Stacking Your Transformers: A Closer Look at Model Growth for Efficient LLM Pre-Training

Neural Information Processing Systems

LLMs are computationally expensive to pre-train due to their large scale. Model growth emerges as a promising approach by leveraging smaller models to accelerate the training of larger ones. However, the viability of these model growth methods in efficient LLM pre-training remains underexplored. This work identifies three critical Obstacles: (O1) lack of comprehensive evaluation, (O2) untested viability for scaling, and (O3) lack of empirical guidelines. To tackle O1, we summarize existing approaches into four atomic growth operators and systematically evaluate them in a standardized LLM pre-training setting.


Look Beneath the Surface: Exploiting Fundamental Symmetry for Sample-Efficient Offline RL

Neural Information Processing Systems

Offline reinforcement learning (RL) offers an appealing approach to real-world tasks by learning policies from pre-collected datasets without interacting with the environment. However, the performance of existing offline RL algorithms heavily depends on the scale and state-action space coverage of datasets. Real-world data collection is often expensive and uncontrollable, leading to small and narrowly covered datasets and posing significant challenges for practical deployments of offline RL. In this paper, we provide a new insight that leveraging the fundamental symmetry of system dynamics can substantially enhance offline RL performance under small datasets. Specifically, we propose a Time-reversal symmetry (T-symmetry) enforced Dynamics Model (TDM), which establishes consistency between a pair of forward and reverse latent dynamics. TDM provides both well-behaved representations for small datasets and a new reliability measure for OOD samples based on compliance with the T-symmetry. These can be readily used to construct a new offline RL algorithm (TSRL) with less conservative policy constraints and a reliable latent space data augmentation procedure. Based on extensive experiments, we find TSRL achieves great performance on small benchmark datasets with as few as 1% of the original samples, which significantly outperforms the recent offline RL algorithms in terms of data efficiency and generalizability. Code is available at: https://github.com/pcheng2/TSRL


Supplemental Materials

Neural Information Processing Systems

We bear all responsibility in case of violation of rights, etc., and confirmation of the data license. This project is licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). This license permits sharing and adapting the work provided it is not used for commercial purposes and appropriate credit is given. Please refer to Section 3 for our hosting plan. Hosting our datasets and code in our institutional repository provides sufficient accessibility, persistence, and traceability, which sufficiently addresses the needs that a DOI typically serves.


CRAG - Comprehensive RAG Benchmark Xiao Yang

Neural Information Processing Systems

Retrieval-Augmented Generation (RAG) has recently emerged as a promising solution to alleviate Large Language Model (LLM)'s deficiency in lack of knowledge. Existing RAG datasets, however, do not adequately represent the diverse and dynamic nature of real-world Question Answering (QA) tasks. To bridge this gap, we introduce the Comprehensive RAG Benchmark (CRAG), a factual question answering benchmark of 4,409 question-answer pairs and mock APIs to simulate web and Knowledge Graph (KG) search. CRAG is designed to encapsulate a diverse array of questions across five domains and eight question categories, reflecting varied entity popularity from popular to long-tail, and temporal dynamisms ranging from years to seconds. Our evaluation of this benchmark highlights the gap to fully trustworthy QA.


EC LipsE: Efficient Compositional Lipschitz Constant Estimation for Deep Neural Networks

Neural Information Processing Systems

The Lipschitz constant plays a crucial role in certifying the robustness of neural networks to input perturbations. Since calculating the exact Lipschitz constant is NP-hard, efforts have been made to obtain tight upper bounds on the Lipschitz constant. Typically, this involves solving a large matrix verification problem, the computational cost of which grows significantly for both deeper and wider networks. In this paper, we provide a compositional approach to estimate Lipschitz constants for deep feed-forward neural networks. We first obtain an exact decomposition of the large matrix verification problem into smaller sub-problems. Then, leveraging the underlying cascade structure of the network, we develop two algorithms. The first algorithm explores the geometric features of the problem and enables us to provide Lipschitz estimates that are comparable to existing methods by solving small semidefinite programs (SDPs) that are only as large as the size of each layer. The second algorithm relaxes these sub-problems and provides a closed-form solution to each sub-problem for extremely fast estimation, altogether eliminating the need to solve SDPs. The two algorithms represent different levels of trade-offs between efficiency and accuracy.



Spectral Graph Pruning Against Over-Squashing and Over-Smoothing

Neural Information Processing Systems

Message Passing Graph Neural Networks are known to suffer from two problems that are sometimes believed to be diametrically opposed: over-squashing and over-smoothing. The former results from topological bottlenecks that hamper the information flow from distant nodes and are mitigated by spectral gap maximization, primarily, by means of edge additions. However, such additions often promote oversmoothing that renders nodes of different classes less distinguishable. Inspired by the Braess phenomenon, we argue that deleting edges can address over-squashing and over-smoothing simultaneously. This insight explains how edge deletions can improve generalization, thus connecting spectral gap optimization to a seemingly disconnected objective of reducing computational resources by pruning graphs for lottery tickets. To this end, we propose a computationally effective spectral gap optimization framework to add or delete edges and demonstrate its effectiveness on the long range graph benchmark and on larger heterophilous datasets.


Exploring Context Window of Large Language Models via Decomposed Positional Vectors

Neural Information Processing Systems

Transformer-based large language models (LLMs) typically have a limited context window, resulting in significant performance degradation when processing text beyond the length of the context window. Extensive studies have been proposed to extend the context window and achieve length extrapolation of LLMs, but there is still a lack of in-depth interpretation of these approaches. In this study, we explore the positional information within and beyond the context window for deciphering the underlying mechanism of LLMs. By using a mean-based decomposition method, we disentangle positional vectors from hidden states of LLMs and analyze their formation and effect on attention. Furthermore, when texts exceed the context window, we analyze the change of positional vectors in two settings, i.e., direct extrapolation and context window extension. Based on our findings, we design two training-free context window extension methods, positional vector replacement and attention window extension. Experimental results show that our methods can effectively extend the context window length.


Exactly Minimax-Optimal Locally Differentially Private Sampling

Neural Information Processing Systems

The sampling problem under local differential privacy has recently been studied with potential applications to generative models, but a fundamental analysis of its privacy-utility trade-off (PUT) remains incomplete. In this work, we define the fundamental PUT of private sampling in the minimax sense, using the f-divergence between original and sampling distributions as the utility measure. We characterize the exact PUT for both finite and continuous data spaces under some mild conditions on the data distributions, and propose sampling mechanisms that are universally optimal for all f-divergences. Our numerical experiments demonstrate the superiority of our mechanisms over baselines, in terms of theoretical utilities for finite data space and of empirical utilities for continuous data space.


CosAE: Learnable Fourier Series for Image Restoration

Neural Information Processing Systems

In this paper, we introduce Cosine Autoencoder (CosAE), a novel, generic Autoencoder that seamlessly leverages the classic Fourier series with a feed-forward neural network. CosAE represents an input image as a series of 2D Cosine time series, each defined by a tuple of learnable frequency and Fourier coefficients. This method stands in contrast to a conventional Autoencoder that often sacrifices detail in their reduced-resolution bottleneck latent spaces.