Searcharxiv⌕ Search

arXiv subjects

Haodong Wen

Publications and source records attributed to Haodong Wen.

11 recordsLinked to original sources

Data Schedules and Learning Rate Schedules Should Be Designed Together

Mid-training (MT) is a training phase in LLM training pipelines that continues pretraining (PT) on smaller datasets of higher value. Taking the PT and MT phases as a whole, training usually starts with only PT data and then transitions completely to MT data. In this work, we view this complete transition from PT to MT data as a special case of a broader class of data mixing schedules that specify how the proportions of PT and MT tokens in each batch change over training. Under fixed total training compute and MT token budgets, we ask: Is this complete transition to MT data optimal for minimizing final MT loss? We show that the answer depends on the learning rate (LR) schedule: one data schedule may outperform another under one LR schedule, but changing only the LR schedule can reverse the outcome. Specifically, when the LR decays throughout continued training, introducing MT data earlier and mixing it with PT data lowers the final MT loss compared with placing all MT data at the end of training, but this result is reversed if we use constant LR. We therefore study how to jointly choose when to introduce MT data and when to decay the LR through ablation studies, and find that it is better to decay the LR only after MT data enters training. Finally, we develop a theoretical model that explains why the optimal data schedule for a fixed MT token budget depends on the LR schedule.

cs.LG↗

Does LLM Pre-Training Typically Occur at the Edge of Stability?

Quadratic approximations provide a common lens for studying neural network optimization, but recent evidence challenges their predictive validity. In full-batch gradient descent with learning rate (LR) $η$, Cohen et al. (2021) observed the Edge of Stability (EoS), where the largest Hessian eigenvalue concentrates near $2/η$, seemingly at odds with classical stability conditions predicted by quadratic approximations. In this work, we revisit the fidelity of quadratic approximations as a model of neural network training dynamics, with a particular focus on their failure modes in LLM training. We first highlight a distinct failure mode of the quadratic approximation, where persistent negative curvature causes the quadratic dynamics to diverge regardless of the LR, even though the real dynamics remain stable. We term this phenomenon the Edge of Convexity (EoC). After removing the effect of this LR-independent instability, we then propose a generalized definition of Edge of Stability (EoS) for stochastic training with adaptive optimizers, based on whether increasing the LR induces significant divergence in the quadratic dynamics while leaving the real dynamics largely unchanged. Across different LLM pretraining runs with various model sizes up to 787M, we find that (1) EoC is always observed across LLM pretraining, and (2) EoS is also prevalent but not universal: it disappears when the LR becomes sufficiently small (e.g., after decay) or when the batch size falls below a critical threshold that has a positive correlation with the critical batch size. Together, these findings characterize when and how quadratic approximations fail and provide a foundation for future work on understanding training dynamics.

cs.LG↗

Tight Transition Time Bounds for Separable Logistic Regression at the Edge of Stability

We study logistic regression on linearly separable data under gradient descent with a large constant stepsize $η$. Such dynamics may exhibit a characteristic Edge of Stability phenomenon, in which the loss initially oscillates before transitioning to a stable phase of monotone decrease. Existing work provides a tight $Θ(1)$ bound in dimension $d=2$ as $η\to \infty$ and conjectures a bound independent of $η$ in arbitrary dimensions $d\geq 2$. In this paper, we disprove this conjecture by showing that, for every fixed sample size $n\geq 2$ and sufficiently small margin $γ$, the worst-case transition time is $$Θ\!\left((\logη)^{\min\{n-2,d-2\}}\right)$$ uniformly over $d\geq2$. The key challenge in establishing a tight bound is that the sample contributing most strongly to the gradient can change repeatedly across iterations. To address this issue, we control such changes by induction on dimension and sample size, and construct matching hard instances.

cs.LG↗

DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression

The widespread adoption of long-horizon agents has made model workloads increasingly input-heavy. Although prior work has substantially reduced the cost of long-context computation, prefill remains computationally expensive, and large KV caches continue to strain HBM and SSD capacity and data-transfer bandwidth. Together, these compute, storage, and bandwidth demands constitute the primary bottleneck to further lowering deployment costs. To address this challenge, we introduce DeepSeek-V4.1-Flash, a multimodal Mixture-of-Experts (MoE) model with 552B backbone parameters and support for contexts of up to one million tokens. With its Causal Encoder-Decoder (CED) architecture, the model activates 16B parameters per token during decode but only 8B parameters during prefill, substantially improving cost efficiency for agentic workloads. To push the limits of KV cache compression, DeepSeek-V4.1-Flash combines cross-layer KV cache reuse in Compressed Sparse Attention 2 (CSA2) with FP4 KV caching. These designs reduce its global KV cache footprint (always in HBM) to 890 bytes per token, roughly 1/4 of the corresponding footprint of DeepSeek-V4-Flash. Further, through a dedicated deployment optimization known as SWA Bounded Replay, DeepSeek-V4.1-Flash reduces its persistent KV cache footprint (always on SSD or in host memory) to roughly 1/8 of that of DeepSeek-V4-Flash. Despite its much smaller KV cache footprint, the model delivers substantially better performance than the baseline. In addition, we streamline the DeepSeek-V4 architecture and introduce several efficient architectural extensions. We pretrain DeepSeek-V4.1-Flash on a multimodal corpus comprising 45T tokens and conduct comprehensive post-training, yielding strong performance across diverse text-based and multimodal agentic scenarios. Model checkpoints are available at https://huggingface.co/deepseek-ai/DeepSeek-V4.1-Flash.

cs.CL↗

Frontier Language Models Struggle to Copy: Text Can Be Better Viewed in 2D

While large language models (LLMs) can solve advanced reasoning problems in seconds, we show that even frontier models fail to perform a much simpler operation: exactly copying an input string that lies well within their context windows. We attribute this failure to positional encodings in Transformer architectures, whose inductive bias favors copying through a shortcut based on matching local contexts rather than carefully locating the corresponding input positions. To address this issue, we introduce 2D-RoPE, which organizes text into a 2D grid rather than a 1D sequence and assigns each token a row ID and a column ID. Under this view, copying becomes simply retrieving input tokens at a fixed column offset, which makes the task easy to learn. In synthetic copy experiments, shallow Transformers with 2D-RoPE achieve perfect copying at input lengths hundreds of times longer than those seen during training, whereas standard positional encodings fall far behind. We further show that the advantage of 2D-RoPE language models on copy tasks consistently holds in large-scale pretraining on DCLM with model sizes up to 1.4B parameters. Overall, our results suggest that viewing text in 2D can benefit language modeling, and we hope this encourages future work to further explore the potential of 2D positional encodings.

cs.CL↗

How Learning Rate Decay Wastes Your Best Data in Curriculum-Based LLM Pretraining

Due to the scarcity of high-quality data, large language models (LLMs) are often trained on mixtures of data with varying quality levels, even after sophisticated data curation. A natural approach to better leverage high-quality data is curriculum-based pretraining, where the model is trained on data sorted in ascending order of quality as determined by a quality metric. However, prior studies have reported limited improvements from such curriculum-based pretraining strategies. This work identifies a critical factor constraining these methods: the incompatibility between the ascending data quality order and the decaying learning rate (LR) schedule. We find that while curriculum-based training substantially outperforms random shuffling when using a constant LR, its advantage diminishes under standard LR decay schedules. Our experiments show this incompatibility can be mitigated by two simple strategies: (1) employing a more moderate LR decay schedule, where the final LR is only moderately smaller than the peak LR, and (2) replacing LR decay with model averaging, i.e., computing a weighted average of the final few checkpoints. By combining these strategies, we improve the average score on a suite of standard benchmarks by 1.64% over random shuffling, without additional data refinement. Validated on 1.5B-parameter models trained over 30B tokens with various data-quality metrics, our findings call for a re-evaluation of curriculum-based LLM pretraining and underscore the potential of co-designing data curricula with optimization methods.

cs.LG↗

Larger Datasets Can Be Repeated More: A Theoretical Analysis of Multi-Epoch Scaling in Linear Regression

While data scaling laws of large language models (LLMs) have been widely examined in the one-pass regime with massive corpora, their form under limited data and repeated epochs remains largely unexplored. This paper presents a theoretical analysis of how a common workaround, training for multiple epochs on the same dataset, reshapes the data scaling laws in linear regression. Concretely, we ask: to match the performance of training on a dataset of size $N$ for $K$ epochs, how much larger must a dataset be if the model is trained for only one pass? We quantify this using the \textit{effective reuse rate} of the data, $E(K, N)$, which we define as the multiplicative factor by which the dataset must grow under one-pass training to achieve the same test loss as $K$-epoch training. Our analysis precisely characterizes the scaling behavior of $E(K, N)$ for SGD in linear regression under either strong convexity or Zipf-distributed data: (1) When $K$ is small, we prove that $E(K, N) \approx K$, indicating that every new epoch yields a linear gain; (2) As $K$ increases, $E(K, N)$ plateaus at a problem-dependent value that grows with $N$ ($Θ(\log N)$ for the strongly-convex case), implying that larger datasets can be repeated more times before the marginal benefit vanishes. These theoretical findings point out a neglected factor in a recent empirical study (Muennighoff et al. (2023)), which claimed that training LLMs for up to $4$ epochs results in negligible loss differences compared to using fresh data at each step, \textit{i.e.}, $E(K, N) \approx K$ for $K \le 4$ in our notation. Supported by further empirical validation with LLMs, our results reveal that the maximum $K$ value for which $E(K, N) \approx K$ in fact depends on the data size and distribution, and underscore the need to explicitly model both factors in future studies of scaling laws with data reuse.

cs.LG↗

Adam Reduces a Unique Form of Sharpness: Theoretical Insights Near the Minimizer Manifold

Despite the popularity of the Adam optimizer in practice, most theoretical analyses study Stochastic Gradient Descent (SGD) as a proxy for Adam, and little is known about how the solutions found by Adam differ. In this paper, we show that Adam implicitly reduces a unique form of sharpness measure shaped by its adaptive updates, leading to qualitatively different solutions from SGD. More specifically, when the training loss is small, Adam wanders around the manifold of minimizers and takes semi-gradients to minimize this sharpness measure in an adaptive manner, a behavior we rigorously characterize through a continuous-time approximation using stochastic differential equations. We further demonstrate how this behavior differs from that of SGD in a well-studied setting: when training overparameterized models with label noise, SGD has been shown to minimize the trace of the Hessian matrix, $\tr(\mH)$, whereas we prove that Adam minimizes $\tr(\Diag(\mH)^{1/2})$ instead. In solving sparse linear regression with diagonal linear networks, this distinction enables Adam to achieve better sparsity and generalization than SGD. Finally, our analysis framework extends beyond Adam to a broad class of adaptive gradient methods, including RMSProp, Adam-mini, Adalayer and Shampoo, and provides a unified perspective on how these adaptive optimizers reduce sharpness, which we hope will offer insights for future optimizer design.

cs.LG↗

A Multi-Power Law for Loss Curve Prediction Across Learning Rate Schedules

Training large models is both resource-intensive and time-consuming, making it crucial to understand the quantitative relationship between model performance and hyperparameters. In this paper, we present an empirical law that describes how the pretraining loss of large language models evolves under different learning rate schedules, such as constant, cosine, and step decay schedules. Our proposed law takes a multi-power form, combining a power law based on the sum of learning rates and additional power laws to account for a loss reduction effect induced by learning rate decay. We extensively validate this law on various model sizes and architectures, and demonstrate that after fitting on a few learning rate schedules, the law accurately predicts the loss curves for unseen schedules of different shapes and horizons. Moreover, by minimizing the predicted final pretraining loss across learning rate schedules, we are able to find a schedule that outperforms the widely used cosine learning rate schedule. Interestingly, this automatically discovered schedule bears some resemblance to the recently proposed Warmup-Stable-Decay (WSD) schedule (Hu et al, 2024) but achieves a slightly lower final loss. We believe these results could offer valuable insights for understanding the dynamics of pretraining and designing learning rate schedules to improve efficiency.

cs.LG↗

Exploring the Robustness of In-Context Learning with Noisy Labels

Recently, the mysterious In-Context Learning (ICL) ability exhibited by Transformer architectures, especially in large language models (LLMs), has sparked significant research interest. However, the resilience of Transformers' in-context learning capabilities in the presence of noisy samples, prevalent in both training corpora and prompt demonstrations, remains underexplored. In this paper, inspired by prior research that studies ICL ability using simple function classes, we take a closer look at this problem by investigating the robustness of Transformers against noisy labels. Specifically, we first conduct a thorough evaluation and analysis of the robustness of Transformers against noisy labels during in-context learning and show that they exhibit notable resilience against diverse types of noise in demonstration labels. Furthermore, we delve deeper into this problem by exploring whether introducing noise into the training set, akin to a form of data augmentation, enhances such robustness during inference, and find that such noise can indeed improve the robustness of ICL. Overall, our fruitful analysis and findings provide a comprehensive understanding of the resilience of Transformer models against label noises during ICL and provide valuable insights into the research on Transformers in natural language processing. Our code is available at https://github.com/InezYu0928/in-context-learning.

cs.CL↗

Rethinking the Graph Polynomial Filter via Positive and Negative Coupling Analysis

Recently, the optimization of polynomial filters within Spectral Graph Neural Networks (GNNs) has emerged as a prominent research focus. Existing spectral GNNs mainly emphasize polynomial properties in filter design, introducing computational overhead and neglecting the integration of crucial graph structure information. We argue that incorporating graph information into basis construction can enhance understanding of polynomial basis, and further facilitate simplified polynomial filter design. Motivated by this, we first propose a Positive and Negative Coupling Analysis (PNCA) framework, where the concepts of positive and negative activation are defined and their respective and mixed effects are analysed. Then, we explore PNCA from the message propagation perspective, revealing the subtle information hidden in the activation process. Subsequently, PNCA is used to analyze the mainstream polynomial filters, and a novel simple basis that decouples the positive and negative activation and fully utilizes graph structure information is designed. Finally, a simple GNN (called GSCNet) is proposed based on the new basis. Experimental results on the benchmark datasets for node classification verify that our GSCNet obtains better or comparable results compared with existing state-of-the-art GNNs while demanding relatively less computational time.

cs.LG↗