Copertina

Thinking Short and Right Over Thinking Long: Serving LLM Reasoning Efficiently and Accurately

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Wang, Yuhang, Jiang, Youhe, Cui, Bin, Fu, Fangcheng
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

SALE : Low-bit Estimation for Efficient Sparse Attention in Long-context LLM Prefilling
di: Ji, Xiaodong, et al.
Pubblicazione: (2025)

BOute: Cost-Efficient LLM Serving with Heterogeneous LLMs and GPUs via Multi-Objective Bayesian Optimization
di: Jiang, Youhe, et al.
Pubblicazione: (2026)

ThinkEdit: Interpretable Weight Editing to Mitigate Overly Short Thinking in Reasoning Models
di: Sun, Chung-En, et al.
Pubblicazione: (2025)

Train Long, Think Short: Curriculum Learning for Efficient Reasoning
di: Hammoud, Hasan Abed Al Kader, et al.
Pubblicazione: (2025)

Efficient Multi-round LLM Inference over Disaggregated Serving
di: He, Wenhao, et al.
Pubblicazione: (2026)

ThunderServe: High-performance and Cost-efficient LLM Serving in Cloud Environments
di: Jiang, Youhe, et al.
Pubblicazione: (2025)

Think Right: Learning to Mitigate Under-Over Thinking via Adaptive, Attentive Compression
di: Singh, Joykirat, et al.
Pubblicazione: (2025)

OverThink: Slowdown Attacks on Reasoning LLMs
di: Kumar, Abhinav, et al.
Pubblicazione: (2025)

Improving Automatic Parallel Training via Balanced Memory Workload Optimization
di: Wang, Yujie, et al.
Pubblicazione: (2023)

Efficient Reasoning with Hidden Thinking
di: Shen, Xuan, et al.
Pubblicazione: (2025)

Efficient Reasoning with Balanced Thinking
di: Li, Yulin, et al.
Pubblicazione: (2026)

OServe: Accelerating LLM Serving via Spatial-Temporal Workload Orchestration
di: Jiang, Youhe, et al.
Pubblicazione: (2026)

Accordion-Thinking: Self-Regulated Step Summaries for Efficient and Readable LLM Reasoning
di: Yang, Zhicheng, et al.
Pubblicazione: (2026)

Think Dense, Not Long: Dynamic Decoupled Conditional Advantage for Efficient Reasoning
di: Peng, Keqin, et al.
Pubblicazione: (2026)

Efficient Serving for Dynamic Agent Workflows with Prediction-based KV-Cache Management
di: Zheng, Haoyu, et al.
Pubblicazione: (2026)

Think-RM: Enabling Long-Horizon Reasoning in Generative Reward Models
di: Hong, Ilgee, et al.
Pubblicazione: (2025)

Atom: Low-bit Quantization for Efficient and Accurate LLM Serving
di: Zhao, Yilong, et al.
Pubblicazione: (2023)

Quantized Reasoning Models Think They Need to Think Longer, but They Do Not
di: Lotfi, Sanae, et al.
Pubblicazione: (2026)

ThinkRouter: Efficient Reasoning via Routing Thinking between Latent and Discrete Spaces
di: Xu, Xin, et al.
Pubblicazione: (2026)

Think-Augmented Function Calling: Improving LLM Parameter Accuracy Through Embedded Reasoning
di: Wei, Lei, et al.
Pubblicazione: (2026)

Deep Think with Confidence
di: Fu, Yichao, et al.
Pubblicazione: (2025)

K2-Think: A Parameter-Efficient Reasoning System
di: Cheng, Zhoujun, et al.
Pubblicazione: (2025)

Good Learners Think Their Thinking: Generative PRM Makes Large Reasoning Model More Efficient Math Learner
di: He, Tao, et al.
Pubblicazione: (2025)

PQCache: Product Quantization-based KVCache for Long Context LLM Inference
di: Zhang, Hailin, et al.
Pubblicazione: (2024)

How Much Thinking is Enough? Quantifying and Understanding Redundancy in LLM Reasoning
di: Zhai, Zhiyuan, et al.
Pubblicazione: (2026)

To Think or Not to Think: Exploring the Unthinking Vulnerability in Large Reasoning Models
di: Zhu, Zihao, et al.
Pubblicazione: (2025)

AdapThink: Adaptive Thinking Preferences for Reasoning Language Model
di: Wan, Xu, et al.
Pubblicazione: (2025)

Think Consistently, Reason Efficiently: Energy-Based Calibration for Implicit Chain-of-Thought
di: Chen, Zhikang, et al.
Pubblicazione: (2025)

Demystifying Cost-Efficiency in LLM Serving over Heterogeneous GPUs
di: Jiang, Youhe, et al.
Pubblicazione: (2025)

DARTS: Distribution-Aware Active Rollout Trajectory Shaping for Accelerating LLM Reinforcement Learning
di: Wang, Yujie, et al.
Pubblicazione: (2026)

Learn to Think: Bootstrapping LLM Reasoning Capability Through Graph Representation Learning
di: Gao, Hang, et al.
Pubblicazione: (2025)

AdaptThink: Reasoning Models Can Learn When to Think
di: Zhang, Jiajie, et al.
Pubblicazione: (2025)

MeTHanol: Modularized Thinking Language Models with Intermediate Layer Thinking, Decoding and Bootstrapping Reasoning
di: Xi, Ningyuan, et al.
Pubblicazione: (2024)

SCORPIO: Serving the Right Requests at the Right Time for Heterogeneous SLOs in LLM Inference
di: Tang, Yinghao, et al.
Pubblicazione: (2025)

Thinking with Knowledge Graphs: Enhancing LLM Reasoning Through Structured Data
di: Wu, Xue, et al.
Pubblicazione: (2024)

OptimalThinkingBench: Evaluating Over and Underthinking in LLMs
di: Aggarwal, Pranjal, et al.
Pubblicazione: (2025)

Think2SQL: Reinforce LLM Reasoning Capabilities for Text2SQL
di: Papicchio, Simone, et al.
Pubblicazione: (2025)

Dynamic Thinking-Token Selection for Efficient Reasoning in Large Reasoning Models
di: Guo, Zhenyuan, et al.
Pubblicazione: (2026)

Mitigating Think-Answer Mismatch in LLM Reasoning Through Noise-Aware Advantage Reweighting
di: Shen, Si, et al.
Pubblicazione: (2025)

Thinking-Free Policy Initialization Makes Distilled Reasoning Models More Effective and Efficient Reasoners
di: Xu, Xin, et al.
Pubblicazione: (2025)