{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:J2FZUJW2B4Z46OFAGGPCGIQ34Z","short_pith_number":"pith:J2FZUJW2","schema_version":"1.0","canonical_sha256":"4e8b9a26da0f33cf38a0319e23221be6724d30bc45726eabe10305f4274a62df","source":{"kind":"arxiv","id":"2502.00745","version":1},"attestation_state":"computed","paper":{"title":"BEEM: Boosting Performance of Early Exit DNNs using Multi-Exit Classifiers as Experts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.LG","authors_text":"Divya Jyoti Bajpai, Manjesh Kumar Hanawal","submitted_at":"2025-02-02T10:35:19Z","abstract_excerpt":"Early Exit (EE) techniques have emerged as a means to reduce inference latency in Deep Neural Networks (DNNs). The latency improvement and accuracy in these techniques crucially depend on the criteria used to make exit decisions. We propose a new decision criterion where exit classifiers are treated as experts BEEM and aggregate their confidence scores. The confidence scores are aggregated only if neighbouring experts are consistent in prediction as the samples pass through them, thus capturing their ensemble effect. A sample exits when the aggregated confidence value exceeds a threshold. The "},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2502.00745","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-02T10:35:19Z","cross_cats_sorted":["cs.CL","cs.CV"],"title_canon_sha256":"82d572907847b5e9f9d2a28fd15aa07c44edcbdcde0c387abbe3c03226ec5e9c","abstract_canon_sha256":"196c67bb662c0126cdf623eb0d92db82802b5a0c12fd36acff643558f1f62af5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:08:37.997098Z","signature_b64":"Xmrzdze3qiWuI1QX1qZ/nu7FP3KoLAOSs+cot+iZkJaJvLloVqmRFECagpspc30n6+X3pmSGkN0PqhGy5yHiAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4e8b9a26da0f33cf38a0319e23221be6724d30bc45726eabe10305f4274a62df","last_reissued_at":"2026-07-05T10:08:37.996694Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:08:37.996694Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BEEM: Boosting Performance of Early Exit DNNs using Multi-Exit Classifiers as Experts","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.CV"],"primary_cat":"cs.LG","authors_text":"Divya Jyoti Bajpai, Manjesh Kumar Hanawal","submitted_at":"2025-02-02T10:35:19Z","abstract_excerpt":"Early Exit (EE) techniques have emerged as a means to reduce inference latency in Deep Neural Networks (DNNs). The latency improvement and accuracy in these techniques crucially depend on the criteria used to make exit decisions. We propose a new decision criterion where exit classifiers are treated as experts BEEM and aggregate their confidence scores. The confidence scores are aggregated only if neighbouring experts are consistent in prediction as the samples pass through them, thus capturing their ensemble effect. A sample exits when the aggregated confidence value exceeds a threshold. The "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.00745","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/2502.00745/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2502.00745","created_at":"2026-07-05T10:08:37.996748+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.00745v1","created_at":"2026-07-05T10:08:37.996748+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.00745","created_at":"2026-07-05T10:08:37.996748+00:00"},{"alias_kind":"pith_short_12","alias_value":"J2FZUJW2B4Z4","created_at":"2026-07-05T10:08:37.996748+00:00"},{"alias_kind":"pith_short_16","alias_value":"J2FZUJW2B4Z46OFA","created_at":"2026-07-05T10:08:37.996748+00:00"},{"alias_kind":"pith_short_8","alias_value":"J2FZUJW2","created_at":"2026-07-05T10:08:37.996748+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/J2FZUJW2B4Z46OFAGGPCGIQ34Z","json":"https://pith.science/pith/J2FZUJW2B4Z46OFAGGPCGIQ34Z.json","graph_json":"https://pith.science/api/pith-number/J2FZUJW2B4Z46OFAGGPCGIQ34Z/graph.json","events_json":"https://pith.science/api/pith-number/J2FZUJW2B4Z46OFAGGPCGIQ34Z/events.json","paper":"https://pith.science/paper/J2FZUJW2"},"agent_actions":{"view_html":"https://pith.science/pith/J2FZUJW2B4Z46OFAGGPCGIQ34Z","download_json":"https://pith.science/pith/J2FZUJW2B4Z46OFAGGPCGIQ34Z.json","view_paper":"https://pith.science/paper/J2FZUJW2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.00745&json=true","fetch_graph":"https://pith.science/api/pith-number/J2FZUJW2B4Z46OFAGGPCGIQ34Z/graph.json","fetch_events":"https://pith.science/api/pith-number/J2FZUJW2B4Z46OFAGGPCGIQ34Z/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/J2FZUJW2B4Z46OFAGGPCGIQ34Z/action/timestamp_anchor","attest_storage":"https://pith.science/pith/J2FZUJW2B4Z46OFAGGPCGIQ34Z/action/storage_attestation","attest_author":"https://pith.science/pith/J2FZUJW2B4Z46OFAGGPCGIQ34Z/action/author_attestation","sign_citation":"https://pith.science/pith/J2FZUJW2B4Z46OFAGGPCGIQ34Z/action/citation_signature","submit_replication":"https://pith.science/pith/J2FZUJW2B4Z46OFAGGPCGIQ34Z/action/replication_record"}},"created_at":"2026-07-05T10:08:37.996748+00:00","updated_at":"2026-07-05T10:08:37.996748+00:00"}