Knowledge catalogue
Disclosure: I work at H2O.ai. We released H2O-Lightning-4B, an open-weight (Apache-2.0) model for the 'decisions API' style of inference that Jev made popular: you send a state plus typed questions (p
arXiv:2509.09311v2 Announce Type: replace Abstract: Vision-language models (VLMs) have enabled strong zero-shot classification through image-text alignment. Yet, their purely visual inference capabili
arXiv:2610.11253v1 Announce Type: new Abstract: Large language model agents are being increasingly deployed as autonomous scientists, designing experiments and inferring mechanistic world models with
arXiv:2610.11794v1 Announce Type: new Abstract: Learning to act in unfamiliar environments requires agents to infer how the world works and revise that understanding as new evidence arrives. Yet limit
LemonSeed Studio is an iPad editor/IDE with on-device inference on an AMD GPU in a Thunderbolt enclosure. It embeds the unmodified upstream Linux amdgpu + amdkfd driver (mac_linuxgpu) as a PCIDriverKi
arXiv:2602.07878v2 Announce Type: replace-cross Abstract: LLM inference is inherently expensive, even a modest slowdown can translate into substantial operating costs and severe availability risks. Re
arXiv:2606.18096v2 Announce Type: replace-cross Abstract: Structured State Space Models (SSMs), including the S4 and S4D architectures, have recently emerged as powerful alternatives to attention-base
arXiv:2610.10625v1 Announce Type: cross Abstract: Looped Language Models (LoopLMs) provide a parameter efficient approach to scaling model capabilities through repeated use of shared parameters across
arXiv:2610.11738v1 Announce Type: new Abstract: Reliable uncertainty quantification (UQ) is essential for deploying neural networks in scientific and high-stakes applications, but full Bayesian infere
arXiv:2610.11333v1 Announce Type: new Abstract: AI services incur inference costs during execution, while revenue may arrive later. Changing API prices, limited upfront capital, and service failures c
arXiv:2610.08849v1 Announce Type: cross Abstract: Purpose: Current abdominal ultrasound (US) simulation methods often require CT-based anatomical references for ray-casting, limiting deformation and p
arXiv:2610.09467v1 Announce Type: cross Abstract: Contextual biasing supplies an ASR system with a list of expected words at inference time, but existing methods rely on word boundaries that Japanese
arXiv:2610.10194v1 Announce Type: cross Abstract: Switching stochastic differential equations (SSDEs) describe continuous-time dynamics whose parameters switch according to a latent regime process tha
arXiv:2610.09221v1 Announce Type: cross Abstract: Flow and diffusion models suffer from slow inference due to computationally expensive numerical integration. Distillation provides a promising way for
arXiv:2609.32701v2 Announce Type: replace-cross Abstract: In security-sensitive applications, language-model agents are often required to coordinate without disclosing confidential information. Yet re
arXiv:2610.09976v1 Announce Type: new Abstract: AI-generated text (AIGT) detection can be sensitive to the decoding choices of the source large language model (LLM). We observe that perturbing next-to
arXiv:2610.09456v1 Announce Type: new Abstract: Reconstructing fine-resolution column-averaged dry-air CO2 (XCO2) fields from sparse satellite observations requires models to infer spatial structure t
arXiv:2610.09788v1 Announce Type: new Abstract: Reward modeling often requires jointly representing and reasoning over multiple evaluation criteria, yet verbalizing this process token by token can inc
arXiv:2610.08811v1 Announce Type: new Abstract: As context windows scale to tens or hundreds of thousands of tokens, KV cache compression has become essential for efficient LLM inference. Existing met
arXiv:2610.09350v1 Announce Type: cross Abstract: Learning from demonstrations (LfD) provides a framework for inferring unknown constraints from locally optimal, constraint-satisfying expert behavior.
arXiv:2610.10138v1 Announce Type: new Abstract: Predicting the future impact of a newly published paper is challenging because it must be inferred from heterogeneous evidence available at publication
arXiv:2610.10528v1 Announce Type: cross Abstract: Real-time robot control demands enough visual history to infer motion and task progress, but processing that history can delay action. We present Long
arXiv:2610.09723v1 Announce Type: new Abstract: Prior artisan mesh generation works largely predict face tokens autoregressively, which makes inference slow. Recent methods instead flow match continuo
arXiv:2610.10457v1 Announce Type: new Abstract: Diffusion Transformers (DiTs) achieve remarkable performance in video synthesis, but their iterative denoising process suffers from high inference laten
arXiv:2610.09498v1 Announce Type: new Abstract: Clinical vision models are often deployed as frozen black boxes with no access to internals, retraining, or ground truth at inference time. We introduce
arXiv:2610.09509v1 Announce Type: new Abstract: Latent actions supervise robot policies with action-like codes inferred from visual transitions, and their usefulness hinges on two questions: what a co
arXiv:2610.09132v1 Announce Type: cross Abstract: In wide Bayesian neural networks, Gaussian mean-field variational inference is prone to 'prior dominance': the Kullback-Leibler (KL) regularization te
arXiv:2610.10049v1 Announce Type: new Abstract: Reasoning models allocate extra computation at inference time and present their answers as the product of deliberate thought. If this deliberation works
Tokenomics, from the chip all the way up. @dylan522p, Founder & CEO of @SemiAnalysis_, joins the Fireworks Forge speaker list on optimizing every layer of AI infrastructure for inference at scale. Joi
arXiv:2610.09118v1 Announce Type: new Abstract: Topology reasoning jointly detects 3D lanes and traffic elements from multi-view images and infers their structural connectivity. Current methods model
arXiv:2610.10411v1 Announce Type: cross Abstract: Speculative decoding accelerates large language model inference by using a low-cost draft model to propose tokens that the full-size target model veri
arXiv:2610.08863v1 Announce Type: new Abstract: A common way for trajectory planning is to leverage generative models trained on large collections of expert trajectories. At inference time, the model
arXiv:2610.09376v1 Announce Type: new Abstract: Acquiring a complete set of magnetic resonance imaging (MRI) contrasts is time-intensive and uncomfortable for patients, despite the diagnostic value of
arXiv:2610.10118v1 Announce Type: cross Abstract: Long-horizon reasoning demands access to earlier information at a manageable generation cost. Full-history attention incurs growing storage and comput
arXiv:2610.09718v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models acquire broad manipulation capabilities via large-scale pretraining, yet eliciting them through language requires
I've been tinkering with GLM-5.3-Flash on two 64GB CMP 170HX cards for a while, and the setup is finally stable enough that I figured I'd share it. I also compared it against the Qwen3.8-Flash-Next se
arXiv:2610.07780v1 Announce Type: new Abstract: Speculative decoding reduces large language model inference latency by drafting multiple tokens before target-model verification, but its effectiveness
arXiv:2610.06854v1 Announce Type: cross Abstract: Synthetic data generation is critical for privacy compliance, machine learning augmentation, and software testing. While single-table evaluation is we
arXiv:2610.07668v1 Announce Type: cross Abstract: Modern cache replacement designs saturate because they operate within fixed representational structures, hand-crafted and heuristic based feature-engi
arXiv:2601.16390v2 Announce Type: replace-cross Abstract: Large language models exhibit strong multilingual capabilities, yet significant performance gaps persist between dominant and non-dominant lan
arXiv:2610.08777v1 Announce Type: new Abstract: Interactive video world models need to generate each video chunk efficiently while responding faithfully to user controls. Many systems use chunk-wise a
arXiv:2610.08344v1 Announce Type: new Abstract: Assistance in collaborative manipulation is often initiated by user instructions, making high-level reasoning request-driven. In fluent human teamwork,
arXiv:2610.07832v1 Announce Type: cross Abstract: Large language models (LLMs) equipped with terminal access have demonstrated strong capabilities in automating software engineering tasks. However, ex
EmbeddingGemma 2 came out this week. It maps images and text into one 768-dim space, so you can search photos by describing them. I ported its text and vision towers to ruNNtime, a WebGPU inference li
arXiv:2610.08475v1 Announce Type: new Abstract: In many-query scenarios, data-driven surrogate models provide an efficient alternative to high-fidelity solvers for simulating physical systems governed
arXiv:2610.07247v1 Announce Type: new Abstract: Large language models have shown strong reasoning capabilities, but their high inference costs make knowledge distillation an important approach for tra
arXiv:2610.08425v1 Announce Type: new Abstract: Vision-language-action (VLA) policies infer grasp actions primarily from visual observations and robot state, but do not explicitly represent the physic
Liquid just dropped d1-omni-600M and it's a decision model! I ported it to runntime, the WebGPU inference library I'm currently working on. It's plain TypeScript on top of TypeGPU with no WASM and vir
arXiv:2610.08075v1 Announce Type: cross Abstract: Conditional neural fields represent signals continuously, but their effectiveness depends on how the conditional latent representations are inferred f
arXiv:2610.08231v1 Announce Type: new Abstract: NVFP4 is an attractive datatype for large language model (LLM) inference, offering compact storage and native tensor-core acceleration. However, preserv
arXiv:2610.08346v1 Announce Type: new Abstract: Polarization imaging provides physical cues beyond intensity imaging but typically requires specialized hardware. Recent methods infer polarization from
arXiv:2610.07863v1 Announce Type: cross Abstract: Long-horizon LLM agents act on an append-only interaction history that is re-sent to the model at every step, so the context and its cost grow with st
arXiv:2610.06955v1 Announce Type: cross Abstract: Humans inherently understand the physical world through an active process. When sensory evidence is insufficient to infer physical properties, we natu
arXiv:2610.08403v1 Announce Type: new Abstract: Large Language Models (LLMs) require substantial computational resources, limiting their deployment on resource-constrained hardware. Ternary LLMs mitig
arXiv:2610.07625v1 Announce Type: cross Abstract: Automated research systems increasingly run LLM agents over long horizons, but more inference does not by itself produce more progress: agents replay
arXiv:2610.06974v1 Announce Type: new Abstract: Knowledge graph embedding (KGE) methods represent entities and predicates in continuous vector spaces to infer missing knowledge. Despite strong benchma
arXiv:2610.06938v1 Announce Type: new Abstract: Medical image segmentation remains fragmented along two axes: segmentation paradigms and data dimensionality. Existing methods are typically developed s
arXiv:2512.13708v2 Announce Type: replace Abstract: Inferring interaction structure from steady-state observations is a central inverse problem when transient trajectories are unavailable. Here we for
arXiv:2610.05835v1 Announce Type: cross Abstract: Tactile Internet (TI) security analytics must balance reliable thresholded decisions with constrained computational and measurement resources. We stud
arXiv:2610.03826v1 Announce Type: new Abstract: Benchmark scores license claims about the capabilities being evaluated. We audit the inference licensed by an evaluation protocol, rather than the predi