Search: “pretraining”

Hesitation Has a Geometry: Entropy-Trained Hyperbolic Probes for Sparse Activation Steering

model-releasesarxiv-cs-cl

arXiv:2610.02391v1 Announce Type: cross Abstract: When a large language model solves a mathematical problem, its reasoning is largely hierarchical, and the solution often branches at a few tokens wher

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Hardware

I2CD: Direct Image-to-Convex Decomposition for Simulation-Ready Collision Geometry

DGX agent

arXiv:2610.03453v1 Announce Type: cross Abstract: Physics simulators and motion planners require convex collision geometry, yet image-to-3D generative models output dense, frequently non-manifold visu

hardwarearxiv-cs-cv

5 Oct 2026

Tutorials

Improving Atomic-Fact Recall via Focused Views in Unstructured Knowledge Editing

arXiv:2610.02772v1 Announce Type: cross Abstract: Large language models (LLMs) increasingly serve as general-purpose interfaces to factual knowledge, but their parameters do not automatically reflect

tutorialsarxiv-cs-ai

Learning Jazz Pianist Style with Cross-Attention Conditioning

arXiv:2610.02918v1 Announce Type: cross Abstract: Jazz pianists develop distinctive traits that experienced listeners can often identify within seconds, yet the features underlying this recognition re

Learning What to Investigate Next: Meta-Reasoning for Long-Horizon Research Agents

arXiv:2610.02525v1 Announce Type: new Abstract: Long-horizon research agents must decide both how to investigate and what to investigate next as evidence accumulates. This is hard to learn because suc

LiteEMG-FM: An Efficient and Deployable Foundation Model for Robust EMG Sensing

arXiv:2610.02497v1 Announce Type: new Abstract: Electromyography (EMG) signals vary substantially across individuals, body regions, recording sessions, and sensing hardware, limiting the generalizatio

local-aiarxiv-cs-lg

Measure Less, Know More: Self-Supervised Test-Time Feature Acquisition

arXiv:2610.03454v1 Announce Type: cross Abstract: Recent progress in multimodal, high-dimensional learning has enabled foundation models to process heterogeneous, large-scale data. However, at test ti

Morpheus: A Morphology-Aware Neural Tokenizer and Word Embedder for Turkish

arXiv:2606.18717v2 Announce Type: replace-cross Abstract: Turkish is agglutinative: meaning is carried by morphemes, yet the subword tokenizers that drive modern language models split words by corpus

Moving Forward with Video Saliency: A New Dataset and Benchmark where Motion Matters

arXiv:2610.03276v1 Announce Type: new Abstract: Video saliency prediction is inherently harder to model than static image saliency due to the additional temporal dimension. Video saliency benchmarks r

MuonIO: Principled Norm-Aware Descent for Embedding Tables and Language Model Heads

arXiv:2610.02705v1 Announce Type: cross Abstract: The Muon optimizer derives its update rule for hidden linear layers by solving a local linearization of the loss penalized by the spectral norm, motiv

Neural Data Needs Semantic Tokenization: Behavioral Events as Boundaries of Session-Transferable Tokens

arXiv:2610.03001v1 Announce Type: new Abstract: Extracellular electrophysiology records a different set of neurons in every session. Neural foundation models embed each neuron and each session into th

NeuroLens: Learning Latent Embeddings of Neural Semantics from Chronic Recordings

arXiv:2610.02864v1 Announce Type: new Abstract: Understanding how neural activity represents higher-order cognition and how these representations evolve over time has long been a central pursuit in ne

Octrees as an Explicit 3D Language

arXiv:2610.02388v1 Announce Type: new Abstract: Existing 3D large language models (LLMs) compromise on two fronts: they compress shapes into latent codebook indices or coordinate text, which removes s

Predicting and Repairing Merge Collapse in Large Language Models

arXiv:2610.03199v1 Announce Type: cross Abstract: Large language models fine-tuned from a shared base can be merged by averaging their task vectors, but some merges collapse far below the base model,

researcharxiv-cs-cl

Applications

Predictor-Guided Latent Space Codon Optimization for Maximizing Protein Expression

arXiv:2610.03098v1 Announce Type: new Abstract: Codon optimization, the process of selecting synonymous codons to improve mRNA translation efficiency and protein expression, is central to therapeutic

applicationsarxiv-cs-ai

ProgressNet: Sketching and Prompting with a Frozen Text-to-Image Model

arXiv:2610.03512v1 Announce Type: new Abstract: Humans draw progressively: a few strokes, a look at the result, a stroke erased, a prompt revised. Image generators do not work this way. They typically

Rethinking World-Action Model for Compositional and In-Context Robotic Manipulation

model-releasesarxiv-cs-ro

arXiv:2610.02368v1 Announce Type: new Abstract: Long-horizon compositional manipulation has become increasingly important for real-world robot deployment, where a single task involves multiple coordin

RoboBridge: A Self-Evolving Embodied Agent Framework for Sim-to-Real Transfer

arXiv:2610.02717v1 Announce Type: new Abstract: A key challenge in bringing embodied intelligence into the real world is transferring capabilities from simulation to reality and enabling agents to con

safetyarxiv-cs-ro

StanceEval 2026: The Second Stance Detection Shared Task

arXiv:2610.03215v1 Announce Type: new Abstract: StanceEval 2026 is the second edition of the StanceEval shared task series on stance detection in Arabic social media text. Stance detection aims to ide

researcharxiv-cs-cl

TACD: Distilling Efficient Text-to-Motion Models via Terminal Amplification Control

arXiv:2610.02867v1 Announce Type: new Abstract: Recent text-to-motion models have improved motion quality and instruction following, yet many-step denoising and large model components make deployment

The Surprising Effectiveness of Shared Memory in Looped Transformers

arXiv:2610.02383v1 Announce Type: cross Abstract: Looped Transformers apply the same layers several times per token, adding compute to improve quality without more parameters. Each recursion, however,

researcharxiv-cs-ai

Applications

Time Series Forecasting Benchmarks Need Scenario-Grounded Stress Testing

arXiv:2610.02608v1 Announce Type: new Abstract: Time series forecasting (TSF) increasingly drives decisions in transportation, energy, finance, healthcare, and infrastructure, yet current evaluation r

applicationsarxiv-cs-ai

Tutorials

Toward Omni Multimodal Graph Foundation Model: A Topology-Driven Binding Approach

arXiv:2610.02881v1 Announce Type: new Abstract: Multimodal graph foundation models (MGFMs) seek to learn generalizable representations from large-scale graphs with heterogeneous node modalities. Howev

tutorialsarxiv-cs-lg

ZAGNet: Zone-Aware Graph Aggregation Network for Patient-Level Lung Ultrasound Diagnosis

arXiv:2610.02263v1 Announce Type: cross Abstract: Patient-level lung ultrasound (LUS) diagnosis requires integrating findings acquired across multiple anatomical zones, yet clinical examinations frequ

ZeroMAG: Zero-Shot Multimodal Adapter Generation for Plug-and-Play EEG Foundation Models

arXiv:2610.03546v1 Announce Type: new Abstract: EEG foundation models (EFMs) capture reusable knowledge from large-scale EEG data, while many EEG recordings also include companion physiological signal

Agents

A framework for auditing grounding claims

arXiv:2512.06205v3 Announce Type: replace Abstract: The symbol grounding problem asks how a token such as cat can be about cats. We propose a framework for auditing grounding claims against a declared

agentsarxiv-cs-ai

AnyJev Technical Report

arXiv:2610.00831v1 Announce Type: new Abstract: A typed decision is a choice among a fixed set of options, returned as a probability rather than as text. Systems that need typed decisions today use mo

Attention Manifolds: Steering or Blocking Language Models by Editing Learned B-Spline Surfaces

arXiv:2610.00257v1 Announce Type: new Abstract: In standard transformer attention, a source token sends the same value vector to every receiver. The query determines how much to attend but not what to

Concept Driven Domain Adaptation: Finding an Abstract Needle in a Haystack

arXiv:2610.00973v1 Announce Type: new Abstract: Science teachers frequently search for documentary excerpts not by describing what appears on screen, but by querying the abstract concepts they intend

Conflicting Supervision Moves Commitment, Not Capability: A 12.29{sigma} arrangement effect that is exactly zero under a convention-agnostic score

arXiv:2610.00234v1 Announce Type: new Abstract: 'Train a model on the same problems written under two incompatible conventions, both correct, and ask what the ordering of that data writes into the par

Counterfactual Auditing of Bias in Open-Source Large Language Models for Clinical Triage

arXiv:2610.01963v1 Announce Type: new Abstract: Emergency department (ED) triage is a high-stakes prioritization task in which demographic, socioeconomic, and system-context information may improperly

Debias Anything: Fairness with Diversity without Supervision in Diffusion Models

arXiv:2610.01815v1 Announce Type: new Abstract: Although diffusion models produce high-quality images, they also reproduce and amplify demographic imbalances in their training data. Debiasing their ge

safetyarxiv-cs-lg

DuoMind: Enabling Distributed Multi-Robot Coordination with Semantic Communication

arXiv:2610.02161v1 Announce Type: cross Abstract: Vision-language models (VLMs) and vision-language-action models (VLAs) have recently driven rapid progress in general-purpose robots, yet most progres

EfficientRollout: System-Aware Self-Speculative Decoding for RL Rollouts

arXiv:2606.18967v2 Announce Type: replace Abstract: Reinforcement learning (RL) has become a representative post-training paradigm for large language models (LLMs), enabling strong reasoning and agent

safetyarxiv-cs-lg

EgoFound3R: End-to-End Egocentric Hand Reconstruction in World Space with Point-Wise Interaction Attributes

arXiv:2610.01210v1 Announce Type: new Abstract: Egocentric video has become a primary source of supervision for embodied models, and its value rests on recovering hand motion in world coordinates, whi

Flow Matching Reinforcement for 3D Mesh Generation via Dynamic Homing Optimization

arXiv:2610.01233v1 Announce Type: new Abstract: Flow matching is central to 3D generation, yet in practice its reinforcement learning (RL) methods are largely adapted from 2D visual generation. Repres

GPEC: Efficient Pre-LLM Gaussian Process Embedding Correction for Cardiac Video Caption Generation

arXiv:2610.00196v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have shown strong potential for video understanding and caption generation, but their performance may decline i

Harnessing Domain Specialists in Multimodal Mixture-of-Experts for Efficient Adaptation

arXiv:2610.02123v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) architectures scale model capacity through sparse computation, routing each token through only a small subset of experts. In th

Increasing Width Allows Greedy Layer-wise Training to Rival End-to-End Backpropagation in Self-Supervised Learning

arXiv:2610.00753v1 Announce Type: cross Abstract: End-to-end backpropagation has been the dominant mode of training in deep learning, allowing for the coordination of parameter updates across layers o

Learning Rate Transfer for Hybrid Transformer-SSM Architectures

arXiv:2610.01172v1 Announce Type: new Abstract: We study learning rate (LR) scaling for hybrid architectures combining Transformer and State-Space Model (SSM) blocks, a class adopted by several recent

Local Support Learning

arXiv:2610.02126v1 Announce Type: cross Abstract: We explore catastrophic forgetting in the context of large pre-trained models. By considering forgetting as a geometric problem in the input space of

MedVL-SAM2: A unified 3D medical vision-language model for multimodal reasoning and prompt-driven segmentation

arXiv:2601.09879v2 Announce Type: replace-cross Abstract: Recent progress in medical vision-language models (VLMs) has achieved strong performance on image-level text-centric tasks such as report gene

Meta-Multi-Agent Reinforcement Learning for Fast Adaptation of Interactive Policies with Applications to Autonomous Driving

arXiv:2610.00705v1 Announce Type: new Abstract: This paper develops a meta-multi-agent reinforcement learning (meta-MARL) framework to enable fast adaptation of interactive policies in a multi-agent s

No One Architecture Fits All: A Cross-Environment Evaluation of Hierarchical Red Team Agents

arXiv:2610.00557v1 Announce Type: cross Abstract: Autonomous red team agents increasingly stress-test AI-enabled cyber defenses by planning strategy and executing multistage attacks. Reinforcement lea

One Basis to Animate Them All: Gaussian Blendshape Distillation for Real-Time Avatars

arXiv:2610.02207v1 Announce Type: cross Abstract: 3D Gaussian avatars support fast rendering, however, their real-time animation is often challenged by the costly neural inference. We address this bot

Agents

OrbitTAMP: Grounding Language Models for Task and Motion Planning in Spacecraft Rendezvous

arXiv:2610.01093v1 Announce Type: cross Abstract: Spacecraft rendezvous and proximity operations (RPO) are currently planned through an expertise-intensive process in which engineers translate high-le

agentsarxiv-cs-ai

T2SPO: Trajectory-to-Step Policy Optimization for Agentic Reinforcement Learning

arXiv:2610.00388v1 Announce Type: cross Abstract: Reinforcement learning enables large language model (LLM) agents to learn multi-step behaviors through interaction with their environments. However, r

TACO: Ternary Absolute-max Column-wise One-sparse Optimizer for LLM Fine-Tuning

arXiv:2610.02199v1 Announce Type: new Abstract: Full-parameter fine-tuning of large language models (LLMs) incurs substantial optimizer state memory overhead, limiting the model sizes that fit on mode