All entries

Beyond Task Success: Stage-Wise Reliability of World Model Planning under Sensing Degradation

arXiv:2609.07126v1 Announce Type: cross Abstract: In world model planning, sensing inputs pass through an encoder and predictor before affecting planner decisions, so final task success alone cannot r

Content type

All Blog X Post Paper YouTube Reddit GitHub

Beyond the Matrix Sign: Quadratic Spectral Descent

DGX agent

arXiv:2609.07597v1 Announce Type: new Abstract: Muon can be interpreted as optimizing a linear local objective over a spectral-norm ball. This gives a matrix-sign update that preserves the singular di

local-aiarxiv-cs-lg

9 Sept 2026

Beyond the Verdict: Evidence-Aligned Evaluation of Visual Prompt-Injection Guardrails

arXiv:2609.05535v1 Announce Type: new Abstract: Verdict-only evaluation does not reveal whether a vision-language model (VLM) used the visual evidence that should support its decision. We study this p

Beyond Worst-Case Coreset Bounds for k-Clustering via Determinantal Sampling

arXiv:2609.06394v1 Announce Type: cross Abstract: Massive datasets in modern machine learning have made data reduction a central challenge, particularly for clustering tasks where memory and computati

Bi-HYCO: Bi-Objective Cooperative Learning for PDE Parameter Identification under Fragmented Observations

arXiv:2609.06511v1 Announce Type: new Abstract: Physical and synthetic models may describe complementary aspects of the same PDE-governed system while receiving different, possibly fragmented, observa

Applications

BiDexGrasp: Coordinated Bimanual Dexterous Grasps across Object Geometries and Sizes

arXiv:2604.06589v2 Announce Type: replace Abstract: Bimanual dexterous grasping is a fundamental and promising area in robotics, yet its progress is constrained by the lack of comprehensive datasets a

applicationsarxiv-cs-ro

BIFTA: Brain-Inspired Few-Shot Tactile Adaptation for Unknown Sensors

arXiv:2609.08673v1 Announce Type: new Abstract: Advances in tactile sensing have made contact-rich perception possible, accelerating progress in robotic manipulation, material understanding, and embod

researcharxiv-cs-ro

Bigger Text Encoders Can Hurt CLIP Zero-Shot Performance

arXiv:2609.05730v1 Announce Type: new Abstract: Contrastive Language-Image Pretraining (CLIP) is a building block of many machine learning applications. Scaling laws have guided resource allocation fo

Tutorials

BinauralVAE: Spatial Audio Reconstruction For World Models

arXiv:2609.06837v1 Announce Type: cross Abstract: Embodied artificial intelligence has historically very much relied on visual perception, leading to a proliferation of multiple vision-centric world m

tutorialsarxiv-cs-lg

.blend URL Viewer

Tool: .blend URL Viewer I'm continuing to have a lot of fun with GPT-6 Astra and Blender (see my TIL). As a big fan of the Imperial Fabergé Easter eggs, I've always thought it would be fun to make som

local-aisimon-willison

BlueprintAgent: Constraint-Triggered Targeted Revisits for Simulation-Ready Generation from Scanned Structural Blueprints

arXiv:2609.07362v1 Announce Type: cross Abstract: Converting in-service reinforced-concrete (RC) building blueprints into simulation-ready models---structured frame representations that support determ

local-aiarxiv-cs-cv

Bottom-up Modeling of Repeated Elements via Single Image Analysis-by-Synthesis

arXiv:2609.07939v1 Announce Type: new Abstract: We address the problem of discovering repeated elements from a single image. In contrast to existing approaches that depend on large annotated datasets,

safetyarxiv-cs-cv

Boundary Voting Network for Ambiguity-Aware Timestamp-Supervised Action Segmentation

arXiv:2609.08167v1 Announce Type: new Abstract: Timestamp-supervised action segmentation aims to segment and classify actions in untrimmed videos with a random frame annotated per action. Precisely lo

BoundarySeg:Boundary-Aware Multi-Task Learning for Semi-Supervised 3D Medical Image Segmentation with Limited Unlabeled Data

arXiv:2505.09829v2 Announce Type: replace Abstract: Data scarcity is a primary bottleneck in 3D medical image segmentation: expert voxel-level annotations are expensive, and privacy regulations often

BrachistoneLR: A Brachistochrone-Inspired Learning-Rate Schedule and a Controlled Benchmark of Scheduling Policies

arXiv:2609.08069v1 Announce Type: new Abstract: The learning-rate schedule is a consequential choice in training deep networks, yet the policies in common use are heuristic, and published comparisons

Branch-Centric Tokenization and Test-Time Augmentation for Skeleton Generation

arXiv:2609.06218v1 Announce Type: new Abstract: Automatic skeleton generation involves predicting both joint positions and skeletal connectivity. However, existing approaches struggle to encode branch

Bridging Language and Physics: Automated Design of Continuum Robots with Large Language Models

model-releasesarxiv-cs-ro

arXiv:2609.08220v1 Announce Type: new Abstract: Large language models (LLMs) have recently emerged as a promising tool for automating robot design from high-level specifications, yet they remain ineff

Broken on Arrival: Silently Defective LLM Artifacts in Public Model Registries and How to Catch Them

arXiv:2609.05881v1 Announce Type: cross Abstract: Developers increasingly run large language models locally by pulling quantized GGUF artifacts from public registries, yet nothing in the distribution

local-aiarxiv-cs-lg

Budgeted Task-Aware Acquisition of Dynamic Networks

arXiv:2609.05862v1 Announce Type: new Abstract: Learning on dynamic graphs is difficult when changes in the underlying network are only partially observed. Acquiring current graph information incurs o

Agents

Built for agents: Omarchy's pipeline moves to DigitalOcean

Omarchy is a keyboard-first Linux desktop built on Arch and Hyprland by David Heinemeier Hansson (DHH), and its production pipeline, packaging builds, PR review agents, and QA testing, now run on Digi

agentsdigitalocean

Bundle Length Tradeoffs in Decentralized Multi-Robot Task Allocation Under Degraded Communications

arXiv:2609.06273v1 Announce Type: new Abstract: Bundle length B is commonly fixed when configuring multi-task multi-robot task allocation (MRTA) algorithms. MinSum and MinMax are known to favor differ

researcharxiv-cs-ro

Cadence: Error-Bounded Lossy Compression of Demand Time Series with a Time-Series Foundation Model

arXiv:2609.06008v1 Announce Type: cross Abstract: We present Cadence, an error-bounded lossy compressor for numeric time series pairing a 330M-parameter time-series foundation model (Google TimesFM-3)

Calendar-SPCA: Interpretable Representation Learning for Multi-Periodic Electricity Consumption Profiles

arXiv:2609.06060v1 Announce Type: new Abstract: Long-term electricity-consumption profiles exhibit several simultaneous periodic structures, including daily, weekly, and annual cycles. This work intro

California Gov. Gavin Newsom signs into law two bills, backed by Anthropic and OpenAI, regulating how outside groups evaluate AI for safety (Christine Mui/Politico)

Christine Mui / Politico: California Gov. Gavin Newsom signs into law two bills, backed by Anthropic and OpenAI, regulating how outside groups evaluate AI for safety — SAN FRANCISCO — California Gov.

safetytechmeme

CALIPER: Clean Scenes Cannot Rank Physical Inference in Pretrained Visual Representations

arXiv:2609.08250v1 Announce Type: cross Abstract: How far a pushed object slides depends on its mass and friction, which no single image reveals. Pretrained visual encoders are increasingly used as th

CALM: Class-wise Agreement and Label-gated Disagreement Modulation for Decentralized Federated Learning

arXiv:2609.05884v1 Announce Type: new Abstract: Conventional federated learning relies on parameter averaging, which forces clients to be doubly homogeneous: all must run an identical architecture, an

CALM: Configuration-Aware Human Intervention Boundaries During Robot Approach

arXiv:2609.07430v1 Announce Type: new Abstract: How robot body configuration shapes human intervention during approach remains underexplored. We conducted a within-participants study with 41 participa

safetyarxiv-cs-ro

CAM: Question Answering on Entity-Centric Videos with Continuous Extraction and Adaptive Querying

arXiv:2609.06504v1 Announce Type: new Abstract: Memory facilitates question answering over long videos by extracting and retrieving facts to fit within the limited context windows of multimodal LLMs (

Can People Distinguish Human and AI Agency in Humanoid Teleoperation? A Preliminary Study of Agency Perception

arXiv:2609.06434v1 Announce Type: cross Abstract: Can people distinguish between human and AI agency in humanoid teleoperation? To explore this question, we developed extit{Ghost-in-the-Loop}, a teleo

researcharxiv-cs-ro

CancerBench: the frontier model cancer cure benchmark. AI lab CEOs keep talking about curing cancer, so I made a benchmark. One metric: how …

model-releasesyohei-nakajima--x

CancerBench: the frontier model cancer cure benchmark. AI lab CEOs keep talking about curing cancer, so I made a benchmark. One metric: how many types of cancer has your model cured? All models are cu

Canonical Color as a Lens into Concept Decodability in Vision Encoders and VLMs

arXiv:2609.09124v1 Announce Type: new Abstract: Visual encoders construct a representation of the image input for Vision-Language models. How much conceptual, as opposed to immediately visible, inform

Tutorials

Capsule Lens: Locating and Tracking Concept Geometry in Model Representations

arXiv:2609.05575v1 Announce Type: cross Abstract: Understanding how concepts are encoded in the internal representations of machine learning models is a central problem in mechanistic interpretability

tutorialsarxiv-cs-cv

CAR-MIL: Counterfactual Attention Regularization for Multiple Instance Learning

arXiv:2609.08419v1 Announce Type: new Abstract: Multiple Instance Learning (MIL) is widely used for weakly supervised learning, particularly in digital pathology, where fine-grained annotations are co

CARDEA: Auditable Reasoning Grounded in Spatial Evidence for End-to-End Coronary Angiography Interpretation

arXiv:2609.06931v1 Announce Type: new Abstract: Invasive coronary angiography (CAG) is the gold standard for diagnosing coronary artery disease, but interpretation varies substantially among observers

safetyarxiv-cs-cv

CARE: Confounder-Aware Aggregation for Reliable LLM Evaluation

arXiv:2603.00039v2 Announce Type: replace Abstract: LLM-as-a-judge ensembles are the standard paradigm for scalable evaluation, but their aggregation mechanisms suffer from a fundamental flaw: they im

CAROL: Context-Aware Online Learning for Fuzzer Scheduling

arXiv:2609.06772v1 Announce Type: cross Abstract: Ensemble fuzzing runs multiple fuzzers on a target while a scheduler allocates CPU time among them. Existing schedulers base these decisions on compac

Hardware

CASCADE: A Spatio-Temporal-Causal Reasoning Representation and Dataset for Driving

arXiv:2609.07094v1 Announce Type: new Abstract: Reasoning is a promising route to the generalization that autonomous driving requires in the long tail, as it can infer how the elements of a scene depe

hardwarearxiv-cs-cv

Cascaded Diffusion Framework for Probabilistic Coarse-to-Fine Hand Pose Estimation

arXiv:2510.00527v2 Announce Type: replace Abstract: Deterministic models for 3D hand pose reconstruction, whether single-staged or cascaded, struggle with pose ambiguities caused by self-occlusions an

CASD: Chunk-Aligned Semantic Distillation for Multi-StageRobot Manipulation

arXiv:2609.08638v1 Announce Type: new Abstract: An action chunk can span several stages of a manipulation task, yet a label for its first step describes only the current stage. We introduce Chunk-Alig

safetyarxiv-cs-ro

CAST: Alternating State-Value Targets and Expanded Policy Gradients for Model-Based Reinforcement Learning

arXiv:2609.08853v1 Announce Type: new Abstract: Model-based reinforcement learning (MBRL) is a family of RL methods that learn a model of the environment and use it for action selection, making it wel

safetyarxiv-cs-ro

Causal DAG Identification for Count Data via Poisson Thinning Structural Equation Models

arXiv:2609.06098v1 Announce Type: cross Abstract: Count-valued variables arise in many scientific and applied settings, yet explicit structural models that allow full identification of causal DAGs fro

CausalBN-Bench: A Comprehensive Benchmark for Causal Learning Capability of LLMs

arXiv:2404.06349v3 Announce Type: replace Abstract: The ability to understand causality significantly impacts the competence of large language models (LLMs) in output explanation and counterfactual re

CausalChapter: Improving Long-Video Chaptering with Interventional Dependency Modeling

arXiv:2609.08686v1 Announce Type: new Abstract: Long-form instructional videos require automatic chaptering to support browsing, navigation, and knowledge access. Recent long-context language models c

CAVEAT: Recurrent Multimodal Diffusion Planning for Mapless Aerial Exploration

arXiv:2609.06623v1 Announce Type: cross Abstract: Can exploratory UAV waypoint sequences be generated from multimodal onboard observations and a fixed-dimensional recurrent internal state without main

local-aiarxiv-cs-lg

CDFlow: Building Invertible Layers with Circulant and Diagonal Matrices

arXiv:2510.25323v4 Announce Type: replace Abstract: Normalizing flows are deep generative models that enable efficient likelihood estimation and sampling through invertible transformations. A key chal

Certain companies train their AIs to plagiarise work by unnamed scholars, extending an AI history permeated by plagiarism and misattribution…

Certain companies train their AIs to plagiarise work by unnamed scholars, extending an AI history permeated by plagiarism and misattributions. It started a long time ago. In the 1900s, the old method

researchdavid-ha--x

Certified Topological Interaction in Neural Representations: Class Disentanglement Is Mostly Pairwise

arXiv:2609.08561v1 Announce Type: new Abstract: Class disentanglement (the separation of a representation's class-conditional point clouds along depth and over training) is usually read off descriptiv

Agents

Certifying cooperation: a novel approach to cooperative multi-agent task generation

arXiv:2609.06586v1 Announce Type: cross Abstract: A shared reward gives agents a common objective, but leaves open when, how and even whether they must cooperate to succeed. We address these questions

agentsarxiv-cs-lg