Search: “pretraining”

Mitigating Long-Tail Bias via Prompt-Controlled Diffusion Augmentation

arXiv:2602.04749v2 Announce Type: replace Abstract: Long-tailed class imbalance remains a fundamental obstacle in semantic segmentation of high-resolution remote-sensing imagery, where dominant classe

safetyarxiv-cs-cv

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Safety

Mixture of Predefined Experts: Maximizing Data Usage on Vertical Federated Learning

DGX agent

arXiv:2602.12708v2 Announce Type: replace Abstract: Vertical Federated Learning (VFL) has emerged as a critical paradigm for collaborative model training in privacy-sensitive domains such as finance a

safetyarxiv-cs-lg

22 Apr 2026

Nexusformer: Nonlinear Attention Expansion for Stable and Inheritable Transformer Scaling

arXiv:2604.19147v1 Announce Type: cross Abstract: Scaling Transformers typically necessitates training larger models from scratch, as standard architectures struggle to expand without discarding learn

researcharxiv-cs-ai

OLLM: Options-based Large Language Models

arXiv:2604.19087v1 Announce Type: new Abstract: We introduce Options LLM (OLLM), a simple, general method that replaces the single next-token prediction of standard LLMs with a extit{set of learned op

ORCA: An Agentic Reasoning Framework for Hallucination and Adversarial Robustness in Vision-Language Models

arXiv:2509.15435v2 Announce Type: replace-cross Abstract: Large Vision-Language Models (LVLMs) exhibit strong multimodal capabilities but remain vulnerable to hallucinations from intrinsic errors and

REVEAL: Multimodal Vision-Language Alignment of Retinal Morphometry and Clinical Risks for Incident AD and Dementia Prediction

arXiv:2604.18757v1 Announce Type: cross Abstract: The retina provides a unique, noninvasive window into Alzheimer's disease (AD) and dementia, capturing early structural changes through morphometric f

safetyarxiv-cs-ai

SAGE-32B: Agentic Reasoning via Iterative Distillation

arXiv:2601.04237v2 Announce Type: replace Abstract: We demonstrate SAGE-32B, a 32 billion parameter language model that focuses on agentic reasoning and long range planning tasks. Unlike chat models t

Storage innovations to accelerate your AI workloads at Next ‘26

model-releasesgoogle-cloud-ai

At Google Cloud Next, we are announcing innovations across every layer of our storage stacks — performance, intelligence, and management — to ensure your data is as fast and as useful as the AI models

Tutorials

TabEmb: Joint Semantic-Structure Embedding for Table Annotation

arXiv:2604.18939v1 Announce Type: new Abstract: Table annotation is crucial for making web and enterprise tables usable in downstream NLP applications. Unlike textual data where learning semantically

tutorialsarxiv-cs-lg

Towards Understanding the Robustness of Sparse Autoencoders

arXiv:2604.18756v1 Announce Type: cross Abstract: Large Language Models (LLMs) remain vulnerable to optimization-based jailbreak attacks that exploit internal gradient structure. While Sparse Autoenco

VCE: A zero-cost hallucination mitigation method of LVLMs via visual contrastive editing

arXiv:2604.19412v1 Announce Type: cross Abstract: Large vision-language models (LVLMs) frequently suffer from Object Hallucination (OH), wherein they generate descriptions containing objects that are

Volume Transformer: Revisiting Vanilla Transformers for 3D Scene Understanding

arXiv:2604.19609v1 Announce Type: new Abstract: Transformers have become a common foundation across deep learning, yet 3D scene understanding still relies on specialized backbones with strong domain p

researcharxiv-cs-cv

A Benchmark Study of Segmentation Models and Adaptation Strategies for Landslide Detection from Satellite Imagery

arXiv:2604.16663v1 Announce Type: new Abstract: Landslide detection from high resolution satellite imagery is a critical task for disaster response and risk assessment, yet the relative effectiveness

Agents

AdaExplore: Failure-Driven Adaptation and Diversity-Preserving Search for Efficient Kernel Generation

arXiv:2604.16625v1 Announce Type: new Abstract: Recent large language model (LLM) agents have shown promise in using execution feedback for test-time adaptation. However, robust self-improvement remai

agentsarxiv-cs-cl

AIM 2025 Rip Current Segmentation (RipSeg) Challenge Report

arXiv:2508.13401v3 Announce Type: replace Abstract: This report presents an overview of the AIM 2025 RipSeg Challenge, a competition designed to advance techniques for automatic rip current segmentati

Tutorials

Aletheia: Physics-Conditioned Localized Artifact Attention (PhyLAA-X) for End-to-End Generalizable and Robust Deepfake Video Detection

arXiv:2604.16486v1 Announce Type: new Abstract: State-of-the-art deepfake detectors achieve near-perfect in-domain accuracy yet degrade under cross-generator shifts, heavy compression, and adversarial

tutorialsarxiv-cs-cv

Beyond the Failures: Rethinking Foundation Models in Pathology

arXiv:2510.23807v5 Announce Type: replace-cross Abstract: Despite their successes in vision and language, foundation models have stumbled in pathology, revealing low accuracy, instability, and heavy c

researcharxiv-cs-cv

BhashaSutra: A Task-Centric Unified Survey of Indian NLP Datasets, Corpora, and Resources

arXiv:2604.18423v1 Announce Type: new Abstract: India's linguistic landscape, spanning 22 scheduled languages and hundreds of marginalized dialects, has driven rapid growth in NLP datasets, benchmarks

researcharxiv-cs-cl

CARI4D: Category Agnostic 4D Reconstruction of Human-Object Interaction

arXiv:2512.11988v3 Announce Type: replace Abstract: Accurate capture of human-object interaction from ubiquitous sensors like RGB cameras is important for applications in human understanding, gaming,

CodePivot: Bootstrapping Multilingual Transpilation in LLMs via Reinforcement Learning without Parallel Corpora

arXiv:2604.18027v1 Announce Type: cross Abstract: Transpilation, or code translation, aims to convert source code from one programming language (PL) to another. It is beneficial for many downstream ap

Common Corpus: The Largest Collection of Ethical Data for LLM Pre-Training

arXiv:2506.01732v2 Announce Type: replace Abstract: Large Language Models (LLMs) are pre-trained on large data from different sources and domains. These datasets often contain trillions of tokens, inc

applicationsarxiv-cs-cl

Continual Safety Alignment via Gradient-Based Sample Selection

arXiv:2604.17215v1 Announce Type: new Abstract: Large language models require continuous adaptation to new tasks while preserving safety alignment. However, fine-tuning on even benign data often compr

safetyarxiv-cs-lg

Tutorials

ControlAudio: Tackling Text-Guided, Timing-Indicated and Intelligible Audio Generation via Progressive Diffusion Modeling

arXiv:2510.08878v3 Announce Type: replace-cross Abstract: Text-to-audio (TTA) generation with fine-grained control signals, e.g., precise timing control or intelligible speech content, has been explor

tutorialsarxiv-cs-cl

Counterfactual Modeling with Fine-Tuned LLMs for Health Intervention Design and Sensor Data Augmentation

arXiv:2601.14590v2 Announce Type: replace Abstract: Counterfactual explanations (CFEs) provide human-centric interpretability by identifying the minimal, actionable changes required to alter a machine

DOSE: Data Selection for Multi-Modal LLMs via Off-the-Shelf Models

arXiv:2604.16979v1 Announce Type: cross Abstract: High-quality and diverse multimodal data are essential for improving vision-language models (VLMs), yet existing datasets often contain noisy, redunda

safetyarxiv-cs-cl

DuConTE: Dual-Granularity Text Encoder with Topology-Constrained Attention for Text-attributed Graphs

arXiv:2604.17411v1 Announce Type: new Abstract: Text-attributed graphs integrate semantic information of node texts with topological structure, offering significant value in various applications such

eCP: Equivariant Conformal Prediction with pre-trained models

arXiv:2602.03986v2 Announce Type: replace Abstract: Conformal prediction, a post-hoc, distribution-free, finite-sample method of uncertainty quantification that offers formal coverage guarantees under

researcharxiv-cs-lg

EmbodiedHead: Real-Time Listening and Speaking Avatar for Conversational Agents

arXiv:2604.17211v1 Announce Type: new Abstract: We present EmbodiedHead, a speech-driven talking-head framework that equips LLMs with real-time visual avatars for conversation. A practical embodied av

researcharxiv-cs-cv

ENTIRE: Learning-based Volume Rendering Time Prediction

arXiv:2501.12119v3 Announce Type: replace-cross Abstract: We introduce ENTIRE, a novel deep learning-based approach for fast and accurate volume rendering time prediction. Predicting rendering time is

From User Recognition to Activity Counting: An Identity-Agnostic Approach to Multi-User WiFi Sensing

arXiv:2604.16572v1 Announce Type: new Abstract: Wi-Fi Channel State Information (CSI) enables device-free human activity recognition, but existing multi-user approaches assume a fixed set of known use

researcharxiv-cs-lg

Frozen Vision Transformers for Dense Prediction on Small Datasets: A Case Study in Arrow Localization

arXiv:2604.16758v1 Announce Type: new Abstract: We present a system for automated detection, localization, and scoring of arrow punctures on 40,cm indoor archery target faces, trained on only 48 annot

applicationsarxiv-cs-cv

Generalizable Face Forgery Detection via Separable Prompt Learning

arXiv:2604.17307v1 Announce Type: new Abstract: Detecting face forgeries using CLIP has recently emerged as a promising and increasingly popular research direction. Owing to its rich visual knowledge

Geometric Stability: The Missing Axis of Representations

arXiv:2601.09173v4 Announce Type: replace-cross Abstract: Representational similarity analysis and related methods have become standard tools for comparing the internal geometries of neural networks a

safetyarxiv-cs-cl

Guardrails in Logit Space: Safety Token Regularization for LLM Alignment

arXiv:2604.17210v1 Announce Type: new Abstract: Fine-tuning well-aligned large language models (LLMs) on new domains often degrades their safety alignment, even when using benign datasets. Existing sa

HORIZON: A Benchmark for In-the-wild User Behaviour Modeling

arXiv:2604.17259v1 Announce Type: cross Abstract: User behavior in the real world is diverse, cross-domain, and spans long time horizons. Existing user modeling benchmarks however remain narrow, focus

Hyperbolic Enhanced Representation Learning for Incomplete Multi-view Clustering

arXiv:2604.16959v1 Announce Type: cross Abstract: Incomplete Multi-View Clustering (IMVC) faces the challenge of learning discriminative representations from fragmentary observations while maintaining

safetyarxiv-cs-cv

In-Context Learning Under Regime Change

arXiv:2604.16988v1 Announce Type: new Abstract: Non-stationary sequences arise naturally in control, forecasting, and decision-making. The data-generating process shifts at unknown times, and models m

applicationsarxiv-cs-lg

Knowledge without Wisdom: Measuring Misalignment between LLMs and Intended Impact

arXiv:2603.00883v2 Announce Type: replace Abstract: LLMs increasingly excel on AI benchmarks, but doing so does not guarantee validity for downstream tasks. This study contrasts LLM alignment on bench

Learning to Control Summaries with Score Ranking

arXiv:2604.17197v1 Announce Type: new Abstract: Recent advances in summarization research focus on improving summary quality across multiple criteria, such as completeness, conciseness, and faithfulne

Live Avatar: Streaming Real-time Audio-Driven Avatar Generation with Infinite Length

arXiv:2512.04677v5 Announce Type: replace Abstract: Audio-driven avatar interaction demands real-time, streaming, and infinite-length generation -- capabilities fundamentally at odds with the sequenti

LIVE: Leveraging Image Manipulation Priors for Instruction-based Video Editing

arXiv:2604.17021v1 Announce Type: new Abstract: Video editing aims to modify input videos according to user intent. Recently, end-to-end training methods have garnered widespread attention, constructi

LLM Hypnosis: Exploiting User Feedback for Unauthorized Knowledge Injection to All Users

arXiv:2507.02850v3 Announce Type: replace Abstract: We describe a vulnerability in language models (LMs) trained with user feedback, whereby a single user can persistently alter LM knowledge and behav

researcharxiv-cs-cl

Low-rank Orthogonalization for Large-scale Matrix Optimization with Applications to Foundation Model Training

arXiv:2509.11983v2 Announce Type: replace Abstract: Neural network (NN) training is inherently a large-scale matrix optimization problem, yet the matrix structure of NN parameters has long been overlo

Mammo-FM: Breast-specific foundational model for Integrated Mammographic Diagnosis, Prognosis, and Reporting

arXiv:2512.00198v2 Announce Type: replace Abstract: Breast cancer is one of the leading causes of death among women worldwide. We introduce Mammo-FM, the first foundation model specifically for mammog

safetyarxiv-cs-cv

MobileAgeNet: Lightweight Facial Age Estimation for Mobile Deployment

arXiv:2604.17007v1 Announce Type: new Abstract: Mobile deployment of facial age estimation requires models that balance predictive accuracy with low latency and compact size. In this work, we present

Motif-Video 2B: Technical Report

arXiv:2604.16503v1 Announce Type: new Abstract: Training strong video generation models usually requires massive datasets, large parameter counts, and substantial compute. In this work, we ask whether

MoVE: Translating Laughter and Tears via Mixture of Vocalization Experts in Speech-to-Speech Translation

arXiv:2604.17435v1 Announce Type: new Abstract: Recent Speech-to-Speech Translation (S2ST) systems achieve strong semantic accuracy yet consistently strip away non-verbal vocalizations (NVs), such as

applicationsarxiv-cs-cl

Multimodal Policy Internalization for Conversational Agents

arXiv:2510.09474v2 Announce Type: replace Abstract: Modern conversational agents like ChatGPT and Alexa+ rely on predefined policies specifying metadata, response styles, and tool-usage rules. As thes

safetyarxiv-cs-cl