Cover Image

When Thinking Fails: The Pitfalls of Reasoning for Instruction-Following in LLMs

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Li, Xiaomin, Yu, Zhou, Zhang, Zhiwei, Chen, Xupeng, Zhang, Ziji, Zhuang, Yingying, Sadagopan, Narayanan, Beniwal, Anurag
Format:	Preprint
Published:	2025
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Semantic Volume: Quantifying and Detecting both External and Internal Uncertainty in LLMs
by: Li, Xiaomin, et al.
Published: (2025)

DARD: A Multi-Agent Approach for Task-Oriented Dialog Systems
by: Gupta, Aman, et al.
Published: (2024)

TOD-ProcBench: Benchmarking Complex Instruction-Following in Task-Oriented Dialogues
by: Ghazarian, Sarik, et al.
Published: (2025)

How and Where to Translate? The Impact of Translation Strategies in Cross-lingual LLM Prompting
by: Gupta, Aman, et al.
Published: (2025)

Multilingual Information Retrieval with a Monolingual Knowledge Base
by: Zhuang, Yingying, et al.
Published: (2025)

REIC: RAG-Enhanced Intent Classification at Scale
by: Zhang, Ziji, et al.
Published: (2025)

Beyond Correctness: Harmonizing Process and Outcome Rewards through RL Training
by: Ye, Chenlu, et al.
Published: (2025)

When Contextual Inference Fails: Cancelability in Interactive Instruction Following
by: Bila, Natalia, et al.
Published: (2026)

Thinking LLMs: General Instruction Following with Thought Generation
by: Wu, Tianhao, et al.
Published: (2024)

Adaptive Deep Reasoning: Triggering Deep Thinking When Needed
by: Wang, Yunhao, et al.
Published: (2025)

AXCEL: Automated eXplainable Consistency Evaluation using LLMs
by: Sreekar, P Aditya, et al.
Published: (2024)

Instruction Tuning and CoT Prompting for Contextual Medical QA with LLMs
by: Le, Chenqian, et al.
Published: (2025)

When to Think, When to Speak: Learning Disclosure Policies for LLM Reasoning
by: Wei, Jiaqi, et al.
Published: (2026)

Right Is Not Enough: The Pitfalls of Outcome Supervision in Training LLMs for Math Reasoning
by: Guo, Jiaxing, et al.
Published: (2025)

When to Continue Thinking: Adaptive Thinking Mode Switching for Efficient Reasoning
by: Zhang, Xiaoyun, et al.
Published: (2025)

Position: On the Methodological Pitfalls of Evaluating Base LLMs for Reasoning
by: Chan, Jason, et al.
Published: (2025)

Think or Not? Exploring Thinking Efficiency in Large Reasoning Models via an Information-Theoretic Lens
by: Yong, Xixian, et al.
Published: (2025)

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

When Safety Fails Before the Answer: Benchmarking Harmful Behavior Detection in Reasoning Chains
by: Kakkar, Ishita, et al.
Published: (2026)

Pseudo-Deliberation in Language Models: When Reasoning Fails to Align Values and Actions
by: Rakshit, Sushrita, et al.
Published: (2026)

Atomic Thinking of LLMs: Decoupling and Exploring Mathematical Reasoning Abilities
by: Kuang, Jiayi, et al.
Published: (2025)

When Thinking LLMs Lie: Unveiling the Strategic Deception in Representations of Reasoning Models
by: Wang, Kai, et al.
Published: (2025)

TRIM: Hybrid Inference via Targeted Stepwise Routing in Multi-Step Reasoning Tasks
by: Kapoor, Vansh, et al.
Published: (2026)

Hierarchical Alignment: Enforcing Hierarchical Instruction-Following in LLMs through Logical Consistency
by: Yang, Shu, et al.
Published: (2026)

LLM CHESS: Benchmarking Reasoning and Instruction-Following in LLMs through Chess
by: Kolasani, Sai, et al.
Published: (2025)

IOPO: Empowering LLMs with Complex Instruction Following via Input-Output Preference Optimization
by: Zhang, Xinghua, et al.
Published: (2024)

Think Only When You Need with Large Hybrid-Reasoning Models
by: Jiang, Lingjie, et al.
Published: (2025)

When Is Thinking Enough? Early Exit via Sufficiency Assessment for Efficient Reasoning
by: Xiang, Yang, et al.
Published: (2026)

Multi-IF: Benchmarking LLMs on Multi-Turn and Multilingual Instructions Following
by: He, Yun, et al.
Published: (2024)

When Instructions Multiply: Measuring and Estimating LLM Capabilities of Multiple Instructions Following
by: Harada, Keno, et al.
Published: (2025)

Larger Language Models Don't Care How You Think: Why Chain-of-Thought Prompting Fails in Subjective Tasks
by: Chochlakis, Georgios, et al.
Published: (2024)

Thinking Economically: A Hierarchical Framework for Adaptive-Complexity Reasoning in LLMs
by: Gao, Yubo, et al.
Published: (2026)

ENCORE: Entropy-guided Reward Composition for Multi-head Safety Reward Models
by: Li, Xiaomin, et al.
Published: (2025)

Scaling Reasoning, Losing Control: Evaluating Instruction Following in Large Reasoning Models
by: Fu, Tingchen, et al.
Published: (2025)

Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs
by: Chen, Xingyu, et al.
Published: (2024)

When Retrieval Succeeds and Fails: Rethinking Retrieval-Augmented Generation for LLMs
by: Wang, Yongjie, et al.
Published: (2025)

Tracking the Limits of Knowledge Propagation: How LLMs Fail at Multi-Step Reasoning with Conflicting Knowledge
by: Feng, Yiyang, et al.
Published: (2026)

Light-IF: Endowing LLMs with Generalizable Reasoning via Preview and Self-Checking for Complex Instruction Following
by: Wang, Chenyang, et al.
Published: (2025)

Call Me When Necessary: LLMs can Efficiently and Faithfully Reason over Structured Environments
by: Cheng, Sitao, et al.
Published: (2024)

The Instruction Gap: LLMs get lost in Following Instruction
by: Tripathi, Vishesh, et al.
Published: (2025)