Knowledge catalogue
Ethan Mollick reported that requesting Claude Opus 4.7 to write sestinas—a complex poetic form with strict structural requirements—frequently triggers the model's safety guardrails, suggesting the AI
arXiv:2604.13268v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) have demonstrated strong cross-modal reasoning capabilities, yet their potential for vision-only tasks remain
Replit shared a YouTube video detailing their company's journey and history of development. The video likely covers key milestones, founding story, product evolution, and the team's vision for their c
arXiv:2602.20913v2 Announce Type: replace Abstract: This paper addresses the critical and underexplored challenge of long video understanding with low computational budgets. We propose LongVideo-R1, a
arXiv:2510.07019v3 Announce Type: replace Abstract: Transformers excel at sequence modeling but face quadratic complexity, while linear attention offers improved efficiency but often compromises recal
arXiv:2604.13919v1 Announce Type: cross Abstract: Computational fluid dynamics (CFD) has become an essential tool for predicting fire behavior, yet maintaining both efficiency and accuracy remains cha
arXiv:2604.14073v1 Announce Type: new Abstract: Resolving and rewriting references is fundamental in programming languages. Motivated by a real-world decompilation task, we abstract reference rewritin
🎬 Ollama Gemma Day Recap: SGLang at the Ollama Gemma 4 Party in Palo Alto 🍾 Last night, @ollama hosted a packed Gemma Day at the Palo Alto office alongside the @GoogleDeepMind Gemma team. SGLang was i
arXiv:2408.10610v5 Announce Type: replace Abstract: We propose an L^2 norm for stationary Autoregressive Moving Average (ARMA) models. We look at ARMA models within the Hilbert space of the past with
Bloomberg: Sources: Upscale AI, which builds AI networking infrastructure, is in talks to raise 180M to 200M at a 2B valuation, its third funding round in seven months — Artificial intelligence startu
arXiv:2604.13847v1 Announce Type: new Abstract: While sparse attention mitigates the computational bottleneck of long-context LLM training, its distributed training process exhibits extreme heterogene
arXiv:2604.01034v2 Announce Type: replace Abstract: We propose a Stein variational distributionally robust controller for nonlinear dynamical systems with latent parametric uncertainty. The method is
this makes sense all else held equal, concentrated fund = less risky bets we've found that 36 companies is the max number of companies we can fit into a fund while claiming that a single company at 1B
arXiv:2604.13552v1 Announce Type: new Abstract: Large language models (LLMs) demonstrate strong reasoning capabilities, but their performance often degrades under distribution shift. Existing test-tim
arXiv:2604.13197v1 Announce Type: new Abstract: Process reward models (PRMs) provide fine-grained reward signals along the reasoning process, but training reliable PRMs often requires step annotations
arXiv:2506.01247v2 Announce Type: replace Abstract: Steering vision foundation models at test time, without updating foundation-model weights or using labeled target data, is a desirable yet challengi
We're proud to be on the @Forbes AI 50 list again! It reflects our focus on building secure, sovereign AI that helps enterprises and governments put their data to work on their terms. Learn more: http
4/5 We upgraded our original 3-line “be correct” prompt → a much more detailed prompt that enforced a hierarchy of constraints for correctness, regression safety, and minimality. Basically, get the LL
arXiv:2603.10559v2 Announce Type: replace Abstract: This paper studies cross-market return predictability through a machine learning framework that preserves economic structure. Exploiting the non-ove
Bloomberg: A TTP analysis finds dozens of nudify apps in Apple and Google app stores via search, despite company policy prohibiting them; the apps earned $122M+ in revenue — Apple Inc. and Google have
Tim Fernholz / TechCrunch: AI cloud infrastructure company Parasail raised a 32M Series A led by Touring Capital and Kindred Ventures, bringing its total funding to 42M — “Give me tokens. Just give me
Six-month-old cybersecurity startup Artemis Global Technologies Inc. today disclosed that it has raised 70 million in funding. The capital arrived in two tranches. Felicis led the largest of the two t
arXiv:2603.27552v2 Announce Type: replace Abstract: Multimodal federated learning (FL) is essential for real-world applications such as autonomous systems and healthcare, where data is distributed acr
Cloudflare's Browser Run is a service that provides AI agents with the ability to control and interact with a real web browser, enabling them to perform tasks such as web scraping, form submission, na
arXiv:2603.28325v3 Announce Type: replace-cross Abstract: Biomedical knowledge resources often either preserve evidence as unstructured text or compress it into flat triples that omit study design, pr
Welcome to the first Cloud CISO Perspectives for April 2026. Today, Thiébaut Meyer and Lia Wertheimer from Google Cloud’s Office of the CISO share Thiébaut’s conversation with Matt Rowe, chief securit
arXiv:2604.11867v1 Announce Type: cross Abstract: We set out to train behavioral dispositions (self-verification, uncertainty acknowledgment, feedback integration) into small language models (0.6B to
arXiv:2510.09536v2 Announce Type: replace Abstract: Large language models (LLMs) are increasingly deployed in multilingual, real-world applications with user inputs -- naturally introducing typographi
arXiv:2604.12776v1 Announce Type: new Abstract: Realizing endogenous narrative evolution in LLM-based multi-agent systems is hindered by the inherent stochasticity of generative emergence. In particul
arXiv:2507.01041v4 Announce Type: replace-cross Abstract: Split learning (SL) is a distributed learning paradigm that can enable computation-intensive artificial intelligence (AI) applications by part
finally: @simonlast + @sarahmsachs on Latent Space! Notion has rebuilt Notion AI 5 times. This is the first time Simon has told the entire story. I've been trying to do this interview for ~3 years. We
arXiv:2511.13790v2 Announce Type: replace-cross Abstract: This paper describes GeoPl@ntNet, an interactive web application designed to make Essential Biodiversity Variables accessible and understandab
arXiv:2604.12229v1 Announce Type: new Abstract: Small language models (SLMs) often struggle with complex mathematical reasoning due to limited capacity to maintain long chains of intermediate steps an
arXiv:2604.11912v1 Announce Type: cross Abstract: While next-token prediction (NTP) has been the standard objective for training language models, it often struggles to capture global structure in reas
arXiv:2604.12665v1 Announce Type: new Abstract: Motion reasoning serves as the cornerstone of multi-object tracking (MOT), as it enables consistent association of targets across frames. However, exist
arXiv:2502.13571v2 Announce Type: replace-cross Abstract: The Influence Maximization (IM) problem aims to find a small set of influential users to maximize their influence spread in a social network.
arXiv:2604.12185v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) enhances large language models by grounding outputs in retrieved knowledge. However, existing RAG methods including
arXiv:2604.12286v1 Announce Type: new Abstract: Live Photo captures both a high-quality key photo and a short video clip to preserve the precious dynamics around the captured moment. While users may c
@lmsysorg @sgl_project @vllm_project @CoreWeave @nebiusai @nscale @togethercompute @Togethercompute enables AI labs and enterprises to run frontier models and agents on the NVIDIA Blackwell platform—g
arXiv:2604.12301v1 Announce Type: cross Abstract: We present a systematic measurement study of seven tactics for reducing cloud LLM token usage when a small local model can act as a triage layer in fr
Meta Platforms Inc. has announced a new deal with Broadcom Inc., extending a partnership that already sees the two companies work together on the design of the social media giant’s in-house artificial
Microsoft Corp.’s push for artificial intelligence independence is gaining traction with today’s release of MAI-Image-2-Efficient, a lean and mean version of its flagship image generation model that d
arXiv:2604.12374v1 Announce Type: cross Abstract: We describe the pre-training, post-training, and quantization of Nemotron 3 Super, a 120 billion (active 12 billion) parameter hybrid Mamba-Attention
arXiv:2604.12163v1 Announce Type: new Abstract: We present Nucleus-Image, a text-to-image generation model that establishes a new Pareto frontier in quality-versus-efficiency by matching or exceeding
arXiv:2604.11994v1 Announce Type: new Abstract: We study offline-online reinforcement learning in linear mixture Markov decision processes (MDPs) under environment shift. In the offline phase, data ar
Parcae is a stable looped language model that matches the quality of a Transformer twice its size — a 770M model reaching 1.3B-level performance. We introduce the first scaling laws for looping and sh
arXiv:2604.12856v1 Announce Type: new Abstract: Audio-driven bimanual piano motion generation requires precise modeling of complex musical structures and dynamic cross-hand coordination. However, exis
arXiv:2604.12384v1 Announce Type: new Abstract: Safety alignment in Large Language Models (LLMs) remains highly fragile during fine-tuning, where even benign adaptation can degrade pre-trained refusal
arXiv:2504.00890v2 Announce Type: replace-cross Abstract: Modern applications increasingly involve highly sensitive network data, where raw edges cannot be shared due to privacy constraints. We propos
Project Think is Cloudflare's initiative to develop next-generation AI agents built on its global network infrastructure, leveraging tools like Workers AI, Durable Objects, and the Workers platform to
Qwen3 is a series of large language models spanning both dense and Mixture-of-Experts (MoE) architectures, with parameter scales ranging from 0.6B to 235B, and a key innovation being the integration o
arXiv:2604.12582v1 Announce Type: new Abstract: Recent Video Large Language Models (Video-LLMs) have demonstrated strong capability in video understanding, yet they still suffer from hallucinations. E
arXiv:2604.12908v1 Announce Type: new Abstract: At its core, robotic manipulation is a problem of vision-to-geometry mapping (f(v) rightarrow G). Physical actions are fundamentally defined by geometri
Salesforce Inc. wants to make conversation the interface for developers, bringing agents to the foreground with a “headless” version of its platform designed to let humans and artificial intelligence
arXiv:2604.12322v1 Announce Type: new Abstract: The push for efficient text to image synthesis has moved the field toward one step sampling, yet existing methods still face a three way tradeoff among
arXiv:2604.12232v1 Announce Type: cross Abstract: Large Language Models (LLMs) are increasingly deployed across diverse domains, yet their vulnerability to jailbreak attacks, where adversarial inputs
arXiv:2602.18109v2 Announce Type: replace Abstract: Real-time schedulers must reason about tight deadlines under strict compute budgets. We present TempoNet, a reinforcement learning scheduler that pa
The front door to the internet hasn't changed -- but the person going through that door has changed from a human browsing 5 blue links, to an agent browsing a similar index in vastly different ways. A
Training code and models are live on Hugging Face. Dan Fu (Together AI's VP of Kernels) led the work. Together AI provided compute. Blog: https://www.together.ai/blog/parcae Paper: https://arxiv.org/a
What if you could get 1.3B Transformer quality from a 770M model? That's not a compression result. It's a different architecture. Parcae, from @realDanFu (Together AI's VP of Kernels) and his lab at U