Search: “models”

From Mechanistic to Compositional Interpretability

arXiv:2605.08934v1 Announce Type: new Abstract: Mechanistic interpretability aims to explain neural model behaviour by reverse-engineering learned computational structure into human-understandable com

GELATO: Generative Entropy- and Lyapunov-based Adaptive Token Offloading for Device-Edge Speculative LLM Inference

DGX agent

arXiv:2605.10124v1 Announce Type: cross Abstract: The recent growth of on-device Large Language Model (LLM) inference has driven significant interest in device-edge collaborative LLM inference. As a p

local-aiarxiv-cs-lg

12 May 2026

Generalization Error Bounds for Picard-Type Operator Learning in Nonlinear Parabolic PDEs

arXiv:2605.10277v1 Announce Type: new Abstract: Operator learning for partial differential equations (PDEs) aims to learn solution operators on infinite-dimensional function spaces from finite-resolut

local-aiarxiv-cs-lg

Global Optimization via Softmin Energy Minimization

arXiv:2509.17815v2 Announce Type: replace Abstract: Global optimization, particularly for non-convex functions with multiple local minima, poses significant challenges for traditional gradient-based m

Hierarchical Reinforced Trader (HRT): A Bi-Level Approach for Optimizing Stock Selection and Execution

arXiv:2410.14927v2 Announce Type: replace-cross Abstract: Automated equity trading requires converting noisy market and news signals into executable portfolio decisions under risk, turnover, and trans

How Instruction and Reasoning Data shape Post-Training: Data Quality through the Lens of Layer-wise Gradients

arXiv:2504.10766v2 Announce Type: replace-cross Abstract: As the post-training of large language models (LLMs) advances from instruction-following to complex reasoning tasks, understanding how differe

researcharxiv-cs-ai

How LLMs Are Persuaded: A Few Attention Heads, Rerouted

arXiv:2605.09314v1 Announce Type: new Abstract: Language models can be persuaded to abandon factual knowledge. This vulnerability is central to AI safety, but its internal mechanism remains poorly und

How Much is Brain Data Worth for Machine Learning?

arXiv:2605.09243v1 Announce Type: new Abstract: If a person can solve a task, can measuring their brain make it easier to train a model to solve that task too? Recent NeuroAI work suggests that supple

Identifying Backdoored Graphs in Graph Neural Network Training: An Explanation-Based Approach with Novel Metrics

arXiv:2403.18136v3 Announce Type: replace-cross Abstract: Graph Neural Networks (GNNs) have gained popularity in numerous domains, yet they are vulnerable to backdoor attacks that can compromise their

Identifying Multi-Hit Cancer Drivers Without Massive Parallelization: A CP, MIP, and Column Generation Framework

arXiv:2602.22551v2 Announce Type: replace-cross Abstract: Cancer is often driven by specific combinations of an estimated two to nine gene mutations, known as multi-hit combinations. Identifying these

Improving Text-to-Image Generation with Intrinsic Self-Confidence Rewards

arXiv:2603.00918v3 Announce Type: replace-cross Abstract: Text-to-image generation powers content creation across design, media, and data augmentation. Post-training of text-to-image generative models

Incorporating rank-free coupling and external field via an incoherent modulated spatial photonic Ising machine

arXiv:2512.21587v2 Announce Type: replace-cross Abstract: Spatial photonic Ising machines offer a novel optical platform for optimization and spin-model simulation, but existing diffraction-based sche

Inpainting physics: self-supervised learning for context-driven fluid simulation

arXiv:2605.08832v1 Announce Type: new Abstract: Neural surrogate models for computational fluid dynamics (CFD) are typically trained as forward operators that map explicit problem specifications, such

local-aiarxiv-cs-lg

Tutorials

Interactive Benchmarks

arXiv:2603.04737v2 Announce Type: replace Abstract: Existing reasoning evaluation paradigms suffer from different limitations: fixed benchmarks are increasingly saturated and vulnerable to contaminati

tutorialsarxiv-cs-ai

Inverse Design for Conditional Distribution Matching

arXiv:2605.09439v1 Announce Type: new Abstract: Generative models are powerful tools for sampling from a learned distribution P(Y mid X), and inverse-design methods invert this map to find an input x

Ister: Linear Transformer for Efficient Multivariate Time Series Forecasting

arXiv:2412.18798v3 Announce Type: replace-cross Abstract: Transformer-based models have achieved remarkable success in multivariate time series forecasting (MTSF) by capturing long-range dependencies.

Lattice Deduction Transformers

arXiv:2605.08605v1 Announce Type: cross Abstract: We introduce the Lattice Deduction Transformer (LDT), a recurrent transformer that approximates logically sound deduction by projecting its latent sta

LEAF-SQL: Level-wise Exploration with Adaptive Fine-graining for Text-to-SQL Skeleton Prediction

arXiv:2605.09295v1 Announce Type: new Abstract: Text-to-SQL translates natural language questions into executable SQL queries, enabling intuitive database access for non-experts. While large language

LITMUS: Benchmarking Behavioral Jailbreaks of LLM Agents in Real OS Environments

arXiv:2605.10779v1 Announce Type: cross Abstract: The rapid proliferation of LLM-based autonomous agents in real operating system environments introduces a new category of safety risk beyond content s

LLM-Agnostic Semantic Representation Attack

arXiv:2605.08898v1 Announce Type: cross Abstract: Large Language Models (LLMs) increasingly employ alignment techniques to prevent harmful outputs. Despite these safeguards, attackers can circumvent t

LLM-FE: Automated Feature Engineering for Tabular Data with LLMs as Evolutionary Optimizers

arXiv:2503.14434v3 Announce Type: replace-cross Abstract: Automated feature engineering plays a critical role in improving predictive model performance for tabular learning tasks. Traditional automate

researcharxiv-cs-ai

LLMs with in-context learning for Algorithmic Theoretical Physics

arXiv:2605.08212v1 Announce Type: cross Abstract: There is an increasing number of algorithmic computations in theoretical physics. These, while conceptually simple, can nevertheless be time-consuming

Log analysis is necessary for credible evaluation of AI agents

arXiv:2605.08545v1 Announce Type: new Abstract: Agent benchmarks typically report only final outcomes: pass or fail. This threatens evaluation credibility in three ways. First, scores may be inflated

M^2E-UAV: A Benchmark and Analysis for Onboard Motion-on-Motion Event-Based Tiny UAV Detection

model-releasesarxiv-cs-cv

arXiv:2605.10496v1 Announce Type: new Abstract: Tiny UAV detection from an onboard event camera is difficult when the observer and target move at the same time. In this motion-on-motion regime, ego-mo

M^3: Reframing Training Measures for Discretized Physical Simulations

arXiv:2605.08843v1 Announce Type: new Abstract: Neural surrogate models for physical simulations are trained on discretized samples of continuous domains, where the induced empirical measure leads to

MARLaaS: Multi-Tenant Asynchronous Reinforcement Learning as a Service

arXiv:2605.08527v1 Announce Type: cross Abstract: Reinforcement Learning from Verifiable Rewards (RLVR) has significantly improved the reasoning capabilities of large language models (LLMs), particula

Max-pooling Network Revisited: Analyzing the Role of Semantic Probability in Multiple Instance Learning for Hallucination Detection

arXiv:2605.08863v1 Announce Type: new Abstract: Hallucination detection has become increasingly important for improving the reliability of large language models (LLMs). Recently, hybrid approaches suc

researcharxiv-cs-cl

MDL-GBG: A Non-parametric and Interpretable Granular-Ball Generation Method for Clustering

arXiv:2605.08759v1 Announce Type: new Abstract: Existing granular-ball generation methods are still mainly driven by handcrafted quality measures and heuristic splitting or stopping criteria, which we

local-aiarxiv-cs-lg

Meet physics-intern🧑‍🎓, our agentic framework for theoretical physics. It takes Gemini 3.1 Pro from 17.7% to 31.4% on CritPt, a new SOTA o…

model-releasesclem-delangue--x

Meet physics-intern🧑‍🎓, our agentic framework for theoretical physics. It takes Gemini 3.1 Pro from 17.7% to 31.4% on CritPt, a new SOTA on one of the hardest benchmarks for LLMs. Theoretical physics

MemPrivacy: Privacy-Preserving Personalized Memory Management for Edge-Cloud Agents

arXiv:2605.09530v1 Announce Type: cross Abstract: As LLM-powered agents are increasingly deployed in edge-cloud environments, personalized memory has become a key enabler of long-term adaptation and u

Meta-reinforcement learning with minimum attention

arXiv:2505.16741v4 Announce Type: replace Abstract: Minimum attention applies the least action principle to changes of control concerning state and time, first proposed by Brockett. The involved regul

safetyarxiv-cs-lg

Metal-Sci: A Scientific Compute Benchmark for Evolutionary LLM Kernel Search on Apple Silicon

arXiv:2605.09708v1 Announce Type: cross Abstract: We present Metal-Sci, a 10-task benchmark of scientific Apple Silicon Metal compute kernels spanning six optimization regimes (stencils, all-pairs in

MicroFuse: Protein-to-Genome Expert Fusion for Microbial Operon Reasoning

arXiv:2605.08815v1 Announce Type: new Abstract: Predicting microbial operon co-membership requires integrating two complementary biological signals: protein-scale molecular identity and genome-context

Minimizing Worst-Case Weighted Latency for Multi-Robot Persistent Monitoring: Theory and RL-Based Solutions

model-releasesarxiv-cs-ro

arXiv:2605.09633v1 Announce Type: new Abstract: We study multi-robot persistent monitoring on weighted graphs, where node weights encode monitoring priorities and edge weights encode travel distances.

MolSight: Molecular Property Prediction with Images

arXiv:2605.10157v1 Announce Type: cross Abstract: Every molecule ever synthesised can be drawn as a 2D skeletal diagram, yet in modern property prediction this universally available representation has

Multimodal Representation Learning Conditioned on Semantic Relations

arXiv:2508.17497v2 Announce Type: replace-cross Abstract: Multimodal representation learning has been largely driven by contrastive models such as CLIP, which learn a shared embedding space by alignin

MULTITEXTEDIT: Benchmarking Cross-Lingual Degradation in Text-in-Image Editing

arXiv:2605.08163v1 Announce Type: cross Abstract: Text-in-image editing has become a key capability for visual content creation, yet existing benchmarks remain overwhelmingly English-centric and often

Muown: Row-Norm Control for Muon Optimization

arXiv:2605.10797v1 Announce Type: new Abstract: Muon has emerged as a strong competitor to AdamW for language model pre-training, yet its behavior at scale is sensitive to weight decay. Recent work ha

Tutorials

MUR: Momentum Uncertainty guided Reasoning

arXiv:2507.14958v2 Announce Type: replace Abstract: Current models have achieved impressive performance on reasoning-intensive tasks, yet optimizing their reasoning efficiency remains an open challeng

tutorialsarxiv-cs-cl

Nautilus Compass: Black-box Persona Drift Detection for Production LLM Agents

arXiv:2605.09863v1 Announce Type: cross Abstract: Production LLM coding agents drift over long sessions: they forget user-specified constraints, slip into mistakes the user already flagged, and confab

Navigating LLM Valley: From AdamW to Memory-Efficient and Matrix-Based Optimizers

arXiv:2605.09176v1 Announce Type: cross Abstract: Training large language models requires optimization algorithms that are not only statistically effective, but also computationally and memory efficie

Neural Posterior Estimation of Terrain Parameters from Radar Sounder Data

arXiv:2605.08179v1 Announce Type: cross Abstract: Radar sounders are electromagnetic instruments that can probe deep into the subsurface of Earth and other planetary bodies by processing the echo of t

Neuroscience-Inspired Analyses of Visual Interestingness in Multimodal Transformers

arXiv:2605.08188v1 Announce Type: cross Abstract: Human attention is the gateway to conscious perception, memory and decision-making. However, its role in modern transformer models remains largely une

researcharxiv-cs-ai

Tutorials

NICE FACT: Diagnosing and Calibrating VLMs in Quantitative Reasoning for Kinematic Physics

arXiv:2605.08452v1 Announce Type: new Abstract: The ability to derive precise spatial and physical insights is a cornerstone of vision-language models (VLMs), yet their poor performances in related sp

tutorialsarxiv-cs-cv

Hardware

Octopus Protocol: One-Shot Hardware Discovery and Control for AI Agents via Infrastructure-as-Prompts

arXiv:2605.09055v1 Announce Type: cross Abstract: Recent agentic-robotics systems, from Code-asPolicies to modern vision-language-action (VLA) foundation models, presuppose that drivers, SDKs, or ROS-

hardwarearxiv-cs-ai

Offline Preference Optimization for Rectified Flow with Noise-Tracked Pairs

arXiv:2605.09433v1 Announce Type: new Abstract: Existing preference datasets for text-to-image models typically store only the final winner/loser images. This representation is insufficient for rectif

safetyarxiv-cs-cv

Optimality of Sub-network Laplace Approximations: New Results and Methods

arXiv:2605.09075v1 Announce Type: cross Abstract: Although the Laplace approximation offers a simple route to uncertainty quantification in deep neural networks, its reliance on inverting large Hessia

Optimised Support Vector Regression for California Housing Price Prediction: The Critical Role of Feature Engineering and Hyperparameter Tuning

arXiv:2605.08660v1 Announce Type: new Abstract: In the recent literature, Support Vector Regression (SVR) has been cited as one of the weakest performers on the California Housing benchmark dataset, w