Performative Thinking? The Brittle Correlation Between CoT Length and Problem Complexity

• URL: http://arxiv.org/abs/2509.07339v1
• Date: Tue, 09 Sep 2025 02:31:16 GMT
• Title: Performative Thinking? The Brittle Correlation Between CoT Length and Problem Complexity
• Authors: Vardhan Palod, Karthik Valmeekam, Kaya Stechly, Subbarao Kambhampati,
• Abstract summary: This work critically examines whether intermediate token sequence length reflects or correlates with problem difficulty.<n>We train transformer models from scratch on derivational traces of the A* search algorithm.<n>We find that even for the simplest tasks, they often produce excessively long reasoning traces and sometimes fail to generate a solution.
• Score: 23.225139930889522
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Intermediate token generation (ITG), where a model produces output before the solution, has been proposed as a method to improve the performance of language models on reasoning tasks. While these reasoning traces or Chain of Thoughts (CoTs) are correlated with performance gains, the mechanisms underlying them remain unclear. A prevailing assumption in the community has been to anthropomorphize these tokens as "thinking", treating longer traces as evidence of higher problem-adaptive computation. In this work, we critically examine whether intermediate token sequence length reflects or correlates with problem difficulty. To do so, we train transformer models from scratch on derivational traces of the A* search algorithm, where the number of operations required to solve a maze problem provides a precise and verifiable measure of problem complexity. We first evaluate the models on trivial free-space problems, finding that even for the simplest tasks, they often produce excessively long reasoning traces and sometimes fail to generate a solution. We then systematically evaluate the model on out-of-distribution problems and find that the intermediate token length and ground truth A* trace length only loosely correlate. We notice that the few cases where correlation appears are those where the problems are closer to the training distribution, suggesting that the effect arises from approximate recall rather than genuine problem-adaptive computation. This suggests that the inherent computational complexity of the problem instance is not a significant factor, but rather its distributional distance from the training data. These results challenge the assumption that intermediate trace generation is adaptive to problem difficulty and caution against interpreting longer sequences in systems like R1 as automatically indicative of "thinking effort".

Related papers

• Explore Briefly, Then Decide: Mitigating LLM Overthinking via Cumulative Entropy Regulation [82.62935304152239]
  Large Language Models (LLMs) have demonstrated remarkable reasoning abilities on complex problems using long Chain-of-Thought (CoT) reasoning.<n>They often suffer from overthinking, meaning generating unnecessarily lengthy reasoning steps for simpler problems.<n>We introduce a novel metric Token Entropy Cumulative Average (TECA), which measures the extent of exploration throughout the reasoning process.
  arXiv  Detail & Related papers  (2025-10-02T17:36:50Z)
• Think Right: Learning to Mitigate Under-Over Thinking via Adaptive, Attentive Compression [68.69801176669843]
  We propose an online post-training RL method that prunes redundant steps and estimates difficulty.<n> TRAAC (Think Right with Adaptive, Attentive Compression) achieves an average absolute accuracy gain of 8.4%.<n>Although our models are trained on math datasets, they show accuracy and efficiency gains on out-of-distribution non-math datasets.
  arXiv  Detail & Related papers  (2025-10-02T02:00:20Z)
• Staying in the Sweet Spot: Responsive Reasoning Evolution via Capability-Adaptive Hint Scaffolding [59.60915947702282]
  Reinforcement learning with verifiable rewards (RLVR) has achieved remarkable success in enhancing the reasoning capabilities of large language models (LLMs)<n>Existing RLVR methods often suffer from exploration inefficiency due to mismatches between the training data's difficulty and the model's capability.<n>We propose SEELE, a novel supervision-aided RLVR framework that dynamically adjusts problem difficulty to stay within the high-efficiency region.
  arXiv  Detail & Related papers  (2025-09-08T17:36:21Z)
• Less is More Tokens: Efficient Math Reasoning via Difficulty-Aware Chain-of-Thought Distillation [82.2288581878096]
  We present a framework for difficulty-aware reasoning that teaches models to dynamically adjust reasoning depth based on problem complexity.<n>We show that models can be endowed with such dynamic inference pathways without any architectural modifications.
  arXiv  Detail & Related papers  (2025-09-05T16:40:13Z)
• On the Bias of Next-Token Predictors Toward Systematically Inefficient Reasoning: A Shortest-Path Case Study [4.319482898846564]
  We study two key factors for improving reasoning in large language models.<n>We train decoder-only transformers on question-trace-answer triples using a custom tokenizer.<n>With the same training-token budget, models trained on inefficient traces generalize better to unseen graphs.
  arXiv  Detail & Related papers  (2025-07-07T18:00:06Z)
• Interpretable Traces, Unexpected Outcomes: Investigating the Disconnect in Trace-Based Knowledge Distillation [14.489157453882767]
  This work aims to address the challenge of evaluating reasoning traces and their correlation with the final performance.<n>We employ a KD method leveraging rule-based problem decomposition to generate interpretable traces.<n>Specifically, we demonstrate this approach on Open Book QA, decomposing the problem into a Classification step and an Information Retrieval step.
  arXiv  Detail & Related papers  (2025-05-20T00:49:19Z)
• Beyond Semantics: The Unreasonable Effectiveness of Reasonless Intermediate Tokens [14.78605805191225]
  We investigate how the semantics of intermediate tokens-often anthropomorphized as "thoughts" or reasoning traces-actually influence model performance.<n>We show that despite significant improvements on the solution-only baseline, models trained on entirely correct traces still produce invalid reasoning traces when arriving at correct solutions.
  arXiv  Detail & Related papers  (2025-05-19T23:29:23Z)
• MetaLadder: Ascending Mathematical Solution Quality via Analogical-Problem Reasoning Transfer [37.81465564673498]
  Large Language Models (LLMs) have demonstrated promising capabilities in solving mathematical reasoning tasks.<n>We propose textbfMetaLadder, a framework that explicitly prompts LLMs to recall and reflect on meta-problems.<n>Our experiments on mathematical benchmarks demonstrate that our MetaLadder significantly boosts LLMs' problem-solving accuracy.
  arXiv  Detail & Related papers  (2025-03-19T04:36:35Z)
• Amortized Inference for Causal Structure Learning [72.84105256353801]
  Learning causal structure poses a search problem that typically involves evaluating structures using a score or independence test. We train a variational inference model to predict the causal structure from observational/interventional data. Our models exhibit robust generalization capabilities under substantial distribution shift.
  arXiv  Detail & Related papers  (2022-05-25T17:37:08Z)
• Generalization of Neural Combinatorial Solvers Through the Lens of Adversarial Robustness [68.97830259849086]
  Most datasets only capture a simpler subproblem and likely suffer from spurious features. We study adversarial robustness - a local generalization property - to reveal hard, model-specific instances and spurious features. Unlike in other applications, where perturbation models are designed around subjective notions of imperceptibility, our perturbation models are efficient and sound. Surprisingly, with such perturbations, a sufficiently expressive neural solver does not suffer from the limitations of the accuracy-robustness trade-off common in supervised learning.
  arXiv  Detail & Related papers  (2021-10-21T07:28:11Z)
• Disentangling Observed Causal Effects from Latent Confounders using Method of Moments [67.27068846108047]
  We provide guarantees on identifiability and learnability under mild assumptions. We develop efficient algorithms based on coupled tensor decomposition with linear constraints to obtain scalable and guaranteed solutions.
  arXiv  Detail & Related papers  (2021-01-17T07:48:45Z)
• Learning Causal Models Online [103.87959747047158]
  Predictive models can rely on spurious correlations in the data for making predictions. One solution for achieving strong generalization is to incorporate causal structures in the models. We propose an online algorithm that continually detects and removes spurious features.
  arXiv  Detail & Related papers  (2020-06-12T20:49:20Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.