Scaling Reasoning, Losing Control: Evaluating Instruction Following in Large Reasoning Models
Instruction-following is essential for aligning large language models (LLMs) with user intent. While recent reasoning-oriented models exhibit impressive performance on complex mathematical problems, their ability to adhere to natural language instructions remains underexplored. In this work, we introduce MathIF, a dedicated benchmark for evaluating instruction-following in mathematical reasoning tasks. Our empirical analysis reveals a consistent tension between scaling up reasoning capacity and maintaining controllability, as models that reason more effectively often struggle to comply with user directives. We find that models tuned on distilled long chains-of-thought or trained with reasoning-oriented reinforcement learning often degrade in instruction adherence, especially when generation length increases. Furthermore, we show that even simple interventions can partially recover obedience, though at the cost of reasoning performance. These findings highlight a fundamental tension in current LLM training paradigms and motivate the need for more instruction-aware reasoning models. We release the code and data at https://github.com/TingchenFu/MathIF.
Code (1)
Tasks
Instruction FollowingMathematical ReasoningSimilar Papers 제목 키워드 기반
$\texttt{ACCORD}$: Closing the Commonsense Measurability Gap
We present $\texttt{ACCORD}$, a framework and benchmark suite for disentangling the commonsense grounding and reasoning abilities of large language models (LLMs) through controlled, multi-hop counterfactuals. $\texttt{AC…
BenchmarkingCommon Sense ReasoningCounterfactual ReasoningLarge Language Model+1IF-VidCap: Can Video Caption Models Follow Instructions?
Although Multimodal Large Language Models (MLLMs) have demonstrated proficiency in video captioning, practical applications require captions that follow specific user instructions rather than generating exhaustive, uncon…
Video CaptioningDense CaptioningReasoning-Trace Collapse: Evaluating the Loss of Explicit Reasoning During Fine-Tuning
Explicit reasoning models are trained to produce intermediate reasoning traces before final answers, but downstream fine-tuning is often performed on ordinary instruction-response data that contains no such traces. We sh…
Length Representations in Large Language Models
Large language models (LLMs) have shown remarkable capabilities across various tasks, that are learned from massive amounts of text-based data. Although LLMs can control output sequence length, particularly in instructio…
NP-Engine: Empowering Optimization Reasoning in Large Language Models with Verifiable Synthetic NP Problems
Large Language Models (LLMs) have shown strong reasoning capabilities, with models like OpenAI's O-series and DeepSeek R1 excelling at tasks such as mathematics, coding, logic, and puzzles through Reinforcement Learning …
Reinforcement LearningInstruction Following