Search: “inference”

Budget-Adaptive Routing: Skipping the Weak When the Strong Answers Anyway

arXiv:2606.30919v1 Announce Type: cross Abstract: Edge-cloud inference collaborations are often designed with a routing estimator that decides whether to offload each frame from weak models at the edg

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

CoLT: Teaching Multi-Modal Models to Think with Chain of Latent Thoughts

DGX agent

arXiv:2606.31986v1 Announce Type: new Abstract: Chain-of-thought (CoT) reasoning has enabled multi-modal large language models (MLLMs) to tackle complex visual reasoning tasks by generating explicit i

model-releasesarxiv-cs-cv

1 Jul 2026

Cross-Resolution Distribution Matching for Diffusion Distillation

arXiv:2603.06136v2 Announce Type: replace Abstract: Diffusion distillation is central to accelerating image and video generation, yet existing methods are fundamentally limited by the denoising proces

ERA: Entropy-Guided Visual Token Pruning with Rectified Attention for Efficient MLLMs

arXiv:2606.31982v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) incur prohibitive inference costs due to long visual token sequences. Training-free visual token reduction prov

safetyarxiv-cs-cv

Get started with the Claude apps gateway for Google Cloud

model-releasesgoogle-cloud-ai

Anthropic's agentic coding tool Claude Code has worked with Google Cloud for a while now. An individual developer could easily point CLAUDE_CODE_USE_VERTEX=1 at a Google Cloud (GCP) project, grant the

Agents

Model subsidies are ending. What do you do now?

Flat-rate AI plans are subsidizing agentic workloads. Learn why LLM inference costs are moving to metered pricing and how evals reveal cost per successful task. The post Model subsidies are ending. Wh

agentsarize-ai

Tools

SwiftAudio: Data-Efficient Caption-Only Distillation for One-Step Text-to-Audio Diffusion-based Generation

arXiv:2606.31259v1 Announce Type: cross Abstract: Diffusion-based text-to-audio (TTA) models achieve impressive synthesis quality but suffer from high inference latency due to iterative multi-step den

The Label Imitation Game: Turing Test Network for Zero-Shot Pseudo-Label Pruning

arXiv:2606.30875v1 Announce Type: cross Abstract: Foundation model pseudo-labeling - labeling data strictly via zero-shot inference - enables massive scale, but performance is undermined by hallucinat

After 18 months of writing, coding, and experimenting, Build a Reasoning Model (From Scratch) is finally out! My first copies just arrived! …

researchsebastian-raschka--x

After 18 months of writing, coding, and experimenting, Build a Reasoning Model (From Scratch) is finally out! My first copies just arrived! 📚 440 full-color pages. Inference scaling, reinforcement lea

b9844

llama.cpp is a C/C++ implementation for LLM inference with minimal setup and state-of-the-art performance on a wide range of hardware. Release b9844 is an intermediate build of the llama.cpp project f

b9846

Release build b9846 is an intermediate build of llama.cpp, an open-source C/C++ project for LLM inference. The llama.cpp project is the main playground for developing new features for the ggml library

b9849

b9849 is a build release from llama.cpp, which enables LLM inference in C/C++ . The release represents an intermediate build version of the llama.cpp library, a widely-used open-source project for run

Behavior Prompting Policy: Demonstrations as Prompts for Manipulation

arXiv:2606.30457v1 Announce Type: new Abstract: We study behavior prompting, a paradigm that enables robots to perform new tasks at inference time given a single human demonstration, which we call a b

safetyarxiv-cs-ro

Closing the Activation-Cone Blind Spot: Response-Time Probing and Unified Defense

model-releasesarxiv-cs-ai

arXiv:2606.29441v1 Announce Type: cross Abstract: Inference-time safety methods for large language models have proliferated, yet no systematic comparison exists. We evaluate five defense paradigms (no

Concept Removal Guidance: Evidence-Calibrated Negative Guidance for Safe Diffusion Sampling

arXiv:2606.29801v1 Announce Type: new Abstract: Text-to-image diffusion models remain vulnerable to adversarial prompts that elicit disallowed content, motivating reliable inference-time controls. A p

safetyarxiv-cs-cv

Data and Evaluation Closed-Loop for Model Capability Enhancement

model-releasesarxiv-cs-ai

arXiv:2606.28471v1 Announce Type: new Abstract: Model capability is the central variable in LLM pre-training, yet is never observed directly: data shapes it prospectively, while evaluation reveals it

How Schrödinger sped up molecular discovery by 4x with Alphaevolve

model-releasesgoogle-cloud-ai

Computational chemistry researchers have traditionally faced a frustrating trade-off when simulating molecular interactions: use fast classical force fields that sacrifice precision or rely on accurat

Invariant Reasoning Directions in Latent Trajectories of Language Models

arXiv:2606.29164v1 Announce Type: cross Abstract: Latent reasoning models perform multi-step inference directly in hidden-state space, yet the structure of these latent reasoning trajectories remains

LeAD-M3D: Leveraging Asymmetric Distillation for Real-Time Monocular 3D Detection

arXiv:2512.05663v3 Announce Type: replace Abstract: Real-time monocular 3D object detection remains challenging due to severe depth ambiguity, viewpoint shifts, and the high computational cost of 3D r

LLMs and their Limited Theory of Mind: Evaluating Mental State Annotations in Situated Dialogue

arXiv:2509.02292v2 Announce Type: replace Abstract: What if large language models could not only infer human mindsets but also expose every blind spot in team dialogue such as discrepancies in the tea

researcharxiv-cs-cl

Agents

MonoSR: Open-Vocabulary Spatial Reasoning from Monocular Images

arXiv:2511.19119v2 Announce Type: replace Abstract: Spatial reasoning (SR), the ability to infer 3D spatial information from 2D inputs, is essential for real-world applications such as embodied AI and

agentsarxiv-cs-cv

On Test-Time Scaling for Vision-Language Models

arXiv:2606.28864v1 Announce Type: new Abstract: Test-time scaling is a paradigm where large models use additional compute at inference to achieve better performance, without changing model weights. Wh

Pondering the Way: Spatial-perceiving World Action Model for Embodied Navigation

arXiv:2606.29908v1 Announce Type: cross Abstract: Existing world model-based planners for visual navigation typically follow a verification-centric paradigm, decoupling goal intent from trajectory syn

Reconsidering Overthinking: Penalizing Internal and External Redundancy in CoT Reasoning

arXiv:2508.02178v3 Announce Type: replace Abstract: Large reasoning models (LRMs) often exhibit overthinking, producing verbose Chain-of-Thought (CoT) traces that increase inference cost and obscure t

Tutorials

Residual-Guided Expert Specialization for Incomplete Multimodal Learning

arXiv:2606.30355v1 Announce Type: cross Abstract: As real-world prediction systems often face missing modalities at inference, incomplete multimodal learning (IML) remains a practical challenge. While

tutorialsarxiv-cs-ai

Resonant Brane Splatting for Arbitrary-Scale Super-Resolution

arXiv:2606.29453v1 Announce Type: cross Abstract: Arbitrary-Scale Super-Resolution (ASR) reconstructs images at continuous magnification factors. Recent methods accelerate inference by replacing compu

Scalable Synthesis of distributed LLM workloads through Symbolic Tensor Graphs

arXiv:2511.10480v3 Announce Type: replace-cross Abstract: Optimizing the performance of large language models (LLMs) on large-scale AI training and inference systems requires a scalable and expressive

Set-Inclusive Uncertainty Modeling for Robust Brain Tumor Segmentation

arXiv:2606.30374v1 Announce Type: cross Abstract: Multimodal MRI is essential for accurate brain tumor segmentation. However, acquiring all modalities at inference is often challenging in practice, wh

Speculative Pre-Positioning: Decoding Stateful Sessions to the Next Decision Point Off the Critical Path

arXiv:2606.29565v1 Announce Type: new Abstract: A stateless inference server (vLLM, SGLang, TensorRT-LLM) idles between requests while the accelerator waits; a stateful session reclaims that idle time

researcharxiv-cs-lg

Hardware

SwarmX: Agentic Scheduling for Low-Latency Agentic Systems

arXiv:2606.21401v2 Announce Type: replace-cross Abstract: Agentic AI applications compose multiple model calls and tool executions, creating new scheduling challenges for GPU-CPU clusters. Their infer

hardwarearxiv-cs-ai

Agents

TacEvo: Self-Evolving Architecture Discovery for Robotic Tactile Perception via LLM-Driven Quality-Diversity Search

arXiv:2606.30109v1 Announce Type: new Abstract: Vision-based tactile sensing converts contact-induced surface deformation into images, enabling robots to infer contact forces and fine surface textures

agentsarxiv-cs-ro

TERC: A Transfer Entropy Redundancy Criterion for State Variable Selection in Reinforcement Learning

arXiv:2401.11512v2 Announce Type: replace-cross Abstract: Identifying the most suitable variables to represent the state is a fundamental challenge in Reinforcement Learning (RL). These variables must

The Alignment Auditor: A Bayesian Framework for Verifying and Refining LLM Objectives

arXiv:2510.06096v3 Announce Type: replace-cross Abstract: The objectives that Large Language Models (LLMs) implicitly optimize remain dangerously opaque, making trustworthy alignment and auditing a gr

safetyarxiv-cs-cl

Tools

Together AI at ICML 2026: frontier research across the full stack

Together AI presented research at ICML 2026 covering advances across the full technology stack, likely including foundational model improvements, inference optimization, and practical deployment solut

toolstogether-ai-blog

Using Large Language Models as Low-Cost Statistical Estimators for Human-Response Data

arXiv:2606.30372v1 Announce Type: new Abstract: Quantitative research across the social and behavioral sciences depends on human subject experiments that are expensive, slow, and subject to sampling b

Wireless Backdoor Attack and Defense for Semantic Communications over Multiple Access Channel

arXiv:2606.30595v1 Announce Type: cross Abstract: Semantic communication (SemCom) aims to preserve semantic meaning and task-oriented information beyond conventional message recovery over wireless cha

researcharxiv-cs-lg

b9840

b9840 is a release of llama.cpp, a tool for LLM inference in C/C++. The release enables running AI models locally on Windows, Linux, and macOS. This specific build tag likely contains bug fixes, perfo

Deployment-Side Adaptiveness in Multi-Horizon Volatility Forecasting

arXiv:2606.27688v1 Announce Type: cross Abstract: In financial forecasting, predictive performance depends not only on which model is trained, but also on how the trained model is deployed. We study t

ET-SAM: Efficient Point Prompt Prediction in SAM for Unified Scene Text Detection and Layout Analysis

arXiv:2603.25168v2 Announce Type: replace Abstract: Previous works based on Segment Anything Model (SAM) have achieved promising performance in unified scene text detection and layout analysis. Howeve

NLL-Guided Full-Attention Layer Selection for Training-Free Sliding-Window Adaptation

arXiv:2606.27791v1 Announce Type: cross Abstract: Hybrid attention models that mix full and sliding-window attention across layers offer a promising approach to efficient long-context inference, but t

Perceptual 3D Simulation With Physical World Modeling

arXiv:2606.27575v1 Announce Type: new Abstract: Predicting how a scene will evolve after a desired 3D transformation from images is a central goal in vision, graphics, and robotics. Yet unlike ideal s

local-aiarxiv-cs-cv

ReFreeKV: Towards Threshold-Free KV Cache Compression

model-releasesarxiv-cs-ai

arXiv:2502.16886v4 Announce Type: replace-cross Abstract: To reduce memory consumption during LLM inference, a handful of methods have been proposed for KV cache pruning. While these techniques can ac

SpikeVLA: Vision-Language-Action Models with Spiking Neural Networks

arXiv:2606.27807v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models have become a dominant paradigm for embodied intelligence. However, most existing approaches are built on large-scal

safetyarxiv-cs-ro

Hardware

The Context-Ready Transformer

arXiv:2606.27538v1 Announce Type: cross Abstract: We introduce the context-ready transformer, a new recurrent neural network architecture built from a D-layer transformer block that pre-contextualizes

hardwarearxiv-cs-ai

Unbiased Diffusion Variational Inversion via Principled Posterior Matching

arXiv:2605.25042v2 Announce Type: replace Abstract: Existing score-based methods for inverse problems often resort to approximate minimization of the KL divergence between the inversion distribution a

VGB for Masked Diffusion Model: Efficient Test-time Scaling for Reward Satisfaction and Sample Editing

arXiv:2606.28301v1 Announce Type: new Abstract: Inference-time scaling is a promising paradigm to improve generative models, especially when outputs must satisfy structural constraints or optimize dow

researcharxiv-cs-lg

b9830

llama.cpp b9830 is a release of the open-source LLM inference project enabling local language model execution with minimal setup . The release likely includes backend improvements and optimizations fo