{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:SXHNF4KOXI5BRPIMSLFC2ME62M","short_pith_number":"pith:SXHNF4KO","canonical_record":{"source":{"id":"2409.03856","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-05T18:38:07Z","cross_cats_sorted":[],"title_canon_sha256":"4bb819b18c1e2b2820d8325ca537ef0ac605cd6e6a0fac34f72684085fd0e743","abstract_canon_sha256":"50d51518ada2a2089fc2913632ede8d5a3930181ddbb3dfb4764a5c42e9ceefb"},"schema_version":"1.0"},"canonical_sha256":"95ced2f14eba3a18bd0c92ca2d309ed32eb9b8a9879d07aab2535d56e25b78c9","source":{"kind":"arxiv","id":"2409.03856","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.03856","created_at":"2026-07-05T09:03:58Z"},{"alias_kind":"arxiv_version","alias_value":"2409.03856v1","created_at":"2026-07-05T09:03:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.03856","created_at":"2026-07-05T09:03:58Z"},{"alias_kind":"pith_short_12","alias_value":"SXHNF4KOXI5B","created_at":"2026-07-05T09:03:58Z"},{"alias_kind":"pith_short_16","alias_value":"SXHNF4KOXI5BRPIM","created_at":"2026-07-05T09:03:58Z"},{"alias_kind":"pith_short_8","alias_value":"SXHNF4KO","created_at":"2026-07-05T09:03:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:SXHNF4KOXI5BRPIMSLFC2ME62M","target":"record","payload":{"canonical_record":{"source":{"id":"2409.03856","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-05T18:38:07Z","cross_cats_sorted":[],"title_canon_sha256":"4bb819b18c1e2b2820d8325ca537ef0ac605cd6e6a0fac34f72684085fd0e743","abstract_canon_sha256":"50d51518ada2a2089fc2913632ede8d5a3930181ddbb3dfb4764a5c42e9ceefb"},"schema_version":"1.0"},"canonical_sha256":"95ced2f14eba3a18bd0c92ca2d309ed32eb9b8a9879d07aab2535d56e25b78c9","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:03:58.187836Z","signature_b64":"vZXO57jTQ8s02zxLIhjlVflQfIbJWHVFdGbCuHV2lozZFHIh0hLr1cdHU+9eZ9ycPuV4Ia7XZIEMDkf66VlVAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"95ced2f14eba3a18bd0c92ca2d309ed32eb9b8a9879d07aab2535d56e25b78c9","last_reissued_at":"2026-07-05T09:03:58.187417Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:03:58.187417Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2409.03856","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":"2026-07-05T09:03:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"E3jlbv7wqKWOrcdQN2I24mF2ncAqacG8YsAidOXi94tXmJOqjTLNOY+qOwDwTXiEKdYjrCZYleuStxnftCyPAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T09:34:34.556660Z"},"content_sha256":"e59bce7a963e5fbc9415308883f0d7d7cbb5a58c360f5952c68890d21ff2dfc7","schema_version":"1.0","event_id":"sha256:e59bce7a963e5fbc9415308883f0d7d7cbb5a58c360f5952c68890d21ff2dfc7"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:SXHNF4KOXI5BRPIMSLFC2ME62M","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Sirius: Contextual Sparsity with Correction for Efficient LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Beidi Chen, Victoria Lin, Yang Zhou, Zhaozhuo Xu, Zhuoming Chen","submitted_at":"2024-09-05T18:38:07Z","abstract_excerpt":"With the blossom of large language models (LLMs), inference efficiency becomes increasingly important. Various approximation methods are proposed to reduce the cost at inference time. Contextual Sparsity (CS) is appealing for its training-free nature and its ability to reach a higher compression ratio seemingly without quality degradation. However, after a comprehensive evaluation of contextual sparsity methods on various complex generation tasks, we find that although CS succeeds in prompt-understanding tasks, CS significantly degrades the model performance for reasoning, deduction, and knowl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.03856","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/2409.03856/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":"2026-07-05T09:03:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"23YWucV9P4pv16TcnMHVYkINUfdEcUfXa21v+t/qYik+RwZ9OhLD+esiRaWzdEc8EKxHbKyuASIRbk3d7u6UDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T09:34:34.557156Z"},"content_sha256":"2420fce7761b5f63b5f4e90ccf51797adb82e8f5d92742be209efc1133798285","schema_version":"1.0","event_id":"sha256:2420fce7761b5f63b5f4e90ccf51797adb82e8f5d92742be209efc1133798285"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/SXHNF4KOXI5BRPIMSLFC2ME62M/bundle.json","state_url":"https://pith.science/pith/SXHNF4KOXI5BRPIMSLFC2ME62M/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/SXHNF4KOXI5BRPIMSLFC2ME62M/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":"2026-08-08T09:34:34Z","links":{"resolver":"https://pith.science/pith/SXHNF4KOXI5BRPIMSLFC2ME62M","bundle":"https://pith.science/pith/SXHNF4KOXI5BRPIMSLFC2ME62M/bundle.json","state":"https://pith.science/pith/SXHNF4KOXI5BRPIMSLFC2ME62M/state.json","well_known_bundle":"https://pith.science/.well-known/pith/SXHNF4KOXI5BRPIMSLFC2ME62M/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:SXHNF4KOXI5BRPIMSLFC2ME62M","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":"50d51518ada2a2089fc2913632ede8d5a3930181ddbb3dfb4764a5c42e9ceefb","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-05T18:38:07Z","title_canon_sha256":"4bb819b18c1e2b2820d8325ca537ef0ac605cd6e6a0fac34f72684085fd0e743"},"schema_version":"1.0","source":{"id":"2409.03856","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2409.03856","created_at":"2026-07-05T09:03:58Z"},{"alias_kind":"arxiv_version","alias_value":"2409.03856v1","created_at":"2026-07-05T09:03:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.03856","created_at":"2026-07-05T09:03:58Z"},{"alias_kind":"pith_short_12","alias_value":"SXHNF4KOXI5B","created_at":"2026-07-05T09:03:58Z"},{"alias_kind":"pith_short_16","alias_value":"SXHNF4KOXI5BRPIM","created_at":"2026-07-05T09:03:58Z"},{"alias_kind":"pith_short_8","alias_value":"SXHNF4KO","created_at":"2026-07-05T09:03:58Z"}],"graph_snapshots":[{"event_id":"sha256:2420fce7761b5f63b5f4e90ccf51797adb82e8f5d92742be209efc1133798285","target":"graph","created_at":"2026-07-05T09:03:58Z","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/2409.03856/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"With the blossom of large language models (LLMs), inference efficiency becomes increasingly important. Various approximation methods are proposed to reduce the cost at inference time. Contextual Sparsity (CS) is appealing for its training-free nature and its ability to reach a higher compression ratio seemingly without quality degradation. However, after a comprehensive evaluation of contextual sparsity methods on various complex generation tasks, we find that although CS succeeds in prompt-understanding tasks, CS significantly degrades the model performance for reasoning, deduction, and knowl","authors_text":"Beidi Chen, Victoria Lin, Yang Zhou, Zhaozhuo Xu, Zhuoming Chen","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-05T18:38:07Z","title":"Sirius: Contextual Sparsity with Correction for Efficient LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.03856","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:e59bce7a963e5fbc9415308883f0d7d7cbb5a58c360f5952c68890d21ff2dfc7","target":"record","created_at":"2026-07-05T09:03:58Z","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":"50d51518ada2a2089fc2913632ede8d5a3930181ddbb3dfb4764a5c42e9ceefb","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-09-05T18:38:07Z","title_canon_sha256":"4bb819b18c1e2b2820d8325ca537ef0ac605cd6e6a0fac34f72684085fd0e743"},"schema_version":"1.0","source":{"id":"2409.03856","kind":"arxiv","version":1}},"canonical_sha256":"95ced2f14eba3a18bd0c92ca2d309ed32eb9b8a9879d07aab2535d56e25b78c9","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"95ced2f14eba3a18bd0c92ca2d309ed32eb9b8a9879d07aab2535d56e25b78c9","first_computed_at":"2026-07-05T09:03:58.187417Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:03:58.187417Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"vZXO57jTQ8s02zxLIhjlVflQfIbJWHVFdGbCuHV2lozZFHIh0hLr1cdHU+9eZ9ycPuV4Ia7XZIEMDkf66VlVAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T09:03:58.187836Z","signed_message":"canonical_sha256_bytes"},"source_id":"2409.03856","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e59bce7a963e5fbc9415308883f0d7d7cbb5a58c360f5952c68890d21ff2dfc7","sha256:2420fce7761b5f63b5f4e90ccf51797adb82e8f5d92742be209efc1133798285"],"state_sha256":"0085a8fc3215503ddf4d2e63263c69040a3eaded96acd1c2f3852610e7dad425"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oknlRHYXd8t1IkeDQQtOphqWQvyyYg4EYfODzBQY526PwRkpZxeWjun4L/Np/zAalQMKj7FGmjTN6QBK91mjAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T09:34:34.560577Z","bundle_sha256":"138012359bedaf2aea1b9c098ed3e61cb374ac55e36a1e42c1dd993db0864423"}}