ec467a32055b06e3674cd1356ea3ecf1df75e58ca85bf9b76e64a0","target":"graph","created_at":"2026-07-05T11:15:57Z","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/2506.04142/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The development of large language models (LLMs) depends on trustworthy evaluation. However, most current evaluations rely on public benchmarks, which are prone to data contamination issues that significantly compromise fairness. Previous researches have focused on constructing dynamic benchmarks to address contamination. However, continuously building new benchmarks is costly and cyclical. In this work, we aim to tackle contamination by analyzing the mechanisms of contaminated models themselves. Through our experiments, we discover that the overestimation of contaminated models is likely due t","authors_text":"Juanzi Li, Jun Zhao, Kejian Zhu, Lei Hou, Shangqing Tu, Zhuoran Jin","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-04T16:33:44Z","title":"Establishing Trustworthy LLM Evaluation via Shortcut Neuron Analysis"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.04142","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:e1fd9f38b69aec52b943213fe154b7e56db48edde67c5702af2b2a4eee04e3fe","target":"record","created_at":"2026-07-05T11:15:57Z","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":"11e23ae88e7a86eb5b08f9ef26428682f8e0b8685c6f7dbdb0089b3a6878b671","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-06-04T16:33:44Z","title_canon_sha256":"ac3a6afab0449f6b01c4b9802bb791f8233d36f1f9df161e2d64de9f7c699653"},"schema_version":"1.0","source":{"id":"2506.04142","kind":"arxiv","version":1}},"canonical_sha256":"e3cadb42fb2446046d55fca05eef9db161edbf2484ea6583875a201a96008d33","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"e3cadb42fb2446046d55fca05eef9db161edbf2484ea6583875a201a96008d33","first_computed_at":"2026-07-05T11:15:57.915129Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:15:57.915129Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"udsPsEroiqos9Npj7yJo+O7IUNvryyz6OrLbYQBiixc+/NfdvTcJAh1UxXvPvaE5qnAlqLjQhpR6vf85fB/HDA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:15:57.915847Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.04142","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e1fd9f38b69aec52b943213fe154b7e56db48edde67c5702af2b2a4eee04e3fe","sha256:aa270b2f61ec467a32055b06e3674cd1356ea3ecf1df75e58ca85bf9b76e64a0"],"state_sha256":"d62ce68bcdea7867f05a110c732c8336446b94a5c21161d7fe2732693fa426a1"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Ni750gvCh2aynkLkKV1Mfq9IdoMy2kHxgl6KaBKQA0Bbq4KVKa+iShCjTG8uD7q6roIYGvZll4RNo0QxNmnKAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T06:59:21.897781Z","bundle_sha256":"8960314d7646e7967a5f9a50ca159a03e743b3994345fc9511b92e8461404b62"}}