Search: “pretraining”

Your Transformer Can Hold Two Thoughts at Once: Evidence of Linear Superposition in LLMs

arXiv:2609.29845v1 Announce Type: cross Abstract: While Large Language Models (LLMs) rely on highly non-linear components, in this work we demonstrate that they exhibit fundamental linearity: when inp

25 Sept 2026

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

Artificial intelligence surrogates for treatment effect estimation with before-and-after data

DGX agent

arXiv:2609.27180v1 Announce Type: cross Abstract: Estimating the causal effects of medical treatments is difficult when clinically important outcomes are costly to measure or require long follow-up. S

safetyarxiv-cs-lg

24 Sept 2026

Distilling Sequential Computation in Transformer Language Models

arXiv:2609.27233v1 Announce Type: new Abstract: Transformer language models process sequences token by token in an autoregressive manner, making growing contexts increasingly expensive. Yet many adjac

researcharxiv-cs-cl

From ECG Signals to Representative-Morphology Heatmaps for Biometric Recognition

arXiv:2609.28183v1 Announce Type: new Abstract: Electrocardiography (ECG) contains subject-specific morphology that supports biometric recognition, yet image-based performance depends on how the wavef

Frozen Flows Forget: Diagnosing and Restoring Lost Motion in a Latent-flow World Model

arXiv:2609.28414v1 Announce Type: cross Abstract: Latent world models that integrate a flow in a frozen self supervised latent space train stably and cheaply, yet silently lose the property manipulati

Agents

Hunyuan-A13B Technical Report

arXiv:2609.27284v1 Announce Type: new Abstract: We present Hunyuan-A13B, an open-source large language model based on a Mixture-of-Experts architecture. It contains 80 billion total parameters but act

agentsarxiv-cs-ai

Learning a Speed-adaptive Hip Exoskeleton Control Policy Via Sim-to-real Reinforcement Learning

arXiv:2609.28027v1 Announce Type: new Abstract: Providing personalized exoskeleton assistance across varying walking speeds remains challenging. Existing online optimization methods are sample-ineffic

Linear RNN Scaling Laws: When Longer Sequences Beat More Sequences

arXiv:2609.27964v1 Announce Type: new Abstract: Empirical scaling laws for autoregressive language models relate prediction loss to model size, data size, and optimization compute, but their theoretic

PhyMo: A Physical-Field Modality for Multimodal AI4Physics

arXiv:2609.27554v1 Announce Type: cross Abstract: Multimodal learning is emerging as a powerful paradigm for AI for Physics (AI4Physics), where predicting physical systems requires the joint interpret

Recursive Uncertainty-Gated Image Registration for Learning-based Algorithms

arXiv:2609.28081v1 Announce Type: cross Abstract: Conventional image registration algorithms are robust to domain shifts and achieve low errors, but they are slow and computationally expensive. Deep-l

Relative Discharge Stage (RDS) Classification: A Practical Indicator of Battery Discharge Progress

arXiv:2609.27986v1 Announce Type: new Abstract: Accurate remaining discharge time (RDT) prediction is challenging in real-world battery applications because future load profiles are unknown and highly

Repurposing Pre-trained LLMs as High Fidelity Continuous Text Autoencoders

arXiv:2609.27248v1 Announce Type: new Abstract: Next-token prediction has enabled highly fluent autoregressive language models, but it represents global structure only indirectly through sequential fa

Self-Improvement as Coherence Optimization: A Theoretical Account

arXiv:2601.13566v2 Announce Type: replace-cross Abstract: Can language models improve their accuracy without external supervision? Methods such as debate, bootstrap, and internal coherence maximizatio

Super-Resolution of Solar Magnetograms via Adaptive Stratified Ensemble Learning with Uncertainty Estimation

arXiv:2609.27131v1 Announce Type: new Abstract: Single-image super-resolution of Sun's photospheric magnetograms enables consistent analysis across heterogeneous space-based instruments and supports l

Track2Art: Motion-Centric Articulated Object Model Recovery from 2D Point Trackers

arXiv:2609.27675v1 Announce Type: new Abstract: Understanding articulated objects is fundamental for robotic interaction, requiring accurate rigid-part discovery and the recovery of their kinematic re

Transfer Learning with Conformalized Quantile Regression for Solar PV Forecasting Under Load-Shedding-Driven Data Scarcity

arXiv:2609.26959v1 Announce Type: new Abstract: Solar photovoltaic (PV) forecasting in regions affected by load shedding is challenging because reliable historical observations are scarce. This study

Transferable Evidence Reconstruction for Longitudinal Glucose Representations

arXiv:2609.28199v1 Announce Type: new Abstract: Long physiological recordings contain many routine measurements, while predictive information is often concentrated in rare events, sustained burden, an

Using Vision Language Foundation Models to Generate Plant Simulation Configurations via In-Context Learning

model-releasesarxiv-cs-ai

arXiv:2603.08930v2 Announce Type: replace-cross Abstract: This paper introduces a benchmark for evaluating whether vision-language models (VLMs) can generate plant simulation configurations from image

VIVAS: Vitalizing Visual Perception in VLM Pre-training via Vision-language Unified Autoregressive Supervision

arXiv:2609.27948v1 Announce Type: new Abstract: While Vision-Language Models (VLMs) demonstrate strong capabilities, they continue to suffer from a critical limitation: insufficient fine-grained visua

Watch, Recall, Act: Always-On Robots in Concurrent Embodied Streams

arXiv:2609.28429v1 Announce Type: new Abstract: An always-on robot faces an endless stream that never resets: instructions arrive and lapse, the scene changes, and its own past actions reshape what it

Zero-Shot Object Removal via Attention Masking, Latent Anchoring, and Refinement

arXiv:2609.28342v1 Announce Type: new Abstract: Removing an object from a real image requires more than synthesizing plausible content within a mask: the method must suppress residual object features,

Beyond Class Marginals: Bounding Rehearsal Gaps without Freezing Class Co-occurrence

arXiv:2609.25735v1 Announce Type: new Abstract: Class-balanced replay controls class frequency but does not determine the interval between successive replay appearances of a class. We study this inter

safetyarxiv-cs-lg

Tutorials

Brain-Inspired Hierarchical Modularity for General Continual Learning

arXiv:2609.25146v1 Announce Type: new Abstract: Continual learning, the ability to learn from sequential experience while retaining and adapting prior knowledge, is central to intelligent systems oper

tutorialsarxiv-cs-lg

Complementary Roles of Radiomics and Foundation Representations in Renal Cell Carcinoma Classification: A Comparative Study of 2D and 3D CT Encodings

arXiv:2609.26463v1 Announce Type: new Abstract: Accurate preoperative subtype classification of renal cell carcinoma (RCC) from contrast-enhanced computed tomography remains clinically challenging. Ra

DreamStream: Towards Policy-Oriented Generative Simulation for End-to-End Driving

model-releasesarxiv-cs-ro

arXiv:2609.26792v1 Announce Type: new Abstract: Faithfully evaluating end-to-end driving policies in simulation requires observations that are not merely photo-realistic, but preserve the scene featur

Fisheye-VLA: Decoupling Coverage and Acuity for Manipulation with a Single Fisheye Camera

arXiv:2609.25750v1 Announce Type: new Abstract: Manipulation requires both broad scene awareness and detailed local feedback, yet conventional camera rigs provide them through separate front and wrist

researcharxiv-cs-ro

ForeDrive: Foresight-Guided End-to-End Autonomous Driving with a Planning-Relevant Latent World Model

arXiv:2609.26299v2 Announce Type: new Abstract: Existing latent world models are typically optimized for future predictability, yet the resulting representations are not necessarily useful for plannin

safetyarxiv-cs-cv

GGUFs in transformers natively!

model-releasesr-localllama

Hey there folks! Aritra here from Hugging Face. I wanted to update you all about the latest changes in `transformers`. We now natively support GGUFs (llama cpp quants). You can use it like so: from tr

HABILIS Brain 0: Geometry-Change Supervision for Vision-Language-Action and Residual Flow Recovery

arXiv:2609.25558v1 Announce Type: cross Abstract: Vision-language-action policies benefit from geometric supervision, but current-frame geometry alone does not explicitly describe the changes associat

safetyarxiv-cs-lg

ImIR: Image-Instruction Tuning for All-in-One Image Restoration

model-releasesarxiv-cs-cv

arXiv:2609.25267v1 Announce Type: new Abstract: Degradations vary widely across images, so a practical restoration system has to handle many degradation types with one model. A recent and effective re

Isolated Sign Language Recognition for Icelandic Sign Language: Experiments in a Low-resource Setting

arXiv:2609.25862v1 Announce Type: new Abstract: We present the first experiments on isolated sign language recognition (ISLR) for Icelandic Sign Language (ITM). We use ITM SignWiki, a dataset derived

researcharxiv-cs-cl

Label-Efficient Learning for Ground-Based Sky-Image Classification: A Benchmark of Transfer Learning, Active Learning, and Pseudo-Labeling on GCD

arXiv:2609.26631v1 Announce Type: new Abstract: Accurate ground-based cloud classification is important for atmospheric monitoring, solar-energy forecasting, aviation weather assessment, and climate o

Latent Commonality Expectation-Maximisation for Box-supervised Tree Crown Instance Segmentation

model-releasesarxiv-cs-cv

arXiv:2609.26549v1 Announce Type: new Abstract: Individual tree crown segmentation from aerial imagery underpins tree-level carbon accounting, biodiversity, and restoration monitoring at landscape sca

Latent Dataset Distillation for Human Motion Prediction

arXiv:2609.26430v1 Announce Type: new Abstract: Dataset distillation (DD) compresses a large training set into a compact synthetic set while preserving downstream training utility. Although DD has bee

MachEmbodied-U0: Unified Understanding and Generation Model for Embodied Intelligence

model-releasesarxiv-cs-ro

arXiv:2609.25627v1 Announce Type: new Abstract: General-purpose robot control requires models to understand task intent, identify where to interact, capture how the scene evolves, and generate precise

Margin-Drop Coordinates for Cross-Budget Robustness Evaluation

arXiv:2609.26081v1 Announce Type: new Abstract: Fixed-budget robustness evaluation can select the wrong frozen vision encoder. An encoder that survives a shallow attack may lose most of that robustnes

MATES: Learning Multi-Agent Interactions by Transforming Observations for Frozen Single-Agent Policies

arXiv:2609.26010v1 Announce Type: cross Abstract: Multi-agent reinforcement learning (MARL) commonly trains decentralized policies from scratch, requiring agents to acquire individual task competence

MIND the Gap: A Geographic Implicit Neural Representation with Adjustable Spatial Scale

arXiv:2609.25454v1 Announce Type: cross Abstract: Geographic measurements are often sparse, leaving large areas without labels for the quantities we want to map. Geographic implicit neural representat

Modular Norm RandOpt: Population-Efficient Ensembling through Architecture-Aware Perturbations

arXiv:2609.25745v1 Announce Type: new Abstract: RandOpt samples weight-perturbed language models and ensembles top-ranked candidates through plurality voting, but its global perturbation scale ignores

Neoadjuvant chemotherapy response prediction using pretreatment diffusion and contrast-enhanced magnetic resonance imaging with clinical variables

arXiv:2609.26105v1 Announce Type: cross Abstract: Prediction of pathological complete response before neoadjuvant chemotherapy may facilitate more tailored therapeutic planning for breast cancer patie

OmniEcho: Spatial Audio Understanding for Embodied Agents

model-releasesarxiv-cs-ai

arXiv:2609.23407v1 Announce Type: cross Abstract: Humans can effortlessly localize the direction of a sound source and integrate it with visual cues for reasoning, yet this remains challenging for emb

PLAT: Sparse Timed Keyframe Motion Tracking for Humanoid Control via Privileged Latent Transition Learning

arXiv:2609.25754v1 Announce Type: new Abstract: Humanoid motion tracking policies rely on dense frame-by-frame references, limiting their use as high-level motion controllers for planning and interact

Radiomics--Foundation Fusion for Interpretable RCC Classification: Internal Benchmarking and Exploratory External Transfer

arXiv:2609.26578v1 Announce Type: new Abstract: Accurate preoperative subtype classification of renal cell carcinoma (RCC) from contrast-enhanced CT remains clinically challenging because clear cell R

RouteRLT: Learning When and Which RL Specialist Should Control a Vision-Language-Action Policy

arXiv:2609.26467v1 Announce Type: new Abstract: Vision-language-action (VLA) models provide broad manipulation competence, but often struggle during the precision-critical stages that dominate contact

SAM-V: Geometry-Aware Segment Anything for Multi-View Instance Segmentation

arXiv:2609.25490v1 Announce Type: cross Abstract: Consistent multi-view object segmentation is critical for 3D perception and robotics, yet remains challenging under severe viewpoint and occlusion cha

Sample, Simulate, Select: Physics-in-the-Loop Text-to-Motion for Humanoids Without Training

arXiv:2609.26420v1 Announce Type: new Abstract: Text-to-motion models generate plausible human motion but do not model a robot's dynamics; whole-body tracking controllers execute robot references reli

Sex Estimation from Footwear Outsole Impressions Using CNN Transfer Learning and Interpretable Image Statistics

arXiv:2609.25386v1 Announce Type: cross Abstract: Footwear outsole impressions are a common form of forensic pattern evidence, yet quantitative methods for estimating wearer attributes from these imag

SPARC: SuperPixel-Aware Region Contrastive Learning for Self-Supervised Dense Prediction

arXiv:2609.25067v1 Announce Type: cross Abstract: Self-supervised learning (SSL) has become an effective approach for learning visual representations without manual annotations. Among SSL approaches,