Knowledge catalogue
arXiv:2610.18642v1 Announce Type: new Abstract: The Key-Value (KV) cache constitutes a dominant source of memory consumption during large language model (LLM) inference, particularly in long-context a
arXiv:2610.14074v1 Announce Type: new Abstract: Large language models typically support different inference budgets through separate model sizes or fixed compute configurations. In multi-turn interact
arXiv:2606.12487v2 Announce Type: replace Abstract: Post-training quantization (PTQ) is essential for efficient large language model (LLM) inference, but reliably quantizing activations remains challe
arXiv:2610.15124v1 Announce Type: new Abstract: Foundation models need to support diverse inference budgets, yet training a separate model for each budget duplicates pretraining effort. Elastic models
arXiv:2610.17530v1 Announce Type: new Abstract: We propose a unified spatial-density signal to score token importance for both inference phases: bipartite merge for prefill and KV eviction for decode,
Data sovereignty is becoming an architectural problem as enterprises push artificial intelligence inference toward their most sensitive information. Pipelines assembled from separate tools struggle to
arXiv:2610.18615v1 Announce Type: new Abstract: Visual agents must turn video observations into timely decisions to track progress and choose what to do next. Building on the emerging System One parad
arXiv:2610.15556v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models commonly generate robot actions token by token through Large Language Model (LLM) decoders, causing inference late
arXiv:2602.01709v3 Announce Type: replace Abstract: Current test-time scaling (TTS) techniques enhance large language model (LLM) performance by allocating additional computation at inference time, ye
arXiv:2610.12702v1 Announce Type: cross Abstract: Aggregating independently computed sub-sample inferences is a well-known challenge in large-scale Bayesian computation, federated learning, and opinio
arXiv:2610.13106v1 Announce Type: new Abstract: Biological function in RNA is encoded across scales, from local sequence motifs to long-range structural dependencies. Conventional RNA foundation model
arXiv:2610.12599v1 Announce Type: new Abstract: Sparse computation can scale the parametric capacity of LLMs without proportionally increasing active computation, but pretraining sparse architectures
arXiv:2610.12850v1 Announce Type: cross Abstract: Bayesian simulation-based inference (SBI) can struggle to recover increasingly concentrated posteriors as observation size grows. We consider learning
Strata vs Infernix on the same model, RTX 5090 — A/B at 262k and 512k context Hi folks, another inference engine post for your feed. I was blown away by Strata, but saw someone else post about Inferni
arXiv:2610.11152v1 Announce Type: cross Abstract: LLM agents increasingly improve at inference time by accumulating experience in context rather than by updating parameters. This process is often desc
arXiv:2610.12189v1 Announce Type: cross Abstract: Generative diffusion models are well-suited for probabilistic precipitation nowcasting, but existing approaches often rely on separately trained compr
arXiv:2510.18239v4 Announce Type: replace-cross Abstract: Scaling large recommendation systems requires advancing three major frontiers: processing longer user histories, expanding candidate sets, and
arXiv:2610.11185v1 Announce Type: new Abstract: Single-cell trajectory inference maps transcriptomic measurements onto developmental continua, yet configurations that fit the data equally well can ass
arXiv:2610.10889v1 Announce Type: new Abstract: Vision-Language (VL) reasoning requires a model to both extract relevant and accurate information from an image (visual reasoning, VR), and to infer the
arXiv:2610.11737v1 Announce Type: cross Abstract: Bayesian physics-informed neural networks (B-PINNs) are a popular framework for parameter and state inference from sparse or noisy observations. They
arXiv:2610.12367v1 Announce Type: new Abstract: Skills, reusable procedural guidance added at inference, can substantially improve LLM downstream performance (Li et al., 2026). Prior work retrieves sk
arXiv:2505.21468v2 Announce Type: replace Abstract: We present Causal Posterior Estimation (CPE), a novel method for Bayesian inference in simulator models, where evaluating the likelihood function is
arXiv:2610.10520v1 Announce Type: new Abstract: GNN-to-MLP distillation aims to retain the predictive accuracy of a message-passing teacher while deploying a graph-free MLP at inference. Existing meth
arXiv:2610.09684v1 Announce Type: new Abstract: Test-time scaling (TTS) improves the reasoning capabilities of large language models by allocating additional inference computation. Existing approaches
arXiv:2610.10453v1 Announce Type: new Abstract: Flow-based policies provide an expressive framework for continuous robot control, but their iterative ODE integration incurs substantial inference cost.
arXiv:2610.09111v1 Announce Type: new Abstract: Deterministic inference is essential for reliable and trustworthy machine learning. Prior studies of text generation have shown that changing factors su
arXiv:2610.09710v1 Announce Type: new Abstract: ANN-to-SNN conversion offers a practical route toward energy-efficient spiking Vision-Language-Action (VLA) models by bypassing the substantial cost of
arXiv:2610.10415v1 Announce Type: cross Abstract: Pretrained text-to-speech (TTS) models can generate expressive speech, but reliable inference-time emotion control remains challenging: prompts and re
arXiv:2610.09074v1 Announce Type: new Abstract: Emerging long-horizon agentic tasks require repeated model calls, worsening the inference cost of already-costly language models. While narrow agentic t
arXiv:2610.08678v1 Announce Type: cross Abstract: Speculative decoding accelerates inference for a large language model (LLM), referred to as the target model, by first using a smaller model, referred
arXiv:2610.08652v1 Announce Type: cross Abstract: Diffusion models are increasingly used as surrogates for expensive simulators in weather prediction, molecular dynamics, and materials design. In thes
arXiv:2610.04300v1 Announce Type: new Abstract: The semantic information of objects in remote sensing images is typically invariant to geometric transformations from the dihedral group D4. However, SA
arXiv:2610.05062v1 Announce Type: cross Abstract: Vision-language-action (VLA) models translate multimodal observations into low-level robot actions. During robot operation, each control period sets a
arXiv:2610.06679v1 Announce Type: new Abstract: Tabular foundation models perform in-context learning (ICL) by conditioning predictions on labeled training examples provided as context. Unlike traditi
arXiv:2610.06726v1 Announce Type: new Abstract: Electroencephalography (EEG) offers millisecond temporal resolution, but inferring underlying neural sources is a severely ill-posed spatial inverse pro
arXiv:2610.04819v1 Announce Type: new Abstract: One-step real-world image super-resolution (Real-ISR) offers efficient inference, but recovering realistic and perceptually rich details often relies on
arXiv:2610.03942v1 Announce Type: new Abstract: When using diffusion models to target image restoration problems, diffusion inversion is typically employed to retain relevant image information from th
arXiv:2610.04283v1 Announce Type: new Abstract: Activation steering exploits interpretable directions in the residual stream to enable inference-time manipulation of model behavior. Composing steering
arXiv:2610.04255v1 Announce Type: cross Abstract: Robotic manipulation often requires inferring task-relevant states from past interactions when the current observation alone is insufficient to determ
arXiv:2610.04703v1 Announce Type: cross Abstract: Recently proposed Drifting Models shift iterative distribution refinement from inference to training, enabling effective one-step generation. However,
arXiv:2610.06813v1 Announce Type: new Abstract: Recent progress in 3D foundation models has enabled rapid 3D reconstruction and camera calibration by leveraging learned 3D priors from vast amount of s
arXiv:2610.05538v1 Announce Type: new Abstract: We introduce Loop Flow Transformers (LiFT), a family of looped generative models that scales computation by repeatedly applying a shared Diffusion Trans
arXiv:2610.05578v1 Announce Type: cross Abstract: Multi-camera streaming perception is increasingly deployed on heterogeneous edge platforms shared with co-resident workloads, yet accelerator placemen
arXiv:2610.05664v1 Announce Type: new Abstract: Vision-and-Language Navigation (VLN) policies increasingly benefit from strong semantic priors provided by large vision-language models (VLMs). However,
arXiv:2610.04335v1 Announce Type: new Abstract: Remote pose estimation of noncooperative Unmanned Aerial Vehicles (UAVs) from imagery is critical, as they cannot be influenced or instrumented in advan
arXiv:2602.10370v3 Announce Type: replace-cross Abstract: Instrumental variable (IV) methods mitigate bias from unobserved confounding in observational causal inference but rely on the availability of
Someone who has always loved looking at the space between different AI techniques, this time I went a little deeper into the memory trade-offs between RNNs, Transformers and SSMs. I found it interesti
arXiv:2610.05025v1 Announce Type: new Abstract: Dual-system Vision-Language-Action (VLA) models improve real-time robotic control by pairing a slow, reasoning-capable generalist with a fast specialist
arXiv:2610.02439v1 Announce Type: new Abstract: Generative graph models are central to understanding and simulating complex networks. However, existing approaches have complementary strengths and limi
Anthropic’s paid subscription plans deliver more than five‑times the perceived value of comparable OpenAI offerings when evaluated by cost per inference unit and overall revenue contribution. The heav
arXiv:2610.02762v1 Announce Type: new Abstract: Dynamic LLM routers promise to cut inference costs by sending each query to the cheapest model that can answer it correctly. We analyze six commercial r
arXiv:2610.02766v1 Announce Type: new Abstract: Retaining language-model prefix states trades recomputation against storage time. Optimizing each cached block independently can overcount savings: a re
arXiv:2610.02675v1 Announce Type: cross Abstract: Single-photon avalanche diode (SPAD) cameras operate fundamentally differently from conventional cameras due to their photon-counting nature. Each fra
Hi folks, LIVE PROJECT PAGE TLDR: Moral of the story. You need better CPU to do actual agentic coding doing real work... I've been on a mission to make my RTX5090 go brrr for past 2 months so much so
arXiv:2509.10756v4 Announce Type: replace-cross Abstract: We show that ensembles of deep neural networks, called deep ensembles, can be used to perform quantum parameter estimation while also providin
The 'old' version of Cowork runs model inference in the cloud, executing tool calls in an Anthropic-provided VM we shipped to your computer. We added the VM for capability, safety, and security reason
arXiv:2610.02491v1 Announce Type: new Abstract: Large language models spend the same amount of computation on every token they generate, regardless of how difficult each token is to produce. Methods s
Hi everyone, Sharing a personal project exploring the minimal bare-metal footprint required to run an autoregressive LLM. Instead of relying on large runtimes or compiler abstractions, I wrote an infe
arXiv:2610.01054v1 Announce Type: cross Abstract: In-context learning (ICL) enables language models to perform new tasks from demonstrations without weight updates. However, every ICL inference requir
arXiv:2610.01133v1 Announce Type: cross Abstract: Scaling reasoning typically spends more compute on reinforcement learning (RL) or on inference. We show that a completed RL training history can yield