Position: On the Methodological Pitfalls of Evaluating Base LLMs for Reasoning

• URL: http://arxiv.org/abs/2511.10381v1
• Date: Fri, 14 Nov 2025 01:48:08 GMT
• Title: Position: On the Methodological Pitfalls of Evaluating Base LLMs for Reasoning
• Authors: Jason Chan, Zhixue Zhao, Robert Gaizauskas,
• Abstract summary: Existing work investigates the reasoning capabilities of large language models (LLMs) to uncover their limitations, human-like biases and underlying processes.<n>We argue that evaluating base LLMs' reasoning capabilities raises inherent methodological concerns that are overlooked in such existing studies.
• Score: 6.916679603940271
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Existing work investigates the reasoning capabilities of large language models (LLMs) to uncover their limitations, human-like biases and underlying processes. Such studies include evaluations of base LLMs (pre-trained on unlabeled corpora only) for this purpose. Our position paper argues that evaluating base LLMs' reasoning capabilities raises inherent methodological concerns that are overlooked in such existing studies. We highlight the fundamental mismatch between base LLMs' pretraining objective and normative qualities, such as correctness, by which reasoning is assessed. In particular, we show how base LLMs generate logically valid or invalid conclusions as coincidental byproducts of conforming to purely linguistic patterns of statistical plausibility. This fundamental mismatch challenges the assumptions that (a) base LLMs' outputs can be assessed as their bona fide attempts at correct answers or conclusions; and (b) conclusions about base LLMs' reasoning can generalize to post-trained LLMs optimized for successful instruction-following. We call for a critical re-examination of existing work that relies implicitly on these assumptions, and for future work to account for these methodological pitfalls.

Related papers

• Revisiting LLM Reasoning via Information Bottleneck [57.519119962528166]
  Large language models (LLMs) have recently demonstrated remarkable progress in reasoning capabilities through reinforcement learning with verifiable rewards (RLVR)<n>We present a theoretical characterization of LLM reasoning grounded in information bottleneck (IB) principle.<n>We propose IB-aware reasoning optimization (IBRO), a framework that encourages reasoning trajectories to be both informative about the final correct answer and generalizable.
  arXiv  Detail & Related papers  (2025-07-24T13:14:25Z)
• LLM Cannot Discover Causality, and Should Be Restricted to Non-Decisional Support in Causal Discovery [30.24849564413826]
  We demonstrate that LLMs' autoregressive, correlation-driven modeling inherently lacks the theoretical grounding for causal reasoning.<n>We show that deliberate prompt engineering could overstate their performance, helping to explain the consistently favorable results reported in much of the current literature.<n>We conclude with a call for the community to shift focus from naively applying LLMs to developing specialized models and training method that respect the core principles of causal discovery.
  arXiv  Detail & Related papers  (2025-06-01T05:38:56Z)
• Guiding Reasoning in Small Language Models with LLM Assistance [23.3038074903744]
  Small Language Models cast doubt suitability for tasks demanding deep, multi-step logical deduction.<n>This paper introduces a framework called Small Reasons, Large Hints, which selectively augments SLM reasoning with targeted guidance from large language models.<n>Our experiments on mathematical reasoning datasets demonstrate that targeted external scaffolding significantly improves performance.
  arXiv  Detail & Related papers  (2025-04-14T06:32:45Z)
• InductionBench: LLMs Fail in the Simplest Complexity Class [53.70978746199222]
  Large language models (LLMs) have shown remarkable improvements in reasoning.<n>Inductive reasoning, where one infers the underlying rules from observed data, remains less explored.<n>We introduce InductionBench, a new benchmark designed to evaluate the inductive reasoning ability of LLMs.
  arXiv  Detail & Related papers  (2025-02-20T03:48:00Z)
• Critical-Questions-of-Thought: Steering LLM reasoning with Argumentative Querying [0.3659498819753633]
  State-of-the-art Large Language models (LLMs) continue to struggle when performing logical and mathematical reasoning.<n>This paper makes use of the notion of critical questions from the literature on argumentation theory, focusing in particular on Toulmin's model of argumentation.<n>We show that employing these critical questions can improve the reasoning capabilities of LLMs.
  arXiv  Detail & Related papers  (2024-12-19T18:51:30Z)
• Evaluating the Correctness of Inference Patterns Used by LLMs for Judgment [53.17596274334017]
  We evaluate the correctness of the detailed inference patterns of an LLM behind its seemingly correct outputs.<n>Experiments show that even when the language generation results appear correct, a significant portion of the inference patterns used by the LLM for the legal judgment may represent misleading or irrelevant logic.
  arXiv  Detail & Related papers  (2024-10-06T08:33:39Z)
• DnA-Eval: Enhancing Large Language Model Evaluation through Decomposition and Aggregation [75.81096662788254]
  Large Language Models (LLMs) are scalable and economical evaluators.<n>The question of how reliable these evaluators are has emerged as a crucial research question.<n>We propose Decompose and Aggregate, which breaks down the evaluation process into different stages based on pedagogical practices.
  arXiv  Detail & Related papers  (2024-05-24T08:12:30Z)
• Unveiling the Misuse Potential of Base Large Language Models via In-Context Learning [61.2224355547598]
  Open-sourcing of large language models (LLMs) accelerates application development, innovation, and scientific progress. Our investigation exposes a critical oversight in this belief. By deploying carefully designed demonstrations, our research demonstrates that base LLMs could effectively interpret and execute malicious instructions.
  arXiv  Detail & Related papers  (2024-04-16T13:22:54Z)
• Beyond Accuracy: Evaluating the Reasoning Behavior of Large Language Models -- A Survey [25.732397636695882]
  Large language models (LLMs) have recently shown impressive performance on tasks involving reasoning. Despite these successes, the depth of LLMs' reasoning abilities remains uncertain.
  arXiv  Detail & Related papers  (2024-04-02T11:46:31Z)
• Beyond Probabilities: Unveiling the Misalignment in Evaluating Large Language Models [24.445829787297658]
  Large Language Models (LLMs) have demonstrated remarkable capabilities across various applications. This study aims to scrutinize the validity of such probability-based evaluation methods within the context of using LLMs for Multiple Choice Questions (MCQs) Our empirical investigation reveals that the prevalent probability-based evaluation method inadequately aligns with generation-based prediction.
  arXiv  Detail & Related papers  (2024-02-21T15:58:37Z)
• LLMs May Perform MCQA by Selecting the Least Incorrect Option [29.202758753639078]
  Large Language Models (LLMs) have markedly enhanced performance across a variety of tasks.<n>The adoption of Multiple Choice Question Answering (MCQA) as a benchmark for assessing LLMs has gained considerable traction.<n>However, concerns regarding the robustness of this evaluative method persist.
  arXiv  Detail & Related papers  (2024-02-02T12:07:00Z)
• Survey on Factuality in Large Language Models: Knowledge, Retrieval and Domain-Specificity [61.54815512469125]
  This survey addresses the crucial issue of factuality in Large Language Models (LLMs) As LLMs find applications across diverse domains, the reliability and accuracy of their outputs become vital.
  arXiv  Detail & Related papers  (2023-10-11T14:18:03Z)

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.