Knowledge catalogue
arXiv:2607.23322v1 Announce Type: new Abstract: Instruction tuning has become the standard method for adapting large language models to follow human intent, yet existing instruction datasets are domin
K3 is a language model that entered the top five most‑liked models on Hugging Face merely 24 hours after its release. The achievement surprised many, placing it ahead of prominent models such as Llama
arXiv:2607.24306v1 Announce Type: new Abstract: Diffusion language models enable flexible arbitrary-order generation, but existing sampling methods are mostly designed for early masked diffusion model
arXiv:2607.23915v1 Announce Type: cross Abstract: We examine how prompt tone affects both accuracy of the LLM answers and inference cost as reflected in output-token consumption. Experiments were perf
arXiv:2601.05148v2 Announce Type: replace Abstract: Pathology foundation models substantially advanced the possibilities in computational pathology --- yet tradeoffs in terms of performance, robustnes
We've joined the alliance. Open-weight models will ensure that we live in a safer digital world, and that America does not get left behind Attackers have frontier AI. Defenders need a frontier AI ecos
Announcing the Claude Code-compatible interface for our new Fugu-Ultra v1.1! 🐡 Put a dynamically coordinated team of frontier models to work inside the coding workflow you already know. Instead of rel
arXiv:2603.11838v2 Announce Type: replace Abstract: Large language models pretrained on internet-scale data risk lookahead bias in forecasting tasks, as they may have already seen the true outcome dur
For my first post, I’m sharing a letter @NVIDIA signed on why open models matter. AI will transform every industry, power every company, and be built by every country. Open models strengthen safety an
arXiv:2607.18218v2 Announce Type: replace-cross Abstract: Foundation models have emerged as a driving force in computational pathology, with the potential to transform cancer diagnosis, prognosis, and
Open models for the win! For my first post, I’m sharing a letter @NVIDIA signed on why open models matter. AI will transform every industry, power every company, and be built by every country. Open mo
The Washington Post processed 1.79B input tokens per month through Together AI, running open models like Llama and Mistral in production with predictable costs and full control over the model stack. T
arXiv:2607.21090v1 Announce Type: cross Abstract: We propose a Reinforcement Learning (RL) method to directly optimize the faithfulness of self-explanations - the extent to which a model's generated r
Arize and Fireworks benchmarked 10 AI models across 2,400 agent runs. Learn why cost per successful task beats token price for model evaluation and routing. The post Cost per successful task: Benchmar
Huge launch from @tryramp. Different steps in an agent workflow can use different models. This can help reduce costs significantly without sacrificing performance. Model routing will become a core par
Who’s Afraid of Chinese Models? Interesting proposal from Ben Thompson that both addresses the hypocrisy of labs outlawing distillation against their models despite training on unlicensed data, and co
arXiv:2607.12884v1 Announce Type: new Abstract: Patients seeking medical information often ask questions that embed incorrect assumptions or misconceptions. In such cases, safe medical communication r
arXiv:2607.13017v1 Announce Type: cross Abstract: World Action Models (WAMs) are able to leverage pretrained video generators for both world modeling and action prediction. However, directly leveragin
arXiv:2607.11889v1 Announce Type: cross Abstract: Large language models trained on unrestricted internet corpora inevitably embed information from the future, introducing lookahead bias that compromis
U.S. open‑source AI models are rapidly gaining popularity, with NVIDIA’s newest model, **Nemotron Ultra**, becoming a prominent entry on the Ollama platform. On Ollama, Nemotron Ultra is quickly scali
Another big reason to use combination of frontier models. Chain-of-thought monitoring is treated as a reliable safety layer for agents. This DeepMind-affiliated study shows the layer can be argued out
arXiv:2510.16165v2 Announce Type: replace Abstract: A key question in benchmarking generative crystal reconstruction models is how the amount and type of crystallographic information provided to a gen
arXiv:2506.07468v4 Announce Type: replace-cross Abstract: Conventional large language model (LLM) safety alignment relies on a reactive, disjoint loop: attackers exploit a static model, then defenders
arXiv:2503.06643v2 Announce Type: replace-cross Abstract: In this paper, we tackle a critical challenge in model evaluation: how to keep code benchmarks useful when models might have already seen them
arXiv:2607.04546v1 Announce Type: cross Abstract: Action-conditioned world models allow robots to predict the future consequences of candidate actions without additional physical interaction, supporti
arXiv:2607.03953v1 Announce Type: cross Abstract: This study independently replicates and extends the Natural Language Tools (NLT) framework of Johnson et al.~(2025), which questions the use of struct
introducing tinyrouter i reverse engineered the routing architecture behind Skana AI's Fugu and built replication for open frontier models. it's a tiny ~10K parameter LLM router that learns which mode
arXiv:2607.01436v1 Announce Type: new Abstract: Diffusion language models, which generate text by denoising a token canvas bidirectionally instead of emitting tokens left to right, have become competi
arXiv:2511.05536v2 Announce Type: replace-cross Abstract: Earth s gravity fundamentally shapes human behaviour. The brain encodes this force as an internal model of gravity, enabling the prediction an
arXiv:2606.16517v2 Announce Type: replace Abstract: Scientific reasoning models for biology combine language models with foundation models trained on multimodal biological data, including DNA, RNA, an
arXiv:2511.08307v2 Announce Type: replace-cross Abstract: Generative models, such as large language models or text-to-image diffusion models, can generate relevant responses to user-given queries. Res
arXiv:2606.30140v1 Announce Type: cross Abstract: Recent breakthroughs in foundation models and Large Language Models (LLMs) have introduced new opportunities for studying and decoding genomic sequenc
arXiv:2606.28962v1 Announce Type: cross Abstract: Model quantization is essential for the efficient deployment of Large Language Models (LLMs), but introduces a critical vulnerability: Quantization-Co
arXiv:2606.29646v1 Announce Type: cross Abstract: Sleeper agents are the canonical model organism of deception: models trained to behave normally but to emit an unsafe behaviour on a specific trigger.
arXiv:2606.30062v1 Announce Type: cross Abstract: While large language models have been dominating the research landscape recently, small language models remain highly relevant across various domains;
arXiv:2606.29247v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models represent a promising direction for embodied intelligence in surgical robotics. Despite the prevalence of VLA benchm
arXiv:2509.19671v3 Announce Type: replace Abstract: Public datasets of Chest X-Rays (CXRs) have long been a popular benchmark for developing machine learning (ML) computer vision models in healthcare.
I put together a new article on setting up local coding agents with open-weight models. Everything runs 100% locally. I thought it might be useful putting this together because many people asked me ab
arXiv:2511.12100v2 Announce Type: replace Abstract: In current visual model training, models often rely on only limited sufficient causes for their predictions, which makes them sensitive to distribut
arXiv:2606.24083v1 Announce Type: cross Abstract: 'Talk short. Drop grammar. Save token.' This caveman style is widely promoted as a way to cut inference cost, but whether it actually saves anything d
arXiv:2606.22606v1 Announce Type: cross Abstract: Large language models (LLMs) achieve strong relation extraction (RE), but their computational demands and reliance on proprietary APIs limit deploymen
AI2 released TMax 27B, a 27 billion parameter terminal agent model available on Hugging Face that achieves 42.7% performance on Terminal Bench 2.0, matching the capabilities of much larger models desp
arXiv:2606.11386v1 Announce Type: cross Abstract: Full-duplex spoken language models (FD-SLMs) enable seamless speech interaction by allowing models to listen and speak simultaneously, yet the interna
arXiv:2603.22017v2 Announce Type: replace Abstract: This work presents a collection of multi-modal domain adapted large language models built upon the instruction tuned variants of open weight models
arXiv:2606.11105v1 Announce Type: cross Abstract: Hallucinations, where language models (LMs) generate factually ungrounded responses, pose serious risks, as users tend to blindly rely on them. This i
arXiv:2606.08948v1 Announce Type: cross Abstract: Comprehensive estimation of dietary micronutrients from food images could improve clinical nutrition care, but training such models requires large mul
arXiv:2606.04811v1 Announce Type: new Abstract: Video generation models have made impressive strides in synthesizing visually compelling content, yet their outputs remain confined to the virtual domai
arXiv:2606.05015v1 Announce Type: new Abstract: World models, learned generative models that predict how an environment evolves, have become a promising tool for sample-efficient robot learning. Yet h
arXiv:2601.06196v3 Announce Type: replace-cross Abstract: Large language models (LLMs) frequently generate factually incorrect or unsupported content, commonly referred to as hallucinations. Prior wor
arXiv:2606.02802v1 Announce Type: new Abstract: Large language models (LLMs) exhibit strong natural-language reasoning abilities for clinical decision support, but struggle to effectively model struct
arXiv:2606.02834v1 Announce Type: cross Abstract: Malware analysis starts with the raw bytes of an executable program, and tools to 'lift' these to higher-level representations, such as assembly, are
arXiv:2606.02535v1 Announce Type: new Abstract: Large-scale generative models have demonstrated remarkable capabilities across image generation and editing tasks. However, their performance in low-lev
arXiv:2606.00566v1 Announce Type: cross Abstract: As language models take on agentic roles that span calling external APIs, reading tool outputs, and acting on instructions embedded in third-party con
arXiv:2606.00053v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models follow a data-driven paradigm and are constrained by the coverage of training data, making them prone to failure on
arXiv:2602.09309v2 Announce Type: replace-cross Abstract: Every generative model for crystalline materials harbors a critical structure size beyond which its outputs become unreliable; we call this th
arXiv:2605.30184v1 Announce Type: new Abstract: While AI weather models excel at short-to-medium range forecasts (up to 15 days), they frequently suffer from ill-defined 'instabilities' when rolled ou
arXiv:2605.28848v1 Announce Type: cross Abstract: Deployed language models are evaluated in a non-stationary environment: model versions, retrieval layers, safety systems, and real-world inputs all ch
arXiv:2605.30018v1 Announce Type: new Abstract: Large language models (LLMs) frequently achieve impressive scores on standardized benchmarks, yet accuracy alone offers a limited view of their capabili
Production traffic from frontier models is a golden data asset. If you can efficiently mine the traces, filter for quality, and fine-tune smaller models on them, you get specialized performance at a f
arXiv:2601.21207v3 Announce Type: replace-cross Abstract: Combinatorial and topological structures, such as graphs, simplicial complexes, and cell complexes, form the foundation of geometric and topol