Knowledge catalogue
I've been working on speeding up DeepSeek-V4-Flash-0731 in Krasis and have now got the long-prompt prefill quite a bit faster on a single RTX PRO 6000 96GB. These are timing-disabled internal Krasis r
https://preview.redd.it/522fsdwvtdhh1.png?width=1200&format=png&auto=webp&s=6a6cf7a467514167a8193029dbd20fb3a9ba4f6c It ranks lower than both Sonnet 4.6 and Luna. I'd wager Luna costs in the same ball
I really like to use this one SQL benchmark when testing new models. I had another post some time ago with my benchmarks, but I decided to post a new one because of how well Deepseek did. I like the b
DeepSeek V4 Flash is now live on Together AI. Frontier agent performance is getting dramatically cheaper. DSV4 on Together AI brings a major jump in coding, tool use, and long-running agent performanc
Just wanted to share my agentic coding benchmark run of DSv4F 0731 at both High and Low reasoning efforts (not Max)... I ran a 109-question subset of Aider Polyglot (the JS/C++/Python languages), base
arXiv:2601.15307v2 Announce Type: replace-cross Abstract: The rapid development of automated survey generation technology has made it increasingly important to establish a comprehensive benchmark to e
arXiv:2410.06814v2 Announce Type: replace Abstract: Over-parameterized models are typically vulnerable to membership inference attacks, which aim to determine whether a specific sample is included in
arXiv:2608.01240v1 Announce Type: new Abstract: Embedded Language Flows (ELF) rely primarily on full non-causal attention for iterative denoising, repeatedly incurring quadratic sequence-mixing cost a
arXiv:2607.15807v2 Announce Type: replace Abstract: Ultra-Wideband (UWB) ranging has become a viable option for industrial Autonomous Mobile Robot (AMR) localization due to improved accuracy and low c
arXiv:2608.01823v1 Announce Type: new Abstract: Hallucination remains a persistent challenge in generative super-resolution (GSR), where reconstructed results may contain visually plausible yet weakly
arXiv:2608.00146v1 Announce Type: new Abstract: We introduce DiffusionGemma, an experimental open-weight language model that uses discrete diffusion to generate text at exceptionally high speed. Rathe
arXiv:2608.00547v1 Announce Type: new Abstract: Contact-rich manipulation remains challenging because successful control depends on physical interaction cues that are often weakly observable from visi
arXiv:2608.01816v1 Announce Type: new Abstract: In this work we investigate how decoder-only transformers resolve lexical ambiguity through layer-by-layer analysis of three models spanning three param
arXiv:2608.00011v1 Announce Type: new Abstract: Current text-to-speech systems face a trade-off: autoregres- sive codec language models produce highly intelligible speech but require large-scale model
arXiv:2607.17999v2 Announce Type: replace-cross Abstract: Spatial understanding is crucial for foundation models (FMs), and maps have long helped humans organize and reason about geographic informatio
arXiv:2608.02112v1 Announce Type: new Abstract: Embedding benchmarks measure standalone model quality, but they do not establish whether a low-cost retriever contributes complementary ranking informat
arXiv:2608.02396v1 Announce Type: new Abstract: Most evidence on the effectiveness of explainable artificial intelligence (XAI) attribution methods has been established on convolutional neural network
arXiv:2608.02235v1 Announce Type: new Abstract: Recent advances in neural text-to-speech (TTS) systems have substantially improved speech naturalness and intelligibility across many languages. However
arXiv:2602.01132v2 Announce Type: replace Abstract: Tasks such as solving arithmetic equations, evaluating truth tables, and completing syllogisms are handled well by large language models (LLMs) in t
arXiv:2608.00548v1 Announce Type: new Abstract: Recent image-generation models and multimodal agents can produce high-quality visuals for increasingly complex visual communication tasks. Yet their ras
arXiv:2608.00086v1 Announce Type: new Abstract: Brain tumor diagnosis is a time-sensitive process in which patients may wait weeks for a finalized pathology report. This problem motivates automated sy
arXiv:2608.01452v1 Announce Type: cross Abstract: Dynamic manipulation is a critical capability for robots operating in complex and dynamic environments, where robots must interact with objects that a
arXiv:2607.17244v2 Announce Type: replace Abstract: Longitudinal T cell receptor repertoires contain signals of clonal expansion, contraction, disappearance, and reappearance after immune perturbation
arXiv:2608.00694v1 Announce Type: new Abstract: Event cameras offer microsecond-level temporal resolution and high dynamic range, potentially facilitating motion-blur-free panoramic imaging from fast
arXiv:2608.02474v1 Announce Type: new Abstract: Audio-driven video generation (A2V) has achieved promising progress in synthesizing temporally coherent and audio-visually aligned videos, yet its infer
arXiv:2601.17883v3 Announce Type: replace-cross Abstract: Electroencephalography (EEG) foundation models (FMs) have recently emerged as a promising paradigm for brain-computer interfaces, aiming to le
arXiv:2511.00382v2 Announce Type: replace-cross Abstract: Organizations increasingly adapt Large Language Models (LLMs) from public repositories such as HuggingFace to downstream tasks. Prior work sho
arXiv:2603.12147v2 Announce Type: replace Abstract: Egocentric video provides a natural modality for studying human behavior, but conventional visual understanding captures mainly observable scenes, o
arXiv:2608.01548v1 Announce Type: cross Abstract: Language can be viewed as a formalized subset of thought: a consequence-governed symbolic structure projected from wider situated cognition. Large lan
arXiv:2608.01979v1 Announce Type: cross Abstract: Visual token pruning reduces the inference cost of multimodal large language models, but a fixed token ratio is poorly matched to text-rich inputs. In
Data privacy engineering company Ethyca Inc. today launched Astralis, a platform that governs how enterprise artificial intelligence models and agents use company data in real time. The company is pit
arXiv:2608.01142v1 Announce Type: new Abstract: Low-Rank Adaptation (LoRA) enables parameter-efficient fine-tuning, but standard LoRA produces a single deterministic model and does not directly suppor
arXiv:2608.01238v1 Announce Type: new Abstract: Vision Language Models (VLMs) have demonstrated exceptional performance across various tasks. However, they have not yet been thoroughly evaluated on mo
arXiv:2608.01948v1 Announce Type: new Abstract: Long-horizon event-based action understanding remains underexplored because existing datasets largely comprise short, trimmed clips, while collecting na
arXiv:2608.01058v1 Announce Type: new Abstract: Artificial Intelligence is increasingly applied to surgical video analysis for phase segmentation, skill assessment, and workflow optimization. A key ch
arXiv:2608.01661v1 Announce Type: new Abstract: The challenge of fair deepfake detection (FDD) has attracted increasing attention. Existing fairness-enhanced detectors often suffer from suboptimal gen
arXiv:2608.02359v1 Announce Type: new Abstract: Attributing quotations to their speakers in literary texts remains an open challenge. Standard methods, which independently predict a speaker mention fo
arXiv:2608.01290v1 Announce Type: new Abstract: Time-series foundation models (TSFMs) such as Chronos have demonstrated strong forecasting capabilities across domains, yet adapting them to institution
arXiv:2608.02071v1 Announce Type: new Abstract: Attention-only dynamical theories model Transformer residual directions as particles aggregating on a sphere. We extend this framework by incorporating
arXiv:2608.00909v1 Announce Type: new Abstract: Can large language models generate not just correct, but fast hardware? This paper investigates the question in financial FPGA design, where 5-10 nanose
arXiv:2608.01917v1 Announce Type: new Abstract: Discounted exponential utility provides a principled criterion for risk-sensitive sequential decision-making, but its nonlinear structure complicates re
arXiv:2608.00500v1 Announce Type: new Abstract: Unified humanoid policies handle agile whole-body motion, yet stumble on a simple demand: staying balanced on one leg. On our single-leg-balance benchma
arXiv:2608.01095v1 Announce Type: new Abstract: Federated learning (FL) enables multiple intelligent devices to collaboratively train a high-accuracy model without sharing raw data. However, due to it
arXiv:2608.01704v1 Announce Type: cross Abstract: A benchmark score means nothing without knowing what a trivial method achieves and what the best possible method could achieve. We construct both boun
arXiv:2607.09842v2 Announce Type: replace-cross Abstract: We investigate whether identity-specifying system prompts produce statistically distinguishable geometric fingerprints in the hidden-state tra
arXiv:2608.00608v1 Announce Type: new Abstract: Visible and near-infrared (vis-NIR) and mid-infrared (MIR) spectroscopy enable rapid, cost-effective prediction of soil properties. Yet, translating hig
arXiv:2507.21449v2 Announce Type: replace-cross Abstract: Degeneracy is an inherent feature of the loss landscape of neural networks, but it is not well understood how stochastic gradient MCMC (SGMCMC
arXiv:2608.02100v1 Announce Type: cross Abstract: As AI increasingly participates in human decision making, understanding how decision-making authority is distributed between humans and AI has become
arXiv:2602.10719v2 Announce Type: replace-cross Abstract: Vision-Language-Action (VLA) driving augments end-to-end (E2E) planning with language-enabled visual backbones, yet it remains unclear how vis
arXiv:2603.03320v2 Announce Type: replace Abstract: Effective communication often relies on aligning a message with an audience's narrative and worldview. Narrative shift involves transforming text to
arXiv:2608.01007v1 Announce Type: new Abstract: Dual-target drug design aims to generate 3D molecules that can simultaneously interact with two target proteins, offering a promising route for discover
Editor's Note: Our Future Mode series will give businesses insight into how Chrome Enterprise is approaching AI in the browser. Stay tuned for more blogs in this series.Future Mode Part 2: The foundat
arXiv:2608.00932v1 Announce Type: new Abstract: The integration of artificial intelligence into medical question-answering systems has advanced rapidly; however, research remains predominantly focused
arXiv:2608.00829v1 Announce Type: new Abstract: We present an autonomous robotic portrait-generation system combining real-time face detection, AI-based sketch generation, and robotic drawing. The sys
arXiv:2608.01686v1 Announce Type: new Abstract: In recent years, generative AI has attracted significant public attention, and its use has been rapidly expanding across a wide range of domains. From c
arXiv:2608.01896v1 Announce Type: new Abstract: Existing generative models for earth observation (EO) predominantly rely on fine-tuning natural image priors, which limits their scalability and introdu
arXiv:2608.02315v1 Announce Type: new Abstract: Geospatial foundation models aim to learn representations that transfer across regions and sensors, yet evaluating them on specific tasks requires large
arXiv:2602.01893v2 Announce Type: replace-cross Abstract: We present a geometric framework for analysing multi-head attention in large language models (LLMs). Without altering the mechanism, we view s
arXiv:2608.02068v1 Announce Type: new Abstract: Monocular depth foundation models, benefiting from large-scale synthetic training data, have demonstrated strong generalization. However, they often hal
Given today, it is surprising how daring Microsoft & Google were initially with AI. Microsoft released GPT-4 before OpenAI, didn't back down after Sydney & got Copilot to market quickly (the 1st profe