Keqian Li

Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri et al. · allen-ai, berkeley

Modern artificial intelligence (AI) systems are powered by foundation models. This paper presents a new set of foundation models, called Llama 3. It is a herd of language models that natively support multilinguality, coding, reasoning, and tool usage. Our largest model is a dense Transformer with 405B parameters and a context window of up to 128K tokens. This paper presents an extensive empirical evaluation of Llama 3. We find that Llama 3 delivers comparable quality to leading language models such as GPT-4 on a plethora of tasks. We publicly release Llama 3, including pre-trained and post-trained versions of the 405B parameter language model and our Llama Guard 3 model for input and output safety. The paper also presents the results of experiments in which we integrate image, video, and speech capabilities into Llama 3 via a compositional approach. We observe this approach performs competitively with the state-of-the-art on image, video, and speech recognition tasks. The resulting models are not yet being broadly released as they are still under development.

Shuzhen Bi, Mengsong Wu, Hao Hao et al.

The transition from monolithic large language models (LLMs) to modular, skill-equipped agents represents a fundamental architectural shift in artificial intelligence deployment. While general-purpose models demonstrate remarkable breadth in declarative knowledge, their utility in autonomous workflows is frequently constrained by insufficient specialized procedural expertise. This report investigates a systematic framework for automated acquisition of high-quality agent skills through mining of open-source repositories on platforms such as GitHub. We focus on the extraction of visualization and educational capabilities from state-of-the-art systems including TheoremExplainAgent and Code2Video, both utilizing the Manim mathematical animation engine. The framework encompasses repository structural analysis, semantic skill identification through dense retrieval, and translation to the standardized SKILL.md format. We demonstrate that systematic extraction from agentic repositories, combined with rigorous security governance and multi-dimensional evaluation metrics, enables scalable acquisition of procedural knowledge that augments LLM capabilities without requiring model retraining. Our analysis reveals that agent-generated educational content can achieve 40\% gains in knowledge transfer efficiency while maintaining pedagogical quality comparable to human-crafted tutorials.

Keqian Li, Yifan Hu, Logan Palanisamy et al.

Accurate understanding of users in terms of predicative segments play an essential role in the day to day operation of modern internet enterprises. Nevertheless, there are significant challenges that limit the quality of data, especially on long tail predictive tasks. In this work, we present MetaCon, our unified predicative segments system with scalable, trillion concepts meta learning that addresses these challenges. It builds on top of a flat concept representation that summarizes entities' heterogeneous digital footprint, jointly considers the entire spectrum of predicative tasks as a single learning task, and leverages principled meta learning approach with efficient first order meta-optimization procedure under a provable performance guarantee in order to solve the learning task. Experiments on both proprietary production datasets and public structured learning tasks demonstrate that MetaCon can lead to substantial improvements over state of the art recommendation and ranking approaches.

Keqian Li, Yifan Hu

We study the problem of user segmentation: given a set of users and one or more predefined groups or segments, assign users to their corresponding segments. As an example, for a segment indicating particular interest in a certain area of sports or entertainment, the task will be to predict whether each single user will belong to the segment. However, there may exist numerous long tail prediction tasks that suffer from data availability and may be of heterogeneous nature, which make it hard to capture using single off the shelf model architectures. In this work, we present SuperCone, our unified predicative segments system that addresses the above challenges. It builds on top of a flat concept representation that summarizes each user's heterogeneous digital footprints, and uniformly models each of the prediction task using an approach called "super learning ", that is, combining prediction models with diverse architectures or learning method that are not compatible with each other. Following this, we provide an end to end approach that learns to flexibly attend to best suited heterogeneous experts adaptively, while at the same time incorporating deep representations of the input concepts that augments the above experts. Experiments show that SuperCone significantly outperform state-of-the-art recommendation and ranking algorithms on a wide range of predicative segment tasks and public structured data learning benchmarks.

Shuzhen Bi, Mingzi Zhang, Zhuoxuan Li et al.

Large language models are increasingly used as educational assistants, yet evaluation of their educational capabilities remains concentrated on question-answering and tutoring tasks. A critical gap exists for multimedia instructional content generation -- the ability to produce coherent, diagram-rich explanations that combine geometrically accurate visuals with step-by-step reasoning. We present EduIllustrate, a benchmark for evaluating LLMs on interleaved text-diagram explanation generation for K-12 STEM problems. The benchmark comprises 230 problems spanning five subjects and three grade levels, a standardized generation protocol with sequential anchoring to enforce cross-diagram visual consistency, and an 8-dimension evaluation rubric grounded in multimedia learning theory covering both text and visual quality. Evaluation of ten LLMs reveals a wide performance spread: Gemini 3.0 Pro Preview leads at 87.8\%, while Kimi-K2.5 achieves the best cost-efficiency (80.8\% at \\$0.12/problem). Workflow ablation confirms sequential anchoring improves Visual Consistency by 13\% at 94\% lower cost. Human evaluation with 20 expert raters validates LLM-as-judge reliability for objective dimensions ($ρ\geq 0.83$) while revealing limitations on subjective visual assessment.

Mengsong Wu, Hao Hao, Shuzhen Bi et al.

While scaling laws for Large Language Models (LLMs) have been extensively studied along dimensions of model parameters, training data, and compute, the scaling behavior of LLM-based educational agents remains unexplored. We propose that educational agent capability scales not merely with the underlying model size, but through structured dimensions that we collectively term the Agent Scaling Law: role definition clarity, skill depth, tool completeness, runtime capability, and educator expertise injection. Central to this framework is AgentProfile, a structured JSON-based specification that serves as the mechanism enabling systematic capability growth of educational agents. We present EduClaw, a profile-driven multi-agent platform that operationalizes this scaling law, demonstrating its effectiveness through the construction and deployment of 330+ educational agent profiles encompassing 1,100+ skill modules across K-12 subjects. Our empirical observations suggest that educational agent performance scales predictably with profile structural richness. We identify two complementary scaling axes -- Tool Scaling and Skill Scaling -- as future directions, arguing that the path to more capable educational AI lies not solely in larger models, but in stronger structured capability systems.

Wei Lu, Shanshan Chen, Keqian Li et al.

Recommender Systems (RS) play a vital role in applications such as e-commerce and on-demand content streaming. Research on RS has mainly focused on the customer perspective, i.e., accurate prediction of user preferences and maximization of user utilities. As a result, most existing techniques are not explicitly built for revenue maximization, the primary business goal of enterprises. In this work, we explore and exploit a novel connection between RS and the profitability of a business. As recommendations can be seen as an information channel between a business and its customers, it is interesting and important to investigate how to make strategic dynamic recommendations leading to maximum possible revenue. To this end, we propose a novel \model that takes into account a variety of factors including prices, valuations, saturation effects, and competition amongst products. Under this model, we study the problem of finding revenue-maximizing recommendation strategies over a finite time horizon. We show that this problem is NP-hard, but approximation guarantees can be obtained for a slightly relaxed version, by establishing an elegant connection to matroid theory. Given the prohibitively high complexity of the approximation algorithm, we also design intelligent heuristics for the original problem. Finally, we conduct extensive experiments on two real and synthetic datasets and demonstrate the efficiency, scalability, and effectiveness our algorithms, and that they significantly outperform several intuitive baselines.