Search: “persistence”

arXiv:2604.21924v1 Announce Type: new Abstract: Long-horizon manipulation remains challenging for vision-language-action (VLA) policies: real tasks are multi-step, progress-dependent, and brittle to c

researcharxiv-cs-ro

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

OpInf-LLM: Parametric PDE Solving with LLMs via Operator Inference

DGX agent

arXiv:2602.01493v2 Announce Type: replace-cross Abstract: Solving diverse partial differential equations (PDEs) is fundamental in science and engineering. Large language models (LLMs) have demonstrate

agentsarxiv-cs-ai

24 Apr 2026

Reasoning Primitives in Hybrid and Non-Hybrid LLMs

arXiv:2604.21454v1 Announce Type: cross Abstract: Reasoning in large language models is often treated as a monolithic capability, but its observed gains may arise from more basic operations. We study

ReFACT: A Benchmark for Scientific Confabulation Detection with Positional Error Annotations

arXiv:2509.25868v3 Announce Type: replace Abstract: The mechanisms underlying scientific confabulation in Large Language Models (LLMs) remain poorly understood. We introduce ReFACT (Reddit False And C

SCM: Sleep-Consolidated Memory with Algorithmic Forgetting for Large Language Models

model-releasesarxiv-cs-lg

arXiv:2604.20943v1 Announce Type: new Abstract: We present SCM (Sleep-Consolidated Memory), a research preview of a memory architecture for large language models that draws on neuroscientific principl

Seeing Isn't Believing: Uncovering Blind Spots in Evaluator Vision-Language Models

arXiv:2604.21523v1 Announce Type: cross Abstract: Large Vision-Language Models (VLMs) are increasingly used to evaluate outputs of other models, for image-to-text (I2T) tasks such as visual question a

SparseGF: A Height-Aware Sparse Segmentation Framework with Context Compression for Robust Ground Filtering Across Urban to Natural Scenes

arXiv:2604.21356v1 Announce Type: new Abstract: High-quality digital terrain models derived from airborne laser scanning (ALS) data are essential for a wide range of geospatial analyses, and their gen

Transient Turn Injection: Exposing Stateless Multi-Turn Vulnerabilities in Large Language Models

arXiv:2604.21860v1 Announce Type: cross Abstract: Large language models (LLMs) are increasingly integrated into sensitive workflows, raising the stakes for adversarial robustness and safety. This pape

VLAA-GUI: Knowing When to Stop, Recover, and Search, A Modular Framework for GUI Automation

arXiv:2604.21375v1 Announce Type: cross Abstract: Autonomous GUI agents face two fundamental challenges: early stopping, where agents prematurely declare success without verifiable evidence, and repet

Behavioral Transfer in AI Agents: Evidence and Privacy Implications

arXiv:2604.19925v1 Announce Type: cross Abstract: AI agents powered by large language models are increasingly acting on behalf of humans in social and economic environments. Prior research has focused

Bridging Mechanistic Interpretability and Prompt Engineering with Gradient Ascent for Interpretable Persona Control

model-releasesarxiv-cs-lg

arXiv:2601.02896v2 Announce Type: replace Abstract: Controlling emergent behavioral personas (e.g., sycophancy, hallucination) in Large Language Models (LLMs) is critical for AI safety, yet remains a

CCTVBench: Contrastive Consistency Traffic VideoQA Benchmark for Multimodal LLMs

arXiv:2604.20460v1 Announce Type: new Abstract: Safety-critical traffic reasoning requires contrastive consistency: models must detect true hazards when an accident occurs, and reliably reject plausib

CHASM: Unveiling Covert Advertisements on Chinese Social Media

arXiv:2604.20511v1 Announce Type: cross Abstract: Current benchmarks for evaluating large language models (LLMs) in social media moderation completely overlook a serious threat: covert advertisements,

Cross-Modal Taxonomic Generalization in (Vision-) Language Models

arXiv:2603.07474v2 Announce Type: replace-cross Abstract: What is the interplay between semantic representations learned by language models (LM) from surface form alone to those learned from more grou

EvoForest: A Novel Machine-Learning Paradigm via Open-Ended Evolution of Computational Graphs

arXiv:2604.19761v1 Announce Type: new Abstract: Modern machine learning is still largely organized around a single recipe: choose a parameterized model family and optimize its weights. Although highly

FeDa4Fair: Client-Level Federated Datasets for Fairness Evaluation

arXiv:2506.21095v4 Announce Type: replace-cross Abstract: Federated Learning (FL) enables collaborative training while preserving privacy, yet it introduces a critical challenge: the 'illusion of fair

LLMs Can Get 'Brain Rot': A Pilot Study on Twitter/X

arXiv:2510.13928v2 Announce Type: replace-cross Abstract: We propose and test the LLM Brain Rot Hypothesis: continual exposure to junk web text induces lasting cognitive decline in large language mode

Prism: An Evolutionary Memory Substrate for Multi-Agent Open-Ended Discovery

arXiv:2604.19795v1 Announce Type: new Abstract: We introduce prism{} (extbf{P}robabilistic extbf{R}etrieval with extbf{I}nformation-extbf{S}tratified extbf{M}emory), an evolutionary memory substrate f

Surrogate modeling for interpreting black-box LLMs in medical predictions

arXiv:2604.20331v1 Announce Type: cross Abstract: Large language models (LLMs), trained on vast datasets, encode extensive real-world knowledge within their parameters, yet their black-box nature obsc

UCCL-Zip: Lossless Compression Supercharged GPU Communication

arXiv:2604.17172v2 Announce Type: replace-cross Abstract: The rapid growth of large language models (LLMs) has made GPU communication a critical bottleneck. While prior work reduces communication volu

AI scientists produce results without reasoning scientifically

arXiv:2604.18805v1 Announce Type: new Abstract: Large language model (LLM)-based systems are increasingly deployed to conduct scientific research autonomously, yet whether their reasoning adheres to t

An AI Agent Execution Environment to Safeguard User Data

arXiv:2604.19657v1 Announce Type: cross Abstract: AI agents promise to serve as general-purpose personal assistants for their users, which requires them to have access to private user data (e.g., pers

AnyRecon: Arbitrary-View 3D Reconstruction with Video Diffusion Model

arXiv:2604.19747v1 Announce Type: new Abstract: Sparse-view 3D reconstruction is essential for modeling scenes from casual captures, but remain challenging for non-generative reconstruction. Existing

Beyond Coefficients: Forecast-Necessity Testing for Interpretable Causal Discovery in Nonlinear Time-Series Models

arXiv:2604.18751v1 Announce Type: cross Abstract: Nonlinear machine-learning models are increasingly used to discover causal relationships in time-series data, yet the interpretation of their outputs

CentaurTA Studio: A Self-Improving Human-Agent Collaboration System for Thematic Analysis

arXiv:2604.18589v1 Announce Type: cross Abstract: Thematic analysis is difficult to scale: manual workflows are labor-intensive, while fully automated pipelines often lack controllability and transpar

FASE : A Fairness-Aware Spatiotemporal Event Graph Framework for Predictive Policing

arXiv:2604.18644v1 Announce Type: cross Abstract: Predictive policing systems that allocate patrol resources based solely on predicted crime risk can unintentionally amplify racial disparities through

Four-Axis Decision Alignment for Long-Horizon Enterprise AI Agents

arXiv:2604.19457v1 Announce Type: new Abstract: Long-horizon enterprise agents make high-stakes decisions (loan underwriting, claims adjudication, clinical review, prior authorization) under lossy mem

Tutorials

How Do Answer Tokens Read Reasoning Traces? Self-Reading Patterns in Thinking LLMs for Quantitative Reasoning

arXiv:2604.19149v1 Announce Type: cross Abstract: Thinking LLMs produce reasoning traces before answering. Prior activation steering work mainly targets on shaping these traces. It remains less unders

tutorialsarxiv-cs-ai

Integrating Anomaly Detection into Agentic AI for Proactive Risk Management in Human Activity

arXiv:2604.19538v1 Announce Type: new Abstract: Agentic AI, with goal-directed, proactive, and autonomous decision-making capabilities, offers a compelling opportunity to address movement-related risk

LLMs Know They're Wrong and Agree Anyway: The Shared Sycophancy-Lying Circuit

arXiv:2604.19117v1 Announce Type: new Abstract: When a language model agrees with a user's false belief, is it failing to detect the error, or noticing and agreeing anyway? We show the latter. Across

safetyarxiv-cs-lg

Local Updates in Distributed Optimization: Provable Acceleration and Topology Effects

arXiv:2601.03442v4 Announce Type: replace-cross Abstract: Inspired by the success of performing multiple local optimization steps between communication rounds in federated learning, incorporating such

applicationsarxiv-cs-lg

MOSA: Motion-Guided Semantic Alignment for Dynamic Scene Graph Generation

arXiv:2604.19631v1 Announce Type: new Abstract: Dynamic Scene Graph Generation (DSGG) aims to structurally model objects and their dynamic interactions in video sequences for high-level semantic under

safetyarxiv-cs-cv

Multi-modal Test-time Adaptation via Adaptive Probabilistic Gaussian Calibration

arXiv:2604.19093v1 Announce Type: cross Abstract: Multi-modal test-time adaptation (TTA) enhances the resilience of benchmark multi-modal models against distribution shifts by leveraging the unlabeled

Position: No Retroactive Cure for Infringement during Training

arXiv:2604.18649v1 Announce Type: cross Abstract: As generative AI faces intensifying legal challenges, the machine learning community has increasingly relied on post-hoc mitigation -- especially mach

ReefNet: A Large-Scale Dataset and Benchmark for Fine-Grained Coral Reef Recognition

arXiv:2510.16822v3 Announce Type: replace-cross Abstract: Coral reefs are rapidly declining under anthropogenic pressures (e.g., climate change), creating an urgent need for scalable and automated mon

Refute-or-Promote: An Adversarial Stage-Gated Multi-Agent Review Methodology for High-Precision LLM-Assisted Defect Discovery

arXiv:2604.19049v1 Announce Type: cross Abstract: LLM-assisted defect discovery has a precision crisis: plausible-but-wrong reports overwhelm maintainers and degrade credibility for real findings. We

RoboWM-Bench: A Benchmark for Evaluating World Models in Robotic Manipulation

arXiv:2604.19092v1 Announce Type: cross Abstract: Recent advances in large-scale video world models have enabled increasingly realistic future prediction, raising the prospect of leveraging imagined v

A Survey of Spatial Memory Representations for Efficient Robot Navigation

arXiv:2604.16482v1 Announce Type: new Abstract: As vision-based robots navigate larger environments, their spatial memory grows without bound, eventually exhausting computational resources, particular

Alexandria: A Multi-Domain Dialectal Arabic Machine Translation Dataset for Culturally Inclusive and Linguistically Diverse LLMs

arXiv:2601.13099v2 Announce Type: replace Abstract: Arabic is a highly diglossic language where most daily communication occurs in regional dialects rather than Modern Standard Arabic (MSA). Despite t

Althea: Human-AI Collaboration for Fact-Checking and Critical Reasoning

arXiv:2602.11161v2 Announce Type: replace-cross Abstract: The web's information ecosystem demands fact-checking systems that are both scalable and epistemically trustworthy. Automated approaches offer

BenchMarker: An Education-Inspired Toolkit for Highlighting Flaws in Multiple-Choice Benchmarks

arXiv:2602.06221v2 Announce Type: replace Abstract: Multiple-choice question answering (MCQA) is standard in NLP, but benchmarks lack rigorous quality control. We present BenchMarker, an education-ins

BengaliMoralBench: A Benchmark for Auditing Moral Reasoning in Large Language Models within Bengali Language and Culture

arXiv:2511.03180v2 Announce Type: replace Abstract: As multilingual Large Language Models (LLMs) gain traction across South Asia, their alignment with local ethical norms, particularly for Bengali, sp

Research

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

Bolzano: Case Studies in LLM-Assisted Mathematical Research

arXiv:2604.16989v1 Announce Type: new Abstract: We report new results on six problems in mathematics and theoretical computer science, produced with the assistance of Bolzano, an open-source multi-age

agentsarxiv-cs-cl

BrainMem: Brain-Inspired Evolving Memory for Embodied Agent Task Planning

arXiv:2604.16331v1 Announce Type: cross Abstract: Embodied task planning requires agents to execute long-horizon, goal-directed actions in complex 3D environments, where success depends on both immedi

agentsarxiv-cs-cv

CDSA-Net:Collaborative Decoupling of Vascular Structure and Background for High-Fidelity Coronary Digital Subtraction Angiography

arXiv:2604.17208v1 Announce Type: new Abstract: Digital subtraction angiography (DSA) in coronary imaging is fundamentally challenged by physiological motion, forcing reliance on raw angiograms clutte

CogDriver: Integrating Cognitive Inertia for Temporally Coherent Planning in Autonomous Driving

arXiv:2509.00789v2 Announce Type: replace Abstract: The pursuit of autonomous agents capable of temporally coherent planning is hindered by a fundamental flaw in current vision-language models (VLMs):

agentsarxiv-cs-cv

Research

Countdown-Code: A Testbed for Studying The Emergence and Generalization of Reward Hacking in RLVR

arXiv:2603.07084v2 Announce Type: replace-cross Abstract: Reward hacking is a form of misalignment in which models overoptimize proxy rewards without genuinely solving the underlying task. Precisely m

researcharxiv-cs-cl