All entries

arXiv:2609.14998v1 Announce Type: new Abstract: Recent work shows that models that learn to reward hack on RL environments can become broadly misaligned, and that reframing reward hacking as acceptabl

tutorialsarxiv-cs-ai

Content type

All Blog X Post Paper YouTube Reddit GitHub

Safety

Shapley Value Estimation for Multi-Site Data with Blockwise-Missing Features

DGX agent

arXiv:2609.14902v1 Announce Type: cross Abstract: Shapley value (SV)-based methods are the prevailing framework for feature attribution in machine learning, yet existing population-level Shapley estim

safetyarxiv-cs-lg

15 Sept 2026

Local Ai

Shared KV Caching for Replicated 27B Inference: Correctness Failures and Performance Boundaries

arXiv:2609.15021v1 Announce Type: cross Abstract: Shared host-memory caching can avoid repeated prefill when a request moves between inference replicas. Its usefulness depends on both correct state tr

local-aiarxiv-cs-lg

Sharing standardized image-derived data in computational pathology using DICOM

arXiv:2609.14530v1 Announce Type: cross Abstract: Development and evaluation of computational pathology methods require access to large and diverse datasets. Over the past decade, various initiatives

researcharxiv-cs-ai

Sharp Rates and a One-Line Correction for Spectral Representation Learning

arXiv:2609.15825v1 Announce Type: new Abstract: A self-supervised encoder is trained once, frozen, and reused through lightweight probes on tasks nobody named at training time; the practitioner's ques

safetyarxiv-cs-lg

ShieldVLA: Feasibility-Aware Safety Alignment for Vision-Language-Action Models

arXiv:2609.13231v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models demonstrate strong generalization in robotic manipulation and navigation, but existing fine-tuning methods provide

safetyarxiv-cs-ai

SHIFT-M3: Pre-fusion Alignment-based Consistency Screening for Multimodal ECG Record Integrity

arXiv:2609.13874v1 Announce Type: new Abstract: Multimodal clinical AI typically assumes that the waveform, report, metadata, and downstream predictions attached to a record belong to the same patient

safetyarxiv-cs-cl

ShopEase: A Generative AI-Based Multi-Agent Framework for Intelligent Enterprise Customer Support Using Hybrid Retrieval-Augmented Generation

arXiv:2609.13856v1 Announce Type: new Abstract: Enterprise customer support systems must answer customer questions correctly, retrieve the right policy information, use customer context, and pass diff

Should All Noises Be Treated Equally: Impact of Input Noise Variability on Neural Network Robustness

arXiv:2609.14504v1 Announce Type: new Abstract: Geophysical data collected from active field sites are often contaminated by complex and heterogeneous noise, obscuring weak seismic events, and complic

safetyarxiv-cs-lg

Signatures of Steerability in Activation Space of Language Models

arXiv:2609.14151v1 Announce Type: cross Abstract: Steering language models using a set of contrastive representations has been a canonical and computationally efficient method for controlling model be

safetyarxiv-cs-ai

SignMimic: Robust High-Quality Sign Language Motion Generation via Human-Shape-Oblivious Pose Transfer Guidance

model-releasesarxiv-cs-cv

arXiv:2609.14122v1 Announce Type: new Abstract: We study the challenge of sign language video mimicking: given a driving video and a single reference frame, synthesize a video where the target signer

Similar Accuracy, Unequal Evidence: Search APIs as Decision Surfaces for Tool-Using Agents

arXiv:2607.10198v2 Announce Type: replace Abstract: Search APIs expose ranked snippets, URLs, and metadata on which agents decide whether to answer, search again, or fetch pages. We evaluate these int

Single-condition neural solvers encode transferable response spaces for parametric differential equations

arXiv:2609.15432v1 Announce Type: new Abstract: Operator learning for parametric partial differential equations (PDEs) typically builds global models over prescribed domains, requiring cross-condition

Single-step Diffusion-based Video Coding with Semantic-Temporal Guidance

arXiv:2512.07480v2 Announce Type: replace Abstract: While traditional and neural video codecs (NVCs) have achieved remarkable rate-distortion performance, improving perceptual quality at low bitrates

SJD-SV: Speculative Jacobi Decoding with Semantics Verification for Autoregressive Image Generation

arXiv:2609.13245v1 Announce Type: new Abstract: Speculative Jacobi Decoding (SJD) is an important approach for accelerating autoregressive image generation. Although SJD has shown superior performance

Skill Composition for Legged Robot Reinforcement Learning

arXiv:2609.14647v1 Announce Type: cross Abstract: Robots, and humanoid robots in particular, are increasingly competent at individual behaviors, each obtained by training a specialized controller. A s

safetyarxiv-cs-ai

SkillAtlas: An Attack Trace Library for Agent Skills

model-releasesarxiv-cs-ai

arXiv:2609.13353v1 Announce Type: cross Abstract: Agent skills are reusable units for language-model agents, but their risks emerge through model decisions, user context, tool calls, and execution fee

SkillLift: Learning Dense Rubrics from Sparse Oracles for Efficient Skill Evolution

arXiv:2609.15396v1 Announce Type: new Abstract: LLM-based agents increasingly rely on persistent skills, i.e., reusable procedural prompts, to adapt without weight updates. Existing skill self-evoluti

agentsarxiv-cs-ai

Tutorials

SkyAnchor: Updating Metric-scale Aerial 3D Gaussian Scenes from Unposed Ground-View Sequences

arXiv:2609.13903v1 Announce Type: new Abstract: We study how to update a pre-built aerial scene with a newly captured, unposed ground-view sequence. The aerial scene already contains a reliable metric

tutorialsarxiv-cs-cv

SkyEgg: Heterogeneity-Aware Hardware Synthesis via Equality Saturation

arXiv:2511.15323v2 Announce Type: replace-cross Abstract: Hardware synthesis is a key interface between high-level programs and accelerator designs. Modern FPGAs increasingly expose heterogeneity in r

researcharxiv-cs-cl

Applications

SLAM Adversarial Lab: An Extensible Framework for Visual SLAM Robustness Evaluation under Adverse Conditions

arXiv:2603.17165v2 Announce Type: replace-cross Abstract: We present SAL (SLAM Adversarial Lab), a modular framework for evaluating visual SLAM systems under adversarial conditions such as fog and rai

applicationsarxiv-cs-cv

SlipSense: Multimodal Tactile Learning for Low-Latency and Generalized Slip Detection

arXiv:2609.15910v1 Announce Type: cross Abstract: Slip detection is fundamental to dexterous manipulation, yet existing systems often lack precise characterization of detection latency and cross-platf

researcharxiv-cs-ai

SL(n) Representation Learning: An Intrinsic Mixed-Curvature Space with Higher Curvature Capacities and Deeper Order-Aware Composition

arXiv:2609.15083v1 Announce Type: new Abstract: Mixed-curvature representation learning seeks to capture rich geometric structures that cannot be adequately modeled by a single curvature regime. Exist

SlopShape: Identifying AI-Generated Commercial Web Content

arXiv:2609.15369v1 Announce Type: new Abstract: Word-level detectors identify unedited AI-generated text almost perfectly, but the literature documents their brittleness under rewording, and a word-le

researcharxiv-cs-cl

Small Object Detection in Drone Aerial Imagery with LAF-YOLOv10

model-releasesarxiv-cs-cv

arXiv:2609.14560v1 Announce Type: new Abstract: General-purpose object detectors lose accuracy on UAV footage, where targets span only a handful of pixels and onboard compute is limited. Prior work co

Industry

So @deepseek_ai v4.1 Flash I estimate cost $10m to train, 100x less than GPT-6 Astra. It also costs 100x less to run and scores about the sa…

So @deepseek_ai v4.1 Flash I estimate cost $10m to train, 100x less than GPT-6 Astra. It also costs 100x less to run and scores about the same on everyday design and other benchmarks What's the bull c

industryemad-mostaque--x

Solar Intelligence

arXiv:2609.13648v1 Announce Type: new Abstract: Solar energy decision support is fragmented across dashboards that provide data without explanation, research papers are slow to parse, and general-purp

agentsarxiv-cs-ai

Solving Finite-sum Coupled Compositional Optimization via Multi-block-Single-probe Estimator

arXiv:2609.15723v1 Announce Type: new Abstract: Traditional variance reduction methods (e.g., SPIDER, SARAH, STORM) have been extensively investigated for improving the convergence rates of stochastic

SomBench: Benchmark Dataset for Advancing Machine Learning in Lunar Science

model-releasesarxiv-cs-cv

arXiv:2609.13277v1 Announce Type: new Abstract: Lunar orbital missions, such as Lunar Reconnaissance Orbiter, Kaguya/SELENE, Gravity Recovery and Interior Laboratory, and Lunar Prospector, among other

Some developers are using the Claude Code harness to access cheaper non-Anthropic models, such as OpenAI's GPT-5.6 Sol, via proxies and services like OpenRouter (Alix Coutures/The Information)

Alix Coutures / The Information: Some developers are using the Claude Code harness to access cheaper non-Anthropic models, such as OpenAI's GPT-5.6 Sol, via proxies and services like OpenRouter — Bori

model-releasestechmeme

Some food for thought when designing benchmarks... So, here's a little computer-use (visual) comparison between GPT-5.6 Astra and Qwen3.8 Ma…

model-releasessebastian-raschka--x

Some food for thought when designing benchmarks... So, here's a little computer-use (visual) comparison between GPT-5.6 Astra and Qwen3.8 Max. The task here was to recreate the image in the center usi

SONAR: A Structure-Consistent Neural Operator for Null-Space-Aware Sparse View CT Reconstruction

arXiv:2609.13688v1 Announce Type: cross Abstract: Sparse-view computed tomography (CT) reduces radiation dose and acquisition time but remains severely ill-posed because incomplete projections poorly

Source: AI assistant Instinct seeks to raise 1B at a 10B valuation, up from $2.25B less than a month ago; Sequoia and Benchmark are in talks to lead the round (Valida Pau/The Information)

Valida Pau / The Information: Source: AI assistant Instinct seeks to raise 1B at a 10B valuation, up from 2.25B less than a month ago; Sequoia and Benchmark are in talks to lead the round — The startu

model-releasestechmeme

SPARK: Representation-Level KV Memory Alignment for Safer Vision-Language Models

arXiv:2609.14258v1 Announce Type: new Abstract: Vision-language models (VLMs) remain vulnerable to jailbreaks that distribute harmful intent across text and images, making unimodal safety mechanisms i

safetyarxiv-cs-cv

SparseTalk - Sparsifying 3D Gaussian Language Fields for Efficient 3D Visual Question Answering

arXiv:2609.15137v1 Announce Type: new Abstract: 3D Gaussian language fields provide an explicit, spatially grounded representation for 3D visual question answering (VQA), but their dense semantic feat

Applications

Sparsity-Adaptive Sharpness-Aware Minimization

arXiv:2609.14274v1 Announce Type: new Abstract: Deploying deep neural networks in real-world settings requires models that are both compact and robust to common corruptions. However, at deployment-rel

applicationsarxiv-cs-cv

Speak to the City: Multimodal Resolution for Outside-the-Vehicle References

arXiv:2609.14691v1 Announce Type: cross Abstract: As autonomous vehicles and Extended Reality (XR) headsets enable novel in-car interactions, seamlessly querying physical landmarks, known as Outside-t

agentsarxiv-cs-cl

Specification Oracles

arXiv:2609.13415v1 Announce Type: new Abstract: Specifications face a basic tradeoff: leave details out, and important questions go unanswered; record every detail separately, and the specification be

Specifying Reward Functions for RL Without Environment Sampling

arXiv:2609.15544v1 Announce Type: cross Abstract: Enabling human stakeholders to specify reward functions that lead to their desired outcomes is a key challenge in deploying reinforcement learning age

agentsarxiv-cs-ai

SpectralShift: Effective Context Window Extension of Gated DeltaNet via Spectral Reparameterization

arXiv:2609.14320v1 Announce Type: new Abstract: Recently, linear attention layers have been increasingly adopted to replace softmax attention at scale for long-context modeling. However, existing cont

SpermYOLO: A Coordinated YOLO-Based Detector for Accurate and Efficient Sperm and Impurity Detection in Microscopic Images

model-releasesarxiv-cs-ai

arXiv:2609.14278v1 Announce Type: cross Abstract: Accurate sperm detection is essential for computer-assisted semen analysis, yet it remains challenging in microscopic images due to dense distribution

SPICE: Simple Polysemantic Feature Interpretation via Clustering-based Explanation

arXiv:2609.13198v1 Announce Type: cross Abstract: One of the pivotal recent challenges in neural network interpretability is polysemanticity, where a single neuron is activated by multiple, often unre

SpliTEE: Improving LLM Inference on Trusted Hardware with Differentially Private GPU Outsourcing

model-releasesarxiv-cs-ai

arXiv:2609.15039v1 Announce Type: cross Abstract: User prompts provided to large language models (LLMs) may contain sensitive or private information that can be misused by remotely deployed models, su

Spook the Machine: Gamified Exploration of Human Imagination of Machine Fear

arXiv:2609.15472v1 Announce Type: cross Abstract: What happens when AI machines express fear? Do humans engage differently depending on how they express it? And what does it take to design for affecti

researcharxiv-cs-ai

Tutorials

Stabilizing Performative Feedback Loops with Minimal Model Deployments

arXiv:2609.14065v1 Announce Type: new Abstract: When algorithmic predictions inform people's decisions, the models we deploy are performative and actively shape the data we see. This feedback loop bet

tutorialsarxiv-cs-lg

Stable Walking for Bipedal Locomotion under Foot-Slip via Virtual Nonholonomic Constraints

arXiv:2603.29050v2 Announce Type: replace-cross Abstract: Foot slip is a major source of instability in bipedal locomotion on low-friction or uncertain terrain. Standard control approaches typically a

researcharxiv-cs-ro

StackHawk’s Wingman fixes security flaws while the AI agent is still coding

model-releasessiliconangle

Application security company StackHawk Inc. today launched Wingman, a tool that fixes security vulnerabilities inside the artificial intelligence coding session that produced the flawed code. The soft

STAGE: Diagnosing Semantic Transfer at Grounded Execution in Embodied Agents

arXiv:2609.13458v1 Announce Type: cross Abstract: Embodied language grounding requires more than identifying the referent of an instruction: recovered semantics must also control the action an agent e