Search: “pretraining”

Periodic Weak Spots: Phase Sensitivity from Chunked KV-Cache Compression

model-releasesarxiv-cs-ai

arXiv:2609.36322v1 Announce Type: cross Abstract: Chunked KV-cache compression reduces the memory and attention costs of long-context inference by compressing windows of consecutive tokens into fewer

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Research

Physical Cross-Modal Masked Autoencoding for Seismic-to-Well Representation Learning

DGX agent

arXiv:2609.36193v1 Announce Type: new Abstract: Learning from scientific measurements often requires aligning modalities with different spatial support and resolution. Subsurface characterization is a

researcharxiv-cs-lg

30 Sept 2026

Point2Part: Unified 3D Partitioning from Point Prompts

arXiv:2609.38180v1 Announce Type: new Abstract: Existing 3D part decomposition methods do not necessarily partition the original shape into non-overlapping parts that collectively cover the entire sha

PreferenceFlow: Test-Time Guidance of Flow-Matching Robot Policies from Human Interventions

arXiv:2609.36872v1 Announce Type: new Abstract: Flow-matching policies can represent complex robot behaviors but remain susceptible to local errors under distribution shift at deployment. Many reinfor

local-aiarxiv-cs-ro

Procedural Core: A Compact Recurrent Initialization for Vision Transformers

arXiv:2609.37631v1 Announce Type: cross Abstract: Transformers are typically trained from random initialization, requiring all their capabilities to emerge from large-scale optimization. Recent work s

Receptive-field-constrained stimulus optimization for human early and intermediate visual cortex

arXiv:2609.36391v1 Announce Type: cross Abstract: An ongoing challenge in sensory neuroscience is to characterize the feature dimensions encoded by cortical populations. Recent approaches probe featur

local-aiarxiv-cs-lg

Recovering the View: Benchmarking Physical Active Vision for Occlusion Recovery in Robotic Manipulation

model-releasesarxiv-cs-ro

arXiv:2609.37292v1 Announce Type: new Abstract: Physical active vision allows robots to change their viewpoint when task-relevant observations become unreliable, yet existing manipulation benchmarks p

Representation by Design in Generation: Cross-View Class-Token Alignment in Diffusion Transformers

arXiv:2609.36348v1 Announce Type: cross Abstract: Generative and representation learning remain asymmetrically connected: semantic representations are used to improve diffusion generation, whereas the

safetyarxiv-cs-ai

Representational Simplicity and Circuit Size Dissociate in a Threshold-Dependent Way: A Controlled Test via Adversarial Training

arXiv:2609.35890v1 Announce Type: new Abstract: Sparse-autoencoder decomposability and concentrated feature attribution are increasingly treated as evidence that a model's computation is easier to rev

researcharxiv-cs-ai

RunyaNER: Auxiliary Language Selection for Runyankore NER

model-releasesarxiv-cs-cl

arXiv:2609.37543v1 Announce Type: new Abstract: Cross-lingual zero-shot transfer and multilingual fine-tuning are promising approaches for NLP tasks such as Named Entity Recognition (NER) in low-resou

Scene-Consistent Illumination Transfer for Inserted Advertising Graphics

arXiv:2609.37951v1 Announce Type: new Abstract: Replacing a visible advertisement in a broadcast frame is geometrically straightforward but photometrically delicate. A pasted graphic can have the corr

researcharxiv-cs-lg

Selecting The Most Informative Tokens in Natural Language Autoencoders

arXiv:2609.37040v1 Announce Type: cross Abstract: Natural language autoencoders translate a language model's internal activations into readable explanations. Explaining every token position is costly.

researcharxiv-cs-ai

Selective Channel Restoration for Backdoored Vision-Language Models

arXiv:2609.37759v1 Announce Type: cross Abstract: Vision-language models (VLMs) exhibit strong multimodal capabilities but remain vulnerable to backdoors implanted through poisoned fine-tuning data. E

Sparse-View Interpretable 3D Animal Behavior Representations for Neural Encoding and Decoding

arXiv:2609.36217v1 Announce Type: new Abstract: A deeper understanding of brain function requires a precise, structured characterization of behavior.Yet, extracting behavioral representations from vid

safetyarxiv-cs-cv

Applications

Support-Set Target Leakage in Relational Foundation Models during In-Context Learning: Model Dependence and Evaluation Reliability

arXiv:2609.36417v1 Announce Type: new Abstract: Relational in-context learning (ICL) uses labeled support examples and their linked relational context to predict labels for new queries. This creates a

applicationsarxiv-cs-ai

T^2Mem: Learning Test-Time Memory for Robotics

model-releasesarxiv-cs-ro

arXiv:2609.36720v1 Announce Type: new Abstract: Memory-dependent robotic manipulation requires policies to use information that is no longer available in the current observation. Retaining history alo

Texture Space Material Diffusion

arXiv:2609.37654v1 Announce Type: new Abstract: We present a method for generating high quality materials for 3D objects entirely in texture space. We finetune a video diffusion transformer for text-g

The Geometry of Inference in Transformer Residual Streams

arXiv:2609.37824v1 Announce Type: new Abstract: Transformer language models build predictions through successive residual updates, but how their representations become specific to an eventual outcome

safetyarxiv-cs-cl

Towards Scalable Context-Aware Single-Cell Spatial Transcriptomics Prediction from Histology Images

arXiv:2609.36429v1 Announce Type: new Abstract: Predicting gene expression from H&E-stained histology images offers a scalable alternative to costly spatial transcriptomics, yet most existing methods

local-aiarxiv-cs-cv

UGO: Unified Architecture for General Multi-Object Tracking by Segmentation

arXiv:2609.37339v1 Announce Type: new Abstract: General multi-object tracking (GMOT) tracks all instances of a user-specified category from a single first-frame exemplar. Prior work relies on bounding

Weights Read and Write Features: Scalable Parameter Decomposition Grounded in Activation Space

arXiv:2609.37731v1 Announce Type: new Abstract: Activation space and parameter space provide complementary views of model computation. Activations represent information, while weights read, transform,

What You Observe Determines How You Identify Causal Effects: Evaluating Causal Models across Observational Views

arXiv:2609.36881v1 Announce Type: new Abstract: Causal foundation models (CFMs) pre-trained on data generated from various structural causal models (SCMs) have been proposed for estimating causal effe

Who Warmed the Archives? LLMs Overestimate Historical Warmth

model-releasesarxiv-cs-cl

arXiv:2609.37499v1 Announce Type: new Abstract: Historical archives are an under-used source for extending the instrumental climate record backward in time, and LLMs offer a way to extract the indices

Xiaomi-OCR-0 Technical Report

arXiv:2609.36136v1 Announce Type: new Abstract: Compact OCR-specific vision-language models achieve strong document parsing performance, but often rely on costly supervision and focus primarily on vis

XU-RS: Explaining Credal Width in Random-Set Language Models

model-releasesarxiv-cs-ai

arXiv:2609.37594v1 Announce Type: new Abstract: Uncertainty estimates tell us how unsure a model is, but not why. Without knowing which parts of an input influences a model's uncertainty, we cannot te

Zero Sum SVD: Balancing Loss Sensitivity for Low Rank LLM Compression

arXiv:2602.02848v2 Announce Type: replace Abstract: Advances in large language models have driven strong performance across many tasks, but their memory and compute costs still hinder deployment. SVD-

researcharxiv-cs-lg

A Comparative Transfer-Learning Study of CNN Backbones for Partial Face Recognition on the SoF Dataset

arXiv:2609.31677v1 Announce Type: new Abstract: Face recognition is widely deployed in surveillance, access control, and forensic workflows, yet accuracy degrades sharply once the face is occluded by

A Surgical Foundation Model Reveals Task-Dependent Label Efficiency

arXiv:2609.31821v1 Announce Type: new Abstract: Developing label-efficient models is a central challenge in surgical AI due to the high cost and scarcity of expert annotation. While self-supervised fo

AD-E2E-JEPA: A Joint-Embedding Predictive Architecture For End-to-End Autonomous Driving

arXiv:2609.34085v1 Announce Type: cross Abstract: Autonomous driving requires extit{world models} that can understand the physical world, reason and plan, and operate safely. In this paper, we first s

Applications

Adaptive Weighted h-Transform Sampling for Coarse-Guided Visual Generation

arXiv:2603.12057v3 Announce Type: replace Abstract: Coarse-guided visual generation, which synthesizes fine visual samples from degraded or low-fidelity coarse references, is essential for various rea

applicationsarxiv-cs-cv

Agents

AI-driven ionic liquid discovery with unified chemical intelligence

arXiv:2511.11257v2 Announce Type: replace-cross Abstract: The rational discovery of functional solvents remains challenging because macroscopic solvent behaviors emerge from complex many-body interact

agentsarxiv-cs-lg

ANaLOG: Anisotropic Native-Latent Operator Guidance for Solving Inverse Problems

arXiv:2609.31933v1 Announce Type: cross Abstract: Native-latent guidance is a recent paradigm for solving inverse problems with latent diffusion models. It replaces repeated evaluations of the image-s

Audio-JEPA: Joint-Embedding Predictive Architecture for Audio Representation Learning

arXiv:2507.02915v2 Announce Type: replace-cross Abstract: Building on the Joint-Embedding Predictive Architecture (JEPA) paradigm, a recent self-supervised learning framework that predicts latent repr

Before Answering: Evidence Sufficiency under Size-Matched Memory Construction

arXiv:2609.32269v1 Announce Type: cross Abstract: Agents that answer questions from compressed or retrieved memory must recognize when the evidence a query needs is no longer in memory. Benchmarks for

Behavioral Foundation Models for Quality Diversity

arXiv:2609.35615v1 Announce Type: new Abstract: Behavioral Foundation Models (BFMs) are an emerging paradigm in reinforcement learning, playing a role analogous to large language models in natural lan

BERT4DTI : BERT-based Model for Predicting Drug-Protein Interactions

arXiv:2609.33254v1 Announce Type: new Abstract: Understanding how drugs interact with protein targets is fundamental to drug discovery, drug repurposing and the early identification of promising thera

Beyond Inpainting: Unleash 3D Understanding for Stable Camera-Controlled Video Generation

arXiv:2601.10214v3 Announce Type: replace Abstract: Camera control has been extensively studied in conditioned video generation; however, performing precisely altering the camera trajectories while fa

Beyond Sparsity: Weight Location and Network Context in Pruned MRI Reconstruction

arXiv:2609.31812v1 Announce Type: cross Abstract: Pruning reduces the number of weights in magnetic resonance imaging (MRI) reconstruction networks. Equal sparsity, however, can retain weights with di

BiMoGen: Bidirectional Motion-Text Generation via Unified Masked Discrete Diffusion

arXiv:2609.35407v1 Announce Type: new Abstract: Text-to-motion generation and motion-to-text captioning are two fundamental tasks in human motion modeling, both grounded in the same underlying motion-

Brain-Conditioned Action Policies for Neural Motor Decoding

arXiv:2609.34561v1 Announce Type: new Abstract: Motor brain-computer interfaces (BCIs) aim to decode motor intention, enabling people with paralysis to control external devices. Neural motor decoding

safetyarxiv-cs-lg

Cache-Aware Conv3D Lowering Across Embedded World-Model Decoders

arXiv:2609.31938v1 Announce Type: new Abstract: Generative world models can provide visual rollouts for embodied planning, yet their feasibility on edge devices depends not only on the learned model b

safetyarxiv-cs-lg

Applications

CAFE: Counterfactual Prediction via Fast Posterior Estimation

arXiv:2609.32167v1 Announce Type: new Abstract: Counterfactual prediction estimates an individual's outcome under an alternative intervention given their factual observations. Such outcomes are genera

applicationsarxiv-cs-lg

Causally Debiased Latent Action Model for Embodied Action-Conditioned World Models

arXiv:2607.09185v2 Announce Type: replace Abstract: Action-conditioned world models (ACWMs) aim to simulate future observations conditioned on embodied actions. However, learning controllable ACWMs re

Control-Geometry Straightening for Sampling-Based Latent Planning

arXiv:2609.35603v1 Announce Type: new Abstract: Joint-embedding predictive architectures enable planning with latent world models, but accurate transition prediction alone does not ensure that the pla

local-aiarxiv-cs-lg

Correcting the Dropout-LayerNorm Expectation Gap Improves Protein Structure Models

arXiv:2609.32062v1 Announce Type: cross Abstract: Although E[dropout(x)] = x, here we show that E[LayerNorm(dropout(x))] is not equal to LayerNorm(x). Accordingly, the pattern of a Dropout layer follo

safetyarxiv-cs-lg

Cost-free Spectral Estimation for Adaptive Newton--Schulz in Matrix Optimizers

arXiv:2609.33047v1 Announce Type: new Abstract: Matrix optimizers such as Muon transform each momentum matrix through an approximate orthogonalization, typically implemented by a small number of Newto

researcharxiv-cs-lg

Cross-Dataset Transfer and Unknown-Class Detection in Imbalanced SAR Ship Classification

arXiv:2609.31658v1 Announce Type: new Abstract: Ship classification from Synthetic Aperture Radar (SAR) imagery is a critical computer vision task, yet the robustness of models under deployment shifts

d-OPD: Future-Aware On-Policy Distillation for Block Diffusion Language Models

arXiv:2609.35362v1 Announce Type: new Abstract: Large language models (LLMs) typically generate text autoregressively (AR), predicting one token at a time. Block diffusion language models (dLLMs) inst