{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:FQZXLZPELNRGOANKRWENL6NZW4","short_pith_number":"pith:FQZXLZPE","schema_version":"1.0","canonical_sha256":"2c3375e5e45b626701aa8d88d5f9b9b71847d91752344be784b6e7516adca722","source":{"kind":"arxiv","id":"2206.04685","version":2},"attestation_state":"computed","paper":{"title":"Predictive Exit: Prediction of Fine-Grained Early Exits for Computation- and Energy-Efficient Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AR","cs.NE"],"primary_cat":"cs.LG","authors_text":"An Zou, Chenfei Lou, Xiangjie Li, Yehan Ma, Yingtao Shen, Yuchi Chen, Zhengping Zhu","submitted_at":"2022-06-09T04:13:55Z","abstract_excerpt":"By adding exiting layers to the deep learning networks, early exit can terminate the inference earlier with accurate results. The passive decision-making of whether to exit or continue the next layer has to go through every pre-placed exiting layer until it exits. In addition, it is also hard to adjust the configurations of the computing platforms alongside the inference proceeds. By incorporating a low-cost prediction engine, we propose a Predictive Exit framework for computation- and energy-efficient deep learning applications. Predictive Exit can forecast where the network will exit (i.e., "},"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":"2206.04685","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2022-06-09T04:13:55Z","cross_cats_sorted":["cs.AR","cs.NE"],"title_canon_sha256":"38c283a87c2ada8fa38f9c3761fcf286608cc74b43738349410692e4a688f7f2","abstract_canon_sha256":"2be8247df3b89b255d281697ecf26f594e204ffda307e54355b05947791eda2b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:28:45.661180Z","signature_b64":"XUjAaubRywVDcJwCy6eLgQuzzO9vtQmvYTmnwzovHmwZbCzHevBKU6JoiSwsfcXo8UY9ba0JV7AtuFZlwCdMCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2c3375e5e45b626701aa8d88d5f9b9b71847d91752344be784b6e7516adca722","last_reissued_at":"2026-07-05T05:28:45.660748Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:28:45.660748Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Predictive Exit: Prediction of Fine-Grained Early Exits for Computation- and Energy-Efficient Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AR","cs.NE"],"primary_cat":"cs.LG","authors_text":"An Zou, Chenfei Lou, Xiangjie Li, Yehan Ma, Yingtao Shen, Yuchi Chen, Zhengping Zhu","submitted_at":"2022-06-09T04:13:55Z","abstract_excerpt":"By adding exiting layers to the deep learning networks, early exit can terminate the inference earlier with accurate results. The passive decision-making of whether to exit or continue the next layer has to go through every pre-placed exiting layer until it exits. In addition, it is also hard to adjust the configurations of the computing platforms alongside the inference proceeds. By incorporating a low-cost prediction engine, we propose a Predictive Exit framework for computation- and energy-efficient deep learning applications. Predictive Exit can forecast where the network will exit (i.e., "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2206.04685","kind":"arxiv","version":2},"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/2206.04685/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":"2206.04685","created_at":"2026-07-05T05:28:45.660802+00:00"},{"alias_kind":"arxiv_version","alias_value":"2206.04685v2","created_at":"2026-07-05T05:28:45.660802+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2206.04685","created_at":"2026-07-05T05:28:45.660802+00:00"},{"alias_kind":"pith_short_12","alias_value":"FQZXLZPELNRG","created_at":"2026-07-05T05:28:45.660802+00:00"},{"alias_kind":"pith_short_16","alias_value":"FQZXLZPELNRGOANK","created_at":"2026-07-05T05:28:45.660802+00:00"},{"alias_kind":"pith_short_8","alias_value":"FQZXLZPE","created_at":"2026-07-05T05:28:45.660802+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.02040","citing_title":"M2R2: Mixture of Multi-Rate Residuals for Efficient Transformer Inference","ref_index":22,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FQZXLZPELNRGOANKRWENL6NZW4","json":"https://pith.science/pith/FQZXLZPELNRGOANKRWENL6NZW4.json","graph_json":"https://pith.science/api/pith-number/FQZXLZPELNRGOANKRWENL6NZW4/graph.json","events_json":"https://pith.science/api/pith-number/FQZXLZPELNRGOANKRWENL6NZW4/events.json","paper":"https://pith.science/paper/FQZXLZPE"},"agent_actions":{"view_html":"https://pith.science/pith/FQZXLZPELNRGOANKRWENL6NZW4","download_json":"https://pith.science/pith/FQZXLZPELNRGOANKRWENL6NZW4.json","view_paper":"https://pith.science/paper/FQZXLZPE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2206.04685&json=true","fetch_graph":"https://pith.science/api/pith-number/FQZXLZPELNRGOANKRWENL6NZW4/graph.json","fetch_events":"https://pith.science/api/pith-number/FQZXLZPELNRGOANKRWENL6NZW4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FQZXLZPELNRGOANKRWENL6NZW4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FQZXLZPELNRGOANKRWENL6NZW4/action/storage_attestation","attest_author":"https://pith.science/pith/FQZXLZPELNRGOANKRWENL6NZW4/action/author_attestation","sign_citation":"https://pith.science/pith/FQZXLZPELNRGOANKRWENL6NZW4/action/citation_signature","submit_replication":"https://pith.science/pith/FQZXLZPELNRGOANKRWENL6NZW4/action/replication_record"}},"created_at":"2026-07-05T05:28:45.660802+00:00","updated_at":"2026-07-05T05:28:45.660802+00:00"}