{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:YBY74ZLM4V6TDIWCOSK2SDRRLH","short_pith_number":"pith:YBY74ZLM","schema_version":"1.0","canonical_sha256":"c071fe656ce57d31a2c27495a90e3159d75976c95b8bd0a62a0419264870a836","source":{"kind":"arxiv","id":"2009.06732","version":3},"attestation_state":"computed","paper":{"title":"Efficient Transformers: A Survey","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV","cs.IR"],"primary_cat":"cs.LG","authors_text":"Dara Bahri, Donald Metzler, Mostafa Dehghani, Yi Tay","submitted_at":"2020-09-14T20:38:14Z","abstract_excerpt":"Transformer model architectures have garnered immense interest lately due to their effectiveness across a range of domains like language, vision and reinforcement learning. In the field of natural language processing for example, Transformers have become an indispensable staple in the modern deep learning stack. Recently, a dizzying number of \"X-former\" models have been proposed - Reformer, Linformer, Performer, Longformer, to name a few - which improve upon the original Transformer architecture, many of which make improvements around computational and memory efficiency. With the aim of helpin"},"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":"2009.06732","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2020-09-14T20:38:14Z","cross_cats_sorted":["cs.AI","cs.CL","cs.CV","cs.IR"],"title_canon_sha256":"0eb54469d6f4c00e3b13abf9774b6772ce4a15c28503f983fa265eb1297f493a","abstract_canon_sha256":"458a847a570df7065c26de18dc613cf8d2ebe4df0d6bb7b1bf06269cd1275a12"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:04:35.720222Z","signature_b64":"mxPhyGXF4iky/0eE5NGZVoITvWh0ndyOVyqAUkCNXZN86BOtqEg6x/M9Fg/p4c6Y4KceUlCyLYsGtdC9YUsfDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c071fe656ce57d31a2c27495a90e3159d75976c95b8bd0a62a0419264870a836","last_reissued_at":"2026-07-05T04:04:35.719710Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:04:35.719710Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Efficient Transformers: A Survey","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.CV","cs.IR"],"primary_cat":"cs.LG","authors_text":"Dara Bahri, Donald Metzler, Mostafa Dehghani, Yi Tay","submitted_at":"2020-09-14T20:38:14Z","abstract_excerpt":"Transformer model architectures have garnered immense interest lately due to their effectiveness across a range of domains like language, vision and reinforcement learning. In the field of natural language processing for example, Transformers have become an indispensable staple in the modern deep learning stack. Recently, a dizzying number of \"X-former\" models have been proposed - Reformer, Linformer, Performer, Longformer, to name a few - which improve upon the original Transformer architecture, many of which make improvements around computational and memory efficiency. With the aim of helpin"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2009.06732","kind":"arxiv","version":3},"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/2009.06732/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":"2009.06732","created_at":"2026-07-05T04:04:35.719779+00:00"},{"alias_kind":"arxiv_version","alias_value":"2009.06732v3","created_at":"2026-07-05T04:04:35.719779+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2009.06732","created_at":"2026-07-05T04:04:35.719779+00:00"},{"alias_kind":"pith_short_12","alias_value":"YBY74ZLM4V6T","created_at":"2026-07-05T04:04:35.719779+00:00"},{"alias_kind":"pith_short_16","alias_value":"YBY74ZLM4V6TDIWC","created_at":"2026-07-05T04:04:35.719779+00:00"},{"alias_kind":"pith_short_8","alias_value":"YBY74ZLM","created_at":"2026-07-05T04:04:35.719779+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":14,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.03569","citing_title":"When Attention Collapses: Stage-Aware Visual Token Pruning from Structure to Semantics","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2202.10873","citing_title":"Ligandformer: A Graph Neural Network for Predicting Compound Property with Robust Interpretation","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15104","citing_title":"From Text to Voice: A Reproducible and Verifiable Framework for Evaluating Tool Calling LLM Agents","ref_index":75,"is_internal_anchor":false},{"citing_arxiv_id":"2407.08608","citing_title":"FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2306.14048","citing_title":"H$_2$O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2404.02258","citing_title":"Mixture-of-Depths: Dynamically allocating compute in transformer-based language models","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2502.13189","citing_title":"MoBA: Mixture of Block Attention for Long-Context LLMs","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2107.14795","citing_title":"Perceiver IO: A General Architecture for Structured Inputs & Outputs","ref_index":81,"is_internal_anchor":false},{"citing_arxiv_id":"2603.01960","citing_title":"TiledAttention: a CUDA Tile SDPA Kernel for PyTorch","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2603.14360","citing_title":"M$^2$RNN: Non-Linear RNNs with Matrix-Valued States for Scalable Language Modeling","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2401.10774","citing_title":"Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads","ref_index":215,"is_internal_anchor":false},{"citing_arxiv_id":"2205.14135","citing_title":"FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness","ref_index":81,"is_internal_anchor":false},{"citing_arxiv_id":"2010.04159","citing_title":"Deformable DETR: Deformable Transformers for End-to-End Object Detection","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20595","citing_title":"An explicit operator explains end-to-end computation in the modern neural networks used for sequence and language modeling","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YBY74ZLM4V6TDIWCOSK2SDRRLH","json":"https://pith.science/pith/YBY74ZLM4V6TDIWCOSK2SDRRLH.json","graph_json":"https://pith.science/api/pith-number/YBY74ZLM4V6TDIWCOSK2SDRRLH/graph.json","events_json":"https://pith.science/api/pith-number/YBY74ZLM4V6TDIWCOSK2SDRRLH/events.json","paper":"https://pith.science/paper/YBY74ZLM"},"agent_actions":{"view_html":"https://pith.science/pith/YBY74ZLM4V6TDIWCOSK2SDRRLH","download_json":"https://pith.science/pith/YBY74ZLM4V6TDIWCOSK2SDRRLH.json","view_paper":"https://pith.science/paper/YBY74ZLM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2009.06732&json=true","fetch_graph":"https://pith.science/api/pith-number/YBY74ZLM4V6TDIWCOSK2SDRRLH/graph.json","fetch_events":"https://pith.science/api/pith-number/YBY74ZLM4V6TDIWCOSK2SDRRLH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YBY74ZLM4V6TDIWCOSK2SDRRLH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YBY74ZLM4V6TDIWCOSK2SDRRLH/action/storage_attestation","attest_author":"https://pith.science/pith/YBY74ZLM4V6TDIWCOSK2SDRRLH/action/author_attestation","sign_citation":"https://pith.science/pith/YBY74ZLM4V6TDIWCOSK2SDRRLH/action/citation_signature","submit_replication":"https://pith.science/pith/YBY74ZLM4V6TDIWCOSK2SDRRLH/action/replication_record"}},"created_at":"2026-07-05T04:04:35.719779+00:00","updated_at":"2026-07-05T04:04:35.719779+00:00"}