research

PhysBrain 1.0 Technical Report

arXiv:2605.15298v1 Announce Type: cross Abstract: Vision-language-action models have advanced rapidly, but robot trajectories alone provide limited coverage for learning broad physical understanding.

Position: Zeroth-Order Optimization in Deep Learning Is Underexplored, Not Underpowered

arXiv:2605.15622v1 Announce Type: new Abstract: Zeroth-order (ZO) optimization, learning from finite differences of function evaluations without backpropagation, has recently regained attention in dee

researcharxiv-cs-lg

18 May 2026

Practical Validity Conditions for Byzantine-Tolerant Federated Learning

arXiv:2605.15887v1 Announce Type: new Abstract: Robust aggregation is the core operation in Byzantine-tolerant federated learning. To ensure the quality of aggregation independently of data distributi

PRB-RUPFormer: A Recursive Unified Probabilistic Transformer for Residual PRB Forecasting

arXiv:2605.15363v1 Announce Type: new Abstract: Accurate forecasting of residual Physical Resource Blocks (PRBs) is critical for proactive network slice provisioning, energy-efficient operation, and s

Pretraining Objective Matters in Extreme Low-Data FGVC: A Backbone-Controlled Study

arXiv:2605.15599v1 Announce Type: cross Abstract: Extreme low-data fine-grained classification is common in expert domains where labeling is expensive, yet practitioners still need principled guidance

Privacy Evaluation of Generative Models for Trajectory Generation

arXiv:2605.15246v1 Announce Type: new Abstract: Trajectory data is fundamental to modern urban intelligence, yet its sensitivity raises significant privacy concerns. Generative models such as Generati

Process Rewards with Learned Reliability

arXiv:2605.15529v1 Announce Type: cross Abstract: Process Reward Models (PRMs) provide step-level feedback for reasoning, but current PRMs usually output only a single reward score for each step. Down

Prompt Stability Scoring for Text Annotation with Large Language Models

arXiv:2407.02039v3 Announce Type: replace Abstract: Researchers are increasingly using language models (LMs) for text annotation. These approaches rely only on a prompt telling the model to return a g

RaPD: Resolution-Agnostic Pixel Diffusion via Semantics-Enriched Implicit Representations

arXiv:2605.15908v1 Announce Type: cross Abstract: Natural images are continuous, yet most generative models synthesize them on discrete grids, limiting resolution-flexible generation. Continuous neura

Reading the Cell, Designing the Cure: Perturbation-Conditioned Molecular Diffusion for Function-Oriented Drug Design

arXiv:2605.15243v1 Announce Type: cross Abstract: When reliable target structures are unavailable at scale or phenotypes arise from dysregulated pathways, transcriptomic perturbations provide a system

Reasoning Models Don't Just Think Longer, They Move Differently

arXiv:2605.15454v1 Announce Type: new Abstract: Reasoning-trained language models often spend more tokens on harder problems, but longer chains of thought do not show whether a model is merely computi

Regret and Sample Complexity of Online Q-Learning via Concentration of Stochastic Approximation with Time-Inhomogeneous Markov Chains

arXiv:2602.16274v2 Announce Type: replace Abstract: We present the first regret bound for classical online Q-learning in infinite-horizon discounted Markov decision processes (MDPs), without relying o

Rethinking Neural Network Learning Rates: A Stackelberg Perspective

arXiv:2605.15530v1 Announce Type: new Abstract: Neural networks are typically trained with a single learning rate across all layers. While recent empirical evidence suggests that assigning layer-speci

RIDE: Retinex-Informed Decoupling for Exposing Concealed Objects

arXiv:2605.15450v1 Announce Type: cross Abstract: Concealed Object Segmentation (COS) encompasses a family of dense-prediction tasks, including camouflaged object detection, polyp segmentation, transp

Robust Prior-Guided Segmentation for Editable 3D Gaussian Splatting

arXiv:2605.16065v1 Announce Type: cross Abstract: 3D Gaussian Splatting (3D-GS) enables real-time 3D scene reconstruction but lacks robust segmentation for editing tasks such as object removal, extrac

RoiMAM: Region-of-Interest Medical Attention Model for Efficient Vision-Language Understanding

arXiv:2605.15561v1 Announce Type: new Abstract: Vision-Language Models (VLMs) facilitate medical visual question answering (MedVQA) by jointly interpreting images and text. However, existing models ty

SAE-RNA: A Sparse Autoencoder Model for Interpreting RNA Language Model Representations

arXiv:2510.02734v2 Announce Type: replace-cross Abstract: Deep learning, particularly with the advancement of Large Language Models, has transformed biomolecular modeling, with protein language models

Sampling-Based Global Optimal Control and Estimation via Semidefinite Programming

arXiv:2507.17572v3 Announce Type: replace Abstract: Global optimization has gained attraction over the past decades, thanks to the development of both theoretical foundations and efficient numerical r

researcharxiv-cs-ro

Sampling-Free Privacy Accounting for Matrix Mechanisms under Random Allocation

arXiv:2601.21636v3 Announce Type: replace Abstract: We study privacy amplification for differentially private model training with matrix factorization under random allocation (also known as the balls-

Seeing is Understanding: Unlocking Causal Attention into Modality-Mutual Attention for Multimodal LLMs

arXiv:2503.02597v3 Announce Type: replace-cross Abstract: Recent Multimodal Large Language Models (MLLMs) have demonstrated significant progress in perceiving and reasoning over multimodal inquiries,

Self-Prompting Diffusion Transformer for Open-Vocabulary Scene Text Editing via In-Context Learning

arXiv:2605.15523v1 Announce Type: new Abstract: Scene text editing aims to modify text in a target region of an image while preserving surrounding background style and texture. Existing methods rely s

Self-Supervised ImageNet Representations for In Vivo Confocal Microscopy: Tortuosity Grading without Segmentation Maps

arXiv:2603.15269v2 Announce Type: replace Abstract: The tortuosity of corneal nerve fibers are used as indication for different diseases. Current state-of-the-art methods for grading the tortuosity he

SFHand: Learning Embodied Manipulation by Streaming Egocentric 3D Hand Forecasting

arXiv:2511.18127v2 Announce Type: replace Abstract: Real-time 3D hand forecasting is a critical component for fluid human-computer interaction in applications like AR and assistive robotics. However,

SGNO: Spectral Generator Neural Operators for Stable Long Horizon PDE Rollouts

arXiv:2602.18801v2 Announce Type: replace Abstract: Autoregressive neural PDE surrogates predict future states by repeatedly applying a learned one-step operator. This is a simple and widely used meth

Sharp Spectral Thresholds for Logit Fixed Points

arXiv:2605.15651v1 Announce Type: cross Abstract: Softmax feedback systems are a common mathematical core of entropy-regularized reinforcement learning, logit game dynamics, population choice, and mea

Small Generalizable Prompt Predictive Models Can Steer Efficient RL Post-Training of Large Reasoning Models

arXiv:2602.01970v2 Announce Type: replace Abstract: Reinforcement learning enhances the reasoning capabilities of large language models but often involves high computational costs due to rollout-inten

Smoothie: Smoothing Diffusion on Token Embeddings for Text Generation

arXiv:2505.18853v2 Announce Type: replace Abstract: Diffusion models have achieved state-of-the-art performance in generating images, audio, and video, but their adaptation to text remains challenging

Sparse Autoencoders enable Robust and Interpretable Fine-tuning of CLIP models

arXiv:2605.15961v1 Announce Type: new Abstract: Large-scale pre-trained vision-language models like CLIP demonstrate remarkable zero-shot performance across diverse tasks. However, fine-tuning these m

Spectral Priors vs. Attention: Investigating the Utility of Attention Mechanisms in EEG-Based Diagnosis

arXiv:2605.15433v1 Announce Type: new Abstract: Electroencephalograph (EEG) timeseries signals are characterized by significant noise and coarse spatial resolution, which complicates the classificatio

Stabilizing Knowledge, Promoting Reasoning: Dual-Token Constraints for RLVR

arXiv:2507.15778v2 Announce Type: replace Abstract: Reinforcement Learning with Verifiable Rewards (RLVR) has become an effective post-training method for improving the reasoning abilities of Large La

Stochastic Compositional Optimization via Hybrid Momentum Frank--Wolfe

arXiv:2605.15350v1 Announce Type: cross Abstract: Stochastic compositional optimization minimizes objectives of the form min_{m{x} in X} F(m{f}(m{x}), m{x}), where m{f} is accessible only through nois

Stochastic Non-Smooth Convex Optimization with Unbounded Gradients

arXiv:2605.15522v1 Announce Type: cross Abstract: Much of the existing theory on first-order non-smooth optimization is built on a restrictive assumption that the gradients of the objective function a

Stock Market Prediction Using Node Transformer Architecture Integrated with BERT Sentiment Analysis

arXiv:2603.05917v3 Announce Type: replace-cross Abstract: Stock market prediction presents considerable challenges for investors, financial institutions, and policymakers operating in complex market e

Structured Jacobian Construction for Motion Optimization with High-Order Time Derivatives in Multi-Link Systems

arXiv:2605.15845v1 Announce Type: new Abstract: This paper presents a novel framework for Jacobian computation in motion optimization problems involving multi-link systems, where physical quantities a

researcharxiv-cs-ro

Symplectic Neural Operators for Learning Infinite Dimensional Hamiltonian Systems

arXiv:2605.15881v1 Announce Type: cross Abstract: The modeling and simulation of infinite-dimensional Hamiltonian systems are central problems in mathematical physics and engineering, however they pos

Syntax Without Semantics: Teaching Large Language Models to Code in an Unseen Language

arXiv:2605.15607v1 Announce Type: new Abstract: Large language models (LLMs) achieve high pass rates on code generation benchmarks, yet whether they can transfer this ability to languages absent from

Testing properties of trees in graphical models with covariance queries

arXiv:2605.15996v1 Announce Type: cross Abstract: We consider the problem of testing properties of graphs underlying high-dimensional graphical models. We adopt the model of covariance queries introdu

Text-RSIR: A Text-Guided Framework for Efficient Remote Sensing Image Transmission and Reconstruction

arXiv:2605.15558v1 Announce Type: cross Abstract: High-resolution remote sensing imagery is critical for environmental monitoring, urban mapping, and land cover analysis, but its transmission is often

TFZ-Tree: An Ultra-Lightweight Waveform Classification Framework for Resource-Constrained Devices

arXiv:2605.15656v1 Announce Type: cross Abstract: Under the trend of multi-waveform coexistence in 6G IoT, intelligent receivers must first identify physical-layer waveform types before performing cor

The Download: Musk v. Altman week 3, and Trump’s tech trading

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. Musk v. Altman week 3: Musk and Altman traded blows over each

researchmit-tech-review

The OncoReach Stylet for Brachytherapy: Design Evaluation and Pilot Study

arXiv:2601.13529v2 Announce Type: replace Abstract: Cervical cancer accounts for a significant portion of the global cancer burden among women. Interstitial brachytherapy (ISBT) is a standard procedur

researcharxiv-cs-ro

The Privacy Price of Tail-Risk Learning: Effective Tail Sample Size in Differentially Private CVaR Optimization

arXiv:2605.16219v1 Announce Type: new Abstract: Differential privacy changes the effective sample size governing CVaR learning. For tail mass au, the privacy-relevant sample size is not n, but nau; eq

This is the single most important chart. If AI were driving prices, you'd see a cluster top-right. You don't. States with huge load growth (…

This is the single most important chart. If AI were driving prices, you'd see a cluster top-right. You don't. States with huge load growth (VA, TX, NV, ND, IA) sit at ~0c change in 5y. States with mas

researchyann-lecun--x

Tighter Regret Bounds for Contextual Action-Set Reinforcement Learning

arXiv:2605.15692v1 Announce Type: new Abstract: We study episodic reinforcement learning with fixed reward and transition functions, but with episode-dependent admissible action sets that are observed

Time-Varying Deep State Space Models for Sequences with Switching Dynamics

arXiv:2605.15311v1 Announce Type: new Abstract: The identification and modeling of time-varying systems is a fundamental challenge in signal processing and system identification. To address this chall

Toward LLMs Beyond English-Centric Development

arXiv:2605.15613v1 Announce Type: new Abstract: Through an analysis of sequences generated by open-weight large language models (LLMs), we demonstrate that LLMs are heavily biased toward English. Whil

Towards Foundation Models for Relational Databases with Language Models and Graph Neural Networks

arXiv:2605.16085v1 Announce Type: cross Abstract: Relational databases store much of the world's structured information, and they are essential for driving complex predictive applications. However, de

Transformer-like Inference from Optimal Control

arXiv:2605.15608v1 Announce Type: new Abstract: Decoder-only transformers compute the conditional probability of the next token from a sequence of past observations. This paper derives, from first pri