Resource Consumption Threats in Large Language Models
Not on the map. Why?
Abstract (arXiv)
Given limited and costly computational infrastructure, resource efficiency is a key requirement for large language models (LLMs). Efficient LLMs increase service capacity for providers and reduce latency and API costs for users. Recent resource consumption threats induce excessive generation, degrading model efficiency and harming both service availability and economic sustainability. This survey presents a systematic review of threats to resource consumption in LLMs. We further establish a unified view of this emerging area by clarifying its scope and examining the problem along the full pipeline from threat induction to mechanism understanding and mitigation. Our goal is to clarify the problem landscape for this emerging area, thereby providing a clearer foundation for characterization and mitigation.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 73 of its 100 references that are in the map. First posted on arXiv 17 Mar 2026.
Builds on
73 of 100 references.
References · 73
- 2017Attention Is All You Needcited by 59,932
- 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
- 2023A Survey on Large Language Model based Autonomous Agentscited by 2,203
- 2023The Rise and Potential of Large Language Model Based Agents: A Surveycited by 1,422
- 2023Reasoning with Language Model is Planning with World Modelcited by 846
- 2024Taming Throughput-Latency Tradeoff in LLM Inference with Sarathi-Servecited by 507
- 2019Neural Text Generation with Unlikelihood Trainingcited by 489
- 2023LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compressioncited by 443
- 2023AutoAgents: A Framework for Automatic Agent Generationcited by 211
- 2023From Words to Watts: Benchmarking the Energy Costs of Large Language Model Inferencecited by 168
- 2025CoT-Valve: Length-Compressible Chain-of-Thought Tuningcited by 167
- 2025SafeChain: Safety of Language Models with Long Chain-of-Thought Reasoning Capabilitiescited by 133
38 more
- 2020Sponge Examples: Energy-Latency Attacks on Neural Networkscited by 108
- 2024Concise Thoughts: Impact of Output Length on LLM Reasoning and Costcited by 108
- 2024A Review on Edge Large Language Models: Design, Execution, and Applicationscited by 98
- 2025OverThink: Slowdown Attacks on Reasoning LLMscited by 94
- 2022Learning to Break the Loop: Analyzing and Mitigating Repetitions for Neural Text Generationcited by 90
- 2023SkipDecode: Autoregressive Skip Decoding with Batching and Caching for Efficient LLM Inferencecited by 89
- 2020A Theoretical Analysis of the Repetition Problem in Text Generationcited by 83
- 2023Dynamic Context Pruning for Efficient and Interpretable Autoregressive Transformerscited by 67
- 2023Auto-Regressive Next-Token Predictors are Universal Learnerscited by 65
- 2020A Panda? No, It's a Sloth: Slowdown Attacks on Adaptive Multi-Exit Neural Network Inferencecited by 61
- 2024Make Every Penny Count: Difficulty-Adaptive Self-Consistency for Cost-Efficient Reasoningcited by 61
- 2025Energy Considerations of Large Language Model Inference and Efficiency Optimizationscited by 54
- 2022NICGSlowDown: Evaluating the Efficiency Robustness of Neural Image Caption Generation Modelscited by 47
- 2024Induction Heads as an Essential Mechanism for Pattern Matching in In-context Learningcited by 46
- 2023Repetition In Repetition Out: Towards Understanding Neural Text Degeneration from the Data Perspectivecited by 39
- 2024From Self-Attention to Markov Models: Unveiling the Dynamics of Generative Transformerscited by 34
- 2024Crabs: Consuming Resource via Auto-generation for LLM-DoS Attack under Black-box Settingscited by 33
- 2024An Engorgio Prompt Makes Large Language Model Babble oncited by 30
- 2022Energy-Latency Attacks via Sponge Poisoningcited by 29
- 2022Phantom Sponges: Exploiting Non-Maximum Suppression to Attack Deep Object Detectorscited by 28
- 2025Efficient Reasoning for Large Reasoning Language Models via Certainty-Guided Reflection Suppressioncited by 23
- 2024Repetition Neurons: How Do Language Models Produce Repetitions?cited by 22
- 2025ThinkTrap: Denial-of-Service Attacks against Black-box LLM Services via Infinite Thinkingcited by 22
- 2021Straight to the Gradient: Learning to Use Novel Tokens for Neural Text Generationcited by 19
- 2025Excessive Reasoning Attack on Reasoning LLMscited by 17
- 2025Parasites in the Toolchain: A Large-Scale Analysis of Attacks on the MCP Ecosystemcited by 15
- 2024Energy-Latency Manipulation of Multi-modal Large Language Models via Verbose Samplescited by 14
- 2025CoIn: Counting the Invisible Reasoning Tokens in Commercial Opaque LLM APIscited by 14
- 2026Beyond Max Tokens: Stealthy Resource Amplification via Tool Calling Chains in LLM Agentscited by 14
- 2026Clawdrain: Exploiting Tool-Calling Chains for Stealthy Token Exhaustion in OpenClaw Agentscited by 14
- 2023Overload: Latency Attacks on Object Detection for Edge Devicescited by 13
- 2024From Loops to Oops: Fallback Behaviors of Language Models Under Uncertaintycited by 13
- 2025LAPO: Internalizing Reasoning Efficiency via Length-Adaptive Policy Optimizationcited by 11
- 2023Energy-Latency Attacks to On-Device Neural Networks via Sponge Poisoningcited by 10
- 2025ExtendAttack: Attacking Servers of LRMs via Extending Reasoningcited by 10
- 2021The Hammer and the Nut: Is Bilevel Optimization Really Needed to Poison Linear Classifiers?cited by 9
- 2025Repetitions are not all alike: distinct mechanisms sustain repetition in language modelscited by 9
- 2025LingoLoop Attack: Trapping MLLMs via Linguistic Context and State Entrapment into Endless Loopscited by 9
50 most cited of 73 All 73 (JSON)
Built on by
Citing papers per year: 2026: 2.
2 papers.