Related papers: A Lightweight Method for Generating Multi-Tier JIT…
In this paper, we design and train a Generative Image-to-text Transformer, GIT, to unify vision-language tasks such as image/video captioning and question answering. While generative models provide a consistent network architecture between…
Recent efforts in open-source GPU research are opening new avenues in a domain that has long been tightly coupled with a few commercial vendors. Emerging open GPU architectures define SIMT functionality through their own ISAs, but executing…
A compiler's optimizer operates over abstract syntax trees (ASTs), continuously applying rewrite rules to replace subtrees of the AST with more efficient ones. Especially on large source repositories, even simply finding opportunities for a…
Evaluating LLMs with a single prompt has proven unreliable, with small changes leading to significant performance differences. However, generating the prompt variations needed for a more robust multi-prompt evaluation is challenging,…
We introduce Token Merging (ToMe), a simple method to increase the throughput of existing ViT models without needing to train. ToMe gradually combines similar tokens in a transformer using a general and light-weight matching algorithm that…
The growing adoption of RISC-V in high-performance and scientific computing has increased the need for performance-portable code targeting the RISC-V Vector (RVV) extension. However, current compiler infrastructures provide limited…
We have developed a novel compiler called the Multiple-Architecture Compiler for Advanced Computing Hardware (MACH) designed specifically for massively-parallel, spatial, dataflow architectures like the Wafer Scale Engine. Additionally,…
Programs implemented in various programming languages form the foundation of software applications. To alleviate the burden of program migration and facilitate the development of software systems, automated program translation across…
This work proposes a compilation flow using open-source compiler passes to build a framework to achieve ninja performance from a generic linear algebra high-level abstraction. We demonstrate this flow with a proof-of-concept MLIR project…
Neutral-atom quantum computing is among the most promising platforms for scalable quantum computation, and compilation toolchains are crucial for leveraging capabilities such as qubit shuttling and parallel gate execution. An important…
Realistic and efficient 3D garment generation remains a longstanding challenge in computer vision and digital fashion. Existing methods typically rely on large vision- language models to produce serialized representations of 2D sewing…
Visual grounding is a common vision task that involves grounding descriptive sentences to the corresponding regions of an image. Most existing methods use independent image-text encoding and apply complex hand-crafted modules or…
Humans can quickly generalize handwriting styles from a single example by intuitively separating content from style. Machines, however, struggle with this task, especially in low-data settings, often missing subtle spatial and stylistic…
Casting semantic segmentation of outdoor LiDAR point clouds as a 2D problem, e.g., via range projection, is an effective and popular approach. These projection-based methods usually benefit from fast computations and, when combined with…
Compilers are essential for the performance and correct execution of software and hold universal relevance across various scientific disciplines. Despite this, there is a notable lack of tools for testing and evaluating them, especially…
Training models with varying capacities can be advantageous for deploying them in different scenarios. While high-capacity models offer better performance, low-capacity models require fewer computing resources for training and inference. In…
Coarse Grained Reconfigurable Arrays (CGRAs) present both high flexibility and efficiency, making them well-suited for the acceleration of intensive workloads. Nevertheless, a key barrier towards their widespread adoption is posed by CGRA…
While LLMs can effectively help prototype single ML functionalities, many real-world applications involve complex tasks that cannot be easily handled via a single run of an LLM. Recent work has found that chaining multiple LLM runs together…
Spatial dataflow architectures such as reconfigurable dataflow accelerators (RDA) can provide much higher performance and efficiency than CPUs and GPUs. In particular, vectorized reconfigurable dataflow accelerators (vRDA) in recent…
Text-to-motion generation has advanced rapidly, yet two challenges persist. First, existing motion autoencoders compress each frame into a single monolithic latent vector, entangling trajectory and per-joint rotations in an unstructured…