A Sober Look at Progress in Language Model Reasoning: Pitfalls and Paths to Reproducibility

• URL: http://arxiv.org/abs/2504.07086v1
• Date: Wed, 09 Apr 2025 17:58:17 GMT
• Title: A Sober Look at Progress in Language Model Reasoning: Pitfalls and Paths to Reproducibility
• Authors: Andreas Hochlehnert, Hardik Bhatnagar, Vishaal Udandarao, Samuel Albanie, Ameya Prabhu, Matthias Bethge,
• Abstract summary: Reasoning has emerged as the next major frontier for language models (LMs)<n>We conduct a comprehensive empirical study and find that current mathematical reasoning benchmarks are highly sensitive to subtle implementation choices.<n>We propose a standardized evaluation framework with clearly defined best practices and reporting standards.
• Score: 29.437125712259046
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reasoning has emerged as the next major frontier for language models (LMs), with rapid advances from both academic and industrial labs. However, this progress often outpaces methodological rigor, with many evaluations relying on benchmarking practices that lack transparency, robustness, or statistical grounding. In this work, we conduct a comprehensive empirical study and find that current mathematical reasoning benchmarks are highly sensitive to subtle implementation choices - including decoding parameters, random seeds, prompt formatting, and even hardware and software-framework configurations. Performance gains reported in recent studies frequently hinge on unclear comparisons or unreported sources of variance. To address these issues, we propose a standardized evaluation framework with clearly defined best practices and reporting standards. Using this framework, we reassess recent methods and find that reinforcement learning (RL) approaches yield only modest improvements - far below prior claims - and are prone to overfitting, especially on small-scale benchmarks like AIME24. In contrast, supervised finetuning (SFT) methods show consistently stronger generalization. To foster reproducibility, we release all code, prompts, and model outputs, for reasoning benchmarks, establishing more rigorous foundations for future work.

Related papers

• Position: There are no Champions in Long-Term Time Series Forecasting [0.0]
  We focus on the need to shift focus away from pursuing ever-more complex models and towards enhancing benchmarking practices.<n>To support our claim, we first perform a broad, thorough, and reproducible evaluation of the top-performing models on the most popular benchmark by training 3,500+ networks over 14 datasets.
  arXiv  Detail & Related papers  (2025-02-19T19:08:37Z)
• Scoring Verifiers: Evaluating Synthetic Verification for Code and Reasoning [59.25951947621526]
  We propose an approach which can transform existing coding benchmarks into scoring and ranking datasets to evaluate the effectiveness of synthetic verifiers.<n>We release four new benchmarks (HE-R, HE-R+, MBPP-R, and MBPP-R+), and analyzed synthetic verification methods with standard, reasoning-based, and reward-based LLMs.<n>Our experiments show that reasoning can significantly improve test case generation and that scaling the number of test cases enhances the verification accuracy.
  arXiv  Detail & Related papers  (2025-02-19T15:32:11Z)
• BRiTE: Bootstrapping Reinforced Thinking Process to Enhance Language Model Reasoning [78.63421517563056]
  Large Language Models (LLMs) have demonstrated remarkable capabilities in complex reasoning tasks.<n>We present a unified probabilistic framework that formalizes LLM reasoning through a novel graphical model.<n>We introduce the Bootstrapping Reinforced Thinking Process (BRiTE) algorithm, which works in two steps.
  arXiv  Detail & Related papers  (2025-01-31T02:39:07Z)
• The Vulnerability of Language Model Benchmarks: Do They Accurately Reflect True LLM Performance? [1.3810901729134184]
  Large Language Models (LLMs) excel at standardized tests while failing to demonstrate genuine language understanding and adaptability. Our systematic analysis of NLP evaluation frameworks reveals pervasive vulnerabilities across the evaluation spectrum. We lay the groundwork for new evaluation methods that resist manipulation, minimize data contamination, and assess domain-specific tasks.
  arXiv  Detail & Related papers  (2024-12-02T20:49:21Z)
• A Preliminary Study of Multilingual Code Language Models for Code Generation Task Using Translated Benchmarks [0.0]
  We evaluate the performance of Poly-Coder, a pioneering open-source, multilingual CLM built for code generation. Our results suggest that the outcomes observed in these translated benchmarks align well with evaluation metrics used during the training phase. These initial insights highlight the need for more comprehensive empirical studies.
  arXiv  Detail & Related papers  (2024-11-23T06:40:47Z)
• HELMET: How to Evaluate Long-Context Language Models Effectively and Thoroughly [34.205934899868346]
  We introduce HELMET, a comprehensive benchmark encompassing seven diverse, application-centric categories. We find that synthetic tasks like NIAH do not reliably predict downstream performance. While most LCLMs achieve perfect NIAH scores, open-source models significantly lag behind closed ones when tasks require full-context reasoning.
  arXiv  Detail & Related papers  (2024-10-03T17:20:11Z)
• Hierarchical Reinforcement Learning for Temporal Abstraction of Listwise Recommendation [51.06031200728449]
  We propose a novel framework called mccHRL to provide different levels of temporal abstraction on listwise recommendation. Within the hierarchical framework, the high-level agent studies the evolution of user perception, while the low-level agent produces the item selection policy. Results observe significant performance improvement by our method, compared with several well-known baselines.
  arXiv  Detail & Related papers  (2024-09-11T17:01:06Z)
• Alice in Wonderland: Simple Tasks Showing Complete Reasoning Breakdown in State-Of-the-Art Large Language Models [13.532180752491954]
  Large Language Models (LLMs) are often described as instances of foundation models that possess strong generalization obeying scaling laws.<n>We demonstrate here a dramatic breakdown of generalization and basic reasoning of all SOTA models claiming strong function.<n>We also observe strong overconfidence in the wrong solutions, expressed in form of plausible sounding explanation-like confabulations.
  arXiv  Detail & Related papers  (2024-06-04T07:43:33Z)
• Disperse-Then-Merge: Pushing the Limits of Instruction Tuning via Alignment Tax Reduction [75.25114727856861]
  Large language models (LLMs) tend to suffer from deterioration at the latter stage ofSupervised fine-tuning process. We introduce a simple disperse-then-merge framework to address the issue. Our framework outperforms various sophisticated methods such as data curation and training regularization on a series of standard knowledge and reasoning benchmarks.
  arXiv  Detail & Related papers  (2024-05-22T08:18:19Z)
• Quantifying Contamination in Evaluating Code Generation Capabilities of Language Models [27.24738197172374]
  Large language models have achieved remarkable performance on various code generation benchmarks. There have been growing concerns regarding potential contamination of these benchmarks as they may be leaked into pretraining and finetuning data. We show that there are substantial overlap between popular code generation benchmarks and open training corpus, and models perform significantly better on the subset of the benchmarks where similar solutions are seen during training.
  arXiv  Detail & Related papers  (2024-03-06T21:45:35Z)
• Don't Make Your LLM an Evaluation Benchmark Cheater [142.24553056600627]
  Large language models(LLMs) have greatly advanced the frontiers of artificial intelligence, attaining remarkable improvement in model capacity. To assess the model performance, a typical approach is to construct evaluation benchmarks for measuring the ability level of LLMs. We discuss the potential risk and impact of inappropriately using evaluation benchmarks and misleadingly interpreting the evaluation results.
  arXiv  Detail & Related papers  (2023-11-03T14:59:54Z)
• Large-scale Pre-trained Models are Surprisingly Strong in Incremental Novel Class Discovery [76.63807209414789]
  We challenge the status quo in class-iNCD and propose a learning paradigm where class discovery occurs continuously and truly unsupervisedly. We propose simple baselines, composed of a frozen PTM backbone and a learnable linear classifier, that are not only simple to implement but also resilient under longer learning scenarios.
  arXiv  Detail & Related papers  (2023-03-28T13:47:16Z)

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.