In-Context Credit Assignment via the Core
arXiv:2605.06920v1 Announce Type: cross Abstract: We propose incentive-aligned mechanisms for in-context credit assignment: the task of assigning credit for AI-generated content (e.g. code, news artic
Knowledge catalogue
arXiv:2605.06920v1 Announce Type: cross Abstract: We propose incentive-aligned mechanisms for in-context credit assignment: the task of assigning credit for AI-generated content (e.g. code, news artic
🚨 In his testimony just now, at the Musk-OpenAI trial, Satya Nadella came off as shrewd, calm, and (mostly) honest, an impressive leader – but also selectively blind. How? He seemed unable to believe
In modern ML accelerators, FLOPS have absolutely exploded. Often though, the bottleneck is not FLOPS but memory bandwidth. Similarly, model intelligence has exploded, causing the bottleneck to be huma
arXiv:2605.07010v1 Announce Type: new Abstract: Identifying vulnerable transmission lines in power grids before a cascading failure occurs is challenging: existing methods can learn inter-line failure
arXiv:2605.07433v1 Announce Type: cross Abstract: Qualitative models provide crucial instruments for modelling complex biological systems. While advances in automated reasoning and symbolic encodings
arXiv:2605.07456v1 Announce Type: new Abstract: Inference-time controllable generation is essential for real-world applications of unconditional diffusion models. However, most existing techniques foc
arXiv:2605.07631v1 Announce Type: new Abstract: Causal probing methods aim to test and control how internal representations influence the behavior of generative models. In causal probing, an intervent
arXiv:2605.07099v1 Announce Type: new Abstract: Cross-view geo-localization (CVGL) is fundamental for precise localization and navigation in GPS-denied environments, aiming to match ground or UAV imag
arXiv:2605.06710v1 Announce Type: cross Abstract: Information theory plays a central role in establishing fundamental limits on what any learning or estimation algorithm can -- and cannot -- achieve,
The changes may be less perceptible than in smartphones, tablets, or wearables, but chargers have also been quietly reinvented over the last decade. At one time a bulky mix of tangled cables and conne
arXiv:2605.07930v1 Announce Type: cross Abstract: Differential privacy (DP) is widely employed in machine learning to protect confidential or sensitive training data from being revealed. As data owner
arXiv:2605.07402v1 Announce Type: new Abstract: Human insertion aims to naturally place specific individuals into a target background. Although existing image editing models may have such ability, the
arXiv:2605.07284v1 Announce Type: new Abstract: Recent interpretability work has identified model-internal handles on post-trained behavior, including refusal directions, assistant/persona axes, and s
arXiv:2605.07057v1 Announce Type: new Abstract: Online reinforcement learning (RL) relies on the Markov property for guaranteed performance, but real-world applications often lack well-defined states
arXiv:2605.07733v1 Announce Type: cross Abstract: Accurate truck-to-shipment matching using GPS data is foundational for full truckload supply chain visibility, enabling real-time tracking and accurat
arXiv:2605.07613v1 Announce Type: new Abstract: Conversational news recommendation requires grounding each suggestion in a rapidly evolving article corpus while addressing implicit user intents that l
arXiv:2605.06832v1 Announce Type: cross Abstract: Accurately understanding the intent behind speech, conversation, and writing is crucial to the development of helpful Large Language Model (LLM) assis
arXiv:2605.07037v1 Announce Type: new Abstract: Robot systems for teleoperation commonly use a spring-like force pulling the follower robot towards the leader's position to track their movements. With
Jensen-Shannon divergence is a method of measuring the similarity between two probability distributions. It is based on the Kullback-Leibler divergence, with the notable difference that it is symmetri
arXiv:2605.07768v1 Announce Type: cross Abstract: We investigate interactive trajectory planning subject to uncertainty in the decisions of surrounding agents. To control the ego-agent, we aim to firs
arXiv:2603.01586v3 Announce Type: replace Abstract: Emerging unified editing models have demonstrated strong capabilities in general object editing tasks. However, it remains a significant challenge t
arXiv:2605.07510v1 Announce Type: cross Abstract: Existing benchmarks for multimodal agentic search evaluate multimodal search and visual browsing, but visual evidence is either confined to the input
arXiv:2605.08007v1 Announce Type: new Abstract: Susceptibilities are a technique for neural network interpretability that studies the response of posterior expectation values of observables to perturb
arXiv:2603.03096v2 Announce Type: replace-cross Abstract: How do speech models trained through self-supervised learning structure their representations? Previous studies have looked at how information
arXiv:2510.08638v3 Announce Type: replace-cross Abstract: DINOv2 is routinely deployed to recognize objects, scenes, and actions; yet the nature of what it perceives remains unknown. As a working base
Today, we're excited to announce the general availability of Claude Platform on AWS. Claude Platform on AWS is a new service that gives customers direct access to Anthropic's native Claude Platform ex
Introducing Daybreak: frontier AI for cyber defenders. Daybreak brings together the most capable OpenAI models, Codex, and our security partners to accelerate cyber defense and continuously secure sof
Introducing the OpenAI Deployment Company, which will help businesses maximally succeed with their deployments of AI. Starting with 150 Forward Deployed Engineers and Deployment Specialists, and $4 bi
arXiv:2509.21172v2 Announce Type: replace Abstract: Inverse reinforcement learning (IRL) aims to infer rewards from observed behavior, but rewards are not identified from the policy alone: many reward
arXiv:2510.01569v3 Announce Type: replace Abstract: We present InvThink, a training and prompting framework that requires the model to enumerate, analyze, and constrain potential failures before gener
iOS 26.5 introduces end-to-end encryption for RCS messages exchanged between iPhone and Android users, with encryption requiring both participants to have a carrier that supports the feature. The upda
arXiv:2512.23032v2 Announce Type: replace-cross Abstract: Recent work, using the Biasing Features metric, labels a CoT as unfaithful if it omits a prompt-injected hint that affected the prediction. We
arXiv:2605.07622v1 Announce Type: new Abstract: Gender bias in large language models has primarily been investigated for English, while languages with grammatical or morphological gender remain compar
arXiv:2605.07514v1 Announce Type: cross Abstract: World Action Models (WAMs) enable decision-making through imagined rollouts by predicting future observations and actions. However, the reliability of
arXiv:2510.22944v2 Announce Type: replace-cross Abstract: Large language models (LLMs) have become indispensable for automated code generation, yet the quality and security of their outputs remain a c
CTech: Israeli startup Frame Security, which protects organizations from AI-powered social engineering attacks, emerges from stealth with $50M led by Index and others — The Israeli firm aims to modern
It is insane that we allow big tech to race toward superintelligence without any oversight. This is a grave national security risk. Industry will not regulate itself. Get trained with Torchbearer to a
It is one reason why I think the push for smaller, local models is more complicated than people think. If you want good answers, especially good answers to unexpected problems, frontier models will ge
arXiv:2605.07972v1 Announce Type: cross Abstract: Neural posterior estimation has emerged as a powerful tool for amortized inference, with growing adoption across scientific and applied domains. In ma
'It would be a mistake for any country to try to slow down open source. A country that leads open source is a country that can lead AI in general' @ClementDelangue, co-founder & CEO @HuggingFace Watch
HTML is being used beyond traditional web development as a versatile tool for planning, specifications, exploratory work, code review documentation, and report generation. This approach leverages HTML
My guest today is longtime friend of the show Joanna Stern. You all know Joanna: she is the former senior personal technology columnist for The Wall Street Journal, a former Decoder guest host, one of
arXiv:2605.06883v1 Announce Type: cross Abstract: The Maximum Mean Discrepancy (MMD) is a cornerstone statistic for nonparametric two-sample testing, but its test power is dictated entirely by the cho
Kimi K2.6 was a hit during the launch promotion - so it's staying free for Pro, Teams, and Max users for another month. Enjoy! Kimi K2.6 is now available in Windsurf! Available for free for the next 2
Kimi K2.6 was a hit during the launch promotion - so it's staying free for Pro, Teams, and Max users for another month. Enjoy! Kimi K2.6 is now available in Windsurf! Available for free for the next 2
arXiv:2605.07865v1 Announce Type: cross Abstract: On-Policy Distillation (OPD) has emerged as a dominant post-training paradigm for large language models, especially for reasoning domains. However, OP
arXiv:2605.05957v2 Announce Type: replace Abstract: LLMs reliably correct false claims when presented in isolation, yet when the same claims are embedded in task-oriented requests, they often comply r
arXiv:2605.06859v1 Announce Type: cross Abstract: Vision foundation models are increasingly moving beyond 2D to volumetric domains such as 3D medical imaging, where unified pretraining across differen
arXiv:2602.02832v3 Announce Type: replace Abstract: Forecasting physical systems over long horizons from irregularly sampled observations demands models that are stable, computationally efficient, and
arXiv:2605.06955v1 Announce Type: cross Abstract: Denoising score matching (DSM) provides a way to learn data distributions by training a neural network to recover the score function, defined as the g
arXiv:2605.07740v1 Announce Type: new Abstract: Mining operations are of utmost importance to the economy of some nations. However, such operations result in land-use change, very high energy consumpt
LangSmith settling the CLI vs MCP debate by simply offering both LangSmith is now a remote MCP server. Point any MCP-capable client at it, sign in with your LangSmith account, and your assistant can r
arXiv:2605.07517v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) enhances the factual grounding of Large Language Models by conditioning their outputs on external documents. Howe
arXiv:2512.15840v2 Announce Type: replace-cross Abstract: General-purpose robots require decision-making models that generalize across diverse tasks and environments. Recent works build robot foundati
Last week, the United States refused to participate in the UN’s review of the Global Compact on Migration. The United States objects to the Global Compact on Migration and UN efforts to facilitate rep
arXiv:2605.07304v1 Announce Type: new Abstract: Bandit algorithms solve diverse sequential decision-making problems, but are often too sample-inefficient for from-scratch personalization. To substanti
arXiv:2602.01166v2 Announce Type: replace Abstract: Vision-Language-Action (VLA) models benefit from chain-of-thought (CoT) reasoning, but existing approaches incur high inference overhead and rely on
arXiv:2602.09034v2 Announce Type: replace-cross Abstract: Neuroimaging does not observe causal variables directly: hemodynamics and volume conduction distort signals so that statistical dependence nee
arXiv:2605.07315v1 Announce Type: new Abstract: Chain-of-thought (CoT) reasoning improves large language models (LLMs) on difficult tasks, but it also makes inference expensive because every intermedi
Launch rehearsal complete. During a flight-like countdown, more than 5,000 metric tonnes (11+ million pounds) of propellant were loaded on the fully stacked Starship and Super Heavy V3 vehicles for th