Image de couverture de livre

The Coupling Tax: How Shared Token Budgets Undermine Visible Chain-of-Thought Under Fixed Output Limits

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Nie, Wenhua, Liu, Junlin, Wu, Jianan, Meng, Zijie, Fan, Yilong, Zijian, Zhang, Zheng, Haoran, Jang, Jyh-Shing Roger
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Gradient Starvation in Binary-Reward GRPO: Why Group-Mean Centering Fails and Why the Simplest Fix Works
par: Nie, Wenhua, et autres
Publié: (2026)

Identified-Set Geometry of Distributional Model Extraction under Top-$K$ Censored API Access
par: Nie, Wenhua, et autres
Publié: (2026)

Equilibrium Residuals Expose Three Regimes of Matrix-Game Strategic Reasoning in Language Models
par: Nie, Wenhua, et autres
Publié: (2026)

Future Validity is the Missing Statistic: From Impossibility to $Φ$-Estimation for Grammar-Faithful Speculative Decoding
par: Nie, Wenhua, et autres
Publié: (2026)

BLAPose: Enhancing 3D Human Pose Estimation with Bone Length Adjustment
par: Hsu, Chih-Hsiang, et autres
Publié: (2024)

Gaming the Judge: Unfaithful Chain-of-Thought Can Undermine Agent Evaluation
par: Khalifa, Muhammad, et autres
Publié: (2026)

CodaRAG: Connecting the Dots with Associativity Inspired by Complementary Learning
par: Li, Cheng-Yen, et autres
Publié: (2026)

A Scaling Law for Token Efficiency in LLM Fine-Tuning Under Fixed Compute Budgets
par: Lagasse, Ryan, et autres
Publié: (2025)

Balanced Budget Multiplier with Indirect Taxes Under Imperfect Competition
par: Ramón J. Torregrosa Montaner
Publié: (2003)

Singer separation for karaoke content generation
par: Lin, Hsuan-Yu, et autres
Publié: (2021)

Output Supervision Can Obfuscate the Chain of Thought
par: Drori, Jacob, et autres
Publié: (2025)

Lost in Taxation: Misunderstanding the Progressive Tax System Undermines Support for Redistribution
par: Silvia Filippi, et autres
Publié: (2026)

Dynamic Grid Trading Strategy: From Zero Expectation to Market Outperformance
par: Chen, Kai-Yuan, et autres
Publié: (2025)

Reasoning Models Sometimes Output Illegible Chains of Thought
par: Jose, Arun
Publié: (2025)

Symbolic Analysis of Grover Search Algorithm via Chain-of-Thought Reasoning and Quantum-Native Tokenization
par: Chen, Min, et autres
Publié: (2025)

Reasoning in Token Economies: Budget-Aware Evaluation of LLM Reasoning Strategies
par: Wang, Junlin, et autres
Publié: (2024)

Chain-of-Thought Tokens are Computer Program Variables
par: Zhu, Fangwei, et autres
Publié: (2025)

Singing Voice Graph Modeling for SingFake Detection
par: Chen, Xuanjun, et autres
Publié: (2024)

Chain-of-Thought and Compressed Looped Transformers: A Memory-Budget Separation
par: Zhang, Haozhou
Publié: (2026)

Output‐Feedback Stabilization for a Type of Stochastic Nonlinear Systems With Unknown Output Function and External Disturbances
par: Xiaoyan Qin, et autres
Publié: (2025)

TokenSkip: Controllable Chain-of-Thought Compression in LLMs
par: Xia, Heming, et autres
Publié: (2025)

Pretraining with Token-Level Adaptive Latent Chain-of-Thought
par: Zeng, Boyi, et autres
Publié: (2026)

Shape of Thought: Progressive Object Assembly via Visual Chain-of-Thought
par: Huo, Yu, et autres
Publié: (2026)

Adversarial Speaker Distillation for Countermeasure Model on Automatic Speaker Verification
par: Liao, Yen-Lun, et autres
Publié: (2022)

Capabilities and Fundamental Limits of Latent Chain-of-Thought
par: Zou, Jiaxuan, et autres
Publié: (2026)

Aligning Tree-Search Policies with Fixed Token Budgets in Test-Time Scaling of LLMs
par: Miyamoto, Sora, et autres
Publié: (2026)

The Token Tax: Systematic Bias in Multilingual Tokenization
par: Lundin, Jessica M., et autres
Publié: (2025)

Rethinking Token-Level Policy Optimization for Multimodal Chain-of-Thought
par: Li, Yunheng, et autres
Publié: (2026)

Eliciting Chain-of-Thought in Base LLMs via Gradient-Based Representation Optimization
par: Wang, Zijian, et autres
Publié: (2025)

SD-E$^2$: Semantic Exploration for Reasoning Under Token Budgets
par: Mishra, Kshitij, et autres
Publié: (2026)

Neural Codec-based Adversarial Sample Detection for Speaker Verification
par: Chen, Xuanjun, et autres
Publié: (2024)

Knowledge Retrieval Based on Generative AI
par: Yang, Te-Lun, et autres
Publié: (2025)

Internalizing ASR with Implicit Chain of Thought for Efficient Speech-to-Speech Conversational LLM
par: Yuen, Robin Shing-Hei, et autres
Publié: (2024)

A Fair Share of Tax
par: Björklund Larsen, Lotta
Publié: (2018)

The Curse of CoT: On the Limitations of Chain-of-Thought in In-Context Learning
par: Zheng, Tianshi, et autres
Publié: (2025)

Deep Electromagnetic Structure Design Under Limited Evaluation Budgets
par: Zheng, Shijian, et autres
Publié: (2025)

Token Signature: Predicting Chain-of-Thought Gains with Token Decoding Feature in Large Language Models
par: Liu, Peijie, et autres
Publié: (2025)

Chain-of-Thought Prompting of Large Language Models for Discovering and Fixing Software Vulnerabilities
par: Nong, Yu, et autres
Publié: (2024)

Streaming Hallucination Detection in Long Chain-of-Thought Reasoning
par: Lu, Haolang, et autres
Publié: (2026)

From Thinking to Output: Chain-of-Thought and Text Generation Characteristics in Reasoning Language Models
par: Liu, Junhao, et autres
Publié: (2025)