21 accepted papers
Converting pretrained attention modules such as *grouped-query attention* (GQA) into *multi-head latent attention* (MLA) can improve expressivity without increasing KV-cache cost, making it attractive for efficient inference. However, existing conversion methods typically apply naïve singular value…
An increasing number of LLMs employ Mixture-of-Experts (MoE) architectures where the feed-forward layer is replaced by a pool of experts and each token only activates a small subset of them. During autoregressive generation, these models often enter a memory-bound regime even for moderate batch size…
Test-time scaling for complex reasoning tasks shows that leveraging inference-time compute, for example by independently sampling and aggregating multiple solutions, results in significantly better task outcomes. However, a critical bottleneck is _verification_: sampling is only effective if correct…
We investigate the challenge of applying Large Language Models (LLMs) to long texts. We propose a theoretical framework that distinguishes the failure modes of long context tasks into three categories: cross-chunk dependence (task noise), confusion that grows with context size (model noise), and the…
Speculative decoding can significantly accelerate LLM serving, but its real-world benefits often erode due to training–serving mismatch and non-stationary traffic. Unlike previous systems that decouple speculator training from inference, we present a unified training–serving system, Aurora, that clo…
Building helpful and harmless large language models (LLMs) requires effective model alignment approach based on human instructions and feedback, which necessitates high-quality human-labeled data. Constructing such datasets is often expensive and hard to scale, and may face potential limitations on…
Large language model inference is both memory-intensive and time-consuming, often requiring distributed algorithms to efficiently scale. Various model parallelism strategies are used in multi-gpu training and inference to partition computation across multiple devices, reducing memory load and comput…
Recent advances in large language models (LLMs) demonstrate substantial capabilities in natural language understanding and generation tasks. With the growing number of LLMs, how to harness the collective expertise of multiple LLMs is an exciting open direction. Toward this goal, we propose a new app…
Large Language Models (LLMs) struggle with long-context reasoning, not only due to the quadratic scaling of computational complexity with sequence length but also because of the scarcity and expense of annotating long-context data. There has been barely any open-source work that systematically ablat…
Activation sparsity can enable practical inference speedups in large language models (LLMs) by reducing the compute and memory-movement required for matrix multiplications during the forward pass. However, existing methods face limitations that inhibit widespread adoption. Some approaches are tail…
Verifiers can improve language model (LM) capabilities by providing feedback or selecting the best response from a pool of generated candidates. Currently, high-quality verifiers are either unscalable (e.g., humans) or limited in utility (e.g., tools like Lean for formal proofs). While LM judges and…
In our study, we present bifurcated attention, a method developed for language model inference in single-context batch sampling contexts. This approach aims to reduce redundant memory IO costs, a significant factor in latency for high batch sizes and long context lengths. Bifurcated attention achiev…
A diverse array of reasoning strategies has been proposed to elicit the capabilities of large language models. However, in this paper, we point out that traditional evaluations which focus solely on performance metrics miss a key factor: the increased effectiveness due to additional compute. By over…
Large language models are increasingly becoming a cornerstone technology in artificial intelligence, the sciences, and society as a whole, yet the optimal strategies for dataset composition and filtering remain largely elusive. Many of the top-performing models lack transparency in their dataset cur…
Generative models, widely utilized in various applications, can often struggle with prompts corresponding to partial tokens. This struggle stems from tokenization, where partial tokens fall out of distribution during inference, leading to incorrect or nonsensical outputs. This paper examines a techn…
We present two new benchmarks, MBXP and Multilingual HumanEval, designed to evaluate code completion models in over 10 programming languages. These datasets are generated using a conversion framework that transpiles prompts and test cases from the original MBPP and HumanEval datasets into the corres…
Compositional reasoning tasks such as multi-hop question answering require models to learn how to make latent decisions using only weak supervision from the final answer. Crowdsourced datasets gathered for these tasks, however, often contain only a slice of the underlying task distribution, which ca…
We propose a new framework, Translation between Augmented Natural Languages (TANL), to solve many structured prediction language tasks including joint entity and relation extraction, nested named entity recognition, relation classification, semantic role labeling, event extraction, coreference resol…
Presently the most successful approaches to semi-supervised learning are based on consistency regularization, whereby a model is trained to be robust to small perturbations of its inputs and parameters. To understand consistency regularization, we conceptually explore how loss geometry interacts wit…
By representing words with probability densities rather than point vectors, proba- bilistic word embeddings can capture rich and interpretable semantic information and uncertainty (Vilnis & McCallum, 2014; Athiwaratkun & Wilson, 2017). The uncertainty information can be particularly meaningful in ca…
Malicious software, or malware, continues to be a problem for computer users, corporations, and governments. Previous research [1] has explored training file-based, malware classifiers using a two-stage approach. In the first stage, a malware language model is used to learn the feature representatio…