We introduce MLE-bench, a benchmark for measuring how well AI agents perform at machine learning engineering. To this end, we curate 75 ML engineering-related competitions from Kaggle, creating a diverse set of challenging tasks that test real-world ML engineering skills such as training models, preparing datasets, and running experiments. We establish human baselines for each competition using Kaggle's publicly available leaderboards. We use open-source agent scaffolds to evaluate several frontier language models on our benchmark, finding that the best-performing setup--OpenAI's o1-preview with AIDE scaffolding--achieves at least the level of a Kaggle bronze medal in 16.9% of competitions. In addition to our main results, we investigate various forms of resource scaling for AI agents and the impact of contamination from pre-training. We open-source our benchmark code (github.com/openai/mle-bench/) to facilitate future research in understanding the ML engineering capabilities of AI agents.

Enterprises possess a vast array of API assets scattered across various functions, forming the backbone of existing business processes. By leveraging these APIs as functional tools, enterprises can design diverse, scenario-specific agent applications, driven by on-premise function-calling models as the core engine. However, generic models often fail to meet enterprise requirements in terms of computational efficiency, output accuracy, and stability, necessitating scenario-specific adaptation. In this paper, we propose a training pipeline for function-calling capabilities tailored to real-world business scenarios. This pipeline includes the synthesis and augmentation of scenario-specific function-calling data, model fine-tuning, and performance evaluation and analysis. Using this pipeline, we generated 1,260 fully AI-generated samples and 1,035 augmented manually-labeled samples in digital HR agent scenario. The Qwen2.5-Coder-7B-Instruct model was employed as the base model and fine-tuned using the LoRA method on four GPUs with 24GB VRAM. Our fine-tuned model demonstrated outstanding performance in evaluations and practical applications, surpassing GPT-4 and GPT-4o in accuracy on the test set. These results validate the reliability of the proposed pipeline for training scenario-specific function-calling models.

The discovery of novel mechanical metamaterials, whose properties are dominated by their engineered structures rather than chemical composition, is a knowledge-intensive and resource-demanding process. To accelerate the design of novel metamaterials, we present MetaScientist, a human-in-the-loop system that integrates advanced AI capabilities with expert oversight with two primary phases: (1) hypothesis generation, where the system performs complex reasoning to generate novel and scientifically sound hypotheses, supported with domain-specific foundation models and inductive biases retrieved from existing literature; (2) 3D structure synthesis, where a 3D structure is synthesized with a novel 3D diffusion model based on the textual hypothesis and refined it with a LLM-based refinement model to achieve better structure properties. At each phase, domain experts iteratively validate the system outputs, and provide feedback and supplementary materials to ensure the alignment of the outputs with scientific principles and human preferences. Through extensive evaluation from human scientists, MetaScientist is able to deliver novel and valid mechanical metamaterial designs that have the potential to be highly impactful in the metamaterial field.

Recently, discrete diffusion language models have demonstrated promising results in NLP. However, there has been limited research on integrating Pretrained Language Models (PLMs) into discrete diffusion models, resulting in underwhelming performance in downstream NLP generation tasks. This integration is particularly challenging because of the discrepancy between step-wise denoising strategy of diffusion models and single-step mask prediction approach of MLM-based PLMs. In this paper, we introduce Diffusion-EAGS, a novel approach that effectively integrates PLMs with the diffusion models. Furthermore, as it is challenging for PLMs to determine where to apply denoising during the diffusion process, we integrate an entropy tracking module to assist them. Finally, we propose entropy-based noise scheduling in the forward process to improve the effectiveness of entropy-adaptive sampling throughout the generation phase. Experimental results show that Diffusion-EAGS outperforms existing diffusion baselines in downstream generation tasks, achieving high text quality and diversity with precise token-level control. We also show that our model is capable of adapting to bilingual and low-resource settings, which are common in real-world applications.

We address the problem of reward hacking, where maximising a proxy reward does not necessarily increase the true reward. This is a key concern for Large Language Models (LLMs), as they are often fine-tuned on human preferences that may not accurately reflect a true objective. Existing work uses various tricks such as regularisation, tweaks to the reward model, and reward hacking detectors, to limit the influence that such proxy preferences have on a model. Luckily, in many contexts such as medicine, education, and law, a sparse amount of expert data is often available. In these cases, it is often unclear whether the addition of proxy data can improve policy learning. We outline a set of sufficient conditions on proxy feedback that, if satisfied, indicate that proxy data can provably improve the sample complexity of learning the ground truth policy. These conditions can inform the data collection process for specific tasks. The result implies a parameterisation for LLMs that achieves this improved sample complexity. We detail how one can adapt existing architectures to yield this improved sample complexity.

In this era of technological advancements, several cutting-edge techniques are being implemented to enhance Autonomous Driving (AD) systems, focusing on improving safety, efficiency, and adaptability in complex driving environments. However, AD still faces some problems including performance limitations. To address this problem, we conducted an in-depth study on implementing the Multi-modal Large Language Model. We constructed a Virtual Question Answering (VQA) dataset to fine-tune the model and address problems with the poor performance of MLLM on AD. We then break down the AD decision-making process by scene understanding, prediction, and decision-making. Chain of Thought has been used to make the decision more perfectly. Our experiments and detailed analysis of Autonomous Driving give an idea of how important MLLM is for AD.

This pilot study presents the development of the InfoTech Assistant, a domain-specific, multimodal chatbot engineered to address queries in bridge evaluation and infrastructure technology. By integrating web data scraping, large language models (LLMs), and Retrieval-Augmented Generation (RAG), the InfoTech Assistant provides accurate and contextually relevant responses. Data, including textual descriptions and images, are sourced from publicly available documents on the InfoTechnology website and organized in JSON format to facilitate efficient querying. The architecture of the system includes an HTML-based interface and a Flask back end connected to the Llama 3.1 model via LLM Studio. Evaluation results show approximately 95 percent accuracy on domain-specific tasks, with high similarity scores confirming the quality of response matching. This RAG-enhanced setup enables the InfoTech Assistant to handle complex, multimodal queries, offering both textual and visual information in its responses. The InfoTech Assistant demonstrates strong potential as a dependable tool for infrastructure professionals, delivering high accuracy and relevance in its domain-specific outputs.

Background: Recent advances in large language models highlight the need for high-quality multilingual medical datasets. While Japan leads globally in CT scanner deployment and utilization, the lack of large-scale Japanese radiology datasets has hindered the development of specialized language models for medical imaging analysis. Objective: To develop a comprehensive Japanese CT report dataset through machine translation and establish a specialized language model for structured finding classification. Additionally, to create a rigorously validated evaluation dataset through expert radiologist review. Methods: We translated the CT-RATE dataset (24,283 CT reports from 21,304 patients) into Japanese using GPT-4o mini. The training dataset consisted of 22,778 machine-translated reports, while the validation dataset included 150 radiologist-revised reports. We developed CT-BERT-JPN based on "tohoku-nlp/bert-base-japanese-v3" architecture for extracting 18 structured findings from Japanese radiology reports. Results: Translation metrics showed strong performance with BLEU scores of 0.731 and 0.690, and ROUGE scores ranging from 0.770 to 0.876 for Findings and from 0.748 to 0.857 for Impression sections. CT-BERT-JPN demonstrated superior performance compared to GPT-4o in 11 out of 18 conditions, including lymphadenopathy (+14.2%), interlobular septal thickening (+10.9%), and atelectasis (+7.4%). The model maintained F1 scores exceeding 0.95 in 14 out of 18 conditions and achieved perfect scores in four conditions. Conclusions: Our study establishes a robust Japanese CT report dataset and demonstrates the effectiveness of a specialized language model for structured finding classification. The hybrid approach of machine translation and expert validation enables the creation of large-scale medical datasets while maintaining high quality.

This paper addresses the challenges in developing language models for less-represented languages, with a focus on Luxembourgish. Despite its active development, Luxembourgish faces a digital data scarcity, exacerbated by Luxembourg's multilingual context. We propose a novel text generation model based on the T5 architecture, combining limited Luxembourgish data with equal amounts, in terms of size and type, of German and French data. We hypothesise that a model trained on Luxembourgish, German, and French will improve the model's cross-lingual transfer learning capabilities and outperform monolingual and large multilingual models. To verify this, the study at hand explores whether multilingual or monolingual training is more beneficial for Luxembourgish language generation. For the evaluation, we introduce LuxGen, a text generation benchmark that is the first of its kind for Luxembourgish.

Large Vision-Language Models (LVLMs) excel in cross-model tasks but experience performance declines in long-context reasoning due to overreliance on textual information and reduced visual dependency. In this study, we empirically analyze LVLMs in long-context reasoning, revealing that increased context length leads to a higher dependence on language at the expense of visual dependency. To address this issue, we propose a novel training-free context pruning method that selectively removes less critical textual information. Our approach enhances visual dependency and reduces textual noise, thereby improving LVLM performance in long-context reasoning. We validate our method by constructing a long-context dataset, demonstrating its effectiveness across various LVLMs. Moreover, further analysis confirms the robustness of different token pruning strategies and preliminary explores scaling laws between pruning rates and context length.