Search: “inference”

Towards Near-Real-Time Telemetry-Aware Routing with Neural Routing Algorithms

arXiv:2604.02927v2 Announce Type: replace Abstract: Routing algorithms are crucial for efficient computer network operations, and in many settings they must be able to react to traffic bursts within m

local-aiarxiv-cs-lg

14 Apr 2026

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

A novel hybrid approach for positive-valued DAG learning

DGX agent

arXiv:2604.08935v1 Announce Type: cross Abstract: Causal discovery from observational data remains a fundamental challenge in machine learning and statistics, particularly when variables represent inh

researcharxiv-cs-lg

13 Apr 2026

Accelerating Local LLMs on Resource-Constrained Edge Devices via Distributed Prompt Caching

model-releasesarxiv-cs-lg

arXiv:2602.22812v2 Announce Type: replace Abstract: Since local LLM inference on resource-constrained edge devices imposes a severe performance bottleneck, this paper proposes distributed prompt cachi

Bandwidth-constrained Variational Message Encoding for Cooperative Multi-agent Reinforcement Learning

arXiv:2512.11179v3 Announce Type: replace Abstract: Graph-based multi-agent reinforcement learning (MARL) enables coordinated behavior under partial observability by modeling agents as nodes and commu

agentsarxiv-cs-lg

EGLOCE: Training-Free Energy-Guided Latent Optimization for Concept Erasure

arXiv:2604.09405v1 Announce Type: new Abstract: As text-to-image diffusion models grow increasingly prevalent, the ability to remove specific concepts-mostly explicit content and many copyrighted char

safetyarxiv-cs-cv

Mamba-Based Graph Convolutional Networks: Tackling Over-smoothing with Selective State Space

model-releasesarxiv-cs-lg

arXiv:2501.15461v4 Announce Type: replace Abstract: Graph Neural Networks (GNNs) have shown great success in various graph-based learning tasks. However, it often faces the issue of over-smoothing as

Memory-Efficient Transfer Learning with Fading Side Networks via Masked Dual Path Distillation

arXiv:2604.09088v1 Announce Type: new Abstract: Memory-efficient transfer learning (METL) approaches have recently achieved promising performance in adapting pre-trained models to downstream tasks. Th

On Divergence Measures for Training GFlowNets

arXiv:2410.09355v2 Announce Type: cross Abstract: Generative Flow Networks (GFlowNets) are amortized inference models designed to sample from unnormalized distributions over composable objects, with a

safetyarxiv-cs-ai

Yes, But Not Always. Generative AI Needs Nuanced Opt-in

arXiv:2604.09413v1 Announce Type: cross Abstract: This paper argues that a one-size-fits-all approach to specifying consent for the use of creative works in generative AI is insufficient. Real-world o

DailyArt: Discovering Articulation from Single Static Images via Latent Dynamics

arXiv:2604.07758v1 Announce Type: new Abstract: Articulated objects are essential for embodied AI and world models, yet inferring their kinematics from a single closed-state image remains challenging

researcharxiv-cs-cv

DYCP: Dynamic Context Pruning for Long-Form Dialogue with LLMs

arXiv:2601.07994v4 Announce Type: replace Abstract: Large Language Models (LLMs) increasingly operate over long-form dialogues with frequent topic shifts. While recent LLMs support extended context wi

researcharxiv-cs-cl

Efficient Quantization of Mixture-of-Experts with Theoretical Generalization Guarantees

arXiv:2604.06515v1 Announce Type: cross Abstract: Sparse Mixture-of-Experts (MoE) allows scaling of language and vision models efficiently by activating only a small subset of experts per input. While

researcharxiv-cs-ai

FBS: Modeling Native Parallel Reading inside a Transformer

arXiv:2601.21708v2 Announce Type: replace Abstract: Large language models (LLMs) excel across many tasks, yet inference is still dominated by strictly token-by-token autoregression. Existing accelerat

researcharxiv-cs-ai

Making Room for AI: Multi-GPU Molecular Dynamics with Deep Potentials in GROMACS

arXiv:2604.07276v1 Announce Type: cross Abstract: GROMACS is a de-facto standard for classical Molecular Dynamics (MD). The rise of AI-driven interatomic potentials that pursue near-quantum accuracy a

MCLR: Improving Conditional Modeling via Inter-Class Likelihood-Ratio Maximization and Unifying Classifier-Free Guidance with Alignment Objectives

arXiv:2603.22364v2 Announce Type: replace-cross Abstract: Diffusion models have achieved state-of-the-art performance in generative modeling, but their success often relies heavily on classifier-free

safetyarxiv-cs-cv

PLUME: Latent Reasoning Based Universal Multimodal Embedding

arXiv:2604.02073v2 Announce Type: replace Abstract: Universal multimodal embedding (UME) maps heterogeneous inputs into a shared retrieval space with a single model. Recent approaches improve UME by g

Probabilistic Language Tries: A Unified Framework for Compression, Decision Policies, and Execution Reuse

arXiv:2604.06228v1 Announce Type: cross Abstract: We introduce probabilistic language tries (PLTs), a unified representation that makes explicit the prefix structure implicitly defined by any generati

safetyarxiv-cs-ai

SensorPersona: An LLM-Empowered System for Continual Persona Extraction from Longitudinal Mobile Sensor Streams

arXiv:2604.06204v1 Announce Type: cross Abstract: Personalization is essential for Large Language Model (LLM)-based agents to adapt to users' preferences and improve response quality and task performa

Space Filling Curves is All You Need: Communication-Avoiding Matrix Multiplication Made Simple

arXiv:2601.16294v2 Announce Type: replace-cross Abstract: General Matrix Multiplication (GEMM) is the cornerstone of HPC workloads and Deep Learning. State-of-the-art vendor libraries tune tensor layo

local-aiarxiv-cs-ai

Variational Feature Compression for Model-Specific Representations

model-releasesarxiv-cs-lg

arXiv:2604.06644v1 Announce Type: cross Abstract: As deep learning inference is increasingly deployed in shared and cloud-based settings, a growing concern is input repurposing, in which data submitte

Hardware

AbridgeKV: Compressing Long Reasoning into the KV Cache of Learned Pause Tokens

arXiv:2610.20641v1 Announce Type: new Abstract: Long chain-of-thought decoding is memory-bound: the KV cache of the generated tokens grows linearly with the test-time compute being spent. Evicting cac

hardwarearxiv-cs-cl

Activation-Adjacent Layer Normalization: Effects on Approximation Power

arXiv:2610.21527v1 Announce Type: new Abstract: Layer normalization (LN) and root mean square normalization (RMSNorm) depend on their inputs and remain nonlinear at inference. Inserting either next to

researcharxiv-cs-lg

An Agentic Framework for Proactive Robots with Procedural Assistance Memory

arXiv:2610.21211v1 Announce Type: new Abstract: Proactive robots should anticipate user needs and provide assistance before being explicitly asked. A straightforward approach is to infer human intenti

agentsarxiv-cs-ro

APRiL: An Open-ended Benchmark for Chinese Social Awareness in Large Language Models

model-releasesarxiv-cs-cl

arXiv:2610.20785v1 Announce Type: new Abstract: Large language models are increasingly used in social interactions, where appropriate responses require recognizing implicit cues, inferring mental stat

Applications

CART: Compatible Routing for Multimodal Continual Instruction Tuning

arXiv:2610.22137v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) achieve strong performance through instruction tuning, yet real-world applications often require them to contin

applicationsarxiv-cs-ai

Compression Does Not End at Selection: Post-Selection Positional Geometry for Sparse KV Caches

model-releasesarxiv-cs-cl

arXiv:2610.21126v1 Announce Type: new Abstract: Sparse KV compression reduces long-context inference costs by retaining selected historical states, yet these states typically keep their coordinates fr

Disagreement-Guided Weak-to-Strong Routing

arXiv:2610.22245v1 Announce Type: new Abstract: Small language models are increasingly deployed on edge devices and robotic systems to meet cost and latency constraints, yet their limited capabilities

local-aiarxiv-cs-ai

DropKV: Decoupling Residual-Output Perturbation for Near-Optimal KV-Cache Eviction

arXiv:2610.22130v1 Announce Type: new Abstract: Key-value (KV) cache management has become a critical bottleneck for scaling Large Language Models (LLMs) to long-context scenarios due to linear memory

researcharxiv-cs-ai

FlowBudget: Training-Free Sampling-Budget Estimation for Flow Matching Models

arXiv:2610.20636v1 Announce Type: cross Abstract: Sampling from continuous flow-based generators such as flow matching models requires numerical ODE or SDE integration, making inference cost depend on

researcharxiv-cs-cv

From Long Context to Selective Memory:A Survey of Memory-Centric Compression for Large Language Models

model-releasesarxiv-cs-cl

arXiv:2610.21165v1 Announce Type: new Abstract: Long-context capability has become essential for large language models (LLMs) to handle complex tasks, yet continuously expanding the context window doe

Tutorials

How Much Physics Can Vision Learn from Audio? Probing and Distilling Damping Structure

arXiv:2610.22221v1 Announce Type: new Abstract: Physical understanding is crucial for decision making, yet vision models struggle to infer properties such as damping that govern how materials behave.

tutorialsarxiv-cs-cv

Tutorials

Latent Annealed Posterior Sampling with a Variance-Preserving Bridge for Sparsely Observed Random Fields

arXiv:2610.20659v1 Announce Type: cross Abstract: Posterior inference for random fields from sparse observations presents two major challenges: accurately characterizing complex spatial uncertainty an

tutorialsarxiv-cs-lg

Mallea: Building, Steering, and Memorizing Worlds with Coding Agent

arXiv:2610.21283v1 Announce Type: new Abstract: Code world models represent world dynamics through explicit state and executable rules. We investigate how to coordinate object representations, behavio

Policy-Aligned Predictive Verification for Adaptive VLA Action-Chunk Execution

arXiv:2610.21434v1 Announce Type: cross Abstract: Vision-language-action (VLA) policies generate action sequences that reduce repeated inference, but deciding how much of each sequence to execute is c

safetyarxiv-cs-lg

Sparse Attention for Video Generation Acceleration via Growing Sparsity and Reduced Search

arXiv:2610.21129v1 Announce Type: new Abstract: Diffusion Transformers (DiTs) dominate video generation but suffer from high inference cost due to the quadratic complexity of self-attention. Sparse at

researcharxiv-cs-cv

Applications

T2-z: Test-Time Optimization of Environment Latents for Robotic World Models

arXiv:2610.22364v1 Announce Type: cross Abstract: Predicting how the world responds to an action requires more than what can be seen: robots must account for environmental factors that are neither dir

applicationsarxiv-cs-ai

Where to Steer: Input-Dependent Layer Selection for Steering Improves LLM Alignment

arXiv:2604.03867v2 Announce Type: replace Abstract: Steering vectors have emerged as a lightweight and effective approach for aligning large language models (LLMs) at inference time, enabling modulati

safetyarxiv-cs-lg

A Reference Architecture for Integrating Agentic AI into Enterprise Software: Responsibilities, Interactions, and Variability

arXiv:2610.14818v1 Announce Type: new Abstract: Agentic AI moves generative AI applications toward systems that coordinate reasoning, enterprise context, model inference, and external actions. Integra

Action Jets: Learning the Local Shape of the Action-conditioned Data Manifold

arXiv:2610.19007v1 Announce Type: new Abstract: World models are a powerful means of reasoning and planning, but many of these action-conditioned predictive models are high dimensional and computation

Alignment Faking - the Train -> Deploy Asymmetry: Through a Game-Theoretic Lens with Bayesian-Stackelberg Equilibria

arXiv:2511.17937v2 Announce Type: replace Abstract: Alignment faking is a form of strategic deception in AI in which models selectively comply with training objectives when they infer that they are in

Beyond Successful Play: Diagnosing Opponent Modeling in Large Language Models

arXiv:2610.17310v1 Announce Type: new Abstract: Language-model agents are expected to adapt to others whose behavior changes through interaction. Yet adaptation may rely on learned behavioral patterns

DD-GBP: Anytime Multi-Arm Planning and Continuous Execution via Distributed Dynamic Gaussian Belief Propagation

arXiv:2610.20460v1 Announce Type: new Abstract: Safely executing long-horizon multi-arm tasks online in complex dynamic environments remains challenging. We propose a new distributed dynamic Gaussian

local-aiarxiv-cs-ro

Delta Tokenizer: Reference-Identical Tokenization of Session Deltas and Segment-Parallelism for Long Prompts on Multicore CPUs

arXiv:2610.19801v1 Announce Type: new Abstract: An agent that drives an LLM sends the model the same conversation again and again, and each turn adds only a little text at the end. Inference engines r

agentsarxiv-cs-cl

ECMR: Evidence-Aware Deliberation Control for Multi-Agent LLM Reasoning

arXiv:2610.17869v1 Announce Type: new Abstract: Multi-agent Large Language Model (LLM) systems typically commit to a deliberation protocol before inference. A fixed protocol cannot allocate computatio

GesBench: When and Where Do Multimodal Language Models Perceive Social Meaning?

arXiv:2610.15129v1 Announce Type: new Abstract: Human social perception is incremental: we infer intent before an utterance is complete, revise our interpretation as new cues arrive, and attend to spe

Applications

Have You Been Convinced? A Neuro-symbolic Argumentation Reasoning Approach to Detecting Persuasiveness in ChangeMyView Discussions

arXiv:2610.15091v1 Announce Type: cross Abstract: Argument structures represent how individual arguments interact within a discourse. However, leveraging these structures to infer meaningful knowledge

applicationsarxiv-cs-ai

Instructions Are Evidence, Not Objectives: Measuring Outcome Alignment of Agents

arXiv:2610.17757v1 Announce Type: new Abstract: Large language model agents can follow complex human instructions to solve real-world tasks. In practice, however, initial instructions are often incomp

KiToke: Distribution-Aware Token Compression for Video Large Language Models

arXiv:2604.03414v2 Announce Type: replace Abstract: Video Large Language Models (Video LLMs) achieve strong performance on video understanding tasks but suffer from high inference costs due to the lar