Search: “model-releases”

lots of very interesting items here. They talk about pretty much everything from how to accelerate capabilities to concerns about agent safety. I'm devastated to inform doomers that 'full stack open s

New Anthropic research: Teaching Claude why. Last year we reported that, under certain experimental conditions, Claude 4 would blackmail users. Since then, we’ve completely eliminated this behavior. H

OpenAI Group PBC has developed a version of GPT-5.5 that is specifically optimized for cybersecurity research. GPT‑5.5‑Cyber, as the model is called, made its debut on Thursday. It’s available in limi

Pinecone has launched its first serverless cloud region in Singapore and rolled out Nexus, KnowQL, a Marketplace and a USD $20 Builder tier. #CloudComputing #Singapore #Serverless #AI https://techday.

DeepSeek-V4 makes million-token context a serving-systems problem. Together AI explores the inference work behind V4 on NVIDIA HGX B200, including compressed KV layouts, prefix caching, kernel maturit

With AI coding tools like Antigravity and Claude Code, I can build a working web app in record time. But deploying it? That's where I'd historically lose the rest of the afternoon to Dockerfiles, IAM

Code with Claude San Francisco was a community event that concluded successfully, with positive feedback from attendees. The event was organized by Anthropic's Claude development team and attracted de

The Pentagon released declassified UFO/UAP files following President Trump's directive, with the initial batch containing 162 documents outlining decades-old investigations of reported unidentified ae

Training models involves many technical and social processes, so prevention of CoT grading has to be built into the process. We’re improving real-time CoT-grading detection, safeguards against acciden

Using Claude Code: The Unreasonable Effectiveness of HTML Thought-provoking piece by Thariq Shihipar (on the Claude Code team at Anthropic) advocating for HTML over Markdown as an output format to req

This release candidate merges updates from the upstream main branch into the llama-runner-phase-0 development branch, likely incorporating bug fixes, improvements, and new features from the main Ollam

Boris Cherny shared a link to free Code with Claude stickers available through Sticker Mule, offering them to followers on a first-come, first-served basis. The stickers appear to be promotional merch

We started by investigating why Claude chose to blackmail. We believe the original source of the behavior was internet text that portrays AI as evil and interested in self-preservation. Our post-train

we'd like to help companies secure themselves and we think it's important to start work on this quickly Today, we're rolling out GPT‑5.5‑Cyber in limited preview to defenders responsible for securing

Anthropic is hosting hackathons in San Francisco where developers can build projects using Claude, Anthropic's AI model. The announcement invites the developer community to participate in these upcomi

When I started to work on the alignment problem more than 10 years ago, we had no idea how AGI was going to be built or how to make it safe. The field had maybe a dozen people who were working on it a

arXiv:2605.04813v1 Announce Type: new Abstract: With the rapid development of cloud computing and Web services, Quality of Service (QoS) has become a key criterion for service selection and recommenda

arXiv:2605.04885v1 Announce Type: new Abstract: This paper compares a PyCaret AutoML branch and a CNN-BiLSTM branch for binary hate speech detection on Indonesian Twitter using the HS label from the c

A few weeks ago @simonw got Claude to port LiteParse to the browser. Today, we are launching that work as a complete guide in our docs! https://developers.llamaindex.ai/liteparse/guides/browser-usage/

arXiv:2605.04076v1 Announce Type: new Abstract: U.S. financial institutions deploying AI-based fraud detection face a fragmented compliance landscape spanning four regulatory frameworks -- OCC Bulleti

arXiv:2601.20634v2 Announce Type: replace Abstract: Virtual sensors replace expensive physical sensors in critical applications through machine learning by predicting target signals from available mea

arXiv:2605.04055v1 Announce Type: new Abstract: Adaptive optimizers like AdamW apply uniform hyperparameters across all parameter groups, ignoring heterogeneous optimization dynamics across layers and

arXiv:2605.05079v1 Announce Type: new Abstract: Video sequence capturing through refractive dynamic media, such as a turbulent air or water surface, often suffer from severe geometric distortions and

arXiv:2601.15486v2 Announce Type: replace Abstract: The use of artificial intelligence (AI) for drone control can have a transformative impact on drone capabilities, especially when real world informa

arXiv:2605.04948v1 Announce Type: new Abstract: This paper presents a comparative study of parameter-efficient fine-tuning (PEFT) methods, including LoRA and QLoRA, applied to the task of adapting lar

arXiv:2507.23501v2 Announce Type: replace Abstract: Ensembles are ubiquitous in off-policy actor-critic learning, yet their efficacy depends critically on how they are aggregated. Current methods typi

OpenAI announced new voice intelligence models available through its API, expanding capabilities for developers to integrate advanced voice processing and understanding features into their application

arXiv:2605.05155v1 Announce Type: new Abstract: As 3D Gaussian Splatting (3DGS) gains attention in immersive media and digital content creation, assessing the aesthetics of 3D scenes becomes important

A benchmark-driven look at why agent harnesses need adaptive finish logic as model behavior changes across Claude, GPT-4o, and Gemma. The post Agent harnesses have an expiration date appeared first on

arXiv:2605.05000v1 Announce Type: cross Abstract: Windows Component Object Model (COM) services run with elevated privileges and are widely accessible to authenticated users, making race conditions in

arXiv:2404.06230v3 Announce Type: replace Abstract: In federated learning (FL), profiling and verifying each client is inherently difficult, which introduces a significant security vulnerability: mali

Anthropic PBC said today it’s giving its AI agents the ability to “dream” and remember past interactions and work they’ve performed so they can identify recurring mistakes and improve over time. In an

Anthropic just shipped sleep into agents. When you sleep, your hippocampus replays the day's neural sequences to the cortex during 150-220 Hz bursts called sharp-wave ripples. The replay runs about 20

Anthropic: Anthropic researchers detail “natural language autoencoders”, which convert LLM activations, the numbers encoding a model's thoughts, into natural language text — When you talk to an AI mod

arXiv:2605.04177v1 Announce Type: new Abstract: As LLMs enter conflict monitoring, understanding systematic distortions in their outputs is critical for humanitarian accountability. We evaluate four v

arXiv:2605.04098v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) have demonstrated promise on publicly available dermatology benchmarks. However, benchmark performance may not

arXiv:2603.25956v2 Announce Type: replace Abstract: Time-series anomaly detection (TSAD) is a critical component in monitoring complex systems, yet modern deep learning-based detectors are often highl

arXiv:2605.04857v1 Announce Type: new Abstract: This paper presents the development and validation of an eye-tracking dataset designed to investigate how second-language (L2) learners process idiomati

arXiv:2605.04083v1 Announce Type: new Abstract: Much of the focus in RL today is on evaluation design: building meaningful evals that serve simultaneously as benchmarks and as well-defined reward sign

arXiv:2605.04472v1 Announce Type: new Abstract: Automating formal proofs of combinatorial identities is challenging for LLM-based provers, as long-horizon proof planning is required and unconstrained

arXiv:2605.03339v1 Announce Type: new Abstract: Solving large-scale CVRP (LSCVRP) with hundreds to thousands of nodes remains difficult for even state-of-the-art solvers. Divide-and-conquer can scale

arXiv:2602.19041v2 Announce Type: replace Abstract: A recurring challenge in preference fine-tuning (PFT) is handling extit{intransitive} (i.e., cyclic) preferences. Intransitive preferences often ste

arXiv:2502.02625v2 Announce Type: replace Abstract: Parameter shift rules (PSRs) are key techniques for efficient gradient estimation in variational quantum eigensolvers (VQEs). In this paper, we prop

Behind the Scenes Hardening Firefox with Claude Mythos Preview Fascinating, in-depth details on how Mozilla used their access to the Claude Mythos preview to locate and then fix hundreds of vulnerabil

arXiv:2605.04556v1 Announce Type: cross Abstract: The Massive Sound Embedding Benchmark (MSEB) has emerged as a standard for evaluating the functional breadth of audio models. While initial baselines

arXiv:2605.04576v1 Announce Type: new Abstract: This paper presents the first benchmark for the task of automatic part-of-speech (POS) tagging for the Tajik language. Despite the existence of multilin