Search: “evaluation”

VISPATH: Visual-Intent-Guided Path Reasoning for Multimodal Knowledge Graph Question Answering

model-releasesarxiv-cs-cl

arXiv:2609.20843v1 Announce Type: new Abstract: Knowledge graph question answering (KGQA) enables models to answer natural-language questions through structured graph reasoning and has achieved substa

21 Sept 2026

Human

Content type

All Blog X Post Paper YouTube Reddit GitHub

Clear filters

Research

ZeroTouch: Tactile-Supervised Visual Contact Estimation for Contact-Rich Manipulation

DGX agent

arXiv:2609.21726v1 Announce Type: new Abstract: Reliable robotic grasping benefits from estimating the evolving physical interaction and selecting a grasp-dependent compression target. Tactile sensors

researcharxiv-cs-ro

21 Sept 2026

jev as a Judge proves to be a cheaper and more precise alternative to LLM as a judge for online evals. great guide from Sean and Daniel on w…

jev as a Judge proves to be a cheaper and more precise alternative to LLM as a judge for online evals. great guide from Sean and Daniel on why Jev is great for evals, an experiment vs other LLMs, and

agentsharrison-chase--x

20 Sept 2026

They don’t have a fucking clue. Not even a theory about how make their stuff safe.

They don’t have a fucking clue. Not even a theory about how make their stuff safe. The simple truth: not even Anthropic has a plan for how to make AI safe in advance. There is no theory. I've asked! O

agentsgary-marcus--x

20 Sept 2026

A Functional Pilot for Certified Freshness-Aware Semantic--Spatial Range Retrieval

arXiv:2609.19855v1 Announce Type: cross Abstract: Geographic applications need every object inside a radius that satisfies a semantic threshold, yet embedding indexes return approximate top-ranked lis

applicationsarxiv-cs-ai

AURORA: A Natural Language-Driven Agentic Framework for Understanding, Reasoning, and Orchestrating Reliable Air-Ground Co-Simulation

arXiv:2609.19527v1 Announce Type: cross Abstract: Air-ground transportation research increasingly relies on co-simulation, yet constructing scenarios remains labor-intensive and difficult to validate.

AVTrace: Diagnosing Audio-Visual Temporal Reasoning in Omni Models

arXiv:2609.19991v1 Announce Type: cross Abstract: Omni models can describe video content, but can they locate events in time, preserve event order, and judge audio-visual synchronization? We introduce

Before the Arrest: Benchmarking LLMs on Criminal Profiling from Incomplete Evidence

arXiv:2609.19965v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly applied to legal and criminal justice tasks, yet existing work focuses almost exclusively on post-arrest s

applicationsarxiv-cs-cl

Benchmarking LLM Compliance with China AI Generated Content Regulations

model-releasesarxiv-cs-cl

arXiv:2609.19989v1 Announce Type: new Abstract: The widespread adoption of LLMs has led to escalating content compliance risks. Prior works have contributed to addressing these risks in the English co

Benchmarking MLLMs via Cognitive Expected Scene Graph for Safety-Critical Visual Negation Understanding

model-releasesarxiv-cs-cv

arXiv:2609.19767v1 Announce Type: new Abstract: True machine intelligence requires transcending passive pixel registration to master top-down functional reasoning over absent information via visual ne

Beyond Flattened Tokens: Structure-Preserving EEG Decoding with Reusable TriDim Blocks

model-releasesarxiv-cs-lg

arXiv:2609.19842v1 Announce Type: new Abstract: Effective EEG decoding requires representations that preserve organization among channels, local waveform dynamics, and long-range temporal context. Exi

Beyond Patch Removal: Persistent Adversarial Effects in Vision-Language-Action Policies

arXiv:2609.19669v1 Announce Type: new Abstract: Adversarial patches to Vision-Language-Action (VLA) policies can cause both immediate action corruption and persistent state effects that remain after t

Boreas Road Trip: A Multi-Sensor Autonomous Driving Dataset on Challenging Roads

model-releasesarxiv-cs-ro

arXiv:2602.16870v2 Announce Type: replace Abstract: The Boreas Road Trip (Boreas-RT) dataset extends the multi-season Boreas dataset to new and diverse locations that pose challenges for modern autono

Bridging Modalities on the Cortex: Surface-based MRI to PET Translation with a Diffusion Bridge

arXiv:2609.20147v1 Announce Type: cross Abstract: Cortical hypometabolism measured by Fluorodeoxyglucose Positron Emission Tomography (FDG-PET) is a highly sensitive biomarker for dementia diagnosis.

researcharxiv-cs-ai

CoMa: Contextual Massing Generation with Vision-Language Models

arXiv:2601.08464v2 Announce Type: replace-cross Abstract: Context-aware building massing is an important early-stage design task: given a site for buildings, a generated massing should not only fit th

researcharxiv-cs-ai

Continual Enterprise World Model Discovery in Dynamic Systems

arXiv:2609.19551v1 Announce Type: new Abstract: In an enterprise system, updating one field can set another, create a record, or start an approval. These effects are produced by business rules that ar

Correct Now, Insufficient Later: Auditing Update Sufficiency in Context Compression

arXiv:2609.20045v1 Announce Type: cross Abstract: A memory can answer a current query correctly while discarding distinctions required by a later update. We investigate this failure with a paired-hist

DexTouch-WM: Learning Action-Conditioned Tactile World Models from Human Touch for Dexterous Robot Manipulation

arXiv:2609.20649v1 Announce Type: cross Abstract: Learning predictive models of contact-rich dexterous manipulation requires dense tactile interaction, but such data are costly to scale on real robots

safetyarxiv-cs-cv

Do AI Agents Understand Computer Architecture?

arXiv:2609.19387v1 Announce Type: new Abstract: Agents are increasingly asked to design hardware, and increasingly reported to succeed. Such reports establish that a design improved; they cannot estab

Effective and Efficient Threat Hunting with Small Language Models

arXiv:2512.06660v3 Announce Type: replace-cross Abstract: Analysts in Security Operations Centers query massive telemetry streams using Kusto Query Language (KQL), but writing correct KQL demands spec

EliGSiR: Continual RGB-D Mapping with Gaussian Splatting under Bounded Compute

arXiv:2609.20348v1 Announce Type: new Abstract: Conventional 3D Gaussian Splatting assumes a closed set of observations and long optimization schedules. Continual RGB-D mapping in contrast poses the p

Exploring Sparsity and Smoothness of Arbitrary Lp Norms in Adversarial Attacks

arXiv:2602.06578v2 Announce Type: replace-cross Abstract: Adversarial attacks against deep neural networks are commonly constructed under ell_p norm constraints, most often using p=1, p=2 or p=infty,

FCA-Guided Counterfactual Explanations for Multi-Modal Breast Cancer Diagnosis: A Framework Achieving Perfect Validity with Emergent Sparsity

arXiv:2609.20067v1 Announce Type: new Abstract: Deep learning models for multi-modal breast cancer diagnosis achieve high predictive accuracy but remain clinically unacceptable without actionable, cou

Local Ai

FORGE: Forensic Reasoning with Grounded Evidence

arXiv:2503.15867v4 Announce Type: replace-cross Abstract: Forensic deepfake analysis demands more than binary classification: investigators need region-grounded natural language explanations they can

local-aiarxiv-cs-ai

From Intent to Action: Benchmarking LLM Safety in Vehicle Voice Command Authorization

arXiv:2609.19630v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly integrated into vehicle voice assistants. But linking natural-language requests to vehicle functions creat

Local Ai

GeLaCo: An Evolutionary Approach to Layer Compression

arXiv:2507.10059v2 Announce Type: replace Abstract: Large Language Models have achieved remarkable performance across a large number of tasks, but face critical deployment and usage barriers due to su

local-aiarxiv-cs-cl

GPT-6-Astra in a Navigation Workflow: Behavioral Analysis in Zero-Shot Vision-and-Language Navigation in Continuous Environments

arXiv:2609.20116v1 Announce Type: new Abstract: We study GPT-6-Astra in a zero-shot Vision-and-Language Navigation in Continuous Environments (VLN-CE) system, where it interprets instructions, assesse

agentsarxiv-cs-ro

GR2PO: Group Relative Return Policy Optimization for Continuous Robot Control

arXiv:2609.19850v1 Announce Type: new Abstract: Actor-critic architecture has been widely used in continuous robot control. However, they rely on learning a value network, introducing additional compu

safetyarxiv-cs-ro

HyperAMS-Net: Adaptive Multi-Scale Spatial Hypergraph Network for Brain Disorder Classification

model-releasesarxiv-cs-cv

arXiv:2609.19755v1 Announce Type: cross Abstract: Accurate classification of brain disorders from neuroimaging data remains challenging because of substantial inter-subject heterogeneity and the compl

Imagine-TAMP: Imagination-Guided Task and Motion Planning in Partial Observability

arXiv:2609.20396v1 Announce Type: new Abstract: Robots operating in cluttered environments must often manipulate objects whose locations are only partially observable. A central challenge is deciding

researcharxiv-cs-ro

Instance Segmentation and Fine-grained Classification for Urban Buildings with Adaptive Region Dividing and Spatially-Supervised Contrastive Learning

arXiv:2609.19631v1 Announce Type: new Abstract: Accurate instance-level and functional understanding of urban buildings in large-scale point clouds is essential for digital city modeling and urban ana

applicationsarxiv-cs-cv

Is It Still Worth Training a Classical Model in the Era of LLMs? A Crossover Benchmark on Tabular Data

arXiv:2609.20218v1 Announce Type: cross Abstract: Large language models can label a tabular row from a plain-English description with no training - a capability now shipping in mainstream spreadsheet

Ischemic Stroke Segmentation and Net Water Uptake Quantification on Multicenter Non-Contrast CT Using Supervised Target-Domain Adaptation

arXiv:2609.20151v1 Announce Type: new Abstract: Objectives: Quantitative assessment of infarct hypodensity on non-contrast computed tomography (NCCT), including net water uptake (NWU), requires manual

LLM-as-an-Improver: Turning Verification into Better Candidates

arXiv:2609.19515v1 Announce Type: new Abstract: Verifier-based selection improves LLM performance by generating multiple candidate solutions and using a verifier to select the most promising one. Howe

researcharxiv-cs-ai

MAGS: Multi-agent Auto-formalization Guarantees Safety for Agentic Outputs

arXiv:2609.19391v1 Announce Type: new Abstract: LLM coding agents now generate complex programs at a scale that makes thorough human review increasingly difficult, raising the risk of safety and secur

safetyarxiv-cs-ai

Mammography Foundation Models for Opportunistic Prediction of Major Adverse Cardiovascular Events

arXiv:2609.19385v1 Announce Type: cross Abstract: Cardiovascular disease (CVD) remains the leading cause of death among women, yet cardiovascular risk assessment often relies on clinical variables tha

Music Hallucination in Audio-Language Models: A Hierarchical Formulation and Empirical Study

arXiv:2609.20195v1 Announce Type: cross Abstract: Audio-language models increasingly generate confident music descriptions that are unsupported by the input audio. We present, to our knowledge, the fi

safetyarxiv-cs-ai

Performance of Machine Learning Classification in Sonomammogram Images using BI-RADS

arXiv:2311.08493v2 Announce Type: replace-cross Abstract: This research aims to investigate the classification accuracy of various state-of-the-art image classification models across different categor

PetriBench: Benchmarking LLM Reasoning over Dynamic State Spaces

arXiv:2609.19883v1 Announce Type: cross Abstract: Characterizing LLM reasoning remains an open challenge, as many existing benchmarks isolate specific reasoning skills, rely on external knowledge, or

Placement Is Free, Composition Is Not: The Latin Square as a Provably-Balanced Construction for Heterogeneous Sequence-Mixer Stacks

model-releasesarxiv-cs-lg

arXiv:2609.20269v1 Announce Type: new Abstract: Since GPT, most Transformers have repeated the same attention mechanism at every layer. Yet this design is largely a convention rather than a tested con

Predict Before You Deploy: Offline Prediction of Quantization-Induced Task Degradation for World Action Models

arXiv:2609.19441v1 Announce Type: cross Abstract: World action models (WAMs) rely on video-generation backbones, requiring substantial memory and compute for deployment. Post-training quantization red

PROVIA: Procedure State Tracking for Online Mistake Detection in Egocentric Videos

arXiv:2609.20638v1 Announce Type: new Abstract: An assistant watching egocentric video should notice a mistake from past frames alone, before the next step begins, and keep working once the person rec

QCPruner: Query-Conditioned Population Coverage for Visual Token Pruning

model-releasesarxiv-cs-cv

arXiv:2609.19990v1 Announce Type: new Abstract: The high visual-token load in multimodal large language models (MLLMs) motivates training-free pruning to reduce later-layer computation, but under a fi

Quantifying Overclaiming Propensity in Frontier LLM Agents

arXiv:2609.20812v1 Announce Type: cross Abstract: Frontier coding agents are increasingly trusted to work autonomously for long periods, yet an agent's final response is often the only account of that

RAUL: Reference-Assisted Ureteroscopy Localization for Skill Assessment

arXiv:2609.19236v1 Announce Type: new Abstract: Objective: Incomplete navigation of anatomy during ureteroscopic kidney stone surgeries can contribute to repeat interventions. While skilled surgeons h

Reach or Solve? Attributing Agentic RL Gains with Checkpoint Handoffs

arXiv:2609.19636v1 Announce Type: new Abstract: Reinforcement learning now trains language-model agents that act over dozens of steps in live environments. The gains are large, and they are read as be

Red-Teaming Auto Mode: Improving Blocking Classifiers Against Malign Coding Agents

arXiv:2609.19587v1 Announce Type: cross Abstract: To keep coding agents from going off the rails, production systems now review each proposed action with a blocking monitor that can reject it before i

RISC-V and machine learning: a survey

arXiv:2609.20677v1 Announce Type: new Abstract: The intersection of open-source processor architectures and machine learning is driving the demand for customizable, efficient, and accessible hardware.

applicationsarxiv-cs-lg