{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:JI4XPD72YNIW3IJG5YIK7ZLZBA","short_pith_number":"pith:JI4XPD72","canonical_record":{"source":{"id":"2505.21670","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-27T18:48:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ed8ca33f4df631d315b70adc9147ab0f8e2c17e1d07a33492319336cbcb657c3","abstract_canon_sha256":"e9e942d94113c6ea6c44c233c20d488f4c1eceed58a4da8c834a8be5bca92675"},"schema_version":"1.0"},"canonical_sha256":"4a39778ffac3516da126ee10afe579081989209a4824a8aa2e0548da01da3eab","source":{"kind":"arxiv","id":"2505.21670","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.21670","created_at":"2026-07-05T11:11:01Z"},{"alias_kind":"arxiv_version","alias_value":"2505.21670v1","created_at":"2026-07-05T11:11:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21670","created_at":"2026-07-05T11:11:01Z"},{"alias_kind":"pith_short_12","alias_value":"JI4XPD72YNIW","created_at":"2026-07-05T11:11:01Z"},{"alias_kind":"pith_short_16","alias_value":"JI4XPD72YNIW3IJG","created_at":"2026-07-05T11:11:01Z"},{"alias_kind":"pith_short_8","alias_value":"JI4XPD72","created_at":"2026-07-05T11:11:01Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:JI4XPD72YNIW3IJG5YIK7ZLZBA","target":"record","payload":{"canonical_record":{"source":{"id":"2505.21670","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-27T18:48:40Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ed8ca33f4df631d315b70adc9147ab0f8e2c17e1d07a33492319336cbcb657c3","abstract_canon_sha256":"e9e942d94113c6ea6c44c233c20d488f4c1eceed58a4da8c834a8be5bca92675"},"schema_version":"1.0"},"canonical_sha256":"4a39778ffac3516da126ee10afe579081989209a4824a8aa2e0548da01da3eab","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:11:01.886740Z","signature_b64":"iM43fNO+uP2oA9bOWKznfLxBxbeX/7IhMGn2KtJiZcvh1x8Nts3UXvPuN14mfQpiv+fjJWQDUl2biArtefjPDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4a39778ffac3516da126ee10afe579081989209a4824a8aa2e0548da01da3eab","last_reissued_at":"2026-07-05T11:11:01.886265Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:11:01.886265Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2505.21670","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":"2026-07-05T11:11:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Xslony2/98gKjoaRqDbEV0fCzUCfXO7OCfdASLgJKdYrI7hN+nom6SbzZRWwlv0G11+k7gko+V2XNfHEZXfLBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T02:14:21.186776Z"},"content_sha256":"a4263546cc92edddee698ccea57f68c09203239f159fc918140d5db5b1091f8c","schema_version":"1.0","event_id":"sha256:a4263546cc92edddee698ccea57f68c09203239f159fc918140d5db5b1091f8c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:JI4XPD72YNIW3IJG5YIK7ZLZBA","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Rethinking the Outlier Distribution in Large Language Models: An In-depth Study","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Khushi Sharma, Rahul Raman, Sai Qian Zhang","submitted_at":"2025-05-27T18:48:40Z","abstract_excerpt":"Investigating outliers in large language models (LLMs) is crucial due to their significant impact on various aspects of LLM performance, including quantization and compression. Outliers often cause considerable quantization errors, leading to degraded model performance. Identifying and addressing these outliers can enhance the accuracy and efficiency of the quantization process, enabling smoother deployment on edge devices or specialized hardware. Recent studies have identified two common types of outliers in LLMs: massive activations and channel-wise outliers. While numerous quantization algo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21670","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/2505.21670/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":"2026-07-05T11:11:01Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"aOrqHB2+wc60NK0ZAyOtIkBGnA9c6TBE0KQR+w3WHSoYet1F+GC9lOYNFJn4DavicVMsQ6bupxgQd7FFKHm1Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-10T02:14:21.187597Z"},"content_sha256":"c6cd94faf55429a4fc6466ff2a91cc8d9b22722dbf5159c40405fd7f9897b2ca","schema_version":"1.0","event_id":"sha256:c6cd94faf55429a4fc6466ff2a91cc8d9b22722dbf5159c40405fd7f9897b2ca"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JI4XPD72YNIW3IJG5YIK7ZLZBA/bundle.json","state_url":"https://pith.science/pith/JI4XPD72YNIW3IJG5YIK7ZLZBA/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JI4XPD72YNIW3IJG5YIK7ZLZBA/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":"2026-08-10T02:14:21Z","links":{"resolver":"https://pith.science/pith/JI4XPD72YNIW3IJG5YIK7ZLZBA","bundle":"https://pith.science/pith/JI4XPD72YNIW3IJG5YIK7ZLZBA/bundle.json","state":"https://pith.science/pith/JI4XPD72YNIW3IJG5YIK7ZLZBA/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JI4XPD72YNIW3IJG5YIK7ZLZBA/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:JI4XPD72YNIW3IJG5YIK7ZLZBA","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":"e9e942d94113c6ea6c44c233c20d488f4c1eceed58a4da8c834a8be5bca92675","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-27T18:48:40Z","title_canon_sha256":"ed8ca33f4df631d315b70adc9147ab0f8e2c17e1d07a33492319336cbcb657c3"},"schema_version":"1.0","source":{"id":"2505.21670","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2505.21670","created_at":"2026-07-05T11:11:01Z"},{"alias_kind":"arxiv_version","alias_value":"2505.21670v1","created_at":"2026-07-05T11:11:01Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.21670","created_at":"2026-07-05T11:11:01Z"},{"alias_kind":"pith_short_12","alias_value":"JI4XPD72YNIW","created_at":"2026-07-05T11:11:01Z"},{"alias_kind":"pith_short_16","alias_value":"JI4XPD72YNIW3IJG","created_at":"2026-07-05T11:11:01Z"},{"alias_kind":"pith_short_8","alias_value":"JI4XPD72","created_at":"2026-07-05T11:11:01Z"}],"graph_snapshots":[{"event_id":"sha256:c6cd94faf55429a4fc6466ff2a91cc8d9b22722dbf5159c40405fd7f9897b2ca","target":"graph","created_at":"2026-07-05T11:11:01Z","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/2505.21670/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Investigating outliers in large language models (LLMs) is crucial due to their significant impact on various aspects of LLM performance, including quantization and compression. Outliers often cause considerable quantization errors, leading to degraded model performance. Identifying and addressing these outliers can enhance the accuracy and efficiency of the quantization process, enabling smoother deployment on edge devices or specialized hardware. Recent studies have identified two common types of outliers in LLMs: massive activations and channel-wise outliers. While numerous quantization algo","authors_text":"Khushi Sharma, Rahul Raman, Sai Qian Zhang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-27T18:48:40Z","title":"Rethinking the Outlier Distribution in Large Language Models: An In-depth Study"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.21670","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:a4263546cc92edddee698ccea57f68c09203239f159fc918140d5db5b1091f8c","target":"record","created_at":"2026-07-05T11:11:01Z","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":"e9e942d94113c6ea6c44c233c20d488f4c1eceed58a4da8c834a8be5bca92675","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-05-27T18:48:40Z","title_canon_sha256":"ed8ca33f4df631d315b70adc9147ab0f8e2c17e1d07a33492319336cbcb657c3"},"schema_version":"1.0","source":{"id":"2505.21670","kind":"arxiv","version":1}},"canonical_sha256":"4a39778ffac3516da126ee10afe579081989209a4824a8aa2e0548da01da3eab","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4a39778ffac3516da126ee10afe579081989209a4824a8aa2e0548da01da3eab","first_computed_at":"2026-07-05T11:11:01.886265Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:11:01.886265Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"iM43fNO+uP2oA9bOWKznfLxBxbeX/7IhMGn2KtJiZcvh1x8Nts3UXvPuN14mfQpiv+fjJWQDUl2biArtefjPDg==","signature_status":"signed_v1","signed_at":"2026-07-05T11:11:01.886740Z","signed_message":"canonical_sha256_bytes"},"source_id":"2505.21670","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a4263546cc92edddee698ccea57f68c09203239f159fc918140d5db5b1091f8c","sha256:c6cd94faf55429a4fc6466ff2a91cc8d9b22722dbf5159c40405fd7f9897b2ca"],"state_sha256":"86354d2d9478e02bfa31e20ed40e062875b588b227ebc7a4acbb69471bbbabec"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pp7G+0luF2Y8DkwsutiIQ7h14B+oWTyL/1+OEHj8B4MDv9zeq62RrjiMuEfBugTl1LuvJUROjni02PaXoWOtDg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-10T02:14:21.193385Z","bundle_sha256":"32657524e0303863e6e0975ac7c82783ab2712a86a6328eb44b573e4a34295f7"}}