Related papers: WaterSIC: information-theoretically (near) optimal…

How can we accurately quantize a pre-trained Vision Transformer model? Quantization algorithms compress Vision Transformers (ViTs) into low-bit formats, reducing memory and computation demands with minimal accuracy degradation. However,…

Model quantization has become essential for efficient large language model deployment, yet existing approaches involve clear trade-offs: methods such as GPTQ and AWQ achieve practical compression but are lossy, while lossless techniques…

Large Language Models (LLMs) have experienced rapid advancements, with applications spanning a wide range of fields, including sentiment classification, review generation, and question answering. Due to their efficiency and versatility,…

We demonstrate that while the current approaches for language model watermarking are effective for open-ended generation, they are inadequate at watermarking LM outputs for constrained generation tasks with low-entropy output spaces.…

Quantization is the key method for reducing inference latency, power and memory footprint of generative AI models. However, accuracy often degrades sharply when activations are quantized below eight bits. Recent work suggests that…

Since numbers in the computer are represented with a fixed number of bits, loss of accuracy during calculation is unavoidable. At high precision where more bits (e.g. 64) are allocated to each number, round-off errors are typically small.…

Despite their remarkable achievements, modern Large Language Models (LLMs) face exorbitant computational and memory footprints. Recently, several works have shown significant success in training-free and data-free compression (pruning and…

The deployment of Large Language Models (LLMs) on edge devices is increasingly important to enhance on-device intelligence. Weight quantization is crucial for reducing the memory footprint of LLMs on devices. However, low-bit LLMs…

The paper is focused on the tradeoff between performance and decoding complexity per iteration for LDPC codes in terms of their gap (in rate) to capacity. The study of this tradeoff is done via information-theoretic bounds which also enable…

As machine learning inferences increasingly move to edge devices, adapting to diverse computational capabilities, hardware, and memory constraints becomes more critical. Instead of relying on a pre-trained model fixed for all future…

Reasoning models excel at complex tasks such as coding and mathematics, yet their inference is often slow and token-inefficient. To improve the inference efficiency, post-training quantization (PTQ) usually comes with the cost of large…

The Wasserstein metric has become increasingly important in many machine learning applications such as generative modeling, image retrieval and domain adaptation. Despite its appeal, it is often too costly to compute. This has motivated…

Deployment of Large Language Models (LLMs) has major computational costs, due to their rapidly expanding size. Compression of LLMs reduces the memory footprint, latency, and energy required for their inference. Post-training Quantization…

Transformers, the standard implementation for large language models (LLMs), typically consist of tens to hundreds of discrete layers. While more layers can lead to better performance, this approach has been challenged as far from efficient,…

In this paper, we consider the information-theoretic characterization of the set of achievable rates and distortions in a broad class of multiterminal communication scenarios with general continuous-valued sources and channels. A framework…

Masking-based post-hoc explanation methods, such as KernelSHAP and LIME, estimate local feature importance by querying a black-box model under randomized perturbations. This paper formulates this procedure as communication over a query…

In this paper, the sum capacity of the Gaussian Multiple Input Multiple Output (MIMO) Cognitive Radio Channel (MCC) is expressed as a convex problem with finite number of linear constraints, allowing for polynomial time interior point…

Matrix completion is a fundamental problem that comes up in a variety of applications like the Netflix problem, collaborative filtering, computer vision, and crowdsourcing. The goal of the problem is to recover a k-by-n unknown matrix from…

Watermark algorithms for large language models (LLMs) have achieved extremely high accuracy in detecting text generated by LLMs. Such algorithms typically involve adding extra watermark logits to the LLM's logits at each generation step.…

Deep convolutional neural networks have made outstanding contributions in many fields such as computer vision in the past few years and many researchers published well-trained network for downloading. But recent studies have shown serious…