Knowledge catalogue
Derek B. Johnson / CyberScoop: Anthropic says red team tests of Fable 5 found no universal jailbreaks, and it will keep first- and third-party user traffic on Mythos-class models for 30 days — Claude
arXiv:2606.09028v1 Announce Type: cross Abstract: Latent world models are increasingly used for control and goal-conditioned planning, yet assessing whether their learned representations are useful fo
arXiv:2606.08881v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models have demonstrated strong generalization in robotic manipulation, yet existing evaluations are primarily conducted
arXiv:2606.07533v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) effectively integrate text and audio to interpret context in complex interactive dialogues. However, the inte
arXiv:2512.13869v3 Announce Type: replace Abstract: Training object detectors demands extensive, task-specific annotations, yet this requirement becomes impractical in UAV-based human detection due to
arXiv:2606.09142v1 Announce Type: cross Abstract: Egocentric vision offers a first-person view of human perception and decision making, yet its potential for traffic-safety prediction remains underexp
arXiv:2606.08974v1 Announce Type: new Abstract: Large reasoning models (LRMs) have attracted increasing attention for their ability to solve complex mathematical problems by generating extended reason
arXiv:2606.09646v1 Announce Type: cross Abstract: We study whether pretrained video foundation models encode intuitive-physics information in their frozen representations, and how this information var
arXiv:2606.09043v1 Announce Type: new Abstract: Reward models trained from pairwise preferences often exploit superficial shortcut cues rather than learning true response quality. We propose DynaCF, a
arXiv:2606.09803v1 Announce Type: new Abstract: We present extbf{Echo-Memory}, a controlled study of memory mechanisms in action-conditioned world models. These models generate multi-segment videos fr
arXiv:2606.09525v1 Announce Type: cross Abstract: During instruction fine-tuning (IFT), large language models (LLMs) learn to follow instructions by using the provided context to answer a query. While
arXiv:2601.09285v2 Announce Type: replace Abstract: Metal-organic frameworks (MOFs) are porous crystalline materials with broad applications such as carbon capture and drug delivery, yet accurately pr
Fable is a step-change in models, and I hope it changes how you work with Claude. More to come in a series of posts on how it’s reshaped our work, but the TLDR: it’s time to be more ambitious. Claude
arXiv:2606.07703v1 Announce Type: cross Abstract: Long-context prefill remains expensive because full/GQA layers still score the historical sequence, even in hybrid models with local, sparse, linear,
arXiv:2510.06492v2 Announce Type: replace Abstract: Latent world models are a promising approach for learning state representations and dynamics directly from high-dimensional observations, enabling r
arXiv:2412.13858v2 Announce Type: replace Abstract: We investigate diffusion models to solve the Traveling Salesman Problem. Building on the recent DIFUSCO and T2TCO approaches, we propose IDEQ. IDEQ
arXiv:2606.09169v1 Announce Type: new Abstract: In recent years, unified multimodal models (UMMs) have emerged to support both understanding and generation within a single framework. Mastering dynamic
arXiv:2606.09430v1 Announce Type: cross Abstract: Online task-free continual learning (TFCL) requires intelligent agents to sequentially accumulate knowledge from an unbounded, non-stationary data str
arXiv:2606.09131v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) commonly inherit the deep, symmetric Transformer backbone designed for unimodal text modeling, and apply the sa
arXiv:2606.09189v1 Announce Type: cross Abstract: EEG foundation-model releases are usually audited one endpoint at a time: raw-reconstruction, membership inference, identity linkage, or DP-SGD on the
arXiv:2606.08926v1 Announce Type: new Abstract: Knowledge graph completion (KGC) models are commonly evaluated using rank-based metrics such as MRR and Hits@K, despite different users often requiring
arXiv:2606.09740v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models demonstrate strong perfor-1 mance on language-conditioned robotic manipulation within their training dis-2 tribution
arXiv:2606.07590v1 Announce Type: cross Abstract: Pathology foundation models are pretrained on large streams of WSI-derived patches, while supervision during data construction is often slide-level, s
arXiv:2606.08451v1 Announce Type: cross Abstract: Safety-aligned large language models often exhibit sycophancy, which is the tendency to affirm users' opinions regardless of factual accuracy. Althoug
arXiv:2606.09499v1 Announce Type: cross Abstract: World models have recently seen a rapid growth in both their popularity and capability as more data efficient tools for generating robot training data
arXiv:2509.09151v2 Announce Type: replace-cross Abstract: Research in video understanding has advanced rapidly, driven by increasingly diverse datasets and more powerful model architectures. While exi
arXiv:2606.07100v1 Announce Type: new Abstract: Visual-language action (VLA) models enable robots to predict actions directly from observations and language instructions, but their performance depends
arXiv:2605.21706v2 Announce Type: replace Abstract: Safety-aligned language models are trained to refuse harmful requests, yet refusal behavior can be suppressed by steering their internal representat
arXiv:2606.06907v1 Announce Type: cross Abstract: Large audio language models (LALMs) extend large language models with an audio encoder and large-scale audio data. However, the scarcity of high-quali
arXiv:2602.02600v3 Announce Type: replace-cross Abstract: Diffusion language models (DLMs) have recently emerged as a competitive alternative to autoregressive (AR) models, offering parallel decoding,
arXiv:2606.07172v1 Announce Type: cross Abstract: Geospatial understanding is a critical yet underexplored dimension in the development of machine learning systems for tasks such as image geolocation
arXiv:2603.22278v2 Announce Type: replace Abstract: Many multimodal tasks, such as image captioning and visual question answering, require vision-language models (VLMs) to bind objects with their prop
arXiv:2508.06249v3 Announce Type: replace-cross Abstract: Fine-tuning lets practitioners repurpose aligned large language models (LLMs) for new domains, yet recent work reveals emergent misalignment (
arXiv:2606.05378v1 Announce Type: cross Abstract: We test whether a single screen-and-ablate recipe -- identify attention-head circuits by task-pattern selectivity, then verify by causal ablation agai
Also, a lot depends on Chinese labs continuing to ship open weights models. If they stop, the frontier falls further and further behind to those who want to use local/fine-tuned models. I think this i
arXiv:2410.13056v4 Announce Type: replace Abstract: Large Language Models (LLMs) have demonstrated remarkable success across a wide range of language tasks, but their deployment on edge devices remain
arXiv:2606.05486v1 Announce Type: new Abstract: Prompt ambiguity is a common source of failure in large language models, but is difficult to localize because it is a latent property of the prompt, whi
Grace Kay / The Information: Sources say xAI used Claude models for distillation and training, including using personal accounts and the intermediary service Blackbox AI after being cut off — SpaceX's
arXiv:2606.05328v1 Announce Type: cross Abstract: Modern video diffusion models generate increasingly realistic and temporally coherent videos, motivating their use as candidate world simulators. Yet
arXiv:2510.23497v3 Announce Type: replace Abstract: Training vision-language models (VLMs) for complex reasoning remains a challenging task, i.a. due to the scarcity of high-quality image-text reasoni
arXiv:2606.05085v1 Announce Type: cross Abstract: The title of a research paper conveys its primary idea and, occasionally, its conclusions in a clear and concise manner. Choosing an appropriate title
arXiv:2606.04130v1 Announce Type: new Abstract: We introduce CLAW, a fully end-to-end self-supervised framework for learning a world model jointly with continuous latent action representations directl
arXiv:2602.23312v3 Announce Type: replace-cross Abstract: Leader-follower interaction is an important paradigm in human-robot interaction (HRI). Yet, assigning roles in real time remains challenging f
arXiv:2603.09493v2 Announce Type: replace-cross Abstract: The adaptation of large-scale vision-language models (VLMs) to downstream tasks with limited labeled data remains a significant challenge. Whi
arXiv:2501.09876v3 Announce Type: replace-cross Abstract: Generative modeling aims to generate new data samples that resemble a given dataset. When using diffusion models for this task, one of the mai
Introducing two NVIDIA Nemotron models on Together AI: Nemotron 3 Ultra for high-throughput agentic workloads and Nemotron 3.5 ASR for low-latency multilingual speech recognition. AI natives can now b
arXiv:2606.05045v1 Announce Type: cross Abstract: We present in this paper a framework for the identification of control-affine reduced-order models (ROMs). The proposed method utilizes autoencoders (
arXiv:2606.04672v1 Announce Type: cross Abstract: Continuous-time dynamic graphs (CTDGs) provide a richer framework to capture fine-grained temporal patterns in evolving relational data. Long-range in
arXiv:2606.04994v1 Announce Type: new Abstract: Accurate computational prediction of T cell receptor (TCR) antigen specificity would transform the study of T cell biology and enable scalable immune en
Pulled the trigger today and switched 100% of Lindy traffic to DeepSeek v4, churning from Anthropic models. Saves us millions of $ and we're actually seeing an *increase* in performance on many core u
That's a badass title, and it's true! Every day, it gets harder and harder to create tests that AI models can't beat. Reality is humanity's real last exam. Andon Labs' Real-World AI Evals: Claude call
arXiv:2602.19101v2 Announce Type: replace-cross Abstract: Value alignment of Large Language Models (LLMs) requires us to empirically measure these models' actual, acquired representation of value. Amo
Microsoft Foundry Managed Compute is a new GPU platform-as-a-service for hosting open-source and custom AI models behind the same endpoint, SDKs, and bill as frontier models. The post Announcing Found
arXiv:2606.03347v1 Announce Type: cross Abstract: Score-based diffusion models have emerged as prominent deep generative models; however, their application to tabular data remains challenging because
arXiv:2606.03624v1 Announce Type: new Abstract: Large Reasoning Models (LRMs) have demonstrated impressive capabilities in many tasks, yet they struggle with reliably following multiple instructions,
arXiv:2606.03157v1 Announce Type: new Abstract: Large language models (LLMs) have been widely adopted in healthcare, yet they still encounter significant challenges in complex clinical decision-making
arXiv:2606.03780v1 Announce Type: new Abstract: Causal tracing of factual recall has been studied predominantly in dense transformer language models, where interventions localize information flow to l
Frontier models are powerful advisors. On @harvey's Legal Agent Benchmark, a GLM 5.1 worker using Claude Opus 4.7 as a sparse advisor reached 18/100 all-pass versus 14/100 for Opus alone, at 39% of th
arXiv:2606.03437v1 Announce Type: new Abstract: Prior work has shown that instruction-tuned large language models (LLMs) are less well calibrated than their base pre-trained counterparts. However, lit
arXiv:2603.20508v2 Announce Type: replace-cross Abstract: Reasoning language models (RLMs) and the intermediate chains of thought they emit play an increasingly central role in multi-agent setups such