All entries

Usage share of OpenAI grew vs Anthropic yesterday despite Mythos 5 / Fable 5 launch Multiple power users at SemiAnalysis tried Mythos / Fable Got refusals for nonsensical reasons Got pissed off at Ant

hardwaredylan-patel--x

Safety

Using Probabilistic Programs to Train Inductive Reasoning in Large Language Models

DGX agent

arXiv:2606.09856v1 Announce Type: cross Abstract: Post-training Large Language Models (LLMs) for reasoning typically focuses on deductive tasks such as mathematics and coding where correctness is veri

safetyarxiv-cs-ai

10 Jun 2026

Using the YOLOv12 Model for Verifying the Correct Color Sequence of Wires in Network Cables (Patch Cords) on the Production Line

arXiv:2606.10699v1 Announce Type: cross Abstract: In the production process of network cables, ensuring the correct color sequence of wire pairs inside the standard connector plays a critical role in

applicationsarxiv-cs-ai

UXBench: Benchmarking User Experience in AI Assistants

model-releasesarxiv-cs-cl

arXiv:2606.09570v2 Announce Type: replace Abstract: As AI assistants serve millions of users daily, evaluating user experience (UX) beyond general model capability has become increasingly important. W

V-REX: Benchmarking Exploratory Visual Reasoning via Chain-of-Questions

arXiv:2512.11995v2 Announce Type: replace-cross Abstract: While many vision-language models (VLMs) are developed to answer well-defined, straightforward questions with highly specified targets, as in

Validation-Stage Combinatorial Fusion Analysis for Imbalanced Credit-Card Fraud Detection

arXiv:2606.10393v1 Announce Type: new Abstract: Credit-card fraud detection is difficult because fraudulent transactions are rare, costly, and unevenly distributed. Strong gradient-boosted tree models

Valve kills its retail gift card program due to scammers

Valve discontinued its physical Steam gift card retail program, with all retailers expected to be fully out of stock by the end of 2026. The company cited persistent scammer abuse of the cards; despit

industryars-technica

Local Ai

Vector Map as Language: Toward Unified Remote Sensing Vector Mapping

arXiv:2606.10701v1 Announce Type: new Abstract: Remote sensing vector mapping aims to generate structured maps of geospatial entities, such as buildings, roads, and water bodies, from remote sensing i

local-aiarxiv-cs-cv

Vehicle Prediction Model for Enhanced MPC Path Tracking in Formula Student Driverless

arXiv:2606.10732v1 Announce Type: new Abstract: Autonomous race cars, such as in Formula Student Driverless, operate close to their physical handling limits. The resulting highly nonlinear vehicle beh

agentsarxiv-cs-ro

VeriSpace: Spatially Grounded Action Verification for Vision-Language-Action Models

arXiv:2606.10568v1 Announce Type: new Abstract: Vision-language-action (VLA) models have shown strong promise for robotic manipulation, but their reliability at test time remains limited by one-shot a

applicationsarxiv-cs-ro

VFUSE: Virulent Feature Understanding with Sparse autoEncoders

arXiv:2606.10080v1 Announce Type: cross Abstract: Generative models have shown remarkable progress in a variety of domains such as protein design, but such power enables the opaque generation of hazar

researcharxiv-cs-ai

Visa partners with OpenAI to let AI agents make payments for users

Visa Inc. has struck a deal with OpenAI Group PBC to let artificial intelligence agents make payments for users, bringing one of the world’s largest payment networks into ChatGPT’s push toward agentic

agentssiliconangle

Tutorials

Vision-Assisted Foundation Model for Solving Multi-Task Vehicle Routing Problems

arXiv:2606.10431v1 Announce Type: cross Abstract: Multi-task vehicle routing problems play a critical role in enhancing efficiency across various industries and service sectors. These problems consist

tutorialsarxiv-cs-ai

VISTA: A Versatile Interactive User Simulation Toolkit for Agent Evaluation

arXiv:2606.11079v1 Announce Type: new Abstract: Evaluation remains a critical bottleneck for interactive agent development. Existing evaluation methods often rely on static benchmarks, which fail to c

agentsarxiv-cs-cl

Visual-TCAV: Concept-based Attribution and Saliency Maps for Post-hoc Explainability in Image Classification

arXiv:2411.05698v3 Announce Type: replace-cross Abstract: Convolutional Neural Networks (CNNs) have shown remarkable performance in image classification. However, interpreting their predictions is cha

safetyarxiv-cs-ai

Want to learn more about the Far-Field ASR benchmark? Join Treble's webinar tomorrow, June 11th, with @shinjiw_at_cmu, Cohere's @Julianfmack…

Want to learn more about the Far-Field ASR benchmark? Join Treble's webinar tomorrow, June 11th, with @shinjiw_at_cmu, Cohere's @Julianfmack, and other industry leaders discussing the future of far-fi

model-releasescohere--x

Warren to SEC, lightly paraphrased: “Do your f’ing job, and don’t let retail investors get screwed”

Senator Elizabeth Warren criticized the SEC for insufficient enforcement and investor protection, urging the agency to strengthen oversight and prevent harm to retail investors. The post, shared by AI

safetygary-marcus--x

We need more real time data on how AI may be impacting the economy - this is a really useful addition.

model-releasesethan-mollick--x

We need more real time data on how AI may be impacting the economy - this is a really useful addition. Today, the Stanford @DigEconLab launches the AI Economic Indicators, a new platform for tracking

We raised $6M led by Sequoia to build the future of travel. Watch me plan a perfect trip to Mexico City in 3 minutes. Flights, hotels and a …

We raised $6M led by Sequoia to build the future of travel. Watch me plan a perfect trip to Mexico City in 3 minutes. Flights, hotels and a full itinerary that matches my preferences. All bookable on

agentsharrison-chase--x

We see the tower of Jesus Christ illuminated for the first time! The light show, starting from the base up to the illumination of the cross,…

We see the tower of Jesus Christ illuminated for the first time! The light show, starting from the base up to the illumination of the cross, culminated with a composition of lights guided by drones th

industryelon-musk--x

WebChallenger: A Reliable and Efficient Generalist Web Agent

model-releasesarxiv-cs-cl

arXiv:2606.10423v1 Announce Type: new Abstract: Autonomous web navigation remains challenging for LLM agents, and the strongest generalist systems rely on proprietary reasoning models whose inference

Tools

We're closing out SHIP LDN with @rauchg and @HarryStebbings for an exclusive fireside chat on building, backing, and what's next. Sign up fo…

We're closing out SHIP LDN with @rauchg and @HarryStebbings for an exclusive fireside chat on building, backing, and what's next. Sign up for your spot on the waitlist today: https://vercel.com/ship/l

toolsvercel--x

What Demonstration Curation Metrics Do to Your Policy

arXiv:2606.10229v1 Announce Type: cross Abstract: We study whether demonstration-curation metrics that detect defective training episodes also improve the downstream behavior-cloning policy that train

What Do Deepfake Speech Detectors Actually Hear?

arXiv:2606.10912v1 Announce Type: cross Abstract: Deepfake speech detectors often output a single score without explaining why an audio sample is flagged, where in the signal the evidence lies, or wha

researcharxiv-cs-ai

What Fits (Into Few Tokens) Doesn't Overfit: Compression and Generalization in ML Research Agents

arXiv:2606.11045v1 Announce Type: new Abstract: Reusing a held-out benchmark adaptively should, in principle, invite overfitting. Yet benchmark-driven machine learning (ML) has produced surprisingly l

What if, Germany locked itself out of the LLM race and • Had students who actually learned things in high school, instead of turning in prom…

What if, Germany locked itself out of the LLM race and • Had students who actually learned things in high school, instead of turning in prompt outputs they barely read • Emerged from the sea of slop •

safetygary-marcus--x

Local Ai

What is the best open sourced image model?

local-air-stablediffusion

The best open-source image generation models in 2026 include FLUX.1 [schnell], Stable Diffusion 3.5 Large, HiDream-I1-Full, SANA-Sprint 1.6B, and HunyuanImage-3.0 . FLUX.1 [dev] holds the crown for ph

What makes a harness a harness: necessary and sufficient conditions for an agent harness

arXiv:2606.10106v1 Announce Type: cross Abstract: The term agent harness now circulates widely in software engineering with generative artificial intelligence. It names the layer that wraps a language

What Matters in Orchestrating Robot Policies: A Systematic Study of Hierarchical VLA Agents

arXiv:2606.10267v1 Announce Type: cross Abstract: Hierarchical vision-language-action (Hi-VLA) systems have emerged as a promising paradigm for complex robot manipulation, by using high-level VLM plan

What Really Matters for Table LLMs? A Meta-Evaluation of Model and Data Effects

arXiv:2501.14717v2 Announce Type: replace Abstract: Table modeling has progressed for decades. In this work, we revisit this trajectory and highlight emerging challenges in the LLM era, particularly t

What Should a Skill Remember? Quality--Cost Trade-offs in Cost-Aware Skill Rewriting for Language Model Agents

arXiv:2606.09421v2 Announce Type: replace Abstract: Large language model agents increasingly rely on skills: reusable procedural documents encoding workflows, tool use, implementation patterns, valida

safetyarxiv-cs-cl

What Sowell said is exactly right and needs to stop.

What Sowell said is exactly right and needs to stop. @elonmusk “What multiculturalism boils down to is that you can praise any culture in the world except Western culture. And you cannot blame any cul

industryelon-musk--x

What Spatial Memory Must Store: Occlusion as the Test for Language-Agent Memory

arXiv:2606.10299v1 Announce Type: new Abstract: Language-agent 'memory palace' systems anchor each memory to a world coordinate, on the intuition that geometry adds something text cannot. We make that

agentsarxiv-cs-ai

What to expect at the Databricks Data + AI Summit: Join theCUBE June 16

This new era of artificial intelligence is all about getting the most value out of data, and Databricks Inc. is at the center of that development. Following competitor Snowflake Inc.’s annual summit l

applicationssiliconangle

What to expect during Pure Accelerate: Join theCUBE June 17

Enterprise storage is being repositioned for the artificial intelligence era, and one of the leading players in the field has actively pursued an agenda aimed at operationalizing data. Everpure Inc.,

applicationssiliconangle

Local Ai

When Attribution Patching Lies: Diagnosis and a Second-Order Correction

arXiv:2606.09899v1 Announce Type: cross Abstract: A central goal of mechanistic interpretability is to identify which internal components causally drive a language model's behavior. Because these impo

local-aiarxiv-cs-ai

When Claude Fable kicks off a workflow, the tokens can go very quickly (these aren't Fable tokens, obviously)

model-releasesethan-mollick--x

Claude Fable efficiently processes tokens at high speed when initiating workflows, demonstrating rapid token consumption during execution. This observation from Ethan Mollick highlights the computatio

When Design Rules Break: Benchmark Composition Determines Whether Label Informativeness Predicts GNN Aggregator Choice

arXiv:2606.10249v1 Announce Type: new Abstract: We examine whether graph neural network (GNN) design rules generalize across benchmark families by studying aggregator selection (sum, mean, max) on 24

When Distance Distracts: Representation Distance Bias in BT-Loss for Reward Models

arXiv:2512.06343v3 Announce Type: replace-cross Abstract: Reward models are central to Large Language Model (LLM) alignment within the framework of RLHF. The standard objective used in reward modeling

safetyarxiv-cs-ai

When Do Autoregressive Sequence Models Forecast Physical Wavefields? A Controlled Study on Synthetic Seismograms

arXiv:2606.10868v1 Announce Type: new Abstract: Long-horizon autoregressive forecasting of oscillatory physical signals, such as seismograms, gravitational-wave strain, and similar wavefields is limit

When I speak in support of the SAVE America Act, some Republicans tell me to stop, insisting that it’s a lost cause and we have to move on O…

When I speak in support of the SAVE America Act, some Republicans tell me to stop, insisting that it’s a lost cause and we have to move on Others chime in to say “remove Thune or your words mean nothi

industryelon-musk--x

When Metrics Disagree: A Meta-Analysis of Knowledge-Graph-Completion Model Benchmarking

arXiv:2606.10287v1 Announce Type: cross Abstract: Evaluating Knowledge Graph Completion (KGC) models remains challenging because standard assessment relies on isolated rank-based metrics such as MRR,

When RL Fails after SFT: Rejuvenating Model Plasticity for Robust SFT-to-RL Handoff

arXiv:2606.09932v1 Announce Type: cross Abstract: Supervised Fine-Tuning (SFT) followed by Reinforcement Learning (RL) has become a standard pipeline for Large Language Model (LLM) post-training. SFT

When the Chain of Thought Knows Better: Failure Modes in Multi-Turn Reasoning Models

arXiv:2606.10740v1 Announce Type: new Abstract: Failures in multi-turn reasoning models are largely invisible to terminal-score evaluation. A model can lock onto an unsafe stance early in a long dialo

safetyarxiv-cs-ai

When to Align, When to Predict: A Phase Diagram for Multimodal Learning

arXiv:2606.11190v1 Announce Type: new Abstract: Cross-modal alignment (CA) and cross-modal prediction (CP) are the dominant paradigms for multimodal representation learning, yet there is no systematic

safetyarxiv-cs-lg

When you hear AI 'safety' you should hear 'censorship' and 'control' instead. All of us surveilled and spied by safeguards of loving grace. …

When you hear AI 'safety' you should hear 'censorship' and 'control' instead. All of us surveilled and spied by safeguards of loving grace. Today it's intelligent Terms of Service control. You can't d

safetyyann-lecun--x

Where You Inject Diversity Matters: A Unified Framework for Diverse Generation

arXiv:2606.10302v1 Announce Type: new Abstract: Open-ended generation tasks often require a set of meaningfully different outputs, yet large language models often produce similar generations. Existing

Which LoRA? An Empirical Study on the Effectiveness of LoRA Techniques During Multilingual Instruction Tuning

arXiv:2606.10428v1 Announce Type: new Abstract: We investigate whether commonly available LoRA variants have an advantage over basic LoRA in multilingual instruction tuning. Experiments involving LoRA