Large Language Models (LLMs) often excel in specific domains but fall short in others due to the limitations of their training. Thus, enabling LLMs to solve problems collaboratively by integrating their complementary knowledge promises to improve their performance across domains. To realize this potential, we introduce a novel Collaborative Speculative Decoding (CoSD) algorithm that enables efficient LLM knowledge fusion at test time without requiring additional model training. CoSD employs a draft model to generate initial sequences and an easy-to-learn rule or decision tree to decide when to invoke an assistant model to improve these drafts. CoSD not only enhances knowledge fusion but also improves inference efficiency, is transferable across domains, and offers greater explainability. Experimental results demonstrate that CoSD improves accuracy by up to 10% across benchmarks compared to existing methods, providing a scalable and effective solution for LLM-based applications. Our code has been released at https://github.com/ATP-1010/CoSD.

Aligning Large Language Models to integrate and reflect human values, especially for tasks that demand intricate human oversight, is arduous since it is resource-intensive and time-consuming to depend on human expertise for context-specific guidance. Prior work has utilized predefined sets of rules or principles to steer the behavior of models (Bai et al., 2022; Sun et al., 2023). However, these principles tend to be generic, making it challenging to adapt them to each individual input query or context. In this work, we present Situated-PRInciples (SPRI), a framework requiring minimal or no human effort that is designed to automatically generate guiding principles in real-time for each input query and utilize them to align each response. We evaluate SPRI on three tasks, and show that 1) SPRI can derive principles in a complex domain-specific task that leads to on-par performance as expert-crafted ones; 2) SPRI-generated principles lead to instance-specific rubrics that outperform prior LLM-as-a-judge frameworks; 3) using SPRI to generate synthetic SFT data leads to substantial improvement on truthfulness.

The recent advances in Graph Neural Networks (GNN) are poised to improve machine learning of IoT systems at the edge. Particularly, GNNs allow modeling the topology of distributed systems, including their physical laws, from sensors data. However, one of the main limitations of using GNNs arises from their adjacency matrix. The adjacency matrix of GNNs needs to be defined a priori and represents the connectivity between the edges of a network. Usually, the adjacency matrix of GNNs consists of binary values that are equal to 1 when two edges are physically connected and 0 otherwise. This representation considers connectivity in terms of proximity and assumes that they are of equal significance. However, in certain applications, areas that are not physically connected can share more properties than physically connected areas. This necessitates new methods for devising the adjacency matrix and leads us to propose an efficient approach for determining the adjacency matrix of GNNs. Our approach extends GNNs in two ways. First, we employ a mechanism that utilizes the time series data at the edges to determine the eigenvalues and eigenvectors associated with each edge, allowing us to compute the proportion of variance. Subsequently, we use the proportion of variance to construct our adjacency matrix. Second, we utilize Dynamic Time Warping (DTW) to cluster related time series at the edge and construct our adjacency matrix. We then integrate the newly derived adjacency matrix into the GNN operating with a sequence to sequence learner to infer insights at the edges. Through extensive experiments, we demonstrate the strength and performance of our proposed GNN approach.

Agriculture faces unprecedented challenges due to climate change, population growth, and water scarcity. These challenges highlight the need for efficient resource usage to optimize crop production. Conventional techniques for forecasting hydrological response features, such as soil moisture, rely on physics-based and empirical hydrological models, which necessitate significant time and domain expertise. Drawing inspiration from traditional hydrological modeling, a novel temporal graph convolution neural network has been constructed. This involves grouping units based on their time-varying hydrological properties, constructing graph topologies for each cluster based on similarity using dynamic time warping, and utilizing graph convolutions and a gated recurrent neural network to forecast soil moisture. The method has been trained, validated, and tested on field-scale time series data spanning 40 years in northeastern United States. Results show that using domain-inspired clustering with time series graph neural networks is more effective in forecasting soil moisture than existing models. This framework is being deployed as part of a pro bono social impact program that leverages hybrid cloud and AI technologies to enhance and scale non-profit and government organizations. The trained models are currently being deployed on a series of small-holding farms in central Texas.