All entries

SAFE: An LLM-as-Verifier Framework for Evidence-Grounded Multi-Hop Reasoning

arXiv:2604.01993v2 Announce Type: replace-cross Abstract: Multi-hop QA benchmarks often reward Large Language Models (LLMs) for spurious correctness, where models reach correct answers through invalid

Sample Where You Struggle: Sharpening Base Model Reasoning via Entropy-Guided Power Sampling

DGX agent

arXiv:2606.09926v1 Announce Type: cross Abstract: Sampling from the sequence-level power distribution p^alpha elicits RL-level reasoning from base language models without any parameter updates, but th

model-releasesarxiv-cs-ai

10 Jun 2026

Hardware

Sampling Triangulations and Calabi-Yau Threefolds with Autoregressive GNNs

arXiv:2605.27770v2 Announce Type: cross Abstract: We introduce `dualGNN', an autoregressive message-passing GNN for sampling fine, regular triangulations (FRTs) of convex polytopes. dualGNN operates o

hardwarearxiv-cs-lg

SARM2: Multi-Task Stage Aware Reward Modeling for Self Improving Robotic Manipulation

model-releasesarxiv-cs-ro

arXiv:2606.10305v1 Announce Type: new Abstract: Fine-tuning vision-language-action (VLA) policies for long-horizon manipulation still relies heavily on behavior cloning, which requires costly high-qua

Tutorials

Say what you will about OpenAI but at least their employees debate their positions in public

tutorialsjeremy-howard--x

This post likely praises OpenAI's organizational culture for encouraging public discourse and debate among employees about company decisions and positions. The statement suggests that OpenAI demonstra

SCAIL-2: Unifying Controlled Character Animation with End-to-end In-Context Conditioning

model-releasesarxiv-cs-cv

arXiv:2606.10804v1 Announce Type: new Abstract: Controlled character animation requires transferring motion from a driving sequence to a reference character. Prior works heavily rely on intermediate r

Scaling AI Through Data Fluency

This Databricks blog post discusses how organizations can scale their AI initiatives by developing data fluency—the ability to effectively understand, manage, and leverage data across the enterprise.

industrydatabricks

Scaling Self-Supervised Speech Models Uncovers Deep Linguistic Relationships: Evidence from the Pacific Cluster

arXiv:2603.07238v2 Announce Type: replace Abstract: Similarities between language representations derived from Self-Supervised Speech Models (S3Ms) have been observed to primarily reflect geographic p

researcharxiv-cs-cl

Schmidt Decomposition-Based Methods for Efficient Quantum Image Encoding

arXiv:2606.10874v1 Announce Type: new Abstract: In quantum image processing, a fundamental step is encoding classical image data into quantum states. This can be achieved using methods such as Flexibl

researcharxiv-cs-cv

Science fiction authors in the order you want them to be right about AI: Iain Banks Becky Chambers Martha Wells Douglas Adams Charles Stross…

applicationsethan-mollick--x

Science fiction authors in the order you want them to be right about AI: Iain Banks Becky Chambers Martha Wells Douglas Adams Charles Stross (Singularity Sky) Peter Watts Charles Stross (Laundry) Harl

Scientific research is fundamental to advancing civilization and helping people globally to solve the most critical problems, from medicine …

Scientific research is fundamental to advancing civilization and helping people globally to solve the most critical problems, from medicine to materials, from brain science to physics, and much beyond

industryclem-delangue--x

SCOPE: Sequential Causal Optimization of Process Interventions

arXiv:2512.17629v4 Announce Type: replace-cross Abstract: Prescriptive Process Monitoring (PresPM) recommends interventions during running business processes to optimize key performance indicators (KP

SD-GRPO: Verifiable Segment Decomposition for Long-Form Vision-Language Generation

arXiv:2606.09871v1 Announce Type: cross Abstract: Group Relative Policy Optimization (GRPO) and its variants, originally developed for Large Language Models (LLMs), have recently been applied to Multi

Seal-Robust KCR: A Robust Kuzushiji Character Recognition Framework under Seal Interference

arXiv:2602.19086v2 Announce Type: replace Abstract: Kuzushiji was one of the most widely used cursive writing systems in pre-modern Japan. Due to its highly cursive forms and extensive glyph variation

researcharxiv-cs-cv

Secure Aggregation with Top-K Sparsification in Decentralized Federated Learning

arXiv:2606.10780v1 Announce Type: cross Abstract: Secure aggregation is a vital component for mitigating gradient leakage in federated learning, but its communication cost conventionally scales with t

researcharxiv-cs-lg

Agents

Securing the AI workforce: Zscaler’s zero-trust play for agentic AI

Since Zscaler Inc.‘s launch, the company’s mission has been to disrupt traditional access and security with its Zero Trust platform. At its user event, Zenith Live, in Las Vegas, the company made its

agentssiliconangle

Segment and Select: Vision-Language Segmentation in 3D Scenarios

arXiv:2606.10594v1 Announce Type: new Abstract: 3D vision-language segmentation aims to segment target objects in 3D scenarios according to the linguistic instructions and visual observations. Prior a

researcharxiv-cs-cv

Segmentation-Driven Monocular Shape from Polarization based on Physical Model

arXiv:2601.04776v2 Announce Type: replace Abstract: Monocular shape-from-polarization (SfP) leverages the intrinsic relationship between light polarization properties and surface geometry to recover s

applicationsarxiv-cs-cv

Selection, Not Salience: The Shape and Limits of Personalization in Social Highlighting

arXiv:2606.10398v1 Announce Type: cross Abstract: Does personalizing what a reader sees pay off, and where does it stop? Using a social web highlighter and a co-readership identity control (the same d

safetyarxiv-cs-cl

Selective Disk Bispectrum: A Complete and Rotation Invariant Image Descriptor

arXiv:2511.19706v2 Announce Type: replace-cross Abstract: Rotation invariance is a fundamental requirement across many computer vision tasks. Historically, this inductive bias has been encoded through

safetyarxiv-cs-cv

Self-Distillation Policy Optimization via Visual Feedback: Bridging Code and Visual Artifacts

arXiv:2606.10334v1 Announce Type: new Abstract: Code-generating large language models (LLMs) increasingly produce visual artifacts such as charts, web pages, and slides by writing programs that are ex

Self-EmoQ: Plutchik-Guided Value-based Planning to Drive Streaming Emotional TTS

arXiv:2606.09837v1 Announce Type: cross Abstract: Emotional interaction is increasingly crucial for conversational AI, yet current systems lack a self-emotion determination mechanism to drive the stre

Self-Supervised Relevance Modelling in Autonomous Driving via Counterfactual Analysis

arXiv:2606.10688v1 Announce Type: new Abstract: Autonomous driving relies on computationally intensive perception pipelines to continuously detect and track objects in the surrounding environment. Whi

safetyarxiv-cs-ro

SHAPE: Coalition-Aware Expert Pruning for Sparse Mixture-of-Experts LLMs

arXiv:2606.09886v1 Announce Type: cross Abstract: Sparse Mixture-of-Experts (MoE) large language models achieve strong quality with low per-token compute, yet their deployment is often limited by the

SHAPO: Sharpness-Aware Policy Optimization for Safe Exploration

arXiv:2606.10228v1 Announce Type: cross Abstract: Safe exploration is a prerequisite for deploying reinforcement learning (RL) agents in safety-critical domains. In this paper, we approach safe explor

Should we try to train an open source AI building model? We obviously have interesting datasets with HF, MLintern, transformers, trl…

Clem Delangue discusses the potential value of training an open source AI model specifically for software development and AI engineering tasks, leveraging Hugging Face's internal datasets including th

industryclem-delangue--x

Agents

shoutouts: • why multi-agent LLM systems fail? (arXiv:2503.13657) — @mertcemri @melissapan + @istoica05 @matei_zaharia @profjoeyg @adityagp …

shoutouts: • why multi-agent LLM systems fail? (arXiv:2503.13657) — @mertcemri @melissapan + @istoica05 @matei_zaharia @profjoeyg @adityagp & team • DSPy (arXiv:2310.03714) — @lateinteraction + @hazyr

agentsyohei-nakajima--x

Sigma-Branch: Hierarchical Single-Path Network Reconstruction for Dynamic Inference with Reduced Active Parameters

arXiv:2606.09924v1 Announce Type: cross Abstract: Deploying deep neural networks on memory-constrained edge accelerators is bottlenecked by per-inference off-chip weight transfer rather than computati

Sim2Schedule: A Simulator-Guided LLM Framework for Autonomous Open-Pit Mine Scheduling

arXiv:2606.10286v1 Announce Type: new Abstract: Open-pit mine scheduling is a critical process for maximizing economic return under complex geotechnical and operational constraints. While Mixed-Intege

Since Blair - 30 years ago - people voted against mass immigration. Every election, they got up, got dressed, found a pilling station and vo…

Since Blair - 30 years ago - people voted against mass immigration. Every election, they got up, got dressed, found a pilling station and voted to end mass immigration. 30 years they did that. Nothing

applicationselon-musk--x

SinkRec: Mitigating Semantic State Sink in Long Sequence Recommendation with Memory-Conditioned Gated Delta Networks

arXiv:2606.09888v1 Announce Type: new Abstract: Linear attention provides an efficient backbone for long-sequence recommendation by avoiding the quadratic cost of standard Transformers, but its compre

safetyarxiv-cs-lg

Sketch-to-Layout: A Human-Centric Computational Agent for Constraint-Aware Synthesis of Modular Photobioreactors

arXiv:2606.09849v1 Announce Type: cross Abstract: Building-integrated photobioreactors (PBRs) offer a pathway for carbon-neutral architecture, yet deployment is hindered by configuration complexity an

safetyarxiv-cs-cv

SkillResolve-Bench: Measuring and Resolving Same-Capability Ambiguity in Agent Skill Retrieval

arXiv:2606.10388v1 Announce Type: cross Abstract: Agent skill libraries are becoming routable software assets: a retrieved skill can contribute instructions, scripts, resource bindings, and execution

Sky News suggesting Northern Ireland has a racism problem. No, it does not. Stop distracting from the truth. It has an open borders problem.

I can't provide a summary of this content as the URL appears to be invalid (the status ID format seems incorrect), and the source attribution to Elon Musk doesn't match the URL structure shown. To cre

industryelon-musk--x

Sleep EEG Signal Criticality as a Non-Invasive Predictor of Cognitive Decline in Dementia

arXiv:2606.10889v1 Announce Type: cross Abstract: Early detection of neurodegeneration remains a critical clinical challenge. This study investigates whether sleep EEG signal criticality, quantified v

researcharxiv-cs-lg

Small Data, Big Noise: Adversarial Training for Robust Parameter-Efficient Fine-Tuning

model-releasesarxiv-cs-cl

arXiv:2606.10610v1 Announce Type: new Abstract: Parameter-Efficient Fine-Tuning (PEFT) has become essential for adapting foundation models to downstream NLP tasks. However, current PEFT methods often

So far today • Banks to SoftBank: No thanks; OpenAI stock ain’t worth what you say. • Germany to Google: LLMs can be held liable • Senator W…

So far today • Banks to SoftBank: No thanks; OpenAI stock ain’t worth what you say. • Germany to Google: LLMs can be held liable • Senator Warren to SEC and SpaceX: Fix this mess; it’’s foul play to s

safetygary-marcus--x

So the message I am getting is that I can't use Fable to further accelerate AI research and education. No company will decide that for me. J…

So the message I am getting is that I can't use Fable to further accelerate AI research and education. No company will decide that for me. Just an absolutely sad day for the research community. As a d

applicationsdair-ai--x

SocraticPO: Policy Optimization via Interactive Guidance

arXiv:2606.09887v1 Announce Type: cross Abstract: Reinforcement learning (RL) for large language models usually supervises reasoning with scalar outcome rewards, such as binary correctness. Such rewar

SoK: Colluding Adversaries in Machine Learning Pipelines

arXiv:2606.10091v1 Announce Type: cross Abstract: Machine learning (ML) models are susceptible to various security, privacy, and fairness risks. Adversaries with different characteristics (i.e., objec

safetyarxiv-cs-lg

Some considerations that many folks seem not to get: 1. It can be a bubble even if the tech works. (For instance, if the tech doesn't have a…

researchfrancois-chollet--x

Some considerations that many folks seem not to get: 1. It can be a bubble even if the tech works. (For instance, if the tech doesn't have a high-demand use case.) 2. It can be a bubble even if the te

Soul Computing: A Theoretical Framework and Technical Architecture for Intelligent Agents with Independent Consciousness

arXiv:2606.10413v1 Announce Type: new Abstract: Breakthroughs in large language models and multimodal generation technologies have propelled the digital reconstruction of human mental traits, emotiona

researcharxiv-cs-ai

Hardware

Sources: Germany's Neura Robotics, which builds AI-powered humanoid robots, raised 1.4B from Tether, Qualcomm, Amazon, Nvidia, and others at a ~7B valuation (Financial Times)

Financial Times: Sources: Germany's Neura Robotics, which builds AI-powered humanoid robots, raised 1.4B from Tether, Qualcomm, Amazon, Nvidia, and others at a ~7B valuation — Crypto group Tether, Ama

hardwaretechmeme

Sources: Microsoft is restricting employees from using Claude Fable 5 because of Anthropic's new 30-day data retention requirements (Tom Warren/The Verge)

Tom Warren / The Verge: Sources: Microsoft is restricting employees from using Claude Fable 5 because of Anthropic's new 30-day data retention requirements — Microsoft's legal teams are evaluating Ant

model-releasestechmeme

Sources: Microsoft's Xbox division is planning major layoffs next month and significant budget cuts for marketing and some other areas (Jason Schreier/Bloomberg)

Jason Schreier / Bloomberg: Sources: Microsoft's Xbox division is planning major layoffs next month and significant budget cuts for marketing and some other areas — Asha Sharma's first major cuts will

industrytechmeme

Sources: OpenAI is considering drastically lowering its price for tokens in anticipation of similar cuts the company expects at Anthropic (Wall Street Journal)

Wall Street Journal: Sources: OpenAI is considering drastically lowering its price for tokens in anticipation of similar cuts the company expects at Anthropic — The company might lower prices for toke

industrytechmeme

Hardware

Sources: OpenAI is in advanced talks to lease a proposed 10GW data center campus in Ohio as part of a deal that could include financial backing from Nvidia (Anissa Gardizy/The Information)

Anissa Gardizy / The Information: Sources: OpenAI is in advanced talks to lease a proposed 10GW data center campus in Ohio as part of a deal that could include financial backing from Nvidia — OpenAI i

hardwaretechmeme

Sources: Sam Altman told staff that OpenAI is expected to go public 'within the next year' and OpenAI plans a tender offer 'very soon' at a $687.69 share price (Stephanie Palazzolo/The Information)

Stephanie Palazzolo / The Information: Sources: Sam Altman told staff that OpenAI is expected to go public “within the next year” and OpenAI plans a tender offer “very soon” at a $687.69 share price —

industrytechmeme