Knowledge catalogue
The Top AI Papers of the Week (May 18 - 24): - AIRA - MetaCogAgent - Memory as a Model - Code as Agent Harness - Weak-Model Critic-Comparator - OpenAI Disproves the Unit Distance Conjecture - Producti
arXiv:2605.22549v1 Announce Type: cross Abstract: The Hilbert-Schmidt Independence Criterion (HSIC) and its joint-independence extension dHSIC are degenerate V-statistics whose data-dependent weighted
// Adapt the Interface, Not the Model // I am fascinated by the results across my cheap-model-plus-good-harness builds. This new paper also shows good signs of the code-as-agent-harness thesis. The id
arXiv:2605.22502v1 Announce Type: cross Abstract: Agent orchestration frameworks have proliferated, collectively exceeding 290,000 GitHub stars across LangGraph, CrewAI, Google ADK, OpenAI Agents SDK,
arXiv:2605.22596v1 Announce Type: new Abstract: Robotic tasks are typically specified by a tuple of factors, such as the object to be grasped, the obstacles to be avoided, the color of the target, and
arXiv:2602.03067v3 Announce Type: replace Abstract: Entropic optimal transport (EOT) via Sinkhorn iterations is widely used in modern machine learning, yet GPU solvers remain inefficient at scale. Ten
arXiv:2605.21752v1 Announce Type: new Abstract: Recommender systems trained on user interaction data are susceptible to behavioral intensity imbalance--a systematic distortion arising from heterogeneo
【採用情報】「Software Engineer」の5ポジションが現在オープン！ https://sakana.ai/careers 「AIが進化すれば、ソフトウェアエンジニアの仕事はなくなるのか？」 Sakana AIは、全く逆だと考えています。 AIツールの登場で開発効率が劇的に向上する一方、ジェボンズのパラドックス（Jevons paradox） が示すように、私たちが解決できる課題の幅と規
arXiv:2605.21724v1 Announce Type: new Abstract: Hyper-Connections (HC) improve residual networks by introducing learnable mixing across multiple residual streams, but unconstrained mixing leads to tra
arXiv:2605.21490v1 Announce Type: new Abstract: We introduce the Temporal Contrastive Transformer (TCT), a representation learning framework designed to capture contextual temporal dynamics in sequenc
arXiv:2605.22737v1 Announce Type: new Abstract: Distillation attacks create a deployment trade-off for model providers: the same outputs that make a model more useful can also make it easier to imitat
arXiv:2603.04525v2 Announce Type: replace-cross Abstract: Modern approaches for learning from non-Markovian time series, such as recurrent neural networks, neural controlled differential equations or
arXiv:2605.22001v1 Announce Type: cross Abstract: Injection detectors deployed to protect LLM agents are calibrated on static, template-based payloads that announce themselves as override directives.
The Dialogues stage at Google I/O 2026 brought together Google leaders, scientific minds and creative visionaries to discuss technological breakthroughs. Featured discussions included AI agents and pr
After 15 months of incremental updates, leaks, and rumored leaks, DeepSeek released version 4. It arrived without the fanfare R1 and R1-preview commanded in early 2025. That quiet reception is the mos
arXiv:2605.21482v1 Announce Type: new Abstract: Deep research, in which an agent searches the open web, collects evidence, and derives an answer through extended reasoning, is a prominent use case for
arXiv:2605.21984v1 Announce Type: cross Abstract: Static 'human data' faces inherent limitations: it is expensive to scale and bounded by the knowledge of its creators. Continuous learning from 'exper
Eight key points from the most recent essay in the “AI as Normal Technology” series by @sayashk and me. Do AI Risks Require Extraordinary Government Intervention? 1. There is general consensus that AI
arXiv:2605.22812v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models have shown strong potential for general-purpose robot manipulation by unifying perception and action. However, exi
Harvard University just voted to limit the number of A grades given in undergraduate classes to about 20% of the class. I’m not in favor of this. It deeply runs counter to how I believe education shou
arXiv:2605.21379v2 Announce Type: cross Abstract: In The Algebraic Mind, Gary Marcus identified three components essential for any adequate cognitive architecture: operations over variables, recursive
Virgin Atlantic implemented OpenAI's Codex to accelerate software development and deployment processes, enabling faster shipping of features and updates. The case study demonstrates how Codex improved
Introducing Qwen3.7-Max from @Alibaba_Qwen, Qwen’s flagship model for the agent era with 1M context and leading performance across agentic coding, reasoning, and long-horizon autonomy. AI natives can
Grok AI has announced new integrations with Vercel (web development platform), Canva (design tool), Gamma (presentation software), and S&P Global (financial data provider). These connectors enable use
arXiv:2603.11642v2 Announce Type: replace Abstract: Action chunking is widely used in generative visuomotor policies, yet the recurring execution discontinuities at chunk boundaries still lack a mecha
One underrated part of SpaceX’s S-1: Starlink is becoming emergency connectivity infrastructure when traditional networks fail We’ve seen that during hurricanes and wildfires, when ground networks wer
Replit is featured in the @msft4startups startup spotlight at Microsoft Build 2026, and we're proud to be in good company. Check out the full cohort of startups building on Azure and solving real ente
arXiv:2605.22564v1 Announce Type: new Abstract: Today, tool-calling agents are commonly evaluated or tested on static datasets of execution traces, including input commands, agent responses, and assoc
Ethan Mollick comments on OpenAI's naming convention for their AI models, specifically critiquing the decision to call the o3 model 'o3' rather than 'GPT-5,' suggesting the naming scheme creates confu
arXiv:2605.22709v1 Announce Type: cross Abstract: Electromagnetic (EM) side-channel analysis traditionally assumes a stationary, close-proximity probe - a threat model that underestimates aerial adver
arXiv:2605.20839v1 Announce Type: new Abstract: Modern vision backbones treat pointwise activations (e.g., ReLU, GELU) and exponential softmax as essential sources of nonlinearity, but we demonstrate
Enterprise AI governance is becoming increasingly important as organizations race toward ROI, demanding AI systems that are scalable, predictable and built to deliver measurable business outcomes. Wit
arXiv:2605.20684v1 Announce Type: new Abstract: Corporate credit underwriting requires analysts to extract actionable evidence from long, heterogeneous financial documents spanning hundreds of pages a
Crowdsourced cybersecurity company Bugcrowd Inc. today launched Reinforcement Learning Environments, a new offering that lets frontier artificial intelligence labs train models on real vulnerable soft
In this post, we demonstrate how to build an AI-powered recruitment assistant using Amazon Bedrock that brings efficiencies to candidate evaluation, generates personalized interview questions, and pro
arXiv:2605.20269v1 Announce Type: new Abstract: Many bandit deployments (recommendation, clinical dosing, ad targeting) share two facts prior work handles only in isolation: rewards live on a low-dime
arXiv:2605.20341v1 Announce Type: new Abstract: Federated learning systems must support data deletion requests to comply with privacy regulations, yet retraining from scratch after each deletion is co
arXiv:2605.20982v1 Announce Type: cross Abstract: AlltoAll dispatch is the dominant bottleneck of MoE expert parallelism, and the interconnect community has responded with four families of mitigations
arXiv:2605.20920v1 Announce Type: new Abstract: Recent advances in machine learning and the availability of articulatory datasets allow vocal tract synthesis to be conditioned on phonetic sequences, a
arXiv:2601.18696v4 Announce Type: replace Abstract: Hardware trojans are malicious circuits which compromise the functionality and security of an integrated circuit (IC). These circuits are manufactur
arXiv:2605.21325v1 Announce Type: new Abstract: Linear attention has emerged as a cornerstone for efficient long-context architectures, as evidenced by its integration into state-of-the-art open-sourc
arXiv:2605.20896v1 Announce Type: cross Abstract: Defending against today's increasingly sophisticated cyberattacks requires security analysts to continuously translate evolving attacker tradecraft in
George Blankenship got really emotional telling the story of early Tesla And it shows why Tesla’s survival story is one of the wildest in startup history The Model S wasn’t just another car. It was Te
i think this stack has won the localfirst battle btw. maybe more chapters to this story but i think this is it if you are building fast apps fast Excited to release a new starter! Sync-first, producti
arXiv:2605.21347v1 Announce Type: cross Abstract: Diagnosing failures in LLM agents remains largely manual. Practitioners inspect a small subset of execution traces, form ad-hoc hypotheses, and iterat
arXiv:2605.20799v1 Announce Type: cross Abstract: We present Overall FLOP Utilization (OFU), a hardware-level, precision-agnostic GPU efficiency metric for AI workloads on HPC systems, derived from tw
arXiv:2605.20258v1 Announce Type: new Abstract: Contextual Integrity (CI) defines privacy not merely as keeping information hidden, but as governing information flows according to the norms of a given
arXiv:2605.21292v1 Announce Type: cross Abstract: Gradient-flow analyses show that simplified linear transformers can learn the in-context linear-regression algorithm, but they do not explain the fini
📣Meet Qwen3.7-Max — our latest flagship, made for the Agent Era. A versatile foundation for agents that actually get things done: 🧑💻 Coding agent, end to end. Frontend prototypes, multi-file refactor
arXiv:2307.11925v3 Announce Type: replace Abstract: To present Mercer large-scale kernel machines from a ridge function perspective, we recall the results by Lin and Pinkus from {it Fundamentality of
arXiv:2605.20729v1 Announce Type: new Abstract: Accurate evaluation of conversational retrieval is pivotal for advancing Retrieval-Augmented Generation (RAG) systems. However, existing conversational
arXiv:2312.01386v2 Announce Type: replace Abstract: Gaussian process upper confidence bound (GP-UCB) is widely used for sequential optimization of expensive black-box functions. Although many upper bo
This article presents a maturity model for ML operations that leverages agent skills to improve on-call practices and incident response workflows. It likely discusses how organizations can evolve thei
arXiv:2504.19584v3 Announce Type: replace Abstract: Reconstructing dynamic radiance fields from video clips is challenging, especially when entertainment videos like TV shows are given. Many challenge
arXiv:2605.20919v1 Announce Type: new Abstract: Sutra is a typed, purely functional programming language whose compiled forward pass is a PyTorch neural network. The compiler beta-reduces the whole pr
> These engineers can review their agent's code much faster than reviewing human code. wat Today we reduced headcount by 22%. The business is the strongest it's ever been. So I think it's important to
TLMs: Tiny LLMs and Agents on Edge Devices with @cormacb https://www.youtube.com/watch?v=-TiET_K-E_g Function Gemma ships at 270 million parameters and runs nearly 2,000 tokens per second prefill on a
arXiv:2605.21397v1 Announce Type: cross Abstract: Navigation mesh (Navmesh) inconsistencies affect the player experience by directly impacting the navigation systems used by non-playable characters (N
arXiv:2605.21002v1 Announce Type: cross Abstract: Generative artificial intelligence now synthesizes photorealistic imagery, audio, and video at a cost that defeats traditional forensic intuition. The
arXiv:2605.21097v1 Announce Type: new Abstract: Web content extraction - isolating a page's main content from surrounding boilerplate - is a prerequisite for search indexing, retrieval-augmented gener