Related papers: Free Energy Mixer

As an effective tool for eliciting the power of Large Language Models (LLMs), prompting has recently demonstrated unprecedented abilities across a variety of complex tasks. To further improve the performance, prompt ensemble has attracted…

Federated learning (FL) enables collaborative model training across distributed devices while preserving data privacy. However, balancing energy efficiency and fair participation while ensuring high model accuracy remains challenging in…

Our work combines aspects of three promising paradigms in machine learning, namely, attention mechanism, energy-based models, and associative memory. Attention is the power-house driving modern deep learning successes, but it lacks clear…

The aggregated unfitted finite element method (AgFEM) is a methodology recently introduced in order to address conditioning and stability problems associated with embedded, unfitted, or extended finite element methods. The method is based…

Efficient attention mechanisms enable long-context transformers but often miss globally important tokens, degrading modeling quality. We introduce a pre-scoring framework that assigns a query-independent global importance prior to keys…

To support long-term interaction in complex environments, LLM agents require memory systems that manage historical experiences. Existing approaches either retain full interaction histories via passive context extension, leading to…

Linearizing pretrained large language models (LLMs) primarily relies on intra-layer hybrid attention mechanisms to alleviate the quadratic complexity of standard softmax attention. Existing methods perform token routing based on…

We introduce the attention-indexed model (AIM), a theoretical framework for analyzing learning in deep attention layers. Inspired by multi-index models, AIM captures how token-level outputs emerge from layered bilinear interactions over…

Improved EM strategies, based on the idea of efficient data augmentation (Meng and van Dyk 1997, 1998), are presented for ML estimation of mixture proportions. The resulting algorithms inherit the simplicity, ease of implementation, and…

Deploying Large Language Models (LLMs) in streaming applications such as multi-round dialogue, where long interactions are expected, is urgently needed but poses two major challenges. Firstly, during the decoding stage, caching previous…

A Content Addressable Memory (CAM) is a memory primarily designed for high speed search operation. Parallel search scheme forms the basis of CAM, thus power reduction is the challenge associated with a large amount of parallel active…

In this work, we explore joint energy-based model (EBM) training during the finetuning of pretrained text encoders (e.g., Roberta) for natural language understanding (NLU) tasks. Our experiments show that EBM training can help the model…

Standard attention scales quadratically with sequence length. Efficient attention methods reduce this O(n^2) cost, but when retrofitted into pretrained models, they often degrade perplexity, downstream accuracy, or both. We introduce Focus,…

While Transformers have revolutionized deep learning, their quadratic attention complexity hinders their ability to process infinitely long inputs. We propose Feedback Attention Memory (FAM), a novel Transformer architecture that leverages…

Entity Matching (EM), which aims to identify whether two entity records from two relational tables refer to the same real-world entity, is one of the fundamental problems in data management. Traditional EM assumes that two tables are…

While attention-based encoder-decoder (AED) models have been successfully extended to the online variants for streaming automatic speech recognition (ASR), such as monotonic chunkwise attention (MoChA), the models still have a large label…

There is increasing interest in the use of the LEArnable Front-end (LEAF) in a variety of speech processing systems. However, there is a dearth of analyses of what is actually learnt and the relative importance of training the different…

We show the formal equivalence of linearised self-attention mechanisms and fast weight controllers from the early '90s, where a ``slow" neural net learns by gradient descent to program the ``fast weights" of another net through sequences of…

Energy management systems (EMS) rely on (non)-intrusive load monitoring (N)ILM to monitor and manage appliances and help residents be more energy efficient and thus more frugal. The robustness as well as the transfer potential of the most…

This paper proposes a Learnable Multiplicative absolute position Embedding based Conformer (LMEC). It contains a kernelized linear attention (LA) module called LMLA to solve the time-consuming problem for long sequence speech recognition as…