Search: “inference”

arXiv:2608.01885v1 Announce Type: new Abstract: While deep learning models have achieved state-of-the-art performance in time series anomaly detection, their complex architectures incur substantial in

applicationsarxiv-cs-lg

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Agents

Practical Online KV Cache Compaction for LLM Agents: An Empirical Study

DGX agent

arXiv:2608.00902v1 Announce Type: new Abstract: LLM agents accumulate long trajectories of reasoning steps, tool calls, and environment feedback, making the KV cache a major inference bottleneck. KV c

agentsarxiv-cs-cl

4 Aug 2026

RamanPFN: learning from Raman spectral structure with a tabular foundation model

model-releasesarxiv-cs-lg

arXiv:2608.02157v1 Announce Type: new Abstract: Raman spectroscopy enables non-destructive, label-free molecular characterization across materials science, biomedicine and process monitoring. Predicti

RSC-GestureNet: Reliability-Aware Selective Causal Recognition of Chinese Traffic Police Gestures

arXiv:2608.02200v1 Announce Type: new Abstract: Traffic police gestures are safety-critical perception cues for autonomous driving. A deployable recognizer must infer commands causally from continuous

ScrambleToolBench: Agents Search Exhaustively Even When Their Own Map Points to the Next Step

model-releasesarxiv-cs-cl

arXiv:2608.02358v1 Announce Type: new Abstract: To operate robustly in open-world environments, autonomous agents should be able to infer the behavior of unfamiliar systems through interaction alone,

UniSim-SLAM: Feed-Forward SLAM with Unified Sim(3) Optimization

arXiv:2608.01706v1 Announce Type: new Abstract: Recent geometric foundation models enable feed-forward inference for SLAM, but their predictions are strongly dependent on the input view set, which lea

researcharxiv-cs-cv

ResKV: Reconstructing Omitted Attention Contributions for Fixed-Budget KV Cache Compression

arXiv:2607.29591v1 Announce Type: new Abstract: KV cache compression is essential for efficient long-context inference. Existing eviction methods permanently discard unselected tokens and consequently

researcharxiv-cs-cl

3 Aug 2026

Safe Vision Language Action Models via Barrier Enhanced Flow Matching

model-releasesarxiv-cs-ro

arXiv:2607.29569v1 Announce Type: new Abstract: This article presents a modular inference framework that integrates Flow Matching generative models with formal Control Barrier Function (CBF) safety gu

3 Aug 2026

Hardware

AgenticCANN: Automated Ascend C Operator Generation via Knowledge-Augmented Agentic Evolution

arXiv:2607.26661v1 Announce Type: new Abstract: Ascend C operator optimization is critical for NPU (Neural Processing Unit) inference performance but requires deep hardware expertise.While large langu

hardwarearxiv-cs-ai

Calibrate Before Reason: Robust Visual Token Reduction against Semantic Drift in VLMs

arXiv:2607.27700v1 Announce Type: new Abstract: Large Vision-Language Models (VLMs) suffer from prohibitive inference overhead due to long sequences of visual tokens. However, existing visual token re

researcharxiv-cs-cv

Corrigible Assistance in One Round: Pragmatic-Pedagogic Best Response

arXiv:2607.27508v1 Announce Type: new Abstract: Assistance games formalize human-robot collaboration under asymmetric information: the human knows the goal, while the robot must infer it from observat

safetyarxiv-cs-ro

Tutorials

PhiZero: A World Model Built Around Physical Language

arXiv:2607.28624v1 Announce Type: new Abstract: We introduce PhiZero, a physical world model built around physical language, a compact discrete representation of world-state transitions. Existing phys

tutorialsarxiv-cs-cv

Hardware

ROCS: Request-Oriented Compute Sharing for Efficient Large-Scale Recommendation

arXiv:2607.27744v1 Announce Type: new Abstract: Modern recommendation models gain prediction quality by scaling feature-interaction and sequence modules, but production cost constraints cap how far sy

hardwarearxiv-cs-lg

SVR: Self-Verifying Refinement via Joint Verdict-Confidence Reinforcement Learning for Adaptive Test-Time Compute

arXiv:2607.28457v1 Announce Type: cross Abstract: Scaling test-time computation can improve language-model reasoning, but uniform budgets waste computation on easy inputs, while verifier-guided refine

safetyarxiv-cs-cl

Tycho: Active Abstraction with Programmatic World Models for ARC-AGI-3

arXiv:2607.28287v1 Announce Type: cross Abstract: ARC-AGI-3 turns abstraction into an interactive problem of skill acquisition. A player must infer an unfamiliar game's rules, hidden state, and goal w

Write-Safe Flow Field Mapping under Ambiguous Onboard Sensing and Localization Drift

arXiv:2607.27713v1 Announce Type: new Abstract: Mobile robots can infer local flow structure from onboard sensing, but a locally plausible estimate is not always safe to write into a global map. Simil

local-aiarxiv-cs-ro

ARC-Encoder: learning compressed text representations for large language models

model-releasesarxiv-cs-cl

arXiv:2510.20535v2 Announce Type: replace Abstract: Recent techniques such as retrieval-augmented generation or chain-of-thought reasoning have led to longer contexts and increased inference costs. Co

30 Jul 2026

Hardware

BATS: Resource-Efficient Volumetric Segmentation with Boundary-Aware Mixed-Resolution Tokens

arXiv:2607.26829v1 Announce Type: new Abstract: Many high-performing volumetric segmentation models maintain dense multi-scale feature maps, leading to high activation memory and inference cost. We pr

hardwarearxiv-cs-cv

30 Jul 2026

Minimizing Targeted Activations: Input-Only Suppression of Evaluation-Awareness Latents in Large Language Models

model-releasesarxiv-cs-ai

arXiv:2607.25907v1 Announce Type: cross Abstract: Activation steering controls model behavior by editing internal activations at inference time. We study its input-side dual: optimizing a fluent promp

Neuromorphic Diffusion Language Models: Addressing Compute and Memory Bottlenecks via Sparsity and Block Denoising

model-releasesarxiv-cs-cl

arXiv:2607.24841v1 Announce Type: new Abstract: Autoregressive (AR) large language models (LLMs) are inherently inefficient at inference time because each generated token requires accessing the full s

Stemma: Induced Decision Regions Reveal LLM Provenance

arXiv:2607.25880v1 Announce Type: cross Abstract: LLM provenance testing asks whether a suspect LLM belongs to the same lineage as a source. Existing black-box methods largely infer this relationship

Towards real-time surrogate-free Bayesian inversion for neutron reflectometry

arXiv:2509.06924v3 Announce Type: replace Abstract: Neutron reflectometry (NR) is a key enabling technology for many areas of scientific development. Although the forward reflectivity model is well-kn

researcharxiv-cs-lg

Agents

Beyond Success Rate: Cost-Aware Evaluation of Offensive and Defensive Security Agents

arXiv:2607.15263v3 Announce Type: replace-cross Abstract: Security-agent evaluations commonly measure peak offensive capability under generous inference budgets, emphasizing vulnerability discovery, e

agentsarxiv-cs-ai

CADER: Confidence-Aware Dynamic Evidence Reasoning for Long-Video Understanding

arXiv:2607.24582v1 Announce Type: cross Abstract: Long-video understanding increasingly relies on large vision-language models and tool-augmented reasoning, but most systems apply the same inference p

local-aiarxiv-cs-ai

Intuitionistic j-Do-Calculus in Topos Causal Models

arXiv:2510.17944v2 Announce Type: replace-cross Abstract: In this paper, we generalize Pearl's do-calculus to an Intuitionistic setting called j-stable causal inference inside a topos of sheaves. Our

local-aiarxiv-cs-ai

LIBMoE: A Library for comprehensive benchmarking Mixture of Experts in Large Language Models

model-releasesarxiv-cs-ai

arXiv:2411.00918v5 Announce Type: replace-cross Abstract: Mixture of experts (MoE) architectures have become a cornerstone for scaling up and are a key component in most large language models such as

Tools

Many AI tools rent intelligence. Kimi K3 on Fireworks is different. Fine-tune it on your own data, serve from US-hosted endpoints, and own t…

Many AI tools rent intelligence. Kimi K3 on Fireworks is different. Fine-tune it on your own data, serve from US-hosted endpoints, and own the weights. Zero data retention. ICYMI yesterday, start buil

toolsfireworks-ai--x

microsoft/VibeVoice-ASR-BitNet

VibeVoice-ASR-BitNet is a compressed variant of VibeVoice-ASR optimized for real-time inference on edge CPUs — no GPU required. Through heterogeneous quantization, the model is compressed from 4.62 GB

local-air-localllama

Tools

The apps you use swap AI models constantly. Doing that without downtime or a bad rollout reaching users is still often manual. We rebuilt th…

The apps you use swap AI models constantly. Doing that without downtime or a bad rollout reaching users is still often manual. We rebuilt that workflow based on what we’ve learned serving more than 40

toolstogether-ai--x

xMIx: High-Performance Serving-Time Platform for Mechanistic Interpretability Apps

arXiv:2607.22595v1 Announce Type: new Abstract: Mechanistic interpretability (MI) has emerged as a powerful approach for analyzing and intervening in inference computations, with a growing number of a

safetyarxiv-cs-ai

Tools

You can now fine-tune Kimi K3 on Fireworks. Conduct supervised fine-tuning, preference tuning, and reinforcement learning via Training API. …

You can now fine-tune Kimi K3 on Fireworks. Conduct supervised fine-tuning, preference tuning, and reinforcement learning via Training API. Run across dedicated, and serverless training. The first ope

toolsfireworks-ai--x

The Lift Spectrum: How Measurement-to-Space Adaptivity Shapes Robustness in Image-Free Single-Pixel Sensing

arXiv:2607.22077v1 Announce Type: cross Abstract: Single-pixel sensing encodes a scene as a short sequence of coded measurements, and image-free methods infer the task directly from that sequence. Rem

27 Jul 2026

Tutorials

Visual Saliency Steering Distillation for Multimodal Chain-of-Thought Reasoning

arXiv:2607.22013v1 Announce Type: new Abstract: Multimodal chain-of-thought (CoT) reasoning integrates visual and textual cues through step-by-step inference. In small models with limited token budget

tutorialsarxiv-cs-cv

27 Jul 2026

I released Inflect v2: two ultra-tiny complete TTS models under 4M and 10M parameters

model-releasesr-localllama

I’ve spent the past month trying to find the point where an extremely small TTS model stops feeling like a size experiment and starts feeling genuinely useful. Today I’m releasing Inflect v2, with two

25 Jul 2026

Adaptive Depth Sparse Framework: Similarity-Driven Resource Allocation for Pre-Trained LLMs

arXiv:2607.21291v1 Announce Type: new Abstract: Large language models (LLMs) achieve strong generation and reasoning performance, but the Transformer architecture incurs high inference cost. Existing

safetyarxiv-cs-cl

Beyond Independent Optimization: Compression, MoE Routing, and Quantization Interactions in Multimodal Edge Intelligence

arXiv:2607.20981v1 Announce Type: new Abstract: Efficient multimodal inference is increasingly constrained not only by model quality or FLOP count, but also by the cost of preserving, moving, routing,

Applications

Generative Bayesian Filtering for State Estimation

arXiv:2607.20521v1 Announce Type: new Abstract: The state of a dynamic system evolves over time, switching among several latent modes that govern its observable behavior. Filtering methods infer the l

applicationsarxiv-cs-lg

Test-Time Scaling via Error Localization

arXiv:2607.21453v1 Announce Type: new Abstract: Scaling inference-time computation has emerged as a reliable method to improve the performance of large language models on complex reasoning and program

local-aiarxiv-cs-lg

VibeVoice-ASR-BitNet Technical Report

arXiv:2607.21075v1 Announce Type: cross Abstract: We present VibeVoice-ASR-BitNet, a compressed variant of VibeVoice-ASR optimized for real-time inference on edge CPUs. We apply heterogeneous quantiza

local-aiarxiv-cs-cl

HijackKV: New Threat in Position-Independent KV Cache Reuse

arXiv:2607.19957v1 Announce Type: cross Abstract: Key-Value (KV) cache reduces inference latency in large language models (LLMs). Traditional prefix-based reuse has low cache hit rates across inferenc

MeetingToM: Evaluating Multimodal LLMs on Theory-of-Mind Reasoning in Multi-Party Meetings

arXiv:2607.19235v1 Announce Type: cross Abstract: Theory of Mind (ToM), the ability to infer other's beliefs, intentions, and states of knowledge, is central to social interaction, yet remains challen

No Training, Better Flights: Test-Time Scaled VLMs for UAV Navigation

arXiv:2607.19288v1 Announce Type: new Abstract: Test-time scaling offers a promising method to improve the inference performance of Vision-Language Models (VLMs) without additional training. Existing

safetyarxiv-cs-cv

Privileged Lesion-Context Relational Distillation for Mask-Free Skin Lesion Classification

arXiv:2607.18773v1 Announce Type: new Abstract: Accurate skin lesion classification can benefit from lesion segmentation masks, but requiring masks or an auxiliary segmentation model during inference

safetyarxiv-cs-cv

Toward Seasonal Guidelines for Robust Deep-Learning Sentinel-2 Building Detection in Different Area Types

arXiv:2607.19994v2 Announce Type: new Abstract: Sentinel-2 imagery offers open access, global coverage, and frequent revisit times, making it attractive for practical building mapping at scale; howeve

researcharxiv-cs-cv

MASPRM: Multi-Agent System Process Reward Model

arXiv:2510.24803v3 Announce Type: replace-cross Abstract: Inference-time search over multi-agent systems (MAS) wastes compute when it cannot identify which agent's intermediate message advanced progre

safetyarxiv-cs-ai

16 Jul 2026

A Replicate-and-Quantize Strategy for Plug-and-Play Load Balancing of Sparse Mixture-of-Experts LLMs

arXiv:2602.19938v2 Announce Type: replace Abstract: Sparse Mixture-of-Experts (SMoE) architectures are increasingly used to scale large language models efficiently, delivering strong accuracy under fi

researcharxiv-cs-lg

15 Jul 2026

Chipmaker Axelera releases Voyager Wingman to speed edge AI development

model-releasessiliconangle

Edge artificial intelligence chip company Axelera AI B.V. today publicly released Voyager Wingman, an AI assistant that lets developers build and troubleshoot applications for its edge chips by typing

15 Jul 2026

Rethinking Reward Models for Multi-Domain Test-Time Scaling

arXiv:2510.00492v3 Announce Type: replace Abstract: The reliability of large language models (LLMs) during test-time scaling is often assessed with external verifiers or reward models that distinguish