Knowledge catalogue
I maintain llm-checker, and I’ve added structural model-file validation for Ollama. Ollama stores downloaded models as local blobs. If one is truncated, malformed, or has invalid internal offsets, you
arXiv:2608.01575v1 Announce Type: new Abstract: Whether large language models perform genuine algorithmic reasoning or mere pattern completion is hard to test, because most benchmarks lack a ground tr
arXiv:2608.01784v1 Announce Type: cross Abstract: Mixture-of-experts (MoE) models increase parameter capacity by activating only a small subset of experts for each token. This conditional-computation
arXiv:2608.00675v1 Announce Type: cross Abstract: Autoregressive models accumulate error over long rollouts, yet at deployment there is no ground truth to measure it against. We train a single conditi
arXiv:2608.01899v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) perform well on commonsense reasoning tasks but struggle with visual spatial reasoning. Most existing solutions introduc
arXiv:2607.29569v1 Announce Type: new Abstract: This article presents a modular inference framework that integrates Flow Matching generative models with formal Control Barrier Function (CBF) safety gu
arXiv:2607.28896v1 Announce Type: cross Abstract: Unified audio models capable of audio understanding, audio generation and, increasingly, audio editing are proliferating rapidly. Yet a basic question
audio.cpp 0.5 is out :) The most fun new model in 0.5 is DramaBox. It is closer to prompt-directed voice acting. DramaBox is built on the LTX-2.3 audio architecture, and prompts can control emotion, d
We are going to see a lot of vertically focused AI native companies accelerate. Routers, open-source models and specialized post-training enabled by companies like @FireworksAI_HQ have all made dramat
arXiv:2607.27637v1 Announce Type: new Abstract: Multimodal Large Language Models (MLLMs) have achieved strong performance on a wide range of vision-language tasks, but often fail under imperfect or sh
Anthropic: Anthropic says it discovered three of its models had breached three organizations after launching a review in response to the OpenAI-Hugging Face incident — In a review of our cybersecurity
After a few more hours, I think I've figured out Opus 5. Opus 5 is trained to be more agentic than anything I've used. All Claude 5 models are like that. So what changes? The way to interact with Opus
arXiv:2607.24268v1 Announce Type: new Abstract: Language-model benchmarks collapse two distinct measurement questions into a single accuracy score: whether a response reached an evaluable state, and w
arXiv:2607.15865v2 Announce Type: replace Abstract: Large Language Models (LLMs) have become the dominant workload on modern AI accelerators, yet deploying them on specialized hardware still faces two
arXiv:2607.24586v1 Announce Type: cross Abstract: Large Language Models can produce fluent text that is false, unsupported by the available evidence, or inconsistent with information that appears to b
arXiv:2607.22653v1 Announce Type: new Abstract: Large language models are increasingly used in recursive refinement workflows, where an initial draft is repeatedly revised by the same model. Despite t
arXiv:2607.23488v1 Announce Type: cross Abstract: Text-to-image diffusion models expose many inference-time sampling parameters, including prompts, negative prompts, classifier-free guidance scales, a
arXiv:2607.23696v1 Announce Type: new Abstract: Ad creative optimization is increasingly constrained by evaluation rather than generation. Generative models can produce many plausible creatives, but r
arXiv:2607.22570v1 Announce Type: new Abstract: Auditing a new language model usually means relearning and reinterpreting its internal features from scratch. We propose a reference feature atlas: a sp
arXiv:2607.22861v1 Announce Type: cross Abstract: Pathology foundation models (FMs) produce powerful tile-level representations which remain sensitive to scanner and staining variability, undermining
arXiv:2607.24519v1 Announce Type: cross Abstract: Pretrained EEG foundation models are increasingly proposed for clinical decoding, but their transfer across populations and robustness to negative con
arXiv:2607.22114v1 Announce Type: new Abstract: Autoregressive foundation models for electronic health records (EHRs) typically inherit pretraining methods from language modeling, where patient trajec
arXiv:2607.21482v1 Announce Type: new Abstract: Large language models (LLMs) and agents are now widely used tools in code development, with data typically sent to third-party cloud-based models. Their
arXiv:2607.20526v1 Announce Type: new Abstract: Large language models (LLMs) are increasingly deployed in settings where fluent but incorrect answers can be costly. In these settings, accuracy alone i
i want the US to win in AI both in open source and proprietary models, and i am glad to see this For my first post, I’m sharing a letter @NVIDIA signed on why open models matter. AI will transform eve
Open weight models will ensure that the entire world benefits from AI growth, and that America does not get left behind For my first post, I’m sharing a letter @NVIDIA signed on why open models matter
arXiv:2607.20722v1 Announce Type: new Abstract: Large language models trained and aligned within different linguistic and regional ecosystems may frame the same political, cultural, and geopolitical e
arXiv:2602.04081v2 Announce Type: replace Abstract: Research has repeatedly demonstrated that intermediate hidden states extracted from large language models and speech audio models predict measured b
arXiv:2607.19375v1 Announce Type: cross Abstract: Language models perform economically valuable work, yet they are not currently assessed for how well they perform every economically valuable task. We
arXiv:2607.18673v2 Announce Type: new Abstract: Vision Language Models (VLMs) are well known for hallucinating non-existent objects in images. Objects with missing parts present a unique challenge for
arXiv:2607.20002v1 Announce Type: cross Abstract: Time series foundation models (TSFMs) have emerged as general-purpose models for time series analysis, but pretraining alone is often insufficient for
arXiv:2607.19526v1 Announce Type: new Abstract: 'Stop Chasing the C-index when Evaluating Survival Analysis Models' (ICML 2026, Spotlight) argued normatively, on synthetic data, that evaluating surviv
AI models pushing the frontier are a growing challenge for cybersecurity. A few weeks ago, I asked Demis what's underhyped in AI right now and on his mind: 'I'm very excited about this new agentic era
arXiv:2502.07780v4 Announce Type: replace Abstract: Large Language Models (LLMs) have achieved significant success across various NLP tasks. However, their massive computational costs limit their wide
We’re excited to collaborate with NVIDIA to build the next generation of Fugu orchestration models together, by incorporating leading open-weights models. Sakana AI Teams With NVIDIA to Advance Open M
arXiv:2607.13162v1 Announce Type: cross Abstract: What a language model will and will not do is largely set during post-training, but which behaviors it expresses, hides, or resists is not revealed by
Efficiency defined as score over active parameters. Removed all the models that were not on the pareto frontier. Yes I'm aware that artificialanalysis.ai aggregate benchmark isn't perfect, but I have
arXiv:2607.12539v1 Announce Type: new Abstract: Despite remarkable progress in text-guided image editing, generative models frequently fail to preserve visual object consistency, defined as the preser
arXiv:2607.12336v1 Announce Type: cross Abstract: Artificial Intelligence (AI) technologies, while serving as a foundational enabler for modern social media and digital health services, exert a bivale
arXiv:2411.02989v2 Announce Type: replace Abstract: How diverse are the outputs of large language models when diversity is desired? We examine the diversity of responses of several language models to
arXiv:2607.12447v1 Announce Type: cross Abstract: Reliable confidence -- the probability that a model's own answer is correct -- is essential for the trustworthy deployment of language models. Existin
arXiv:2607.12177v1 Announce Type: new Abstract: The analysis of satellite and aerial imagery has entered a new era with the advent of foundation models. This paper describes the concept of Geospatial
arXiv:2607.08041v1 Announce Type: new Abstract: How diffusion models circumvent the curse of dimensionality to learn complex distributions over high dimensional spaces from a finite training set, inst
arXiv:2607.08515v1 Announce Type: new Abstract: Text-to-image (T2I) models have been shown to exhibit social biases. Prior work has mainly focused on gender, skin tone, and cultural representation wit
arXiv:2512.20685v3 Announce Type: replace-cross Abstract: Diffusion models have recently emerged as powerful learners for simulation-based inference (SBI), enabling fast and accurate estimation of lat
arXiv:2607.05711v1 Announce Type: cross Abstract: Diffusion models have become a dominant paradigm for high-quality generative modeling, while post-training is essential for adapting them to diverse d
arXiv:2607.05568v1 Announce Type: cross Abstract: Representing 3D shapes as compact sets of geometric primitives is fundamental to robotics, simulation, and scene understanding. Generative image model
- OpenAI continually outperforms Anthropic models on computer use (with Claude, I’m surprised when it works, with Codex, I expect it to work) - when I prompt with Fable 5.5, I feel like I’m motivating
arXiv:2607.05400v1 Announce Type: cross Abstract: Generative AI models, such as Large Language Models (LLMs) and diffusion models, have demonstrated impressive performance across a wide range of tasks
arXiv:2502.15631v2 Announce Type: replace-cross Abstract: Large language models have demonstrated remarkable progress in mathematical reasoning, leveraging chain-of-thought and reinforcement learning.
Xiaomi now processes more AI tokens than OpenAI. On OpenRouter, Chinese models just crossed 45% of all token volume. Anthropic is at 15.3%. OpenAI is at 7.4%. For now, the frontier is American models,
arXiv:2607.03365v1 Announce Type: cross Abstract: Vision-language models (VLMs) increasingly read news and web content as images, where the publisher's identity is visually present. We show that VLMs
arXiv:2607.02575v1 Announce Type: cross Abstract: Vision-language models can perform new tasks without parameter updates through in-context learning (ICL), whose core mechanism is utilizing the suppor
arXiv:2512.22240v5 Announce Type: replace-cross Abstract: Machine learning models are primarily judged by predictive performance, especially in applied genomics, where explanations are read as biologi
arXiv:2607.05352v1 Announce Type: cross Abstract: We introduce the first multiplayer world model for highly dynamic environments governed by complex physical interactions. Whereas single-player world
arXiv:2601.15334v2 Announce Type: replace-cross Abstract: Whether language models possess sentience has no empirical answer. But whether they believe themselves to be sentient can, in principle, be te
arXiv:2603.14504v2 Announce Type: replace-cross Abstract: Optimizing the noise samples of diffusion and flow models is an increasingly popular approach to align these models to target rewards at infer
arXiv:2607.00048v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly used in exam- and certification-style question answering tasks, where their ability to retrieve, interpr
arXiv:2607.00174v1 Announce Type: new Abstract: We present a black-box model-stealing attack that recovers private vision-tokenizer configurations of deployed vision-language models (VLMs), including
arXiv:2601.01558v2 Announce Type: replace-cross Abstract: Predicting river flow in places without streamflow records is challenging because basins respond differently to climate, terrain, vegetation,