Searcharxiv⌕ Search

arXiv subjects

Shane Bergsma

Publications and source records attributed to Shane Bergsma.

15 recordsLinked to original sources

DNeo: Your Parallel Draft Model Needs Better Target Feature Encoding

Speculative-diffusion decoding uses a small parallel drafter model to generate b draft tokens in one forward pass, which then undergo parallel verification via a larger target model. Faster inference can be achieved when multiple draft tokens pass verification simultaneously. Recent parallel draft architectures (DFlash, DSpark) improve draft quality by conditioning on rich features computed by the target, through a mechanism known as feature encoding. However, standard feature encoding mechanisms have three limitations: (1) they are lossy, applying encoding projections that are destructive to the underlying target feature-space via a discovered effect known as residual mixing, (2) they are memory-less, with no record of previously processed proposals, and (3) they utilize features sampled uniformly from across the target layers, omitting important hidden states. To address these issues, we propose Mixture-of-Features Attention (MoFA), a lightweight feature-encoding approach equipped with a working memory that encodes previously rejected proposals. Leveraging MoFA, rejection memory, and a set of further enhancements, including 2-tap causal convolutions, prefix-only cross-attention, and non-uniform target layer selection, the final architecture, DNeo, achieves 1.3x faster generation speed than SoTA (DFlash-2) without impacting drafter footprint or generation quality. DNeo marks a new state of the art in high quality, fast LLM inference.

cs.AI↗

Don't Drop Dropout: Optimizing Layer Sparsity for Efficient LLM Training and Inference

Layer dropout (a.k.a. stochastic depth) has been shown to enable faster training, higher accuracy, and robustness to zero-shot layer pruning in both language and vision transformers. However, as models and datasets have scaled, dropout - particularly layer dropout - has largely disappeared from large language models (LLMs) pre-training recipes. While some prior work has reported that dropout can degrade accuracy, no comprehensive study has quantified, let alone mitigated, this effect. In this study, we show that layer dropout should be used in state-of-the-art LLM training, establishing best practices and scaling analysis for both training and post-training benefits. Concretely, with optimal layer distribution, time schedule, and optimizer hyperparameters, we observe that at the same training FLOPs layer dropout leads to lower loss. For a given number of training steps, LLMs can achieve lower or similar validation loss while saving upto 25% of training FLOPs. Moreover, layer dropout enables significant post-training optimizations, such as early exit, intermediate-layer skipping, and self-speculative decoding, yielding up to 1.5x inference speedup with negligible accuracy loss. Across more than 2400 training experiments, spanning models from 271M to 8.2B parameters and datasets up to 160B tokens, we demonstrate that these findings extend reliably to large-scale training regimes. All pre-training experiments were run on Cerebras CS-3 systems.

cs.AI↗

Unlocking Lossless Speedups in LLMs via Discrete Diffusion

Large Language Models (LLMs) owe much of their success to next-token prediction (NTP), but their autoregressive (AR) structure requires slow, sequential token generation. To overcome this bottleneck, we introduce diffusion-augmented LLMs, a new class of models that defines an AR model distribution while using diffusion to draw multiple tokens in parallel from that distribution. We decouple the parameters of these models into two sets: AR weights, trained using the standard NTP objective, and lightweight diffusion weights, trained to generate multiple tokens simultaneously. The diffusion weights are learned through a simple Diffusion Distillation phase that adds negligible overhead to existing LLM training pipelines. We also introduce $Ψ$-Spec, a family of samplers that enables lossless acceleration and inference-time scaling at a fixed context length. Unlike speculative decoding, our method requires no separate draft model. Unlike diffusion LLMs (d-LLMs), it accelerates generation without sacrificing the quality of the underlying AR model. The resulting models, called Uno, can be trained from scratch or built by augmenting existing open-weight AR LLMs. Uno achieves higher throughput than leading speculative-decoding methods at every evaluated batch size and delivers up to $3\times$ speedups over the base AR model, including at the largest batch size supported by the device. Notably, our 8B Uno model outperforms the leading open d-LLM, the 26B DiffusionGemma, and the proprietary Mercury 2 across all evaluated benchmarks in agentic tool use, coding, and long-context reasoning. We release code and checkpoints at: https://s-sahoo.github.io/uno/

cs.LG↗

Scaling with Collapse: Efficient and Predictable Training of LLM Families

Effective LLM training depends on predictable scaling of key quantities -- such as final loss and optimal hyperparameters -- with model and dataset size. Qiu et al. (2025) recently showed that this predictability can extend beyond scalars: whole training loss curves can *collapse* onto a universal trajectory after a simple normalization. What remains unclear is whether this phenomenon persists for LLM families trained under *practical scaling recipes*, where width, depth, learning rate, batch size, and weight decay are scaled jointly. We show that it does: loss curves collapse across scales precisely when optimization hyperparameters are set optimally for the given data budget, in accordance with recent empirical scaling laws. Collapse therefore emerges as a signature of compute-efficient training. We demonstrate two applications at scale: (1) deviation-from-collapse provides a sensitive, early diagnostic of training pathologies, and (2) predictability of collapsed curves enables early stopping in large-scale hyperparameter tuning. Finally, we train a competitive LLM family, *Celerity*, using these insights, establishing collapse as an effective tool for developing efficient LLMs.

cs.LG↗

Predicting Training Re-evaluation Curves Enables Effective Data Curriculums for LLMs

Data curriculums have become central to successful LLM training, yet principles governing optimal data placement remain unclear. We introduce the *training re-evaluation curve (TREC)*, a diagnostic that retrospectively evaluates training batches *using the final model weights*. The TREC characterizes how well a trained model retains training data as a function of *when* the data was encountered during training. Analyzing TRECs for models from 111M to 3.9B parameters, we show that placing high-quality data at low points on the TREC significantly improves performance. Importantly, while a TREC is initially observable only after training, we demonstrate it can be *predicted in advance* from AdamW's implicit EMA coefficients, enabling proactive curriculum design. By predicting TRECs for published training recipes, we explain prior ablations and reveal suboptimal data placements. We also align high-quality data with TREC minima in order to improve continual pre-training of a 3.9B-parameter LLM trained on 900B tokens.

cs.LG↗

Sparse maximal update parameterization: A holistic approach to sparse training dynamics

Several challenges make it difficult for sparse neural networks to compete with dense models. First, setting a large fraction of weights to zero impairs forward and gradient signal propagation. Second, sparse studies often need to test multiple sparsity levels, while also introducing new hyperparameters (HPs), leading to prohibitive tuning costs. Indeed, the standard practice is to re-use the learning HPs originally crafted for dense models. Unfortunately, we show sparse and dense networks do not share the same optimal HPs. Without stable dynamics and effective training recipes, it is costly to test sparsity at scale, which is key to surpassing dense networks and making the business case for sparsity acceleration in hardware. A holistic approach is needed to tackle these challenges and we propose S$μ$Par as one such approach. For random unstructured static sparsity, S$μ$Par ensures activations, gradients, and weight updates all scale independently of sparsity level. Further, by reparameterizing the HPs, S$μ$Par enables the same HP values to be optimal as we vary both sparsity level and model width. HPs can be tuned on small dense networks and transferred to large sparse models, greatly reducing tuning costs. On large-scale language modeling, S$μ$Par shows increasing improvements over standard parameterization as sparsity increases, leading up to 11.9% relative loss improvement at 99.2% sparsity. A minimal implementation of S$μ$Par is available at https://github.com/EleutherAI/nanoGPT-mup/tree/supar.

cs.LG↗

Don't be lazy: CompleteP enables compute-efficient deep transformers

We study compute efficiency of LLM training when using different parameterizations, i.e., rules for adjusting model and optimizer hyperparameters (HPs) as model size changes. Some parameterizations fail to transfer optimal base HPs (such as learning rate) across changes in model depth, requiring practitioners to either re-tune these HPs as they scale up (expensive), or accept sub-optimal training when re-tuning is prohibitive. Even when they achieve HP transfer, we develop theory to show parameterizations may still exist in the lazy learning regime where layers learn only features close to their linearization, preventing effective use of depth and nonlinearity. Finally, we identify and adopt the parameterization we call CompleteP that achieves both depth-wise HP transfer and non-lazy learning in all layers. CompleteP enables a wider range of model width/depth ratios to remain compute-efficient, unlocking shapes better suited for different hardware settings and operational contexts. Moreover, CompleteP enables 12-34% compute efficiency improvements over the prior state-of-the-art. All experiments were run on Cerebras CS-3 systems. A minimal implementation is available at https://github.com/EleutherAI/nanoGPT-mup/tree/completep.

cs.LG↗

Straight to Zero: Why Linearly Decaying the Learning Rate to Zero Works Best for LLMs

LLMs are commonly trained with a learning rate (LR) warmup, followed by cosine decay to 10% of the maximum (10x decay). In a large-scale empirical study, we show that under an optimal peak LR, a simple linear decay-to-zero (D2Z) schedule consistently outperforms other schedules when training at compute-optimal dataset sizes. D2Z is superior across a range of model sizes, batch sizes, datasets, and vocabularies. Benefits increase as dataset size increases. Leveraging a novel interpretation of AdamW as an exponential moving average of weight updates, we show how linear D2Z optimally balances the demands of early training (moving away from initial conditions) and late training (averaging over more updates in order to mitigate gradient noise). In experiments, a 610M-parameter model trained for 80 tokens-per-parameter (TPP) using D2Z achieves lower loss than when trained for 200 TPP using 10x decay, corresponding to an astonishing 60% compute savings. Models such as Llama2-7B, trained for 286 TPP with 10x decay, could likely have saved a majority of compute by training with D2Z.

cs.LG↗

Power Lines: Scaling Laws for Weight Decay and Batch Size in LLM Pre-training

Efficient LLM pre-training requires well-tuned hyperparameters (HPs), including learning rate $η$ and weight decay $λ$. We study scaling laws for HPs: formulas for how to scale HPs as we scale model size N, dataset size D, and batch size B. Recent work suggests the AdamW timescale, $τ= B/(ηλD)$, should remain constant across training settings, and we verify the implication that optimal $λ$ scales linearly with B, for a fixed N and D. However, as N and D scale, we show optimal $τ$ obeys a precise power law in the tokens-per-parameter ratio, D/N. This law thus provides a method to accurately predict $λ$opt in advance of large-scale training. We also study scaling laws for optimal batch size Bopt (the B enabling lowest loss at a given N,D) and critical batch size Bcrit (the B beyond which further data parallelism becomes ineffective). In contrast to prior work, we find both Bopt and Bcrit scale as power laws in D, independent of model size, N. Finally, we analyze how these findings inform the real-world selection of Pareto-optimal N and D under dual training time and compute objectives. All experiments were run on Cerebras CS-3 systems.

cs.LG↗

PTPP-Aware Adaptation Scaling Laws: Predicting Domain-Adaptation Performance at Unseen Pre-Training Budgets

Continual pre-training (CPT) for domain adaptation must balance target-domain gains with stability on the base domain. Existing CPT scaling laws typically assume a fixed pre-training budget, which limits their ability to forecast adaptation outcomes for models trained at different tokens-per-parameter (PTPP). We present \emph{PTPP-aware} adaptation scaling laws that make the pre-training budget an explicit variable, enabling accurate \emph{prediction} of adaptation loss at unseen \ptpp. On a multilingual setup (English/Arabic $\rightarrow$ French), PTPP-aware formulations trained on early stages (\ptpp{}=\{15,31\}) predict target loss at \ptpp{}=279 and outperform a PTPP-agnostic \dcpt{} transfer baseline on metrics (Huber-on-log, MAE$_\mathrm{rel}$, calibration slope); full diagnostics (RMSE, MAPE) are in the appendix. Beyond forecasting, we show a practical use case: planning replay ratios and adaptation token budgets that satisfy target and forgetting constraints under compute limits.

cs.LG↗

Normalization Layer Per-Example Gradients are Sufficient to Predict Gradient Noise Scale in Transformers

Per-example gradient norms are a vital ingredient for estimating gradient noise scale (GNS) with minimal variance. Observing the tensor contractions required to compute them, we propose a method with minimal FLOPs in 3D or greater tensor regimes by simultaneously computing the norms while computing the parameter gradients. Using this method we are able to observe the GNS of different layers at higher accuracy than previously possible. We find that the total GNS of contemporary transformer models is predicted well by the GNS of only the normalization layers. As a result, focusing only on the normalization layer, we develop a custom kernel to compute the per-example gradient norms while performing the LayerNorm backward pass with zero throughput overhead. Tracking GNS on only those layers, we are able to guide a practical batch size schedule that reduces training time by 18% on a Chinchilla-optimal language model.

cs.LG↗

SutraNets: Sub-series Autoregressive Networks for Long-Sequence, Probabilistic Forecasting

We propose SutraNets, a novel method for neural probabilistic forecasting of long-sequence time series. SutraNets use an autoregressive generative model to factorize the likelihood of long sequences into products of conditional probabilities. When generating long sequences, most autoregressive approaches suffer from harmful error accumulation, as well as challenges in modeling long-distance dependencies. SutraNets treat long, univariate prediction as multivariate prediction over lower-frequency sub-series. Autoregression proceeds across time and across sub-series in order to ensure coherent multivariate (and, hence, high-frequency univariate) outputs. Since sub-series can be generated using fewer steps, SutraNets effectively reduce error accumulation and signal path distances. We find SutraNets to significantly improve forecasting accuracy over competitive alternatives on six real-world datasets, including when we vary the number of sub-series and scale up the depth and width of the underlying sequence models.

cs.LG↗

C2FAR: Coarse-to-Fine Autoregressive Networks for Precise Probabilistic Forecasting

We present coarse-to-fine autoregressive networks (C2FAR), a method for modeling the probability distribution of univariate, numeric random variables. C2FAR generates a hierarchical, coarse-to-fine discretization of a variable autoregressively; progressively finer intervals of support are generated from a sequence of binned distributions, where each distribution is conditioned on previously-generated coarser intervals. Unlike prior (flat) binned distributions, C2FAR can represent values with exponentially higher precision, for only a linear increase in complexity. We use C2FAR for probabilistic forecasting via a recurrent neural network, thus modeling time series autoregressively in both space and time. C2FAR is the first method to simultaneously handle discrete and continuous series of arbitrary scale and distribution shape. This flexibility enables a variety of time series use cases, including anomaly detection, interpolation, and compression. C2FAR achieves improvements over the state-of-the-art on several benchmark forecasting datasets.

cs.LG↗

Capacity Allocation for Clouds with Parallel Processing, Batch Arrivals, and Heterogeneous Service Requirements

Problem Definition: Allocating sufficient capacity to cloud services is a challenging task, especially when demand is time-varying, heterogeneous, contains batches, and requires multiple types of resources for processing. In this setting, providers decide whether to reserve portions of their capacity to individual job classes or to offer it in a flexible manner. Methodology/results: In collaboration with Huawei Cloud, a worldwide provider of cloud services, we propose a heuristic policy that allocates multiple types of resources to jobs and also satisfies their pre-specified service level agreements (SLAs). We model the system as a multi-class queueing network with parallel processing and multiple types of resources, where arrivals (i.e., virtual machines and containers) follow time-varying patterns and require at least one unit of each resource for processing. While virtual machines leave if they are not served immediately, containers can join a queue. We introduce a diffusion approximation of the offered load of such system and investigate its fidelity as compared to the observed data. Then, we develop a heuristic approach that leverages this approximation to determine capacity levels that satisfy probabilistic SLAs in the system with fully flexible servers. Managerial Implications: Using a data set of cloud computing requests over a representative 8-day period from Huawei Cloud, we show that our heuristic policy results in a 20% capacity reduction and better service quality as compared to a benchmark that reserves resources. In addition, we show that the system utilization induced by our policy is superior to the benchmark, i.e., it implies less idling of resources in most instances. Thus, our approach enables cloud operators to both reduce costs and achieve better performance.

stat.AP↗

Gridiron: A Technique for Augmenting Cloud Workloads with Network Bandwidth Requirements

Cloud applications use more than just server resources, they also require networking resources. We propose a new technique to model network bandwidth demand of networked cloud applications. Our technique, Gridiron, augments VM workload traces from Azure cloud with network bandwidth requirements. The key to the Gridiron technique is to derive inter-VM network bandwidth requirements using Amdahl's second law. As a case study, we use Gridiron to generate realistic traces with network bandwidth demands for a distributed machine learning training application. Workloads generated with Gridiron allow datacenter operators to estimate the network bandwidth demands of cloud applications and enable more realistic cloud resource scheduler evaluation.

cs.DC↗