DanLing NestedTensor is a PyTorch tensor abstraction that embeds multi‑ragged structure directly into the tensor, allowing packed values to carry partition information and logical dimension order. This design enables broadcasting, feature transformations, and reductions to automatically respect ragged axes while preserving the same representation through autograd and both eager and compiled execution. Benchmarks on an A100 show significant speedups—up to 3.39× over padding for BERT models and 2.40–4.32× for a Pairformer‑style workload—while dramatically reducing peak memory usage.
By Zhiyuan Chen
arXiv:2608. 00029v1 Announce Type: cross Abstract: The performance of deep learning models at scale relies heavily on how effectively high-level mathematical operations are mapped to underlying physical hardware.
By Adwaid Suresh, Aparna A, Harshini V M, Jona Delcy C A, Killi Uma Maheswara Rao, Ram Charan Golla, Surendra Vendra
arXiv:2610.07742v1 Announce Type: cross
Abstract: Optimized kernels such as FlashAttention and FlashDecoding are crucial for accelerating today's large models. Most of them are handwritten by experts...
By David Pissarra, Jinkun Lin, Haitian Jiang, Aurojit Panda, Jinyang Li
arXiv:2610.15124v1 Announce Type: cross
Abstract: Foundation models need to support diverse inference budgets, yet training a separate model for each budget duplicates pretraining effort. Elastic mod...
By Can Jin, Yan Kang, Yifan Gong, Chongjian Ge, Mingze Xu, Yuanjun Xiong, Dimitris N. Metaxas, Hongwu Peng
The paper investigates how to optimally partition streaming inference models into megakernels and conventional graph components. It introduces a framework that predicts the best megakernel composition without full end‑to‑end runs, showing that multiple megakernels or a fully conventional graph often outperform a single megakernel across various models and batch sizes. The study defines four model categories based on megakernel stability and demonstrates significant latency and throughput gains on an H100 NVL GPU using a three‑megakernel design.
By Aditya Chatterjee, Anand Joseph X M, Sanjay Singh Chauhan
The paper introduces a looped causal encoder-decoder (CED) architecture that increases model depth without enlarging the key-value (KV) cache across layers and rounds. By placing recurrence in the encoder and allowing sparse encoder layers to generate global KV, the design achieves language‑modeling gains while maintaining long‑context scaling. Experiments show that the Looped CED matches the memory footprint of an unlooped CED and recovers most of a looped Transformer’s performance with significantly less KV, enabling higher throughput on large‑token prompts.
By Jiaqi Leng, Minshen Zhang, Zhaofeng Sun, Yu-Yang Qian, Peng Zhao, Hao Zhang
arXiv:2607. 18323v1 Announce Type: cross Abstract: Exhaustive site-by-site interventions on a neural network's computational graph -- activation-patching sweeps, circuit-discovery searches, systematic ablation studies -- mutate the graph at every candidate site, and their cost is dominated by recomputation after each mutation.
By Abdallah Khemais (ISITCOM, University of Sousse)
We address a fundamental question: How far should kernel fusion extend and which boundaries are worth preserving? Unlike the general convention of using one monolith megakernel at lower batch sizes or...
arXiv:2607. 25504v1 Announce Type: cross Abstract: Fine-grained weight pruning and activation sparsification have emerged as effective approaches for reducing the compute and memory cost of inference for Transformer models.
By Bowen Wang, Chi Zhang, Diyou Shen, Renzo Andri, Navaneeth Kunhi Purayil, Luca Benini
The paper presents a method for reliably scheduling operator execution in real‑time inference on small devices, achieving exact peak‑memory predictions and certified optimality. By profiling the generated program and modeling its memory usage, the authors can permute statements to match the modeled peak, guaranteeing bitwise‑identical results on CPUs and characterizing GPU limitations. They also provide efficient lower‑bound traps and analyze selective recomputation, showing it is NP‑hard in general but can reduce peak memory on certain architectures.
By Ivan Fomin, Ivan Koryabkin, Boris Sheludko
arXiv:2608. 09444v1 Announce Type: new Abstract: A main promise of looped language models (LMs) is depth-adaptive inference.
By Kristian Schwethelm, Daniel Rueckert, Georgios Kaissis
PageWeaver introduces a KV‑guided query union strategy for sparse attention, assembling query groups based on selected‑page affinity while preserving each query’s support and output ownership. The design employs a bounded GPU search to generate query IDs and an ID‑aware two‑CTA kernel that consumes them without materializing reordered Q tensors or cross‑page partial outputs. Experiments on the H200 Union8 with FP8 KV show a 1.70× speedup over FlashInfer, with additional latency reductions of 3.26–7.66% on selected captures and modest throughput gains for whole‑model prefill.
By Zhiyuan Li, Zihan Li, Zefang Yuan, Lei Wang, Hao Wang