Search: “inference”

Real-Time Human Reconstruction and Animation using Feed-Forward Gaussian Splatting

arXiv:2604.10259v1 Announce Type: new Abstract: We present a generalizable feed-forward Gaussian splatting framework for human 3D reconstruction and real-time animation that operates directly on multi

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Research

Reason Only When Needed: Efficient Generative Reward Modeling via Model-Internal Uncertainty

DGX agent

arXiv:2604.10072v1 Announce Type: new Abstract: Recent advancements in the Generative Reward Model (GRM) have demonstrated its potential to enhance the reasoning abilities of LLMs through Chain-of-Tho

researcharxiv-cs-cl

14 Apr 2026

SatReg: Regression-based Neural Architecture Search for Lightweight Satellite Image Segmentation

arXiv:2604.10306v1 Announce Type: new Abstract: As Earth-observation workloads move toward onboard and edge processing, remote-sensing segmentation models must operate under tight latency and energy c

hardwarearxiv-cs-cv

Self-Calibrating Language Models via Test-Time Discriminative Distillation

arXiv:2604.09624v1 Announce Type: new Abstract: Large language models (LLMs) are systematically overconfident: they routinely express high certainty on questions they often answer incorrectly. Existin

Sharpness-Aware Surrogate Training for On-Sensor Spiking Neural Networks

arXiv:2604.09696v1 Announce Type: cross Abstract: Spiking neural networks (SNNs) are a natural computational model for on-sensor and near-sensor vision, where event driven processors must operate unde

Simple but Stable, Fast and Safe: Achieve End-to-end Control by High-Fidelity Differentiable Simulation

model-releasesarxiv-cs-ro

arXiv:2604.10548v1 Announce Type: new Abstract: Obstacle avoidance is a fundamental vision-based task essential for enabling quadrotors to perform advanced applications. When planning the trajectory,

Stability of a Generalized Debiased Lasso with Applications to Resampling-Based Variable Selection

arXiv:2405.03063v2 Announce Type: replace-cross Abstract: We propose a generalized debiased Lasso estimator based on a stability principle. When a single column of the design matrix is perturbed, the

Steered LLM Activations are Non-Surjective

arXiv:2604.09839v1 Announce Type: new Abstract: Activation steering is a popular white-box control technique that modifies model activations to elicit an abstract change in output behavior. It has als

safetyarxiv-cs-ai

StreamServe: Adaptive Speculative Flows for Low-Latency Disaggregated LLM Serving

arXiv:2604.09562v1 Announce Type: cross Abstract: Efficient LLM serving must balance throughput and latency across diverse, bursty workloads. We introduce StreamServe, a disaggregated prefill decode s

hardwarearxiv-cs-ai

Structural Gating and Effect-aligned Lag-resolved Temporal Causal Discovery Framework with Application to Heat-Pollution Extremes

arXiv:2604.10371v1 Announce Type: new Abstract: This study proposes Structural Gating and Effect-aligned Discovery for Temporal Causal Discovery (SGED-TCD), a novel and general framework for lag-resol

safetyarxiv-cs-lg

Structured Causal Video Reasoning via Multi-Objective Alignment

arXiv:2604.04415v2 Announce Type: replace Abstract: Human understanding of video dynamics is typically grounded in a structured mental representation of entities, actions, and temporal relations, rath

safetyarxiv-cs-cl

TaleDiffusion: Multi-Character Story Generation with Dialogue Rendering

arXiv:2509.04123v2 Announce Type: replace Abstract: Text-to-story visualization is challenging due to the need for consistent interaction among multiple characters across frames. Existing methods stru

Test-time Scaling over Perception: Resolving the Grounding Paradox in Thinking with Images

arXiv:2604.11025v1 Announce Type: new Abstract: Recent multimodal large language models (MLLMs) have begun to support Thinking with Images by invoking visual tools such as zooming and cropping during

Towards Green Wearable Computing: A Physics-Aware Spiking Neural Network for Energy-Efficient IMU-based Human Activity Recognition

arXiv:2604.10458v1 Announce Type: cross Abstract: Wearable IMU-based Human Activity Recognition (HAR) relies heavily on Deep Neural Networks (DNNs), which are burdened by immense computational and buf

Transactional Attention: Semantic Sponsorship for KV-Cache Retention

arXiv:2604.11288v1 Announce Type: new Abstract: At K=16 tokens (0.4% of a 4K context), every existing KV-cache compression method achieves 0% on credential retrieval. The failure mode is dormant token

Local Ai

UHD-GPGNet: UHD Video Denoising via Gaussian-Process-Guided Local Spatio-Temporal Modeling

arXiv:2604.11014v1 Announce Type: new Abstract: Ultra-high-definition (UHD) video denoising requires simultaneously suppressing complex spatio-temporal degradations, preserving fine textures and chrom

local-aiarxiv-cs-cv

Visual Enhanced Depth Scaling for Multimodal Latent Reasoning

arXiv:2604.10500v1 Announce Type: new Abstract: Multimodal latent reasoning has emerged as a promising paradigm that replaces explicit Chain-of-Thought (CoT) decoding with implicit feature propagation

safetyarxiv-cs-cv

Warm-Started Reinforcement Learning for Iterative 3D/2D Liver Registration

arXiv:2604.10245v1 Announce Type: new Abstract: Registration between preoperative CT and intraoperative laparoscopic video plays a crucial role in augmented reality (AR) guidance for minimally invasiv

safetyarxiv-cs-cv

Woosh: A Sound Effects Foundation Model

model-releasesarxiv-cs-ai

arXiv:2604.01929v2 Announce Type: replace-cross Abstract: The audio research community depends on open generative models as foundational tools for building novel approaches and establishing baselines.

Zero-shot World Models Are Developmentally Efficient Learners

arXiv:2604.10333v1 Announce Type: new Abstract: Young children demonstrate early abilities to understand their physical world, estimating depth, motion, object coherence, interactions, and many other

A Little Rank Goes a Long Way: Random Scaffolds with LoRA Adapters Are All You Need

arXiv:2604.08749v1 Announce Type: new Abstract: How many of a neural network's parameters actually encode task-specific information? We investigate this question with LottaLoRA, a training paradigm in

Accurate and Reliable Uncertainty Estimates for Deterministic Predictions Extensions to Under and Overpredictions

arXiv:2604.08755v1 Announce Type: cross Abstract: Computational models support high-stakes decisions across engineering and science, and practitioners increasingly seek probabilistic predictions to qu

Attention-Based Sampler for Diffusion Language Models

arXiv:2604.08564v1 Announce Type: new Abstract: Auto-regressive models (ARMs) have established a dominant paradigm in language modeling. However, their strictly sequential decoding paradigm imposes fu

Boosting Brain-inspired Path Integration Efficiency via Learning-based Replication of Continuous Attractor Neurodynamics

arXiv:2511.17687v2 Announce Type: replace Abstract: The brain's Path Integration (PI) mechanism offers substantial guidance and inspiration for Brain-Inspired Navigation (BIN). However, the PI capabil

Agents

Building Better Environments for Autonomous Cyber Defence

arXiv:2604.08805v1 Announce Type: cross Abstract: In November 2025, the authors ran a workshop on the topic of what makes a good reinforcement learning (RL) environment for autonomous cyber defence (A

agentsarxiv-cs-ai

Conformal Prediction in Hierarchical Classification with Constrained Representation Complexity

arXiv:2501.19038v3 Announce Type: replace-cross Abstract: Conformal prediction has emerged as a widely used framework for constructing valid prediction sets in classification and regression tasks. In

DDSP-QbE++: Improving Speech Quality for Speech Anonymisation for Atypical Speech

arXiv:2604.09246v1 Announce Type: cross Abstract: Differentiable Digital Signal Processing (DDSP) pipelines for voice conversion rely on subtractive synthesis, where a periodic excitation signal is sh

Tutorials

Degradation-Robust Fusion: An Efficient Degradation-Aware Diffusion Framework for Multimodal Image Fusion in Arbitrary Degradation Scenarios

arXiv:2604.08922v1 Announce Type: new Abstract: Complex degradations like noise, blur, and low resolution are typical challenges in real world image fusion tasks, limiting the performance and practica

tutorialsarxiv-cs-cv

Demystifying the Silence of Correctness Bugs in PyTorch Compiler

arXiv:2604.08720v1 Announce Type: cross Abstract: Performance optimization of AI infrastructure is key to the fast adoption of large language models (LLMs). The PyTorch compiler (torch.compile), a cor

Differentially Private and Federated Structure Learning in Bayesian Networks

arXiv:2512.01708v2 Announce Type: replace-cross Abstract: Learning the structure of a Bayesian network from decentralized data poses two major challenges: (i) ensuring rigorous privacy guarantees for

Dual Mamba for Node-Specific Representation Learning: Tackling Over-Smoothing with Selective State Space Modeling

arXiv:2511.06756v3 Announce Type: replace Abstract: Over-smoothing remains a fundamental challenge in deep Graph Neural Networks (GNNs), where repeated message passing causes node representations to b

Agents

EchoTrail-GUI: Building Actionable Memory for GUI Agents via Critic-Guided Self-Exploration

arXiv:2512.19396v3 Announce Type: replace Abstract: Contemporary GUI agents, while increasingly capable due to advances in Large Vision-Language Models (VLMs), often operate with a critical limitation

agentsarxiv-cs-ai

Efficient Hierarchical Implicit Flow Q-learning for Offline Goal-conditioned Reinforcement Learning

arXiv:2604.08960v1 Announce Type: new Abstract: Offline goal-conditioned reinforcement learning (GCRL) is a practical reinforcement learning paradigm that aims to learn goal-conditioned policies from

ELT: Elastic Looped Transformers for Visual Generation

model-releasesarxiv-cs-cv

arXiv:2604.09168v1 Announce Type: new Abstract: We introduce Elastic Looped Transformers (ELT), a highly parameter-efficient class of visual generative models based on a recurrent transformer architec

EquiformerV3: Scaling Efficient, Expressive, and General SE(3)-Equivariant Graph Attention Transformers

arXiv:2604.09130v1 Announce Type: cross Abstract: As SE(3)-equivariant graph neural networks mature as a core tool for 3D atomistic modeling, improving their efficiency, expressivity, and physical con

Evolutionary Optimization Trumps Adam Optimization on Embedding Space Exploration

arXiv:2511.03913v2 Announce Type: replace-cross Abstract: Deep diffusion models have revolutionized image generation by producing high-quality outputs. However, achieving specific objectives with thes

safetyarxiv-cs-ai

Facet-Level Tracing of Evidence Uncertainty and Hallucination in RAG

model-releasesarxiv-cs-cl

arXiv:2604.09174v1 Announce Type: new Abstract: Retrieval-Augmented Generation (RAG) aims to reduce hallucination by grounding answers in retrieved evidence, yet hallucinated answers remain common eve

Fast Model-guided Instance-wise Adaptation Framework for Real-world Pansharpening with Fidelity Constraints

arXiv:2604.08903v1 Announce Type: new Abstract: Pansharpening aims to generate high-resolution multispectral (HRMS) images by fusing low-resolution multispectral (LRMS) and high-resolution panchromati

hardwarearxiv-cs-cv

FlashLips: 100-FPS Mask-Free Latent Lip-Sync using Reconstruction Instead of Diffusion or GANs

arXiv:2512.20033v2 Announce Type: replace Abstract: We present FlashLips, a two-stage, mask-free lip-sync system that decouples lips control from rendering and achieves real-time performance, with our

hardwarearxiv-cs-cv

Geometry Reinforced Efficient Attention Tuning Equipped with Normals for Robust Stereo Matching

arXiv:2604.09142v1 Announce Type: new Abstract: Despite remarkable advances in image-driven stereo matching over the past decade, Synthetic-to-Realistic Zero-Shot (Syn-to-Real) generalization remains

Growing a Multi-head Twig via Distillation and Reinforcement Learning to Accelerate Large Vision-Language Models

arXiv:2503.14075v3 Announce Type: replace-cross Abstract: Large vision-language models (VLMs) have demonstrated remarkable capabilities in open-world multimodal understanding, yet their high computati

HD-VGGT: High-Resolution Visual Geometry Transformer

arXiv:2603.27222v2 Announce Type: replace Abstract: High-resolution imagery is essential for accurate 3D reconstruction, as many geometric details only emerge at fine spatial scales. Recent feed-forwa

Imitation Learning for Combinatorial Optimisation under Uncertainty

arXiv:2601.05383v4 Announce Type: replace Abstract: Imitation learning (IL) provides a data-driven framework for approximating policies for large-scale combinatorial optimisation problems formulated a

Integrated electro-optic attention nonlinearities for transformers

arXiv:2604.09512v1 Announce Type: new Abstract: Transformers have emerged as the dominant neural-network architecture, achieving state-of-the-art performance in language processing and computer vision

Iterative Identification Closure: Amplifying Causal Identifiability in Linear SEMs

arXiv:2604.09309v1 Announce Type: cross Abstract: The Half-Trek Criterion (HTC) is the primary graphical tool for determining generic identifiability of causal effect coefficients in linear structural

Local Ai

LADR: Locality-Aware Dynamic Rescue for Efficient Text-to-Image Generation with Diffusion Large Language Models

arXiv:2603.13450v2 Announce Type: replace-cross Abstract: Discrete Diffusion Language Models have emerged as a compelling paradigm for unified multimodal generation, yet their deployment is hindered b

local-aiarxiv-cs-cl

Applications

Matrix-Game 3.0: Real-Time and Streaming Interactive World Model with Long-Horizon Memory

arXiv:2604.08995v1 Announce Type: new Abstract: With the advancement of interactive video generation, diffusion models have increasingly demonstrated their potential as world models. However, existing

applicationsarxiv-cs-cv

Agents

Multi-Agent Decision-Focused Learning via Value-Aware Sequential Communication

arXiv:2604.08944v1 Announce Type: new Abstract: Multi-agent coordination under partial observability requires agents to share complementary private information. While recent methods optimize messages

agentsarxiv-cs-lg