{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:IUKHAW67LADQ7FFQ6LTVZJOB77","short_pith_number":"pith:IUKHAW67","schema_version":"1.0","canonical_sha256":"4514705bdf58070f94b0f2e75ca5c1fff2a9d0d3e6d926080884245cb5225207","source":{"kind":"arxiv","id":"2402.16363","version":6},"attestation_state":"computed","paper":{"title":"LLM Inference Unveiled: Survey and Roofline Model Insights","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Beidi Chen, Bingzhe Wu, Chenhao Xue, Guangyu Sun, Kurt Keutzer, Qingyi Gu, Yang Zhou, Yan Yan, Yong Jae Lee, Yuzhang Shang, Zhen Dong, Zhe Zhou, Zhihang Yuan, Zhikai Li","submitted_at":"2024-02-26T07:33:05Z","abstract_excerpt":"The field of efficient Large Language Model (LLM) inference is rapidly evolving, presenting a unique blend of opportunities and challenges. Although the field has expanded and is vibrant, there hasn't been a concise framework that analyzes the various methods of LLM Inference to provide a clear understanding of this domain. Our survey stands out from traditional literature reviews by not only summarizing the current state of research but also by introducing a framework based on roofline model for systematic analysis of LLM inference techniques. This framework identifies the bottlenecks when de"},"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":"2402.16363","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-02-26T07:33:05Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"24d72644baf3ed88052f0e1f3de8c592ce9c0f4c11879e16549bd90652972e74","abstract_canon_sha256":"758dcf64b6cc598e7bf257de6e80333d7fd154e87362bd610e63567d1553c21f"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:14:22.253131Z","signature_b64":"5gAoxoeiITjRLS24T1+F5n1x3cL8we8HTOLqqCuPEpYy1wf0p9PJzdbSxehhHv4CeQs8u61zddC9IBDRka6dBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4514705bdf58070f94b0f2e75ca5c1fff2a9d0d3e6d926080884245cb5225207","last_reissued_at":"2026-07-05T08:14:22.252642Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:14:22.252642Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LLM Inference Unveiled: Survey and Roofline Model Insights","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Beidi Chen, Bingzhe Wu, Chenhao Xue, Guangyu Sun, Kurt Keutzer, Qingyi Gu, Yang Zhou, Yan Yan, Yong Jae Lee, Yuzhang Shang, Zhen Dong, Zhe Zhou, Zhihang Yuan, Zhikai Li","submitted_at":"2024-02-26T07:33:05Z","abstract_excerpt":"The field of efficient Large Language Model (LLM) inference is rapidly evolving, presenting a unique blend of opportunities and challenges. Although the field has expanded and is vibrant, there hasn't been a concise framework that analyzes the various methods of LLM Inference to provide a clear understanding of this domain. Our survey stands out from traditional literature reviews by not only summarizing the current state of research but also by introducing a framework based on roofline model for systematic analysis of LLM inference techniques. This framework identifies the bottlenecks when de"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.16363","kind":"arxiv","version":6},"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/2402.16363/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":"2402.16363","created_at":"2026-07-05T08:14:22.252705+00:00"},{"alias_kind":"arxiv_version","alias_value":"2402.16363v6","created_at":"2026-07-05T08:14:22.252705+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.16363","created_at":"2026-07-05T08:14:22.252705+00:00"},{"alias_kind":"pith_short_12","alias_value":"IUKHAW67LADQ","created_at":"2026-07-05T08:14:22.252705+00:00"},{"alias_kind":"pith_short_16","alias_value":"IUKHAW67LADQ7FFQ","created_at":"2026-07-05T08:14:22.252705+00:00"},{"alias_kind":"pith_short_8","alias_value":"IUKHAW67","created_at":"2026-07-05T08:14:22.252705+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":26,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2607.05876","citing_title":"Think Before You Grid-Search: Floor-First Triage for LLM Serving","ref_index":41,"is_internal_anchor":true},{"citing_arxiv_id":"2607.05876","citing_title":"Think Before You Grid-Search: Floor-First Triage for LLM Serving","ref_index":41,"is_internal_anchor":true},{"citing_arxiv_id":"2606.24957","citing_title":"Dustin: Draft-Augmented Sparse Verification for Efficient Long-Context Generation with Speculative Decoding","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01237","citing_title":"KARA: Efficient Reasoning LLM Serving via Sliding-Window KV Cache Compression","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18967","citing_title":"EfficientRollout: System-Aware Self-Speculative Decoding for RL Rollouts","ref_index":65,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08565","citing_title":"EinSort: Sorting is All We Need for Tensorizing LLM","ref_index":94,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09879","citing_title":"Operator Fusion for LLM Inference on the Tensix Architecture","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.00288","citing_title":"Model-Native Computing Architecture: Envisioning Future System Architecture Through the Lens of Computer Architecture","ref_index":170,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30544","citing_title":"On-Device Generative AI for GDPR-Compliant Visual Monitoring: Natural Language Alerts from Local Object Detection","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2503.19950","citing_title":"LogQuant: Log-Distributed 2-Bit Quantization of KV Cache with Superior Accuracy Preservation","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2312.05821","citing_title":"ASVD: Activation-aware Singular Value Decomposition for Compressing Large Language Models","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18071","citing_title":"KVDrive: A Holistic Multi-Tier KV Cache Management System for Long-Context LLM Inference","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2601.21351","citing_title":"Analytical Provisioning for Attention-FFN Disaggregated LLM Serving under Stochastic Workloads","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2602.05695","citing_title":"SweetSpot: An Analytical Model for Predicting Energy Efficiency of LLM Inference","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2404.14294","citing_title":"A Survey on Efficient Inference for Large Language Models","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2604.03298","citing_title":"ENEC: A Lossless AI Model Compression Method Enabling Fast Inference on Ascend NPUs","ref_index":60,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04987","citing_title":"Cactus: Accelerating Auto-Regressive Decoding with Constrained Acceptance Speculative Sampling","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03109","citing_title":"Gated Subspace Inference for Transformer Acceleration","ref_index":33,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04738","citing_title":"OSAQ: Outlier Self-Absorption for Accurate Low-bit LLM Quantization","ref_index":18,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21399","citing_title":"A Task Decomposition and Planning Framework for Efficient LLM Inference in AI-Enabled WiFi-Offload Networks","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00555","citing_title":"Sim-FA: A GPGPU Simulator Framework for Fine-Grained Asynchronous Pipeline Analysis","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00342","citing_title":"Making Every Verified Token Count: Adaptive Verification for MoE Speculative Decoding","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2604.12358","citing_title":"Why and When Visual Token Pruning Fails? A Study on Relevant Visual Information Shift in MLLMs Decoding","ref_index":49,"is_internal_anchor":false},{"citing_arxiv_id":"2409.19256","citing_title":"HybridFlow: A Flexible and Efficient RLHF Framework","ref_index":96,"is_internal_anchor":false},{"citing_arxiv_id":"2604.07144","citing_title":"Autopoiesis: A Self-Evolving System Paradigm for LLM Serving Under Runtime Dynamics","ref_index":37,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IUKHAW67LADQ7FFQ6LTVZJOB77","json":"https://pith.science/pith/IUKHAW67LADQ7FFQ6LTVZJOB77.json","graph_json":"https://pith.science/api/pith-number/IUKHAW67LADQ7FFQ6LTVZJOB77/graph.json","events_json":"https://pith.science/api/pith-number/IUKHAW67LADQ7FFQ6LTVZJOB77/events.json","paper":"https://pith.science/paper/IUKHAW67"},"agent_actions":{"view_html":"https://pith.science/pith/IUKHAW67LADQ7FFQ6LTVZJOB77","download_json":"https://pith.science/pith/IUKHAW67LADQ7FFQ6LTVZJOB77.json","view_paper":"https://pith.science/paper/IUKHAW67","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2402.16363&json=true","fetch_graph":"https://pith.science/api/pith-number/IUKHAW67LADQ7FFQ6LTVZJOB77/graph.json","fetch_events":"https://pith.science/api/pith-number/IUKHAW67LADQ7FFQ6LTVZJOB77/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IUKHAW67LADQ7FFQ6LTVZJOB77/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IUKHAW67LADQ7FFQ6LTVZJOB77/action/storage_attestation","attest_author":"https://pith.science/pith/IUKHAW67LADQ7FFQ6LTVZJOB77/action/author_attestation","sign_citation":"https://pith.science/pith/IUKHAW67LADQ7FFQ6LTVZJOB77/action/citation_signature","submit_replication":"https://pith.science/pith/IUKHAW67LADQ7FFQ6LTVZJOB77/action/replication_record"}},"created_at":"2026-07-05T08:14:22.252705+00:00","updated_at":"2026-07-05T08:14:22.252705+00:00"}