Search: “benchmark”

Beyond the Mean: Within-Model Reliable Change Detection for LLM Evaluation

arXiv:2604.27405v1 Announce Type: cross Abstract: We adapted the Reliable Change Index (RCI; Jacobson and Truax, 1991) from clinical psychology to item-level LLM version comparison on 2,000 MMLU-Pro i

Safety

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

BicKD: Bilateral Contrastive Knowledge Distillation

DGX agent

arXiv:2602.01265v2 Announce Type: replace Abstract: Knowledge distillation (KD) is a machine learning framework that transfers knowledge from a teacher model to a student model. The vanilla KD propose

safetyarxiv-cs-lg

1 May 2026

Can Tabular Foundation Models Guide Exploration in Robot Policy Learning?

arXiv:2604.27667v1 Announce Type: cross Abstract: Policy optimization in high-dimensional continuous control for robotics remains a challenging problem. Predominant methods are inherently local and of

local-aiarxiv-cs-lg

ChipLingo: A Systematic Training Framework for Large Language Models in EDA

arXiv:2604.27415v1 Announce Type: new Abstract: With the rapid advancement of semiconductor technology, Electronic Design Automation (EDA) has become an increasingly knowledge-intensive and document-d

Compositional Meta-Learning for Mitigating Task Heterogeneity in Physics-Informed Neural Networks

arXiv:2604.26999v1 Announce Type: new Abstract: Physics-informed neural networks (PINNs) approximate solutions of partial differential equations (PDEs) by embedding physical laws into the loss functio

researcharxiv-cs-ai

Applications

ConformaDecompose: Explaining Uncertainty via Calibration Localization

arXiv:2604.27149v1 Announce Type: cross Abstract: Conformal Prediction provides distribution-free prediction intervals with guaranteed coverage, but its reliance on a single global calibration thresho

applicationsarxiv-cs-ai

Contextual Agentic Memory is a Memo, Not True Memory

arXiv:2604.27707v1 Announce Type: new Abstract: Current agentic memory systems (vector stores, retrieval-augmented generation, scratchpads, and context-window management) do not implement memory: they

Continuous-tone Simple Points: An ell_0-Norm of Cyclic Gradient for Topology-Preserving Data-Driven Image Segmentation

arXiv:2604.28159v1 Announce Type: new Abstract: Topological features play an essential role in ensuring geometric plausibility and structural consistency in image analysis tasks such as segmentation a

Cross-Lingual Response Consistency in Large Language Models: An ILR-Informed Evaluation of Claude Across Six Languages

model-releasesarxiv-cs-cl

arXiv:2604.27137v1 Announce Type: new Abstract: This paper introduces a systematic evaluation framework grounded in the Interagency Language Roundtable (ILR) Skill Level Descriptions and applies it to

Tutorials

Crosscoding Through Time: Tracking Emergence & Consolidation Of Linguistic Representations Throughout LLM Pretraining

arXiv:2509.05291v2 Announce Type: replace-cross Abstract: Large language models (LLMs) learn non-trivial abstractions during pretraining, such as detecting irregular plural noun subjects. However, bec

tutorialsarxiv-cs-ai

Safety

Debiasing Reward Models via Causally Motivated Inference-Time Intervention

arXiv:2604.27495v1 Announce Type: cross Abstract: Reward models (RMs) play a central role in aligning large language models (LLMs) with human preferences. However, RMs are often sensitive to spurious

safetyarxiv-cs-ai

DEFault++: Automated Fault Detection, Categorization, and Diagnosis for Transformer Architectures

arXiv:2604.28118v1 Announce Type: cross Abstract: Transformer models are widely deployed in critical AI applications, yet faults in their attention mechanisms, projections, and other internal componen

Differential Subgroup Discovery: Characterizing Where Two Populations Differ, and Why

arXiv:2604.27741v1 Announce Type: new Abstract: We study the problem of understanding where two populations differ within a feature space, which we formalize in the concept of a differential subgroup:

researcharxiv-cs-lg

Do World Action Models Generalize Better than VLAs? A Robustness Study

model-releasesarxiv-cs-ro

arXiv:2603.22078v3 Announce Type: replace Abstract: Robot action planning in the real world is challenging as it requires not only understanding the current state of the environment but also predictin

Dynamic Scaled Gradient Descent for Stable Fine-Tuning for Classifications

arXiv:2604.27987v1 Announce Type: new Abstract: Fine-tuning pretrained models has become a standard approach to adapting pretrained knowledge to improve the accuracy on new sparse, imbalance datasets.

Echo-{alpha}: Large Agentic Multimodal Reasoning Model for Ultrasound Interpretation

arXiv:2604.28011v1 Announce Type: new Abstract: Ultrasound interpretation requires both precise lesion localization and holistic clinical reasoning, yet existing methods typically excel at only one of

local-aiarxiv-cs-cv

EdgeSpike: Spiking Neural Networks for Low-Power Autonomous Sensing in Edge IoT Architectures

arXiv:2604.27004v1 Announce Type: cross Abstract: We propose EdgeSpike, a co-designed spiking neural network (SNN) framework for autonomous low-power sensing in edge Internet of Things (IoT) architect

EDU-CIRCUIT-HW: Evaluating Multimodal Large Language Models on Real-World University-Level STEM Student Handwritten Solutions

arXiv:2602.00095v3 Announce Type: replace-cross Abstract: Multimodal Large Language Models (MLLMs) hold significant promise for revolutionizing traditional education and reducing teachers' workload. H

Effective Prompt Pool Learning for Continual Category Discovery

arXiv:2407.19001v3 Announce Type: replace Abstract: This paper studies effective prompt pool learning for Continual Category Discovery (CCD), a challenging open-world setting where a model must discov

local-aiarxiv-cs-cv

Tutorials

Efficient Sparse Selective-Update RNNs for Long-Range Sequence Modeling

arXiv:2603.02226v2 Announce Type: replace Abstract: Real-world sequential signals, such as audio or video, contain critical information that is often embedded within long periods of silence or noise.

tutorialsarxiv-cs-lg

Applications

Energy-Efficient Plant Monitoring via Knowledge Distillation

arXiv:2604.27178v1 Announce Type: new Abstract: Recent advances in large-scale visual representation learning have significantly improved performance in plant species and plant disease recognition tas

applicationsarxiv-cs-cv

Exploring Interaction Paradigms for LLM Agents in Scientific Visualization

arXiv:2604.27996v1 Announce Type: new Abstract: This paper examines how different types of large language model (LLM) agents perform on scientific visualization (SciVis) tasks, where users generate vi

Faster 3D Gaussian Splatting Convergence via Structure-Aware Densification

arXiv:2604.28016v1 Announce Type: new Abstract: 3D Gaussian Splatting has emerged as a powerful scene representation for real-time novel-view synthesis. However, its standard adaptive density control

Applications

FinCARDS: Card-Based Analyst Reranking for Financial Document Question Answering

arXiv:2601.06992v2 Announce Type: replace-cross Abstract: Financial question answering (QA) over long corporate filings requires evidence to satisfy strict constraints on entities, financial metrics,

applicationsarxiv-cs-ai

FMCL: Class-Aware Client Clustering with Foundation Model Representations for Heterogeneous Federated Learning

arXiv:2604.27510v1 Announce Type: cross Abstract: Federated Learning (FL) enables collaborative model training across distributed clients without sharing raw data, yet its performance deteriorates und

Safety

FP-IRL: Fokker--Planck Inverse Reinforcement Learning -- A Physics-Constrained Approach to Markov Decision Processes

arXiv:2306.10407v3 Announce Type: replace-cross Abstract: Inverse reinforcement learning (IRL) is a powerful paradigm for uncovering the incentive structure that drives agent behavior, by inferring an

safetyarxiv-cs-ai

FreeOcc: Training-Free Embodied Open-Vocabulary Occupancy Prediction

arXiv:2604.28115v1 Announce Type: cross Abstract: Existing learning-based occupancy prediction methods rely on large-scale 3D annotations and generalize poorly across environments. We present FreeOcc,

Gait Recognition via Deep Residual Networks and Multi-Branch Feature Fusion

arXiv:2604.27353v1 Announce Type: new Abstract: Gait recognition has emerged as a compelling biometric modality for surveillance and security applications, offering inherent advantages such as non-int

Generalizable Sparse-View 3D Reconstruction from Unconstrained Images

arXiv:2604.28193v1 Announce Type: new Abstract: Reconstructing 3D scenes from sparse, unposed images remains challenging under real-world conditions with varying illumination and transient occlusions.

Generate Your Talking Avatar from Video Reference

arXiv:2604.27918v1 Announce Type: new Abstract: Existing talking avatar methods typically adopt an image-to-video pipeline conditioned on a static reference image within the same scene as the target g

Geometry-Calibrated Conformal Abstention for Language Models

arXiv:2604.27914v1 Announce Type: new Abstract: When language models lack relevant knowledge for a given query, they frequently generate plausible responses that can be hallucinations, rather than adm

researcharxiv-cs-cl

Global Optimality for Constrained Exploration via Penalty Regularization

arXiv:2604.28144v1 Announce Type: new Abstract: Efficient exploration is a central problem in reinforcement learning and is often formalized as maximizing the entropy of the state-action occupancy mea

Gradient-Based Optimization on Godel Logic as Discrete Local Search

arXiv:2503.01817v2 Announce Type: replace Abstract: A fundamental challenge in neurosymbolic systems is applying continuous gradient-based optimization to discrete logical domains. While fuzzy relaxat

local-aiarxiv-cs-lg

Tutorials

Graph World Models: Concepts, Taxonomy, and Future Directions

arXiv:2604.27895v1 Announce Type: new Abstract: As one of the mainstream models of artificial intelligence, world models allow agents to learn the representation of the environment for efficient predi

tutorialsarxiv-cs-ai

Grounding Agent Memory in Contextual Intent

arXiv:2601.10702v2 Announce Type: replace-cross Abstract: Deploying large language models in long-horizon, goal-oriented interactions remains challenging because similar entities and facts recur under

GuideDog: A Real-World Egocentric Multimodal Dataset for Blind and Low-Vision Accessibility-Aware Guidance

arXiv:2503.12844v2 Announce Type: replace Abstract: For people affected by blindness and low vision (BLV), safe and independent navigation remains a major challenge, impacting over 2.2 billion individ

HERMES++: Toward a Unified Driving World Model for 3D Scene Understanding and Generation

arXiv:2604.28196v1 Announce Type: new Abstract: Driving world models serve as a pivotal technology for autonomous driving by simulating environmental dynamics. However, existing approaches predominant

HighFM: Towards a Foundation Model for Learning Representations from High-Frequency Earth Observation Data

arXiv:2604.04306v2 Announce Type: replace-cross Abstract: The increasing frequency and severity of climate related disasters have intensified the need for real time monitoring, early warning, and info

HiMix: Hierarchical Artifact-aware Mixup for Generalized Synthetic Image Detection

arXiv:2604.27903v1 Announce Type: new Abstract: The rapid evolution of generative models has enabled the creation of highly realistic and diverse synthetic images, posing significant challenges to rel

Hinge Regression Tree: A Newton Method for Oblique Regression Tree Splitting

arXiv:2602.05371v3 Announce Type: replace Abstract: Oblique decision trees combine the transparency of trees with the power of multivariate decision boundaries, but learning high-quality oblique split

local-aiarxiv-cs-lg

How Generative AI Disrupts Search: An Empirical Study of Google Search, Gemini, and AI Overviews

arXiv:2604.27790v1 Announce Type: cross Abstract: Generative AI is being increasingly integrated into web search for the convenience it provides users. In this work, we aim to understand how generativ

Hypencoder Revisited: Reproducibility and Analysis of Non-Linear Scoring for First-Stage Retrieval

arXiv:2604.27037v1 Announce Type: cross Abstract: The Hypencoder, proposed by Killingback et al., is a retrieval framework that replaces the fixed inner-product scoring function used in standard bi-en

researcharxiv-cs-cl

Hyper-Dimensional Fingerprints as Molecular Representations

arXiv:2604.27810v1 Announce Type: new Abstract: Computational molecular representations underpin virtual screening, property prediction, and materials discovery. Conventional fingerprints are efficien

researcharxiv-cs-lg

Agents

ImagineNav++: Prompting Vision-Language Models as Embodied Navigator through Scene Imagination

arXiv:2512.17435v3 Announce Type: replace Abstract: Visual navigation is a fundamental capability for autonomous home-assistance robots, enabling long-horizon tasks such as object search. While recent

agentsarxiv-cs-ro

Improving Graph Few-shot Learning with Hyperbolic Space and Denoising Diffusion

arXiv:2604.27462v1 Announce Type: cross Abstract: Graph few-shot learning, which focuses on effectively learning from only a small number of labeled nodes to quickly adapt to new tasks, has garnered s

InterPartAbility: Text-Guided Part Matching for Interpretable Person Re-Identification

arXiv:2604.27122v1 Announce Type: new Abstract: Text-to-image person re-identification (TI-ReID) relies on natural-language text description to retrieve top matching individuals from a large gallery o

Iterative Definition Refinement for Zero-Shot Classification via LLM-Based Semantic Prototype Optimization

arXiv:2604.27335v1 Announce Type: new Abstract: Web filtering systems rely on accurate web content classification to block cyber threats, prevent data exfiltration, and ensure compliance. However, cla

Iterative Multimodal Retrieval-Augmented Generation for Medical Question Answering

arXiv:2604.27724v1 Announce Type: new Abstract: Medical retrieval-augmented generation (RAG) systems typically operate on text chunks extracted from biomedical literature, discarding the rich visual c