Related papers: Utility-Aware Data Pricing: Token-Level Quality an…
Recent advances in large language models (LLMs) have increasingly relied on reinforcement learning (RL) to improve their reasoning capabilities. Three types of approaches have been widely adopted: The first relies on a deep neural network…
We introduce a simple, yet novel entropy-based framework to drive token efficiency in large language models during reasoning tasks. Our approach uses Shannon entropy from token-level logprobs as a confidence signal to enable early stopping,…
Traditional data quality control methods are based on users experience or previously established business rules, and this limits performance in addition to being a very time consuming process with lower than desirable accuracy. Utilizing…
Current trends in pre-training Large Language Models (LLMs) primarily focus on the scaling of model and dataset size. While the quality of pre-training data is considered an important factor for training powerful LLMs, it remains a nebulous…
Motivated by applications such as cloud platforms allocating GPUs to users or governments deploying mobile health units across competing regions, we study the dynamic allocation of a reusable resource to strategic agents with private…
Traditional ranking systems optimize offline proxy objectives that rely on oversimplified assumptions about user behavior, often neglecting factors such as position bias and item diversity. Consequently, these models fail to improve true…
Reinforcement learning with human feedback (RLHF), as a widely adopted approach in current large language model pipelines, is \textit{bottlenecked by the size of human preference data}. While traditional methods rely on offline preference…
Business processes are fundamental to organizational operations, yet their optimization remains challenging due to the timeconsuming nature of manual process analysis. Our paper harnesses Large Language Models (LLMs) to automate value-added…
Tokenization is a fundamental component of large language models (LLMs), yet its influence on model scaling and performance is not fully explored. In this paper, we introduce Over-Tokenized Transformers, a novel framework that decouples…
The immense success of ML systems relies heavily on large-scale, high-quality data. The high demand for data has led to many paradigms that involve selling, exchanging, and sharing data, motivating the study of economic processes with data…
Interactive large language model (LLM) agents operating via multi-turn dialogue and multi-step tool calling are increasingly used in production. Benchmarks for these agents must both reliably compare models and yield on-policy training…
Reinforcement learning (RL) training of large language models (LLMs) on unverifiable tasks is challenging even when a reasonable-quality reference answer is available. We propose a constrained RL training framework that (i) optimizes a…
Fine-tuning the pre-trained model with active learning holds promise for reducing annotation costs. However, this combination introduces significant computational costs, particularly with the growing scale of pre-trained models. Recent…
Active learning is a promising paradigm to reduce the labeling cost by strategically requesting labels to improve model performance. However, existing active learning methods often rely on expensive acquisition function to compute,…
The conventional Cranfield paradigm struggles to effectively capture user satisfaction due to its weak correlation between relevance and satisfaction, alongside the high costs of relevance annotation in building test collections. To tackle…
Recent advancements have shown that reinforcement learning (RL) can substantially improve the reasoning abilities of large language models (LLMs). The effectiveness of such RL training, however, depends critically on the exploration space…
Collecting labeled datasets in finance is challenging due to scarcity of domain experts and higher cost of employing them. While Large Language Models (LLMs) have demonstrated remarkable performance in data annotation tasks on general…
Process Reward Models (PRMs) have achieved remarkable success in augmenting the reasoning capabilities of Large Language Models (LLMs) within static domains such as mathematics. However, their potential in dynamic data analysis tasks…
Continual instruction tuning enables large language models (LLMs) to learn incrementally while retaining past knowledge, whereas existing methods primarily focus on how to retain old knowledge rather than on selecting which new knowledge to…
Recently, reinforcement learning with verifiable rewards (RLVR) has been widely used for enhancing the reasoning abilities of large language models (LLMs). A core challenge in RLVR involves managing the exchange between entropy and…