{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:O6V6VIEB2AMBWZDLPGGJDPYADD","short_pith_number":"pith:O6V6VIEB","schema_version":"1.0","canonical_sha256":"77abeaa081d0181b646b798c91bf0018e60670eee6755f1acc2d98e9f12b6f74","source":{"kind":"arxiv","id":"1910.10073","version":4},"attestation_state":"computed","paper":{"title":"Depth-Adaptive Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Edouard Grave, Jiatao Gu, Maha Elbayad, Michael Auli","submitted_at":"2019-10-22T16:15:58Z","abstract_excerpt":"State of the art sequence-to-sequence models for large scale tasks perform a fixed number of computations for each input sequence regardless of whether it is easy or hard to process. In this paper, we train Transformer models which can make output predictions at different stages of the network and we investigate different ways to predict how much computation is required for a particular sequence. Unlike dynamic computation in Universal Transformers, which applies the same set of layers iteratively, we apply different layers at every step to adjust both the amount of computation as well as 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":"1910.10073","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2019-10-22T16:15:58Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"00a8643c7e68cf93e46cc7688a9cffb16b4b7e62cb0e7ecebeeb67ec0b4dea31","abstract_canon_sha256":"fc97f2497879a411477606d546cfd5a3414ff79f1b7bca9de964cb01de705b34"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:40:59.715511Z","signature_b64":"i9EYTY1NJX3/3/0CC3SoL3ETYirh0hXd/9IJRH9/x46SN2wzPDItf9z2afj+3slnWbPwQqeEyPmFnZ6lXYdVAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"77abeaa081d0181b646b798c91bf0018e60670eee6755f1acc2d98e9f12b6f74","last_reissued_at":"2026-07-05T00:40:59.715060Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:40:59.715060Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Depth-Adaptive Transformer","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Edouard Grave, Jiatao Gu, Maha Elbayad, Michael Auli","submitted_at":"2019-10-22T16:15:58Z","abstract_excerpt":"State of the art sequence-to-sequence models for large scale tasks perform a fixed number of computations for each input sequence regardless of whether it is easy or hard to process. In this paper, we train Transformer models which can make output predictions at different stages of the network and we investigate different ways to predict how much computation is required for a particular sequence. Unlike dynamic computation in Universal Transformers, which applies the same set of layers iteratively, we apply different layers at every step to adjust both the amount of computation as well as the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1910.10073","kind":"arxiv","version":4},"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/1910.10073/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":"1910.10073","created_at":"2026-07-05T00:40:59.715118+00:00"},{"alias_kind":"arxiv_version","alias_value":"1910.10073v4","created_at":"2026-07-05T00:40:59.715118+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.10073","created_at":"2026-07-05T00:40:59.715118+00:00"},{"alias_kind":"pith_short_12","alias_value":"O6V6VIEB2AMB","created_at":"2026-07-05T00:40:59.715118+00:00"},{"alias_kind":"pith_short_16","alias_value":"O6V6VIEB2AMBWZDL","created_at":"2026-07-05T00:40:59.715118+00:00"},{"alias_kind":"pith_short_8","alias_value":"O6V6VIEB","created_at":"2026-07-05T00:40:59.715118+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.03054","citing_title":"ToolGate: Token-Efficient Pre-Call Control for Tool-Augmented Vision-Language Agents","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26099","citing_title":"Do Language Models Need Sleep? Offline Recurrence for Improved Online Inference","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19376","citing_title":"Generative Recursive Reasoning","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19376","citing_title":"Generative Recursive Reasoning","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2211.17192","citing_title":"Fast Inference from Transformers via Speculative Decoding","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2404.02258","citing_title":"Mixture-of-Depths: Dynamically allocating compute in transformer-based language models","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03263","citing_title":"LPC-SM: Local Predictive Coding and Sparse Memory for Long-Context Language Modeling","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2604.27981","citing_title":"ITS-Mina: A Harris Hawks Optimization-Based All-MLP Framework with Iterative Refinement and External Attention for Multivariate Time Series Forecasting","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.10875","citing_title":"Compute Where it Counts: Self Optimizing Language Models","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2006.16668","citing_title":"GShard: Scaling Giant Models with Conditional Computation and Automatic Sharding","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/O6V6VIEB2AMBWZDLPGGJDPYADD","json":"https://pith.science/pith/O6V6VIEB2AMBWZDLPGGJDPYADD.json","graph_json":"https://pith.science/api/pith-number/O6V6VIEB2AMBWZDLPGGJDPYADD/graph.json","events_json":"https://pith.science/api/pith-number/O6V6VIEB2AMBWZDLPGGJDPYADD/events.json","paper":"https://pith.science/paper/O6V6VIEB"},"agent_actions":{"view_html":"https://pith.science/pith/O6V6VIEB2AMBWZDLPGGJDPYADD","download_json":"https://pith.science/pith/O6V6VIEB2AMBWZDLPGGJDPYADD.json","view_paper":"https://pith.science/paper/O6V6VIEB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1910.10073&json=true","fetch_graph":"https://pith.science/api/pith-number/O6V6VIEB2AMBWZDLPGGJDPYADD/graph.json","fetch_events":"https://pith.science/api/pith-number/O6V6VIEB2AMBWZDLPGGJDPYADD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O6V6VIEB2AMBWZDLPGGJDPYADD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O6V6VIEB2AMBWZDLPGGJDPYADD/action/storage_attestation","attest_author":"https://pith.science/pith/O6V6VIEB2AMBWZDLPGGJDPYADD/action/author_attestation","sign_citation":"https://pith.science/pith/O6V6VIEB2AMBWZDLPGGJDPYADD/action/citation_signature","submit_replication":"https://pith.science/pith/O6V6VIEB2AMBWZDLPGGJDPYADD/action/replication_record"}},"created_at":"2026-07-05T00:40:59.715118+00:00","updated_at":"2026-07-05T00:40:59.715118+00:00"}