Tootfinder

🇺🇦 #NowPlaying on KEXP's #AfternoonShow
Death Valley Girls:
🎵 Fire and Brimstone
#DeathValleyGirls
https://deathvalleygirls.bandcamp.com/album/sisters-of-the-moon-b-w-fire-and-brimstone
https://open.spotify.com/track/4jRq49pjTXmk5QOd6tHmxl

„Ich fahr ja SUV wegen der besseren Übersicht.“
From: @…
https://masto.bike/@doodooavelo/114803362565737616

🇺🇦 #NowPlaying on #BBC6Music's #CerysMatthews
Datblygu:
🎵 Y Teimlad
#Datblygu
https://ankstmusik.bandcamp.com/track/y-teimlad-datblygu
https://open.spotify.com/track/0nSyGg7aHp5lfV5PRrq05o

Mit "Deadpool VR" und "Reach" erscheinen 2025 zwei echte VR-Blockbuster
Meta kündigt mit "Deadpool VR" den nächsten Exklusivtitel für die Quest 3 an, während nDreams mit "Reach" eine VR-Großproduktion für alle VR-Plattformen zeigt.

On the Feasibility of Deduplicating Compiler Bugs with Bisection
Xintong Zhou, Zhenyang Xu, Chengnian Sun
https://arxiv.org/abs/2506.23281 https://

On the Feasibility of Deduplicating Compiler Bugs with Bisection
Random testing has proven to be an effective technique for compiler validation. However, the debugging of bugs identified through random testing presents a significant challenge due to the frequent occurrence of duplicate test programs that expose identical compiler bugs. The process to identify duplicates is a practical research problem known as bug deduplication. Prior methodologies for compiler bug deduplication primarily rely on program analysis to extract bug-related features for duplicate…

VEDA: Efficient LLM Generation Through Voting-based KV Cache Eviction and Dataflow-flexible Accelerator
Zhican Wang, Hongxiang Fan, Haroon Waris, Gang Wang, Zhenyu Li, Jianfei Jiang, Yanan Sun, Guanghui He
https://arxiv.org/abs/2507.00797

VEDA: Efficient LLM Generation Through Voting-based KV Cache Eviction and Dataflow-flexible Accelerator
Large Language Models (LLMs) excel in natural language processing tasks but pose significant computational and memory challenges for edge deployment due to their intensive resource demands. This work addresses the efficiency of LLM inference by algorithm-hardware-dataflow tri-optimizations. We propose a novel voting-based KV cache eviction algorithm, balancing hardware efficiency and algorithm accuracy by adaptively identifying unimportant kv vectors. From a dataflow perspective, we introduce a…

🇺🇦 #NowPlaying on KEXP's #MiddayShow
Death Valley Girls:
🎵 Sisters of the Moon
#DeathValleyGirls
https://deathvalleygirls.bandcamp.com/album/sisters-of-the-moon-b-w-fire-and-brimstone
https://open.spotify.com/track/3IFIlSnjtGwOF4fcpRuOJa

This week's #GrindayFriday is new EP 'Buried Half Dead' by Dayton, Ohio's DEAD FILTH. Super gross song titles here, as you'd expect with goregrind. But there's some real dynamics and interesting riffing going on in some of this. The drums are great. Several spots of unique genius. And you can listen to this whole thing in just a couple minutes.

This https://arxiv.org/abs/2505.08992 has been replaced.
initial toot: https://mastoxiv.page/@arXiv_csAR_…

Dataflow & Tiling Strategies in Edge-AI FPGA Accelerators: A Comprehensive Literature Review
Edge-AI applications demand high-throughput, low-latency inference on FPGAs under tight resource and power constraints. This survey provides a comprehensive review of two key architectural decisions for FPGA-based neural network accelerators: (i) the dataflow (the order and manner in which data is moved and reused on chip), and (ii) the tiling/blocking strategy (how large tensors are partitioned to fit on-chip). We first present a broadened taxonomy of canonical dataflow styles: Weight-Stationa…