All entries

RotRNN: Modelling Long Sequences with Rotations

arXiv:2407.07239v3 Announce Type: replace Abstract: Linear recurrent neural networks, such as State Space Models (SSMs) and Linear Recurrent Units (LRUs), have recently shown state-of-the-art performa

Tools

Routing for serverless servers with Pingora, Envoy, and Spanner

This article discusses routing infrastructure and architectural approaches for serverless computing platforms, examining how systems like Pingora and Envoy handle request routing and load balancing, w

toolsmodal-blog

25 Jun 2026

Run Codex in the cloud – DigitalOcean for Codex is now available

DigitalOcean has launched a cloud-based service enabling users to run OpenAI's Codex, an AI code generation model, on DigitalOcean's infrastructure. This offering allows developers to access Codex cap

industrydigitalocean

Runpod raises $100M to build the leading cloud platform for AI developers

Artificial intelligence developer-centric cloud provider Runpod Inc. announced Tuesday it has raised 100 million in early-stage funding, pushing the valuation of the company to 1 billion. The Series A

industrysiliconangle

RWGBench: Evaluating Scholarly Positioning in Related Work Generation

model-releasesarxiv-cs-ai

arXiv:2606.24894v1 Announce Type: cross Abstract: Large language models have shown strong fluency in scientific writing, yet the evaluation of related work generation (RWG) remains limited. Existing R

S^{2}-FracMix: Label-Preserving Self-Saliency Mixup Augmentation

arXiv:2606.25784v1 Announce Type: new Abstract: Data augmentation is known to improve generalization of deep visual models. Recent methods favor mixup strategies that generate interpolated samples to

Hardware

SA-LIVO: Efficient LiDAR-Inertial-Visual Odometry with Subspace-Aware Degeneracy Handling

arXiv:2606.25699v1 Announce Type: new Abstract: Tightly coupled LiDAR-visual-inertial odometry (LIVO) fuses precise geometric depth with complementary visual measurements, yet its exteroceptive sensor

hardwarearxiv-cs-ro

SAC^2-Net: Semantic Anchoring and Complementary-Consensus Fusion for Multimodal Micro-Expression Recognition

arXiv:2606.25542v1 Announce Type: new Abstract: Micro-expression recognition (MER) is challenging due to subtle facial movements, limited data, and the ambiguous relationship between Action Units (AUs

safetyarxiv-cs-cv

Safe Learning Control with Optimality and Stability Guarantees

arXiv:2501.15373v2 Announce Type: replace-cross Abstract: Merely pursuing performance may adversely affect safety, while a conservative policy for safe exploration will degrade the performance. How to

safetyarxiv-cs-lg

SAGE-Nav: Leveraging LLM Planning and Alignment Fusion for Hierarchical Scene Graph-Guided Navigation

arXiv:2606.25497v1 Announce Type: new Abstract: Object-Goal Navigation (ObjNav) requires embodied agents to autonomously locate specified targets using only egocentric visual observations. Existing mo

safetyarxiv-cs-ro

Hardware

Sail, whose software optimizes how AI models run on existing chips, emerges from stealth with 80M in seed and Series A led by Kleiner at a 450M valuation (Lily Mae Lazarus/Fortune)

Lily Mae Lazarus / Fortune: Sail, whose software optimizes how AI models run on existing chips, emerges from stealth with 80M in seed and Series A led by Kleiner at a 450M valuation — For months, Klei

hardwaretechmeme

Salesforce launches Help Agent to simplify AI customer service deployment

model-releasessiliconangle

Salesforce Inc. is launching a new prepackaged artificial intelligence agent for customer service, enabling organizations to quickly build and deploy AI agents. Today Salesforce announced Help Agent,

Same Evidence, Different Answer: Auditing Order Sensitivity in Multimodal Large Language Models

arXiv:2606.26079v1 Announce Type: new Abstract: Standard benchmarks for multimodal large language models (MLLMs) score each item on one canonical ordering and miss whether order-irrelevant shuffling c

Sample complexity of unbalanced entropic OT

arXiv:2606.24987v1 Announce Type: cross Abstract: Optimal transport (OT) has become a central language for comparing probability measures, but exact balanced OT is often both too rigid for data with m

Sampling Strategies for Robust Universal Quadrupedal Locomotion Policies

model-releasesarxiv-cs-ro

arXiv:2510.07094v2 Announce Type: replace Abstract: This work focuses on sampling strategies of configuration variations for generating robust universal locomotion policies for quadrupedal robots. We

SARA: Unlocking Multilingual Knowledge in Mixture-of-Experts via Semantically Anchored Routing Alignment

arXiv:2606.25821v1 Announce Type: new Abstract: Sparse Mixture-of-Experts (MoE) architectures have emerged as an increasingly influential paradigm as they offer a strategic balance between parameter s

Sarashina2.2-TTS: Tackling Kanji Polyphony in Japanese Speech Generation via Data Scaling and Targeted Data Synthesis

arXiv:2606.25369v1 Announce Type: cross Abstract: While large language model (LLM)-based text-to-speech (TTS) systems have achieved high-quality speech synthesis, most existing systems focus on Englis

Scalable Peptide Design via Memory-Efficient Equivariant Transformer

arXiv:2606.25006v1 Announce Type: new Abstract: Target-specific peptide design requires sequence and structure co-design under full atom geometric constraints. Latent generative frameworks offer an ef

Scale or Reason? A Compute-Equivalent Analysis of Reasoning Distillation

arXiv:2509.22193v2 Announce Type: replace Abstract: Distilling reasoning traces from strong teacher models has become the standard recipe for building capable small language models. Yet reasoning trac

researcharxiv-cs-cl

Scaled Cognition nabs $100M to automate high stakes customer interactions

Artificial intelligence startup Scaled Cognition Inc. today announced that it has raised 100 million in funding. Khosla Ventures led the Series A deal with participation from Genesys Telecommunication

industrysiliconangle

Applications

ScaleHP: Estimating Hand Pose in Metric Space

arXiv:2606.25619v1 Announce Type: new Abstract: Accurate metric-space hand pose estimation (HPE) is essential for immersive human-computer interaction and robotics. However, most existing methods pred

applicationsarxiv-cs-cv

Hardware

Scaling AI Inference Across Multiple GPUs Using NVIDIA TensorRT with Multi-Device Inference Support

NVIDIA TensorRT's multi-device inference feature scales inference across multiple GPUs, enabling deployment of models that exceed single-GPU memory or reducing latency for memory-bound workloads. Each

hardwarenvidia-developer

ScalingAR: Scaling Confidence for Autoregressive Image Generation

arXiv:2509.26376v3 Announce Type: replace Abstract: Test-time strategies have shown remarkable success in improving large language models, but their application to next-token prediction (NTP) autoregr

safetyarxiv-cs-cv

SciRisk-Bench: A Risk-Dimension-Aware Benchmark for AI4Science Safety

model-releasesarxiv-cs-ai

arXiv:2606.18936v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly embedded in AI for Science (AI4Science) workflows, from scientific question answering and literature a

SDE-Driven Spatio-Temporal Hypergraph Neural Networks for Irregular Longitudinal fMRI Connectome Modeling in Alzheimer's Disease

arXiv:2603.20452v2 Announce Type: replace Abstract: Longitudinal neuroimaging is essential for modeling disease progression in Alzheimer's disease (AD), yet irregular sampling and missing visits pose

SE-AGCNet: An End-to-End Framework for Joint Speech Enhancement and Loudness Control in Meeting Scenarios

arXiv:2606.25959v1 Announce Type: cross Abstract: Conventional audio pipelines typically treat speech enhancement (SE) and automatic gain control (AGC) as discrete modules, which often limits overall

researcharxiv-cs-ai

Securing Time Integrity in Energy IoT Against Clock Drift and Y2K38 Failures

arXiv:2601.23147v2 Announce Type: replace Abstract: The integrity of time in distributed Internet of Things (IoT) devices is crucial for reliable operation in energy cyber-physical systems, such as sm

Local Ai

Security and Privacy in Retrieval-Augmented Generation: Architectures, Threats, Defenses, and Future Directions for Building Trustworthy Systems

arXiv:2606.25533v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) has emerged as a dominant paradigm for enhancing large language models with external knowledge. By coupling retri

local-aiarxiv-cs-cl

Local Ai

Seedance 2.0 Mini is live in ComfyUI. Same cinematic toolkit. Same nodes. Lower cost per run. → 480p + 720p output → Multi-shot sequences wi…

Seedance 2.0 Mini is live in ComfyUI. Same cinematic toolkit. Same nodes. Lower cost per run. → 480p + 720p output → Multi-shot sequences with coherent pacing → Stable character identity across shots

local-aicomfyui--x

Self Capacitive Tactile Sensor System designed for Companion Robots

arXiv:2606.25348v1 Announce Type: new Abstract: Tactile sensing is essential for humanoid robots to achieve safe physical interaction, dexterous manipulation, and truly human-like responsiveness. Howe

researcharxiv-cs-ro

Self-Modulating Quantum Fast-Weight Programmers for Efficient Adaptive Sequential Learning

arXiv:2606.24933v1 Announce Type: cross Abstract: Recent advances in quantum machine learning have motivated efficient models for sequential data processing. In this paper, we propose Self-Modulating

Semantic Allocation in Ordered Bottlenecks: Predictive Residual Inference for Visual Representation Learning

arXiv:2606.25232v1 Announce Type: cross Abstract: Ordered bottlenecks aim to provide utility at flexible budgets by assigning coarse information to early tokens and task-relevant detail to later ones.

Semantic Consistency Policy Optimization for Reinforcement Learning of LLM Agents

arXiv:2606.25852v1 Announce Type: new Abstract: Group-based reinforcement learning effectively post-trains LLM agents for long-horizon, sparse-reward tasks by deriving step-level credit from trajector

safetyarxiv-cs-lg

SEMIR: Topology-Preserving Graph Minors for Thin-Structure Segmentation

arXiv:2606.24935v1 Announce Type: new Abstract: Thin-structure segmentation--power lines, cracks, lane markings at 1-3 pixel width--requires preserving connectivity that standard representations precl

SFL-MTSC: Leveraging Semantic Frame-Level Multi-Task Self-Consistency for Robust Multi-Intent Spoken Language Understanding

arXiv:2606.25552v1 Announce Type: new Abstract: Prompt-based spoken language understanding (SLU) with large language models (LLMs) often suffers from inconsistent intent--slot structures due to decodi

Shapley-Inspired Feature Weighting in k-means with No Additional Hyperparameters

model-releasesarxiv-cs-lg

arXiv:2508.07952v2 Announce Type: replace Abstract: Clustering algorithms often assume all features contribute equally to the data structure, an assumption that usually fails in high-dimensional or no

Shift Variant Image Degradation and Restoration Using Singular Value Decomposition

arXiv:2606.25818v1 Announce Type: new Abstract: Shift-variant image degradation is frequently encountered in practical imaging systems where the point spread function (PSF) varies across the image fie

ShutterMuse: Capture-Time Photography Guidance with MLLMs

model-releasesarxiv-cs-cv

arXiv:2606.25763v1 Announce Type: new Abstract: Real-world photography requires capture-time guidance for both camera framing and subject pose. Yet existing aesthetic cropping benchmarks mainly evalua

Silent Failures in Physics-Informed Neural Networks: Parameter Poisoning and the Limits of Loss-Based Validation

model-releasesarxiv-cs-lg

arXiv:2606.25151v1 Announce Type: new Abstract: Physics-informed neural networks (PINNs) embed governing equations in their loss function, enabling mesh-free solutions to partial differential equation

Simplify to Amplify: Achieving Information-Theoretic Bounds with Fewer Steps in Spectral Community Detection

arXiv:2602.17104v2 Announce Type: replace-cross Abstract: We propose a streamlined spectral algorithm for community detection in the two-community stochastic block model (SBM) under constant edge dens

Agents

Skill-MAS: Evolving Meta-Skill for Automatic Multi-Agent Systems

arXiv:2606.18837v2 Announce Type: replace-cross Abstract: Large Language Model (LLM)-based automatic Multi-Agent Systems (MAS) generation has become a crucial frontier for tackling complex tasks. Howe

agentsarxiv-cs-lg

Small edits, large models: How Wikipedia advocacy shapes LLM values

arXiv:2606.24890v1 Announce Type: new Abstract: Can a small group of volunteers shape how AI systems discuss animal welfare, just by editing Wikipedia? We show that they can. Wikipedia appears in near

Small Initialization Matters for Large Language Models

model-releasesarxiv-cs-ai

arXiv:2606.17945v2 Announce Type: replace Abstract: Large language models provide a tractable system for asking how intelligence itself emerges, rather than only how LLMs can be engineered. Although p

SoftBank shares fell 12% after reports that OpenAI may delay its IPO until 2027; expectations of a windfall from OpenAI's debut helped support Softbank's stock (Aya Wagatsuma/Bloomberg)

Aya Wagatsuma / Bloomberg: SoftBank shares fell 12% after reports that OpenAI may delay its IPO until 2027; expectations of a windfall from OpenAI's debut helped support Softbank's stock — SoftBank Gr

industrytechmeme

Agents

SoK: AI Secure Code Generation: Progress, Pitfalls, and Paths Forward

arXiv:2606.25195v1 Announce Type: cross Abstract: The increasing use of AI systems for code generation raises a central security question: what can today's models and coding agents actually do to prod

agentsarxiv-cs-ai

Solving Markov Decision Processes with Future Information via MPC

arXiv:2606.24991v1 Announce Type: cross Abstract: Model Predictive Control (MPC) is widely used in industrial and robotic systems for enforcing constraints and embedding domain knowledge through finit

safetyarxiv-cs-lg

Sony's Bungie says it is laying off a 'significant number of employees', including 'most of the Destiny team'; Bungie has an estimated headcount of 800 (Ethan Gach/Kotaku)

Ethan Gach / Kotaku: Sony's Bungie says it is laying off a “significant number of employees”, including “most of the Destiny team”; Bungie has an estimated headcount of 800 — The studio said the sci-f

industrytechmeme

Soros is Magneto and his son is Mini Magneto

Soros is Magneto and his son is Mini Magneto BREAKING: George Soros is the largest individual donor in the 2026 midterm election cycle with $102 million to Democrats. This is the same George Soros who

industryelon-musk--x