Related papers: Adapter-Augmented Bandits for Online Multi-Constra…
Developing compound Large Language Model (LLM) applications is becoming an increasingly prevalent approach to solving real-world problems. In these applications, an LLM collaborates with various external modules, including APIs and even…
Large language models (LLMs) are typically governed by post-training alignment (e.g., RLHF or DPO), which yields a largely static policy during deployment and inference. However, real-world safety is a full-lifecycle problem: static…
Contemporary generative recommendation systems face significant challenges in handling multimodal data, eliminating algorithmic biases, and providing transparent decision-making processes. This paper introduces an enhanced generative…
We study offline scheduling for large language model (LLM) serving under a fixed KV-cache memory budget, where requests have heterogeneous prompt (prefill) and response (decode) lengths. Prompt tokens determine initial KV usage, and each…
Large language models (LLMs) exhibit complementary strengths across domains and come with varying inference costs, motivating the design of multi-agent LLM systems where specialized models collaborate efficiently. Existing approaches…
In this study, we delve into the Thresholding Linear Bandit (TLB) problem, a nuanced domain within stochastic Multi-Armed Bandit (MAB) problems, focusing on maximizing decision accuracy against a linearly defined threshold under resource…
Multimodal Large Language Models (MLLMs) have been rapidly advancing, enabling cross-modal understanding and generation, and propelling artificial intelligence towards artificial general intelligence. However, existing MLLM inference…
Multimodal large language models (MLLMs) promise enhanced reasoning by integrating diverse inputs such as text, vision, and audio. Yet cross-modal reasoning remains underexplored, with conflicting reports on whether added modalities help or…
Robust deployment of large multimodal models (LMMs) in real-world scenarios requires access to external knowledge sources, given the complexity and dynamic nature of real-world information. Existing approaches such as retrieval-augmented…
Extracting actionable intelligence from distributed, heterogeneous, correlated and high-dimensional data sources requires run-time processing and learning both locally and globally. In the last decade, a large number of meta-learning…
The task adaptation and alignment of Large Multimodal Models (LMMs) have been significantly advanced by instruction tuning and further strengthened by recent preference optimization. Yet, most LMMs still suffer from severe modality…
Large Language Models (LLMs) have achieved remarkable success across a wide range of natural language tasks, and recent efforts have sought to extend their capabilities to multimodal domains and resource-constrained environments. However,…
Aligning Large Language Models (LLM) to address subjectivity and nuanced preference levels requires adequate flexibility and control, which can be a resource-intensive and time-consuming procedure. Existing training-time alignment methods…
LLM search agents increasingly rely on tools at inference time, but their trajectories are often constrained by hard limits on both tool calls and generated tokens. Under such dual budgets, better answers require not only stronger models,…
While large language models (LLMs) are still being adopted to new domains and utilized in novel applications, we are experiencing an influx of the new generation of foundation models, namely multi-modal large language models (MLLMs). These…
LLM-based agents show strong potential for long-horizon reasoning, yet their context size is limited by deployment factors (e.g., memory, latency, and cost), yielding a constrained context budget. As interaction histories grow, this induces…
We consider the Scale-Free Adversarial Multi Armed Bandits(MAB) problem. At the beginning of the game, the player only knows the number of arms $n$. It does not know the scale and magnitude of the losses chosen by the adversary or the…
We consider the classic online learning and stochastic multi-armed bandit (MAB) problems, when at each step, the online policy can probe and find out which of a small number ($k$) of choices has better reward (or loss) before making its…
Large Language Models (LLMs) exhibit potential artificial generic intelligence recently, however, their usage is costly with high response latency. Given mixed LLMs with their own strengths and weaknesses, LLM routing aims to identify the…
Multimodal large language models (MLLMs) have extended the success of large language models (LLMs) to multiple data types, such as image, text and audio, achieving significant performance in various domains, including multimodal…