{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:FC4Y6H4IG6T2PPU2KX5BUNY4GH","short_pith_number":"pith:FC4Y6H4I","schema_version":"1.0","canonical_sha256":"28b98f1f8837a7a7be9a55fa1a371c31c08b2b660fa7a465412a676813c92653","source":{"kind":"arxiv","id":"2310.05736","version":2},"attestation_state":"computed","paper":{"title":"LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chin-Yew Lin, Huiqiang Jiang, Lili Qiu, Qianhui Wu, Yuqing Yang","submitted_at":"2023-10-09T14:10:21Z","abstract_excerpt":"Large language models (LLMs) have been applied in various applications due to their astonishing capabilities. With advancements in technologies such as chain-of-thought (CoT) prompting and in-context learning (ICL), the prompts fed to LLMs are becoming increasingly lengthy, even exceeding tens of thousands of tokens. To accelerate model inference and reduce cost, this paper presents LLMLingua, a coarse-to-fine prompt compression method that involves a budget controller to maintain semantic integrity under high compression ratios, a token-level iterative compression algorithm to better model th"},"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":"2310.05736","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2023-10-09T14:10:21Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"008a55e4d1fe86caa4e7fa2483edc3f00a697b4a5a72de6a0b51c4e2fece3551","abstract_canon_sha256":"42d95f86ed5562276f1b63cd393d67aff72dff851c840f6e522a2d32339176ec"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:20:54.181343Z","signature_b64":"Xc7GySkgRRJP7BeCvWbyR3Teu+J/F0w3VSocYAHNy0jL6W8UtSF6HvJJZFOueE7JnYhd+qbCumt3PiDcQsKBAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"28b98f1f8837a7a7be9a55fa1a371c31c08b2b660fa7a465412a676813c92653","last_reissued_at":"2026-07-05T07:20:54.180822Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:20:54.180822Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LLMLingua: Compressing Prompts for Accelerated Inference of Large Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chin-Yew Lin, Huiqiang Jiang, Lili Qiu, Qianhui Wu, Yuqing Yang","submitted_at":"2023-10-09T14:10:21Z","abstract_excerpt":"Large language models (LLMs) have been applied in various applications due to their astonishing capabilities. With advancements in technologies such as chain-of-thought (CoT) prompting and in-context learning (ICL), the prompts fed to LLMs are becoming increasingly lengthy, even exceeding tens of thousands of tokens. To accelerate model inference and reduce cost, this paper presents LLMLingua, a coarse-to-fine prompt compression method that involves a budget controller to maintain semantic integrity under high compression ratios, a token-level iterative compression algorithm to better model th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2310.05736","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/2310.05736/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":"2310.05736","created_at":"2026-07-05T07:20:54.180884+00:00"},{"alias_kind":"arxiv_version","alias_value":"2310.05736v2","created_at":"2026-07-05T07:20:54.180884+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2310.05736","created_at":"2026-07-05T07:20:54.180884+00:00"},{"alias_kind":"pith_short_12","alias_value":"FC4Y6H4IG6T2","created_at":"2026-07-05T07:20:54.180884+00:00"},{"alias_kind":"pith_short_16","alias_value":"FC4Y6H4IG6T2PPU2","created_at":"2026-07-05T07:20:54.180884+00:00"},{"alias_kind":"pith_short_8","alias_value":"FC4Y6H4I","created_at":"2026-07-05T07:20:54.180884+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":33,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.08032","citing_title":"What to Keep, What to Forget: A Rate--Distortion View of Memory Compaction in LLMs and Agents","ref_index":53,"is_internal_anchor":true},{"citing_arxiv_id":"2606.20295","citing_title":"Token-Operations-Oriented Inference Optimization Techniques for Large Models","ref_index":130,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19613","citing_title":"StaminaBench: Stress-Testing Coding Agents over 100 Interaction Turns","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01241","citing_title":"Mapping Text to Multiplex Graph: Prompt Compression as L\\'evy Walk-Guided Graph Pruning","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01916","citing_title":"ContextSniper: AntTrail's Token-Efficient Code Memory for Repository-Level Program Repair","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10209","citing_title":"Less Context, Better Agents: Efficient Context Engineering for Long-Horizon Tool-Using LLM Agents","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2606.09659","citing_title":"End-to-End Context Compression at Scale","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08151","citing_title":"Decision-Aware Memory Cards: Counterfactual-Inspired Context Selection and Compression for Tool-Using LLM Agents","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05875","citing_title":"QCFuse: Query-Aware Cache Fusion via Compressed View for Efficient RAG Serving","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03979","citing_title":"Language Models Need Sleep: Learning to Self-Modify and Consolidate Memories","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24279","citing_title":"ContextEcho: A Benchmark for Persona Drift in Long Agentic-Coding Sessions","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28349","citing_title":"HMARS: A Hierarchical Multi-Agent Memory System for Long-Context Reasoning","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.26165","citing_title":"Tool-Schema Compression Enables Agentic RAG Under Constrained Context Budgets","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2409.01579","citing_title":"AdaComp: Extractive Context Compression with Adaptive Predictor for Retrieval-Augmented Large Language Models","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2409.06679","citing_title":"E2LLM: Encoder Elongated Large Language Models for Long-Context Understanding and Reasoning","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2602.11574","citing_title":"Learning to Configure Agentic AI Systems","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2602.08324","citing_title":"Towards Efficient Large Language Reasoning Models via Extreme-Ratio Chain-of-Thought Compression","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2602.11574","citing_title":"Learning to Configure Agentic AI Systems","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2605.00505","citing_title":"LLM-Oriented Information Retrieval: A Denoising-First Perspective","ref_index":82,"is_internal_anchor":false},{"citing_arxiv_id":"2412.13171","citing_title":"Compressed Chain of Thought: Efficient Reasoning Through Dense Representations","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2502.11089","citing_title":"Native Sparse Attention: Hardware-Aligned and Natively Trainable Sparse Attention","ref_index":62,"is_internal_anchor":false},{"citing_arxiv_id":"2604.09607","citing_title":"LLM-assisted Agentic Edge Intelligence Framework","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2507.02259","citing_title":"MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agent","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14156","citing_title":"Compressed-Sensing-Guided, Inference-Aware Structured Reduction for Large Language Models","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2311.05232","citing_title":"A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questions","ref_index":144,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FC4Y6H4IG6T2PPU2KX5BUNY4GH","json":"https://pith.science/pith/FC4Y6H4IG6T2PPU2KX5BUNY4GH.json","graph_json":"https://pith.science/api/pith-number/FC4Y6H4IG6T2PPU2KX5BUNY4GH/graph.json","events_json":"https://pith.science/api/pith-number/FC4Y6H4IG6T2PPU2KX5BUNY4GH/events.json","paper":"https://pith.science/paper/FC4Y6H4I"},"agent_actions":{"view_html":"https://pith.science/pith/FC4Y6H4IG6T2PPU2KX5BUNY4GH","download_json":"https://pith.science/pith/FC4Y6H4IG6T2PPU2KX5BUNY4GH.json","view_paper":"https://pith.science/paper/FC4Y6H4I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2310.05736&json=true","fetch_graph":"https://pith.science/api/pith-number/FC4Y6H4IG6T2PPU2KX5BUNY4GH/graph.json","fetch_events":"https://pith.science/api/pith-number/FC4Y6H4IG6T2PPU2KX5BUNY4GH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FC4Y6H4IG6T2PPU2KX5BUNY4GH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FC4Y6H4IG6T2PPU2KX5BUNY4GH/action/storage_attestation","attest_author":"https://pith.science/pith/FC4Y6H4IG6T2PPU2KX5BUNY4GH/action/author_attestation","sign_citation":"https://pith.science/pith/FC4Y6H4IG6T2PPU2KX5BUNY4GH/action/citation_signature","submit_replication":"https://pith.science/pith/FC4Y6H4IG6T2PPU2KX5BUNY4GH/action/replication_record"}},"created_at":"2026-07-05T07:20:54.180884+00:00","updated_at":"2026-07-05T07:20:54.180884+00:00"}