Related papers: Unlocking the Potential of Continual Model Merging…

Open-vocabulary mobile manipulation (OVMM) requires robots to follow language instructions, navigate, and manipulate while updating their world representation under dynamic environmental changes. However, most prior approaches update their…

Omni-modal language models (OLMs) aim to integrate and reason over diverse input modalities--such as text, images, video, and audio--while maintaining strong language capabilities. Despite recent advancements, existing models, especially…

Despite the remarkable capabilities of Language Models (LMs) across diverse tasks, no single model consistently outperforms others, necessitating efficient methods to combine their strengths without expensive retraining. Existing model…

Online continual learning (OCL) seeks to learn new tasks from data streams that appear only once, while retaining knowledge of previously learned tasks. Most existing methods rely on replay, focusing on enhancing memory retention through…

Model Driven Engineering (MDE) is an emerging approach of software engineering. MDE emphasizes the construction of models from which the implementation should be derived by applying model transformations. The Ontology Definition Meta-model…

In the era of large language models, model merging is a promising way to combine multiple task-specific models into a single multitask model without extra training. However, two challenges remain: (a) interference between different models…

Model merging aims to integrate the strengths of multiple fine-tuned models into a unified model while preserving task-specific capabilities. Existing methods, represented by task arithmetic, are typically classified into global- and…

Multimodal Large Language Models (MLLMs) achieve strong performance through instruction tuning, but real-world deployment requires them to continually expand their capabilities, making Multimodal Continual Instruction Tuning (MCIT)…

Real-world natural language processing (NLP) models need to be continually updated to fix the prediction errors in out-of-distribution (OOD) data streams while overcoming catastrophic forgetting. However, existing continual learning (CL)…

Machine learning models often need to adapt to new data after deployment due to structured or unstructured real-world dynamics. The Continual Learning (CL) framework enables continuous model adaptation, but most existing approaches either…

Recent progress in large-scale flow and diffusion models raised two fundamental algorithmic challenges: (i) control-based reward adaptation of pre-trained flows, and (ii) integration of multiple models, i.e., flow merging. While current…

Online Class-Incremental continual Learning (OCIL) addresses the challenge of continuously learning from a single-channel data stream, adapting to new tasks while mitigating catastrophic forgetting. Recently, Mutual Information (MI)-based…

Channel knowledge maps (CKMs) provide a site-specific, location-indexed knowledge base that supports environment-aware communications and sensing in 6G networks. In practical deployments, CKM observations are often noisy and irregular due…

Millimeter wave (mmWave)-based orthogonal frequency-division multiplexing (OFDM) stands out as a suitable alternative for high-resolution sensing and high-speed data transmission. To meet communication and sensing requirements, many works…

This paper discusses the limitations of existing microscopic traffic models in accounting for the potential impacts of on-ramp vehicles on the car-following behavior of main-lane vehicles on highways. We first surveyed U.S. on-ramps to…

Large-scale deep learning models with a pretraining-finetuning paradigm have led to a surge of numerous task-specific models fine-tuned from a common pre-trained model. Recently, several research efforts have been made on merging these…

We present a hybrid transformer architecture that replaces discrete middle layers with a continuous-depth Neural Ordinary Differential Equation (ODE) block, enabling inference-time control over generation attributes via a learned steering…

Unified Multimodal Generative Models (UMGMs) unify visual understanding and image generation within a single autoregressive framework. However, their ability to continually learn new tasks is severely hindered by catastrophic forgetting,…

Controlled generation with pre-trained Diffusion and Flow Matching models has vast applications. One strategy for guiding ODE-based generative models is through optimizing a target loss $R(x_1)$ while staying close to the prior…

Model merging has gained increasing attention due to its intriguing property: interpolating the parameters of different task-specific fine-tuned models leads to multi-task abilities. However, despite its empirical success, the underlying…