Related papers: The Fourth State: Signed-Zero Ternary for Stable L…
Previous works in zero-shot text-to-speech (ZS-TTS) have attempted to enhance its systems by enlarging the training data through crowd-sourcing or augmenting existing speech data. However, the use of low-quality data has led to a decline in…
Simulation models of complex dynamics in the natural and social sciences commonly lack a tractable likelihood function, rendering traditional likelihood-based statistical inference impossible. Recent advances in machine learning have…
We show that the method of distributed noise-shaping beta-quantization offers superior performance for the problem of spectral super-resolution with quantization whenever there is redundancy in the number of measurements. More precisely, if…
Quantum State Tomography (QST) is a fundamental technique in Quantum Information Processing (QIP) for reconstructing unknown quantum states. However, the conventional QST methods are limited by the number of measurements required, which…
We propose Multiplier-less INTeger (MINT) quantization, a uniform quantization scheme that efficiently compresses weights and membrane potentials in spiking neural networks (SNNs). Unlike previous SNN quantization methods, MINT quantizes…
We present a scheme for quantum randomness certification based on quantum steering. The protocol is one-sided device independent, providing high security, but requires only states and measurements that are simple to realise on quantum…
Latent reasoning offers a computation-efficient alternative to Chain-of-Thought but often suffers from performance degradation due to distributional misalignment and ambiguous chain definitions. Ideally, latent reasoning should function as…
We give a new characterization of $\mathsf{NL}$ as the class of languages whose members have certificates that can be verified with small error in polynomial time by finite state machines that use a constant number of random bits, as…
We consider coverless steganography where a Large Language Model (LLM) is used to generate stego-texts in combination with arithmetic coding. An efficient method should embed secret bits in as few language tokens as possible while keeping…
If a measurement is made on one half of a bipartite system, then, conditioned on the outcome, the other half has a new reduced state. If these reduced states defy classical explanation -- that is, if shared randomness cannot produce these…
State-space models (SSMs) have recently gained attention in deep learning for their ability to efficiently model long-range dependencies, making them promising candidates for edge-AI applications. In this paper, we analyze the effects of…
There is currently no unified metric for evaluating the efficiency of quantized neural networks. We propose QuIDE, built around the Intelligence Index I = (C x P)/log_2(T+1), which collapses the compression-accuracy-latency trade-off into a…
We introduce a data-free quantization method for deep neural networks that does not require fine-tuning or hyperparameter selection. It achieves near-original model performance on common computer vision architectures and tasks. 8-bit…
Model quantization reduces the bit-width of weights and activations, improving memory efficiency and inference speed in diffusion models. However, achieving 4-bit quantization remains challenging. Existing methods, primarily based on…
This paper proposes a novel binarized weight network (BT) for a resource-efficient neural structure. The proposed model estimates a binary representation of weights by taking into account the approximation error with an additional term.…
Large Language Models (LLMs) deliver strong performance but are difficult to deploy under tight memory and compute constraints. Low-bit post-training quantization (PTQ) is a promising direction; however, it typically relies on calibration…
Reinforcement Learning (RL) has significantly improved large language model reasoning, but existing RL fine-tuning methods rely heavily on heuristic techniques such as entropy regularization and reweighting to maintain stability. In…
State-of-the-art language models (LMs) represented by long-short term memory recurrent neural networks (LSTM-RNNs) and Transformers are becoming increasingly complex and expensive for practical applications. Low-bit neural network…
A novel scheme is presented for fault-tolerant quantum computation based on the cluster model. Some relevant logical cluster states are constructed in concatenation by post-selection through verification, without necessity of recovery…
Neural network quantization aims to reduce the bit-widths of weights and activations, making it a critical technique for deploying deep neural networks on resource-constrained hardware. Most Quantization-Aware Training (QAT) methods rely on…