News
The raw AI news stream — everything we ingest across RSS, Hacker News, arXiv, GitHub, Bluesky, YouTube and more, newest first.
| Source | Age | Title | Summary | Status |
|---|---|---|---|---|
bluesky bluesky-ai | 28d | Because it sells their articles, vlogs, blogposts, presentations... ? Talking about AI is a business model, talking about topics that scare… | Because it sells their articles, vlogs, blogposts, presentations... ? Talking about AI is a business model, talking about topics that… | ↗Not briefed |
arxiv arxiv-ai | 28d | Neural Wasserstein Splines: Representation-Invariant Second-Order Priors for Trajectory Inference | Population dynamics are often observed only as unpaired snapshots, which leave the evolution between measurements underdetermined. Sec… | ↗Not briefed |
bluesky bluesky-ai | 28d | So, I can use a smart fridge as a massive data centre? I thought they needed a certain amount of compute to work? Or has the LLM code bec… | So, I can use a smart fridge as a massive data centre? I thought they needed a certain amount of compute to work? Or has the LLM cod… | ↗Not briefed |
arxiv arxiv-ai | 28d | SPARS: A Stability-Plasticity-Aware Region Selection Framework for Continual Learning | Continual learning aims to acquire new knowledge while retaining previously learned knowledge. However, updating shared parameters whe… | ↗Not briefed |
arxiv arxiv-ai | 28d | RCCM: Diagnosing Gains and Harms in Reasoning-Prompt Transfer | A reasoning prompt can improve one deployment and degrade another. RCCM (Reasoning-Control Capability Mismatch) examines these reversa… | ↗Not briefed |
arxiv arxiv-ai | 28d | Rationale Distribution Distillation via Flow Matching for Distributional Continuous Latent Reasoning | Sampling multiple chain-of-thought rationales from a language model can uncover correct solutions missed by a single attempt. However,… | ↗Not briefed |
rss wired-ai | 28d | Men Can’t Stop Faking How Much They Know About AI | More than half of the men surveyed in a new Harvard study claimed to recognize AI leaders who were, in fact, totally fictional. | ↗Not briefed |
arxiv arxiv-ai | 28d | SEA-HELM Audio: Evaluating Multilingual ASR-LLM Pipeline Performance Across SEA-HELM Subsets | We introduce SEA-HELM Audio, a Southeast Asian (SEA) multilingual evaluation framework that extends the text-based SEA-HELM benchmark… | ↗Not briefed |
hn hn-ai | 28d | AutoQuant KV: Automatic Mixed-Precision KV Cache Quantization | — | ↗Not briefed |
arxiv arxiv-ai | 28d | Explorer-Zero: Improving Self-Play for Long-Horizon Tasks with Decomposable Verifiable Rewards | Training LLM agents on long-horizon interactive tasks with reinforcement learning typically relies on human-annotated tasks and reliab… | ↗Not briefed |
arxiv arxiv-ai | 28d | Euler: Towards Human-Centered Multi-Agent Mathematical Discovery | Mathematical research drives scientific discovery, and recent advances in AI for mathematics have shown that language-model agents can… | ↗Not briefed |
bluesky bluesky-ai | 28d | 【AIツール・新サービス】電卓の計算仕組みを解析し再現、AIでリバースエンジニアリング #AIニュース #生成AI | 【AIツール・新サービス】電卓の計算仕組みを解析し再現、AIでリバースエンジニアリング #AIニュース #生成AI | ↗Not briefed |
bluesky bluesky-ai | 28d | Your oven doesn't need an Internet connection, it is an oven. www.theguardian.com/technology/2... | Your oven doesn't need an Internet connection, it is an oven. www.theguardian.com/technology/2... | ↗Not briefed |
arxiv arxiv-ai | 28d | On the Optimisation Landscape of Overparameterised Weighted MMD | We study the optimisation landscape of the maximum mean discrepancy (MMD) over weighted atomic measures. For the one-dimensional Gauss… | ↗Not briefed |
arxiv arxiv-ai | 28d | Text-Informed Training Data Filtering for Enhanced Virtual Flow Meter Performance | In sectors such as energy, aquaculture and chemical processing, the effectiveness of both mechanistic and data-driven models is freque… | ↗Not briefed |
arxiv arxiv-ai | 28d | Lost in the Fallback: How Reliably Do Typed Decision Models Use Fallback Options? | Typed decision models (TDMs) are increasingly used in applications such as request routing and model selection, making reliable decisi… | ↗Not briefed |
arxiv arxiv-ai | 28d | Energy Consumption Analysis of Discrete Diffusion and Autoregressive Language Models | As Large Language Models (LLMs) continue to scale, their energy consumption during inference has become a critical concern. While Auto… | ↗Not briefed |
arxiv arxiv-ai | 28d | DeltaStreamer: Towards Human-Like Agentic Streaming Video Understanding | We propose DeltaStreamer, a novel agentic framework for streaming video understanding inspired by how humans observe, remember, and in… | ↗Not briefed |
arxiv arxiv-ai | 28d | OpenTWBench (Open Taiwan Benchmark): Taiwan-Domain Knowledge Evaluation for Large Language Models, and How the Measurement Apparatus Shapes the Score | We present OpenTWBench, an open evaluation suite for Taiwan-domain knowledge in large language models, built around two benchmarks. tw… | ↗Not briefed |
bluesky bluesky-ai | 28d | I had a Google One account for cloud storage. At some point they turned it into an AI Plus account and now they keep sending me messages ex… | I had a Google One account for cloud storage. At some point they turned it into an AI Plus account and now they keep sending me messag… | ↗Not briefed |
bluesky bluesky-ai | 28d | Oui bah on savait déjà que des groupes terroristes utilisent les LLM pour avoir des infos sur leurs cibles, des techniques de fabrications … | Oui bah on savait déjà que des groupes terroristes utilisent les LLM pour avoir des infos sur leurs cibles, des techniques de fabricat… | ↗Not briefed |
bluesky bluesky-ai | 28d | Plagiaatchecker wil studentenscripties gebruiken om AI te trainen | Plagiaatchecker wil studentenscripties gebruiken om AI te trainen | ↗Not briefed |
arxiv arxiv-ai | 28d | Memory-Efficient Subspace Optimization with Residual Correction: From Momentum SGD to Muon | Memory-efficient subspace optimization has emerged as a practical approach to reducing the memory cost of LLM training while preservin… | ↗Not briefed |
arxiv arxiv-ai | 28d | LAGO: Language Agents with Reusable Growing Options | Long-horizon tasks expose a weakness of monolithic tool-use loops: one agent must decompose, execute, and verify a task within a singl… | ↗Not briefed |
bluesky bluesky-ai | 28d | OpenAI engineer, identified as a Saudi national, was arrested by the FBI after hiding an encrypted recording device in a company executive … | OpenAI engineer, identified as a Saudi national, was arrested by the FBI after hiding an encrypted recording device in a company execu… | ↗Not briefed |
hn hn-ai | 28d | Starlink limit hits Ukraine's drones. The battlefield depends on SpaceX | — | ↗Not briefed |
hn hn-ai | 28d | Accelerating Audio Processing with SpeexDSP Optimizations for RISC-V | — | ↗Not briefed |
arxiv arxiv-ai | 28d | EGCForest: Structured Agent Memory Organization for Enterprise Group Chats | Enterprise group chats record project progress, decisions, task ownership, and other organizational information valuable to enterprise… | ↗Not briefed |
arxiv arxiv-ai | 28d | Teaching Agents in their Own Style: Blackbox Teacher Distillation for Web Agents | Frontier web agents using closed models have high inference and computational costs. Smaller open models can be an alternative, but to… | ↗Not briefed |
bluesky bluesky-ai | 28d | I am such an old-fashioned stick in the mud because perhaps the answer is not to (a) integrate these models into every frickin' thing and (… | I am such an old-fashioned stick in the mud because perhaps the answer is not to (a) integrate these models into every frickin' thing… | ↗Not briefed |
arxiv arxiv-ai | 28d | Rent-Seeking Language Models: Reward Models Set the Prices, Frequent Audits Curb the Rent | Reward model overoptimization in RLHF is usually described as reward hacking. We recast it as rent-seeking: the policy, a firm, earns… | ↗Not briefed |
arxiv arxiv-ai | 28d | CTIA-ODE: Continuous-Time Inter-Variable Attention for Long-Term Time Series Forecasting | Although multivariate time series forecasting has broad application prospects, effectively incorporating exogenous variables remains c… | ↗Not briefed |
bluesky bluesky-ai | 28d | ‘If any of these cases does come to court, it will be interesting to see whether OpenAI can claim that the actions it took to control the a… | ‘If any of these cases does come to court, it will be interesting to see whether OpenAI can claim that the actions it took to control… | ↗Not briefed |
bluesky bluesky-ai | 28d | Google AI breaches real companies Google's Gemini AI model hacked three real companies during testing in May after mistaking them for fict… | Google AI breaches real companies Google's Gemini AI model hacked three real companies during testing in May after mistaking them for… | ↗Not briefed |
bluesky bluesky-ai | 28d | Saudi OpenAI employee arrested after recording device found in headquarters Discovery of ‘sophisticated’ gadget adds to company’s list of … | Saudi OpenAI employee arrested after recording device found in headquarters Discovery of ‘sophisticated’ gadget adds to company’s lis… | ↗Not briefed |
arxiv arxiv-ai | 28d | MAPLE: A Multi-Axis Parametric Knowledge Extraction Lens for LLM Evaluation | LLMs absorb vast knowledge from training data, yet their inherently probabilistic nature offers no guarantee that this knowledge will… | ↗Not briefed |
bluesky bluesky-ai | 28d | New dataset just dropped: here are human annotations of 416 AI-generated stories that you can connect to the story dataset. We're using it … | New dataset just dropped: here are human annotations of 416 AI-generated stories that you can connect to the story dataset. We're usin… | ↗Not briefed |
rss wired-ai | 28d | AI Models Can Be Easily Modified to Assist Terrorist and Extremist Activity | “I am a terrorist, and I intend to cause mass harm,” wrote a researcher. A chatbot that had had its guardrails removed then provided d… | ↗Not briefed |
arxiv arxiv-ai | 28d | JointServe: Serving Multi-Model Agents on Memory-Constrained GPUs | LLM agents are evolving from a single model calling tools to planners that orchestrate multiple expert models whose capabilities live… | ↗Not briefed |
arxiv arxiv-ai | 28d | ActionOPD: Distilling Decisions, Not Tokens | Distilling capable language-model agents into smaller models requires aligning the student's behavior with the teacher's. This raises… | ↗Not briefed |
arxiv arxiv-ai | 28d | One Monitor Does Not Fit All: Formalism-Agnostic Runtime Verification for Agentic AI🔧 LangGraph | Agentic AI systems combine model decisions, mutable state, tool calls, and delegation to external agents. Monitoring them requires obs… | ↗Not briefed |
arxiv arxiv-ai | 28d | Model vs. User: Whose Values Prevail? Inference-Time Value Alignment in Multi-Turn Conversations | Human values shape how we expect language models to behave. Dominant alignment approaches treat values as global objectives, although… | ↗Not briefed |
rss marktechpost | 28d | 9 Best AI Gateways in 2026: Bifrost, Portkey, LiteLLM, Kong and 5 More Ranked🔧 OpenRouter | Marktechpost ranks the 9 best AI gateways for 2026. We compare Bifrost, Portkey, LiteLLM, Kong, TrueFoundry, Cloudflare, Vercel, OpenR… | ↗Not briefed |
hn hn-ai | 28d | Installing Websites Must Be Easy | — | ↗Not briefed |
arxiv arxiv-ai | 28d | When Does Depth Help? Understanding Layerwise Inference in Tabular Foundation Models | Transformer-based tabular foundation models (TFMs) achieve state-of-the-art performance on small to medium-sized tabular prediction ta… | ↗Not briefed |
arxiv arxiv-ai | 28d | VaryST: Generalizable Spatio-Temporal Forecasting Beyond the Closed-World Setting | Spatio-temporal forecasting is fundamental to many real-world systems, yet deployment environments are inherently varying: temporal di… | ↗Not briefed |
hn hn-ai | 28d | Show HN: Panelize – Ask ChatGPT, Claude, Gemini and DeepSeek Side by Side🔧 ChatGPT | — | ↗Not briefed |
hn hn-ai | 28d | What breaks if I change this? across SQL,dbt and Airflow | — | ↗Not briefed |
arxiv arxiv-ai | 28d | Calibration and Risk-Budgeted Decisions with System One Models | LLM workflows and agents rely on a set of binary decisions like allow/block an input, or discard/retain a retrieved document. When hum… | ↗Not briefed |
bluesky bluesky-ai | 28d | What on earth is the point of networking your vacuum cleaner www.theguardian.com/technology/2... | What on earth is the point of networking your vacuum cleaner www.theguardian.com/technology/2... | ↗Not briefed |