model releases

arXiv:2604.23750v1 Announce Type: cross Abstract: Hypernetwork-based methods such as Doc-to-LoRA internalize a document into an LLM's weights in a single forward pass, but they fail systematically on

arXiv:2604.24079v1 Announce Type: cross Abstract: Large Language Models (LLMs) reveal inherent and distinctive personas through dialogue. However, most existing persona discovery approaches rely on su

arXiv:2604.24668v1 Announce Type: new Abstract: Given the increased use of LLMs in financial systems today, it becomes important to evaluate the safety and robustness of such systems. One failure mode

arXiv:2604.22771v1 Announce Type: cross Abstract: Language models cannot be random. This paper introduces Entropic Deviation (ED), the normalised KL divergence between a model's token distribution and

arXiv:2601.15485v2 Announce Type: replace-cross Abstract: Federal research funding shapes the direction, diversity, and impact of the US scientific enterprise. Large language models (LLMs) are rapidly

arXiv:2604.23058v1 Announce Type: cross Abstract: Firms are deploying more capable AI systems, but organizational controls often have not kept pace. These systems can generate greater productivity gai

arXiv:2508.09603v2 Announce Type: replace Abstract: Membership inference attacks serves as useful tool for fair use of language models, such as detecting potential copyright infringement and auditing

arXiv:2511.08577v2 Announce Type: replace-cross Abstract: Improving reasoning abilities of Large Language Models (LLMs), especially under parameter constraints, is crucial for real-world applications.

this feels like less a trend and just something we expect... but q1 had a lot of launches that overlapped with recent startups i've seen: - codex desktop for running parallel - managed agents by claud

This is great - @deepseek_ai V4 supports prefill! :D Most other providers have been dropping support for this critically important capability, so wonderful to see at least one company stepping up. htt

Today we’re releasing Laguna XS.2, Poolside’s first open-weight model. It’s a 33B total / 3B active MoE model built for agentic coding and long-horizon tasks. Trained fully in-house on our own stack.

Together AI announced immediate availability of NVIDIA's Nemotron 3 Nano Omni model to developers through its platform on the day of its release. The Nemotron 3 Nano Omni is a lightweight multimodal m

arXiv:2604.01644v2 Announce Type: replace Abstract: Natural language provides an intuitive way to express spatial intent in geospatial applications. While existing localization methods often rely on d

arXiv:2604.24119v1 Announce Type: new Abstract: Topology reasoning is crucial for autonomous driving. Current methods primarily focus on instance-level learning for centerline detection, followed by a

arXiv:2604.23231v1 Announce Type: cross Abstract: Semantic Communication (SC) backdoor attacks aim to utilize triggers to manipulate the system into producing predetermined outputs via backdoored shar

arXiv:2509.26158v2 Announce Type: replace-cross Abstract: The performance of deep neural networks is strongly influenced by the quality of their training data. However, mitigating dataset bias by manu

arXiv:2603.25158v4 Announce Type: replace Abstract: Equipping Large Language Model (LLM) agents with domain-specific skills is critical for tackling complex tasks. Yet, manual authoring creates a seve

NVIDIA Nemotron-3 Nano Omni is now available to try through Together AI's platform, offering access to a compact multimodal model capable of processing both text and audio inputs. This announcement hi

arXiv:2604.23609v1 Announce Type: new Abstract: Contact-rich manipulation is central to many everyday human activities, requiring continuous adaptation to contact uncertainty and external disturbances

arXiv:2604.23460v1 Announce Type: new Abstract: Chain-of-Thought (CoT) reasoning has emerged as a key technique for eliciting complex reasoning in Large Language Models (LLMs). Although interpretable,

arXiv:2604.24525v1 Announce Type: cross Abstract: Automated code review (ACR) bots are increasingly used in industrial software development to assist developers during pull request (PR) review. As ado

arXiv:2604.23145v1 Announce Type: cross Abstract: Video Question Answering (VideoQA) demands models that jointly reason over spatial, temporal, and linguistic cues. However, the task's inherent comple

Carl Franzen / VentureBeat: US startup Poolside debuts its first open-weight model, Laguna XS.2, a 33B-A3B-parameter MoE model, and Laguna M.1, a proprietary 225B-A23B-parameter MoE model — The AI rac

arXiv:2509.14837v2 Announce Type: replace Abstract: Recent advances in causal interpretability have extended from language models to vision-language models (VLMs), seeking to reveal their internal mec

arXiv:2510.08618v2 Announce Type: replace-cross Abstract: Omni-modal large language models (OLLMs) offer a promising end-to-end solution for slide-enhanced speech recognition due to their inherent mul

Want to see which frontier models do the best on document understanding? Check out our ParseBench leaderboard on @kaggle! https://www.kaggle.com/benchmarks/llamaindex-org/parsebench For more details o

arXiv:2604.22834v1 Announce Type: new Abstract: This paper presents webmcu-vision-web, a single-file, zero-install browser application for end-to-end TinyML vision model training and deployment on the

Welcome to the agentic era! Last week, leaders from our public sector customer and partner ecosystem took the stage at Google Cloud Next to share how they are leveraging AI and agents to scale their i

we're doing a lot more of this, hunting down some of the most annoying bugs in Claude Code let me know if you have any white whales In the last four Claude Code CLI releases, we’ve shipped 50+ stabili

arXiv:2604.23942v1 Announce Type: cross Abstract: People increasingly turn to large language models (LLMs) to interpret ambiguous social situations: a delayed text reply, an unusually cold supervisor,

This post is a question about search provider preferences for use with various open-source tools and APIs (openclaw, pi, opencode, etc.), with examples of options like Brave, SerpAPI, and Gemini. The

arXiv:2604.23398v1 Announce Type: new Abstract: We report a reproducible error pattern in GPT-5.4 on OWL~2~DL compliance queries: the model frequently answers ``unknown'' when the reasoner-entailed an

arXiv:2604.23434v1 Announce Type: cross Abstract: Dynamic Tanh (DyT) removes LayerNorm by bounding activations with a learned tanh(alpha x). We show that this bounding is a regime-dependent implicit r

arXiv:2604.22774v1 Announce Type: cross Abstract: Accurate transcription of handwritten mathematics is crucial for educational AI systems, yet current benchmarks fail to evaluate this capability prope

arXiv:2604.23767v1 Announce Type: new Abstract: Deploying machine learning models across diverse well portfolios requires generalisation to wells with design parameters outside the training distributi

This is a Wordle game result posted by Anthropic on X (formerly Twitter), showing the solution was found in 5 out of 6 possible attempts. The emoji grid indicates the letter placement feedback for eac

arXiv:2509.13021v2 Announce Type: replace-cross Abstract: This work introduces xOffense, an AI-driven, multi-agent penetration testing framework that shifts the process from labor-intensive, expert-dr

Yes this is a line from the system prompt. gpt-5.5 prompt for codex seems to have a duplicated line trying to get it to not talk about creatures? Never talk about goblins, gremlins, raccoons, trolls,

arXiv:2604.24479v1 Announce Type: new Abstract: Computer-Aided Design (CAD) models are defined by their construction history: a parametric recipe that encodes design intent. However, existing large-sc

A completely local agent that lives right inside your browser. Powered by Gemma 4 E2B and WebGPU, it uses native tool calling to: 🔍 Search browsing history 📄 Read and summarize pages 🔗 Manage tabs 100

arXiv:2604.22436v1 Announce Type: new Abstract: The rapid growth of AI agent ecosystems is transforming how complex tasks are delegated and executed, creating a new challenge of identifying suitable a

AI has had one safest technology roll-outs in history. Read that again, because it's a fact. It's used by billions with a tiny fraction of a percent of actual problems. And yet it's seen as dangerous

arXiv:2604.22722v1 Announce Type: cross Abstract: Dense vector retrieval is the practical backbone of Retrieval- Augmented Generation (RAG), but similarity search can suffer from precision limitations

arXiv:2604.21936v1 Announce Type: new Abstract: Medical imaging research is increasingly shifting from controlled benchmark evaluation toward real-world clinical deployment. In such settings, applying

arXiv:2604.22139v1 Announce Type: new Abstract: Reliable automated analysis of Optical Coherence Tomography (OCT) imaging is crucial for diagnosing retinal disorders but faces a critical barrier: the

Google DeepMind announced a partnership with South Korea's Ministry of Science and ICT (MSIT) , marking the 10th anniversary of the historic AlphaGo match. Google will establish an AI Campus in Seoul