Densing Law of LLMs

Large Language Models (LLMs) have emerged as a milestone in artificial intelligence, and their performance can improve as the model size increases. However, this scaling brings great challenges to training and inference efficiency, particularly for deploying LLMs in resource-constrained environments, and the scaling trend is becoming increasingly unsustainable. This paper introduces the concept of ``\textit{capacity density}'' as a new metric to evaluate the quality of the LLMs across different scales and describes the trend of LLMs in terms of both effectiveness and efficiency. To calculate the capacity density of a given target LLM, we first introduce a set of reference models and develop a scaling law to predict the downstream performance of these reference models based on their parameter sizes. We then define the \textit{effective parameter size} of the target LLM as the parameter size required by a reference model to achieve equivalent performance, and formalize the capacity density as the ratio of the effective parameter size to the actual parameter size of the target LLM. Capacity density provides a unified framework for assessing both model effectiveness and efficiency. Our further analysis of recent open-source base LLMs reveals an empirical law (the densing law)that the capacity density of LLMs grows exponentially over time. More specifically, using some widely used benchmarks for evaluation, the capacity density of LLMs doubles approximately every three months. The law provides new perspectives to guide future LLM development, emphasizing the importance of improving capacity density to achieve optimal results with minimal computational overhead.

DecorateLM: Data Engineering through Corpus Rating, Tagging, and Editing with Language Models

The performance of Large Language Models (LLMs) is substantially influenced by the pretraining corpus, which consists of vast quantities of unsupervised data processed by the models. Despite its critical role in model performance, ensuring the quality of this data is challenging due to its sheer volume and the absence of sample-level quality annotations and enhancements. In this paper, we introduce DecorateLM, a data engineering method designed to refine the pretraining corpus through data rating, tagging and editing. Specifically, DecorateLM rates texts against quality criteria, tags texts with hierarchical labels, and edits texts into a more formalized format. Due to the massive size of the pretraining corpus, adopting an LLM for decorating the entire corpus is less efficient. Therefore, to balance performance with efficiency, we curate a meticulously annotated training corpus for DecorateLM using a large language model and distill data engineering expertise into a compact 1.2 billion parameter small language model (SLM). We then apply DecorateLM to enhance 100 billion tokens of the training corpus, selecting 45 billion tokens that exemplify high quality and diversity for the further training of another 1.2 billion parameter LLM. Our results demonstrate that employing such high-quality data can significantly boost model performance, showcasing a powerful approach to enhance the quality of the pretraining corpus.

Beyond Forecasting: Compositional Time Series Reasoning for End-to-End Task Execution

In recent decades, there has been substantial advances in time series models and benchmarks across various individual tasks, such as time series forecasting, classification, and anomaly detection. Meanwhile, compositional reasoning in time series is prevalent in real-world applications (e.g., decision-making and compositional question answering) and is in great demand. Unlike simple tasks that primarily focus on predictive accuracy, compositional reasoning emphasizes the synthesis of diverse information from both time series data and various domain knowledge, making it distinct and extremely more challenging. In this paper, we introduce Compositional Time Series Reasoning, a new task of handling intricate multistep reasoning tasks from time series data. Specifically, this new task focuses on various question instances requiring structural and compositional reasoning abilities on time series data, such as decision-making and compositional question answering. As an initial attempt to tackle this novel task, we developed TS-Reasoner, a program-aided approach that utilizes large language model (LLM) to decompose a complex task into steps of programs that leverage existing time series models and numerical subroutines. Unlike existing reasoning work which only calls off-the-shelf modules, TS-Reasoner allows for the creation of custom modules and provides greater flexibility to incorporate domain knowledge as well as user-specified constraints. We demonstrate the effectiveness of our method through a comprehensive set of experiments. These promising results indicate potential opportunities in the new task of time series reasoning and highlight the need for further research.

MiniCPM-V: A GPT-4V Level MLLM on Your Phone

The recent surge of Multimodal Large Language Models (MLLMs) has fundamentally reshaped the landscape of AI research and industry, shedding light on a promising path toward the next AI milestone. However, significant challenges remain preventing MLLMs from being practical in real-world applications. The most notable challenge comes from the huge cost of running an MLLM with a massive number of parameters and extensive computation. As a result, most MLLMs need to be deployed on high-performing cloud servers, which greatly limits their application scopes such as mobile, offline, energy-sensitive, and privacy-protective scenarios. In this work, we present MiniCPM-V, a series of efficient MLLMs deployable on end-side devices. By integrating the latest MLLM techniques in architecture, pretraining and alignment, the latest MiniCPM-Llama3-V 2.5 has several notable features: (1) Strong performance, outperforming GPT-4V-1106, Gemini Pro and Claude 3 on OpenCompass, a comprehensive evaluation over 11 popular benchmarks, (2) strong OCR capability and 1.8M pixel high-resolution image perception at any aspect ratio, (3) trustworthy behavior with low hallucination rates, (4) multilingual support for 30+ languages, and (5) efficient deployment on mobile phones. More importantly, MiniCPM-V can be viewed as a representative example of a promising trend: The model sizes for achieving usable (e.g., GPT-4V) level performance are rapidly decreasing, along with the fast growth of end-side computation capacity. This jointly shows that GPT-4V level MLLMs deployed on end devices are becoming increasingly possible, unlocking a wider spectrum of real-world AI applications in the near future.

PyBench: Evaluating LLM Agent on various real-world coding tasks

The LLM Agent, equipped with a code interpreter, is capable of automatically solving real-world coding tasks, such as data analysis and image editing. However, existing benchmarks primarily focus on either simplistic tasks, such as completing a few lines of code, or on extremely complex and specific tasks at the repository level, neither of which are representative of various daily coding tasks. To address this gap, we introduce \textbf{PyBench}, a benchmark encompassing five main categories of real-world tasks, covering more than 10 types of files. Given a high-level user query and related files, the LLM Agent needs to reason and execute Python code via a code interpreter for a few turns before making a formal response to fulfill the user's requirements. Successfully addressing tasks in PyBench demands a robust understanding of various Python packages, superior reasoning capabilities, and the ability to incorporate feedback from executed code. Our evaluations indicate that current open-source LLMs are struggling with these tasks. Hence, we conduct analysis and experiments on four kinds of datasets proving that comprehensive abilities are needed for PyBench. Our fine-tuned 8B size model: \textbf{PyLlama3} achieves an exciting performance on PyBench which surpasses many 33B and 70B size models. Our Benchmark, Training Dataset, and Model are available at: \href{https://github.com/Mercury7353/PyBench}{https://github.com/Mercury7353/PyBench}

MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies

The burgeoning interest in developing Large Language Models (LLMs) with up to trillion parameters has been met with concerns regarding resource efficiency and practical expense, particularly given the immense cost of experimentation. This scenario underscores the importance of exploring the potential of Small Language Models (SLMs) as a resource-efficient alternative. In this context, we introduce MiniCPM, specifically the 1.2B and 2.4B non-embedding parameter variants, not only excel in their respective categories but also demonstrate capabilities on par with 7B-13B LLMs. While focusing on SLMs, our approach exhibits scalability in both model and data dimensions for future LLM research. Regarding model scaling, we employ extensive model wind tunnel experiments for stable and optimal scaling. For data scaling, we introduce a Warmup-Stable-Decay (WSD) learning rate scheduler (LRS), conducive to continuous training and domain adaptation. We present an in-depth analysis of the intriguing training dynamics that occurred in the WSD LRS. With WSD LRS, we are now able to efficiently study data-model scaling law without extensive retraining experiments on both axes of model and data, from which we derive the much higher compute optimal data-model ratio than Chinchilla Optimal. Additionally, we introduce MiniCPM family, including MiniCPM-DPO, MiniCPM-MoE and MiniCPM-128K, whose excellent performance further cementing MiniCPM's foundation in diverse SLM applications. MiniCPM models are available publicly at https://github.com/OpenBMB/MiniCPM .

A Preliminary Exploration of YouTubers' Use of Generative-AI in Content Creation

Content creators increasingly utilize generative artificial intelligence (Gen-AI) on platforms such as YouTube, TikTok, Instagram, and various blogging sites to produce imaginative images, AI-generated videos, and articles using Large Language Models (LLMs). Despite its growing popularity, there remains an underexplored area concerning the specific domains where AI-generated content is being applied, and the methodologies content creators employ with Gen-AI tools during the creation process. This study initially explores this emerging area through a qualitative analysis of 68 YouTube videos demonstrating Gen-AI usage. Our research focuses on identifying the content domains, the variety of tools used, the activities performed, and the nature of the final products generated by Gen-AI in the context of user-generated content.

Knowledge Graph Completion with Counterfactual Augmentation

Graph Neural Networks (GNNs) have demonstrated great success in Knowledge Graph Completion (KGC) by modeling how entities and relations interact in recent years. However, most of them are designed to learn from the observed graph structure, which appears to have imbalanced relation distribution during the training stage. Motivated by the causal relationship among the entities on a knowledge graph, we explore this defect through a counterfactual question: "would the relation still exist if the neighborhood of entities became different from observation?". With a carefully designed instantiation of a causal model on the knowledge graph, we generate the counterfactual relations to answer the question by regarding the representations of entity pair given relation as context, structural information of relation-aware neighborhood as treatment, and validity of the composed triplet as the outcome. Furthermore, we incorporate the created counterfactual relations with the GNN-based framework on KGs to augment their learning of entity pair representations from both the observed and counterfactual relations. Experiments on benchmarks show that our proposed method outperforms existing methods on the task of KGC, achieving new state-of-the-art results. Moreover, we demonstrate that the proposed counterfactual relations-based augmentation also enhances the interpretability of the GNN-based framework through the path interpretations of predictions.

Real-Time Super-Resolution for Real-World Images on Mobile Devices

Image Super-Resolution (ISR), which aims at recovering High-Resolution (HR) images from the corresponding Low-Resolution (LR) counterparts. Although recent progress in ISR has been remarkable. However, they are way too computationally intensive to be deployed on edge devices, since most of the recent approaches are deep learning-based. Besides, these methods always fail in real-world scenes, since most of them adopt a simple fixed "ideal" bicubic downsampling kernel from high-quality images to construct LR/HR training pairs which may lose track of frequency-related details. In this work, an approach for real-time ISR on mobile devices is presented, which is able to deal with a wide range of degradations in real-world scenarios. Extensive experiments on traditional super-resolution datasets (Set5, Set14, BSD100, Urban100, Manga109, DIV2K) and real-world images with a variety of degradations demonstrate that our method outperforms the state-of-art methods, resulting in higher PSNR and SSIM, lower noise and better visual quality. Most importantly, our method achieves real-time performance on mobile or edge devices.

The Detection of Thoracic Abnormalities ChestX-Det10 Challenge Results

The detection of thoracic abnormalities challenge is organized by the Deepwise AI Lab. The challenge is divided into two rounds. In this paper, we present the results of 6 teams which reach the second round. The challenge adopts the ChestX-Det10 dateset proposed by the Deepwise AI Lab. ChestX-Det10 is the first chest X-Ray dataset with instance-level annotations, including 10 categories of disease/abnormality of 3,543 images. The annotations are located at https://github.com/Deepwise-AILab/ChestX-Det10-Dataset. In the challenge, we randomly split all data into 3001 images for training and 542 images for testing.