Search: “inference”

CACHEFORGE: LLM-Guided End-to-End Generative Cache Replacement Policy for Performance and Hardware Efficiency

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

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

Cross-Lingual Activation Steering for Multilingual Language Models

DGX agent

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

safetyarxiv-cs-ai

7 Oct 2026

CtrlCache: Accelerating Interactive Video World Models with Control-Aware Caching

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

safetyarxiv-cs-cv

Applications

Event-Driven Proactive Robot Assistance through Vision-Language Reasoning

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,

applicationsarxiv-cs-ro

Harness Engineering for Software Engineering via Modular Executable Dev-Primitives

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

Image-text retrieval with EmbeddingGemma 2's vision tower, running in the browser on WebGPU

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

local-air-localllama

Applications

Learning PDE solution operators with variable initial conditions via Latent Dynamics Networks

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

applicationsarxiv-cs-lg

Learning What to Distill: Bilevel Top-K Token Selection for Self-Distillation in Large Language Models

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

safetyarxiv-cs-lg

Applications

MIM-VLA: Learning Physical Interaction Representations from Gripper Motor Feedback

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

applicationsarxiv-cs-ro

omni-d1 600M by Liquid AI running in the browser using WebGPU

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

local-air-localllama

Optimization Encoders: Rethinking Second-Order Meta-Learning for Neural Fields

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

safetyarxiv-cs-cv

Hardware

OSFP4: Joint Optimization of Diagonal Smoothing and Block Scales for NVFP4 Quantization

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

hardwarearxiv-cs-ai

PolarScale: A Physics-Grounded Benchmark for Radiometrically Consistent RGB-to-Stokes Estimation

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

ReFold: Training-Free Reversible Inter-Turn Context Folding for Long-Horizon Agents

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

Tutorials

ROMA: LLM System for Real-World Object-Centric Multi-Sensory Active Perception

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

tutorialsarxiv-cs-cv

SSR: Sparse Segment Reduction for Ternary GEMM Acceleration

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

Agents

Stateless Language Agents: Scaling Long-Horizon Automated Research

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

agentsarxiv-cs-ai

Uncertainty in Representation Learning on Knowledge Graphs

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

UniPro: Unified Multi-Mode Medical Image Segmentation from 2D Images to 3D Volumes via Propagation

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

researcharxiv-cs-cv

Variational Physics-Informed Ansatz for Reconstructing Hidden Interaction Networks from Steady States

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

Adaptive-Shot Hybrid Quantum Anomaly Detection for Tactile Internet Security: Reliability-Aware Measurement Allocation Under Resource Constraints

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

Are We Measuring Anticipation? Auditing Privileged Information in Procedural Video Evaluation

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

Causal Lag Structure Discovery in Confounded Time Series via Orthogonalized Adaptive Estimation

arXiv:2610.05618v1 Announce Type: cross Abstract: Finding which variables cause which others in multivariate time series, and at what lags, is central to science and policy, yet existing methods force

Efficient Test-time Adaptation through Candidate Verification and Divergence Shifts

arXiv:2610.06147v1 Announce Type: new Abstract: Vision-language models (VLMs) achieve strong zero-shot transferability but remain vulnerable to target-domain shifts at inference time. Test-time adapta

Expanding LLM Reasoning

arXiv:2610.05584v1 Announce Type: new Abstract: Extra inference compute is usually spent on sampling more reasoning chains. We study where inside an existing chain an additional continuation should be

Fine-Grained Caching for Diffusion Transformers with Few Calibration Conditions

arXiv:2512.05134v2 Announce Type: replace Abstract: Diffusion transformers require repeated denoiser evaluations, making image and video generation computationally expensive. We propose a training-fre

safetyarxiv-cs-cv

Gefen: Optimized Stochastic Optimizer

arXiv:2606.13894v3 Announce Type: replace-cross Abstract: AdamW is a default optimizer for deep learning, but its moment states add two parameter-sized buffers to training memory, increasing the cost

GPT-6.1 Sol looped 'leak' hints at nested models serving architecture

model-releasesr-localllama

Hello llamas. I am posting this because I believe that, despite it being closed source models, the discussion will bring value to the local AI community. As many of you probably heard, GPT-6 Astra is

In-Context Time Series Classification with Random Convolutional Features

arXiv:2607.19234v2 Announce Type: replace Abstract: Time series classification is central to domains such as medical signal analysis, industrial monitoring, and sensor-based activity recognition, wher

local-aiarxiv-cs-lg

Is Escalation Worth It? On the Depth of LLM Cascades

arXiv:2605.06350v2 Announce Type: replace Abstract: LLM cascades, in which a cheap model defers to an expensive one on low-confidence queries, are widely used to reduce inference cost. Given a pool of

local-aiarxiv-cs-lg

LyapuFlow: Controlling Generative Flows with Lyapunov Feedback for Inverse Problems

arXiv:2610.04326v1 Announce Type: new Abstract: Pretrained flow models are now widely used as generative priors in science and vision, where inference-time guidance enables test-time constraints witho

Hardware

Measuring and Reducing Cross-Vendor Mismatch in Language Models

arXiv:2610.05458v1 Announce Type: new Abstract: Running the same language model on different graphics processing unit (GPU) vendors can produce different logits, even when the model weights and inputs

hardwarearxiv-cs-lg

MiMo V2.6 Pro almost matched GPT-6 Astra at figuring out what a user actually wants

model-releasesr-localllama

Introducing InferBench: A benchmark testing how well frontier LLMs infer a user's priorities from their instructions. We tested 12 LLMs across 20 scenarios with 2.8k conversations to see which models

NAMVIS: Next-Scale Autoregressive Multi-View Image Synthesis

arXiv:2610.04722v1 Announce Type: new Abstract: Sparse-view novel view synthesis is a central problem in 3D content creation, but diffusion-based approaches remain limited by iterative denoising, maki

researcharxiv-cs-cv

P3: Persistent Particle Planning for Constrained Diffusion Control

arXiv:2610.06002v1 Announce Type: cross Abstract: Diffusion models provide expressive priors over trajectories, but adapting these priors to test-time constraints requires maintaining feasibility and

Physics-informed GNN for medium-high voltage AC power flow with edge-aware attention and line search correction operator

arXiv:2509.22458v3 Announce Type: replace Abstract: Physics-informed graph neural networks (PIGNNs) have emerged as fast AC power-flow solvers that can replace the classic NewtonRaphson (NR) solvers,

REBASE: Reference-Background Subspace Elimination for Training-Free In-Context Segmentation

arXiv:2607.09082v2 Announce Type: replace Abstract: Training-free in-context segmentation enables new object categories to be introduced at inference time from a single annotated reference image, elim

Hardware

Reflection AI debuts open-source Beam model with 501B parameters

Reflection AI Inc. today introduced Beam, an open-source large language model with 501 billion parameters. The launch comes a few months after the startup raised funding at a 25 billion valuation. Aro

hardwaresiliconangle

Steering by Influence: Curvature Aware Data Weighting for Activation Steering

arXiv:2610.06383v1 Announce Type: new Abstract: Inference-time steering offers cheap, fine-grained control over a language model's outputs by estimating a concept's representation in activation space

Teaching Agents to Code Reliably

arXiv:2610.03984v1 Announce Type: cross Abstract: Autonomous coding agents solve repository issues by reading code, running commands, editing files, and submitting patches. Extra inference-time comput

safetyarxiv-cs-lg

When to Switch: Reliable Action-Chunk Extension for Vision-Language-Action Models

arXiv:2610.05719v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models serve as unified policies for robotic manipulation, yet their expensive inference forces robots to pause between pol

safetyarxiv-cs-ro

AdaStep: Adaptive Step Credit Weighting for Agentic Reinforcement Learning

arXiv:2610.03223v1 Announce Type: cross Abstract: Long-horizon LLM agents are typically trained with sparse outcome rewards, making trajectory-level objectives too coarse to distinguish the contributi

local-aiarxiv-cs-cl

Automating the Application of HCI Principles: Skills for On-Demand UI Construction, the Human-AI Space to Think, and the Future of HCI

arXiv:2610.02369v1 Announce Type: cross Abstract: Human-computer interaction (HCI) is in the middle of a transition: large language models can now generate functional user interfaces (UIs) on demand f

Bridging Frontier Reasoning and Robot Execution: From Autonomous Demonstration Generation to Dense Language Supervision

arXiv:2610.03615v1 Announce Type: new Abstract: Recent advances in frontier models enable robot manipulation from only a few demonstrations, but high inference latency limits their use for real-time r

local-aiarxiv-cs-ro

Capability Scaling-Down Laws for LLM Compression

arXiv:2610.02462v1 Announce Type: cross Abstract: LLM compression reduces inference costs and memory requirements, but selecting a method and configuration remains largely empirical because comparable

researcharxiv-cs-cl

DriftTTS: Few-Step Text-to-Speech Without Distillation via Distribution-Matching Drift

arXiv:2610.03390v1 Announce Type: cross Abstract: Few-step neural text-to-speech models often rely on short- ened diffusion or flow-matching schedules, or on distillation from pretrained multi-step te

safetyarxiv-cs-ai

ENCORE: Exact Non-equilibrium COntrol with Replica Exchange for Diffusion Generation

arXiv:2610.02538v1 Announce Type: cross Abstract: Inference-time control steers a pretrained generative model towards a target distribution without retraining. We study tilted targets pi_0propto G_0,p

Fed-ADApt: Federated Anytime Depth Adaptation for Resource-Aware Medical Image Segmentation

arXiv:2610.03474v1 Announce Type: new Abstract: Federated learning (FL) enables collaborative training of medical image segmentation models without sharing raw patient data, yet existing approaches as

local-aiarxiv-cs-cv