Knowledge catalogue
arXiv:2610.16160v1 Announce Type: new Abstract: Recent advances in 3D reconstruction enable the extraction of temporally consistent 3D motion from RGB observations; however, current Vision-Language-Ac
arXiv:2610.16744v1 Announce Type: new Abstract: Video large language models (Video LLMs) encode each frame into hundreds of visual tokens, making video inference expensive. Existing methods reduce thi
arXiv:2610.17948v1 Announce Type: new Abstract: Diffusion large language models (dLLMs) have attracted increasing research attention due to their bidirectional attention and arbitrary order decoding.
arXiv:2610.16261v1 Announce Type: cross Abstract: Mamba-based super-resolution models are efficient alternatives to Transformer-based restoration backbones, but their low-bit post-training quantizatio
arXiv:2610.18051v1 Announce Type: new Abstract: This work investigates the application of the Membership Inference Test (MINT) to large language models (LLMs) fine-tuned with LoRA, aiming to determine
arXiv:2610.14449v1 Announce Type: cross Abstract: Experiments with remote decision-making agents depend on more than the policy: the world may change during inference, the simulator determines which a
arXiv:2610.18289v1 Announce Type: new Abstract: Autoregressive video diffusion models have emerged as a foundation for long-horizon world rollout. However, when extended far beyond their training wind
arXiv:2610.16368v1 Announce Type: new Abstract: Looped Language Models (LoopLMs) improve parameter efficiency by recursively reusing shared Transformer modules, yet their repeated computation over evo
arXiv:2610.17877v1 Announce Type: new Abstract: Reusing Transformer blocks offers a way to increase diffusion-model computation at test time, but naive recurrence can degrade generation beyond its tra
arXiv:2610.14117v1 Announce Type: new Abstract: GPT-6 Astra provides robust embodied reasoning, but frequent external queries increase inference time and token costs. We equip a frozen pi_{0.5} policy
arXiv:2610.16252v1 Announce Type: new Abstract: Whole-body human-object interaction (HOI) generation requires coordinating body and finger motion, object trajectories, and time-varying contacts. Exist
arXiv:2610.17202v1 Announce Type: new Abstract: Multi-Head Latent Attention (MLA) compresses the KV cache into a latent representation, reducing inference memory and bandwidth. Existing MHA2MLA/TransM
arXiv:2610.18677v1 Announce Type: new Abstract: Online decision-making faces a tension between efficient inference and long-horizon reasoning. Value-based methods rely on short-horizon bootstrapping t
I've been working on an inference engine called jitllm. The idea came from my background in operating systems and JIT compilers: generate kernels for whatever hardware is there, and treat layers a bit
arXiv:2610.16902v1 Announce Type: cross Abstract: Bayesian Networks (BNs) are established tools for letting agents reason under uncertainty in sequential decision-making. They split a joint distributi
arXiv:2610.18176v1 Announce Type: cross Abstract: Proactive human-robot collaboration requires autonomous systems to infer not only immediate actions but also the underlying hierarchical intentions go
arXiv:2601.14628v2 Announce Type: replace Abstract: Vision-language-action (VLA) models inherit powerful semantic representations from large language models, but carrying over their design priorities
arXiv:2303.07308v3 Announce Type: replace-cross Abstract: We present NeuSE, a novel Neural SE(3)-Equivariant Embedding for objects, and illustrate how it supports object-based Simultaneous Localizatio
arXiv:2610.17347v1 Announce Type: new Abstract: We revisit the statistical mechanics of Transformer attention via modern Hopfield models. We present a free-energy formalism to characterize parameter r
arXiv:2610.17900v1 Announce Type: cross Abstract: Extended chain-of-thought reasoning improves problem-solving performance but increases inference cost and latency, motivating more accurate answers wi
arXiv:2610.15722v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) are good at video understanding and code generation, but it remains unclear whether they can recover the physic
arXiv:2610.14998v1 Announce Type: new Abstract: Tabular foundation models adapt to new datasets through in-context learning (ICL), and, like most transformers, they tie computational depth to the numb
arXiv:2610.17426v1 Announce Type: new Abstract: Clinical decision-making is a sequential process of information collection, clinical inference, and forecasting under partial observability, often witho
arXiv:2512.17977v2 Announce Type: replace-cross Abstract: Sampling from multimodal distributions is a central challenge in Bayesian inference and machine learning. In light of hardness results for sam
arXiv:2610.15514v1 Announce Type: new Abstract: Structured pruning reduces the inference cost of large language models (LLMs) by permanently removing neurons and weight rows, yielding dense, hardware-
arXiv:2610.14968v1 Announce Type: cross Abstract: Large Language Models (LLMs) can generate functionally correct code while introducing exploitable security weaknesses, particularly when developers sp
arXiv:2610.18735v1 Announce Type: new Abstract: Speculative Decoding (SD) has demonstrated its efficacy in accelerating large language model inference, but for Vision-Language Models (VLMs), the diver
arXiv:2610.14400v1 Announce Type: new Abstract: Successor-feature-based zero-shot reinforcement learning methods pre-train a task encoder and successor-feature predictor on offline, reward-free data,
arXiv:2403.01079v2 Announce Type: replace Abstract: We study the challenging problem for inference tasks on large-scale graph datasets of Graph Neural Networks: huge time and memory consumption, and t
arXiv:2610.15098v1 Announce Type: new Abstract: Text-to-video (T2V) models are highly sensitive to prompt specificity, while iterative prompt optimization typically improves generation quality only th
arXiv:2603.18091v2 Announce Type: replace Abstract: Test-time sampling can improve Vision-Language-Action (VLA) policies, but only if the model can efficiently select a good action from multiple candi
arXiv:2610.15076v1 Announce Type: new Abstract: Plasma experiments observe coupled physical states through sparse, heterogeneous, and incomplete diagnostics. We present a proof of concept toward a mul
arXiv:2610.16969v1 Announce Type: cross Abstract: Latent Action Models (LAMs) can capture manipulation priors from action-free videos, but discrete latents usually impose a granularity bottleneck, lim
arXiv:2610.14885v1 Announce Type: new Abstract: Test-time compute (TTC), which scores multiple candidate actions at inference time, is a powerful way to improve the execution performance of vision-lan
Quartermaster (my local inference manager: one server that swaps llama.cpp, image, audio and video models in and out of VRAM) can now help you server and manage decision models: Auto-discovered. Drop
I’ve been building SonicBoom, a native tensor runtime written in C++/CUDA with a C ABI and no Python anywhere in the stack. I just tagged v0.0.1, and the first target is local LLM inference on consume
arXiv:2510.22809v2 Announce Type: replace-cross Abstract: Traditional machine learning relies on explicit models and domain assumptions, often limiting flexibility and interpretability. We introduce a
arXiv:2610.13600v1 Announce Type: cross Abstract: Large Audio Language Models (LALMs) achieve strong performance on speech-text tasks, but standard audio encoders produce hundreds of tokens for even s
llama : report compute graph reallocations during inference (#30329) Add an optional ggml_backend_sched_realloc_callback that user code registers with ggml_backend_sched_set_realloc_callback to be not
arXiv:2610.12530v1 Announce Type: new Abstract: Training-free feature reuse accelerates diffusion transformer inference by replacing full denoising computation at selected steps with cached approximat
arXiv:2610.12570v1 Announce Type: new Abstract: Speculative decoding accelerates large language model inference by using a lightweight drafter to propose multiple tokens that are subsequently verified
Enterprise AI scalability is running into a hard budget wall for some customers. As agentic workloads multiply, cloud token bills are consuming allocations faster than planned, prompting companies to
arXiv:2610.13035v1 Announce Type: cross Abstract: Multi-view representations enable pretrained image and video vision-language models to understand 3D scenes, but their long visual token sequences mak
arXiv:2610.13208v1 Announce Type: new Abstract: Time-series foundation models use the same network depth for every input, although forecasting windows differ in structure and difficulty. We test wheth
Coding agents have changed how fast teams ship, and more of the software they ship now includes agents. A production agent spans model calls, routing decisions, tools, and the infrastructure underneat
Microsoft is announcing its third-generation Maia in-house AI chip today, designed to be the 'primary AI inference system' for Microsoft's products and services. Maia 300 not only improves the cost an
arXiv:2610.13674v1 Announce Type: cross Abstract: Linearizing softmax attention offers a route to efficient Transformer inference, but preserving pretrained attention remains challenging. We identify
arXiv:2610.12879v1 Announce Type: new Abstract: Diffusion language models generate tokens in parallel, but still require multiple sampling steps to produce high-quality text. In this paper, we investi
arXiv:2610.13369v1 Announce Type: new Abstract: Classifier-free guidance (CFG) is a standard technique for improving conditional generation in diffusion models, yet its activation schedule is usually
arXiv:2610.13370v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) models have become the standard architecture for frontier LLMs. The multiple parallel experts in each transformer block theoret
arXiv:2610.13871v1 Announce Type: new Abstract: Many optimization problems provide direct access to only part of the gradient, while the remaining component must be inferred from function evaluations.
arXiv:2610.13660v1 Announce Type: cross Abstract: The Abstraction and Reasoning Corpus (ARC) is increasingly used to evaluate whether neural networks can infer task-specific transformations from few d
arXiv:2610.13300v1 Announce Type: cross Abstract: Modeling human reactions to spatial audio requires understanding both what is heard and where it comes from. Prior work generates motion offline from
What's Changed llama: update llama.cpp to b11521 for decision models and EmbeddingGemma 2 by @jmorganca in #18917 server: reuse verified manifests and loaded models in inference routes by @dhiltgen in
arXiv:2610.12561v1 Announce Type: new Abstract: Memory allows large language model (LLM) agents to use past experience, often stored as textual records, when responding to requests. For embodied agent
Basalt: Blackwell inference engine Hi all! So over the last week I've been working on a Strata fork that's heavily tuned and can achieve throughput up to 2.6x what Strata usually does on the same weig
arXiv:2610.11724v1 Announce Type: new Abstract: Multimodal meme understanding is increasingly used to analyze socially sensitive content, yet existing models often exhibit biased behavior when interpr
arXiv:2610.11853v1 Announce Type: new Abstract: Modern neural networks are heavily over-parameterized. This redundancy incurs substantial compute and memory overhead during training and inference. Exi
arXiv:2610.10859v1 Announce Type: new Abstract: Text-to-image diffusion models are increasingly distilled into few-step variants and being deployed to enable fast inference. However, their ability to
arXiv:2610.11215v1 Announce Type: new Abstract: Reconstructing complete, scene-aligned 3D objects from casual images requires integrating sparse, uncertain observations and inferring surfaces hidden b