model releases

Dental-TriageBench: Benchmarking Multimodal Reasoning for Hierarchical Dental Triage

arXiv:2604.13060v1 Announce Type: new Abstract: Dental triage is a safety-critical clinical routing task that requires integrating multimodal clinical information (e.g., patient complaints and radiogr

Model Releases

Design Space Exploration of Hybrid Quantum Neural Networks for Chronic Kidney Disease

DGX agent

arXiv:2604.13608v1 Announce Type: new Abstract: Hybrid Quantum Neural Networks (HQNNs) have recently emerged as a promising paradigm for near-term quantum machine learning. However, their practical pe

model-releasesarxiv-cs-lg

16 Apr 2026

DF3DV-1K: A Large-Scale Dataset and Benchmark for Distractor-Free Novel View Synthesis

arXiv:2604.13416v1 Announce Type: new Abstract: Advances in radiance fields have enabled photorealistic novel view synthesis. In several domains, large-scale real-world datasets have been developed to

Do We Still Need Humans in the Loop? Comparing Human and LLM Annotation in Active Learning for Hostility Detection

arXiv:2604.13899v1 Announce Type: new Abstract: Instruction-tuned LLMs can annotate thousands of instances from a short prompt at negligible cost. This raises two questions for active learning (AL): c

Document-tuning for robust alignment to animals

arXiv:2604.13076v1 Announce Type: new Abstract: We investigate the robustness of value alignment via finetuning with synthetic documents, using animal compassion as a value that is both important in i

Drowsiness-Aware Adaptive Autonomous Braking System based on Deep Reinforcement Learning for Enhanced Road Safety

arXiv:2604.13878v1 Announce Type: new Abstract: Driver drowsiness significantly impairs the ability to accurately judge safe braking distances and is estimated to contribute to 10%-20% of road acciden

Efficient Multi-View 3D Object Detection by Dynamic Token Selection and Fine-Tuning

arXiv:2604.13586v1 Announce Type: new Abstract: Existing multi-view three-dimensional (3D) object detection approaches widely adopt large-scale pre-trained vision transformer (ViT)-based foundation mo

EmbodiedClaw: Conversational Workflow Execution for Embodied AI Development

model-releasesarxiv-cs-ro

arXiv:2604.13800v1 Announce Type: new Abstract: Embodied AI research is increasingly moving beyond single-task, single-environment policy learning toward multi-task, multi-scene, and multi-model setti

EMGFlow: Robust and Efficient Surface Electromyography Synthesis via Flow Matching

arXiv:2604.13685v1 Announce Type: cross Abstract: Deep learning-based surface electromyography (sEMG) gesture recognition is frequently bottlenecked by data scarcity and limited subject diversity. Whi

Enhancing Confidence Estimation in Telco LLMs via Twin-Pass CoT-Ensembling

arXiv:2604.13271v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly applied to complex telecommunications tasks, including 3GPP specification analysis and O-RAN network troub

ESCAPE: Episodic Spatial Memory and Adaptive Execution Policy for Long-Horizon Mobile Manipulation

arXiv:2604.13633v1 Announce Type: new Abstract: Coordinating navigation and manipulation with robust performance is essential for embodied AI in complex indoor environments. However, as tasks extend o

Evaluating LLM-Based Translation of a Low-Resource Technical Language: The Medical and Philosophical Greek of Galen

arXiv:2602.24119v2 Announce Type: replace Abstract: Purpose: This study evaluates the quality of commercial large language model (LLM) machine translation (MT) for Ancient Greek technical prose and be

Evaluating Supervised Machine Learning Models: Principles, Pitfalls, and Metric Selection

arXiv:2604.13882v1 Announce Type: new Abstract: The evaluation of supervised machine learning models is a critical stage in the development of reliable predictive systems. Despite the widespread avail

Evaluating the Evaluator: Problems with SemEval-2020 Task 1 for Lexical Semantic Change Detection

arXiv:2604.13232v1 Announce Type: new Abstract: This discussion paper re-examines SemEval-2020 Task 1, the most influential shared benchmark for lexical semantic change detection, through a three-part

Evaluating the Formal Reasoning Capabilities of Large Language Models through Chomsky Hierarchy

arXiv:2604.02709v2 Announce Type: replace Abstract: The formal reasoning capabilities of LLMs are crucial for advancing automated software engineering. However, existing benchmarks for LLMs lack syste

EVE: A Domain-Specific LLM Framework for Earth Intelligence

arXiv:2604.13071v1 Announce Type: new Abstract: We introduce Earth Virtual Expert (EVE), the first open-source, end-to-end initiative for developing and deploying domain-specialized LLMs for Earth Int

Exposia: Teaching and Assessment of Academic Writing Skills for Research Project Proposals and Peer Feedback

arXiv:2601.06536v2 Announce Type: replace Abstract: We present Exposia, the first public dataset that connects writing and feedback in higher education, enabling research on educationally grounded com

ExpSeek: Self-Triggered Experience Seeking for Web Agents

arXiv:2601.08605v2 Announce Type: replace Abstract: Experience intervention in web agents emerges as a promising technical paradigm, enhancing agent interaction capabilities by providing valuable insi

F-Actor: Controllable Conversational Behaviour in Full-Duplex Models

arXiv:2601.11329v3 Announce Type: replace Abstract: Spoken conversational systems require more than accurate speech generation to have human-like conversations: to feel natural and engaging, they must

Failure Makes the Agent Stronger: Enhancing Accuracy through Structured Reflection for Reliable Tool Interactions

arXiv:2509.18847v3 Announce Type: replace-cross Abstract: Tool-augmented large language models (LLMs) are usually trained with supervised imitation or coarse-grained reinforcement learning that optimi

Fast training of accurate physics-informed neural networks without gradient descent

arXiv:2405.20836v3 Announce Type: replace-cross Abstract: Solving time-dependent Partial Differential Equations (PDEs) is one of the most critical problems in computational science. While Physics-Info

FieldWorkArena: Agentic AI Benchmark for Real Field Work Tasks

arXiv:2505.19662v3 Announce Type: replace-cross Abstract: This paper introduces FieldWorkArena, a benchmark for agentic AI targeting real-world field work. With the recent increase in demand for agent

FiLM-Nav: Efficient and Generalizable Navigation via VLM Fine-tuning

model-releasesarxiv-cs-ro

arXiv:2509.16445v2 Announce Type: replace Abstract: Enabling robotic assistants to navigate complex environments and locate objects described in free-form language is a critical capability for real-wo

Finally, we’ve been talking to lots of users about helping them make the most of Claude Code. We’ll be sharing more this week on that work, …

Anthropic is working on improvements to help users maximize the utility of Claude Code functionality, with additional details and updates to be announced later in the same week. This announcement come

model-releasesthariq--x

FlexGuard: Continuous Risk Scoring for Strictness-Adaptive LLM Content Moderation

arXiv:2602.23636v3 Announce Type: replace Abstract: Ensuring the safety of LLM-generated content is essential for real-world deployment. Most existing guardrail models formulate moderation as a fixed

Flow-based Generative Modeling of Potential Outcomes and Counterfactuals

arXiv:2505.16051v4 Announce Type: replace-cross Abstract: Predicting potential and counterfactual outcomes from observational data is central to individualized decision-making, particularly in clinica

fMRI-LM: Towards a Universal Foundation Model for Language-Aligned fMRI Understanding

arXiv:2511.21760v3 Announce Type: replace Abstract: Recent advances in multimodal large language models (LLMs) have enabled unified reasoning across images, audio, and video, but extending such capabi

For the developers building with Claude, a direct line from the team. Follow for changelogs, API releases, community updates, and deep dives…

This is an announcement for the official Claude Developers Twitter/X account that serves as a direct communication channel from Anthropic's team to developers using Claude's API. The account provides

For those not seeing the increase, make sure you're using Opus 4.7 with the latest Claude Code

Claude Code users should ensure they are using Opus 4.7 with the latest updates to experience performance improvements or feature enhancements. The post suggests that users not observing expected incr

Free Geometry: Refining 3D Reconstruction from Longer Versions of Itself

arXiv:2604.14048v1 Announce Type: new Abstract: Feed-forward 3D reconstruction models are efficient but rigid: once trained, they perform inference in a zero-shot manner and cannot adapt to the test s

From Feelings to Metrics: Understanding and Formalizing How Users Vibe-Test LLMs

arXiv:2604.14137v1 Announce Type: new Abstract: Evaluating LLMs is challenging, as benchmark scores often fail to capture models' real-world usefulness. Instead, users often rely on ``vibe-testing'':

From Plausibility to Verifiability: Risk-Controlled Generative OCR with Vision-Language Models

arXiv:2603.19790v3 Announce Type: replace Abstract: Modern vision-language models (VLMs) can act as generative OCR engines, yet open-ended decoding can expose rare but consequential failures. We ident

From Weights to Activations: Is Steering the Next Frontier of Adaptation?

arXiv:2604.14090v1 Announce Type: new Abstract: Post-training adaptation of language models is commonly achieved through parameter updates or input-based methods such as fine-tuning, parameter-efficie

Functional Emotions or Situational Contexts? A Discriminating Test from the Mythos Preview System Card

arXiv:2604.13466v1 Announce Type: cross Abstract: The Claude Mythos Preview system card deploys emotion vectors, sparse autoencoder (SAE) features, and activation verbalisers to study model internals

Gaslight, Gatekeep, V1-V3: Early Visual Cortex Alignment Shields Vision-Language Models from Sycophantic Manipulation

arXiv:2604.13803v1 Announce Type: new Abstract: Vision-language models are increasingly deployed in high-stakes settings, yet their susceptibility to sycophantic manipulation remains poorly understood

Gemini can now create personalized AI images by digging around in Google Photos

model-releasesars-technica

Gemini now uses user interests and Google Photos to create personalized AI images without requiring long descriptions, allowing users to simply ask for pictures of themselves or family members. This f

Gemini can now pull from Google Photos to generate personalized images

model-releasesthe-verge-ai

Google's Personal Intelligence feature, which lets Gemini pull data from apps like Google Photos to offer responses tailored to you, can now use that data and its Nano Banana 2 image model to create i

GeoBridge: A Semantic-Anchored Multi-View Foundation Model Bridging Images and Text for Geo-Localization

arXiv:2512.02697v3 Announce Type: replace Abstract: Cross-view geo-localization infers a location by retrieving geo-tagged reference images that visually correspond to a query image. However, the trad

GLM-5.1 Tool Calling Issue Fix & Chat Template Update If you are running GLM-5.1 with vLLM/SGLang and using tool calling, please update your…

model-releaseszhipu-ai--x

GLM-5.1 Tool Calling Issue Fix & Chat Template Update If you are running GLM-5.1 with vLLM/SGLang and using tool calling, please update your chat template. http://huggingface.co/zai-org/GLM-5.1/blob/m

Go from blank slate to analysis with BigQuery Studio notebook gallery templates

model-releasesgoogle-cloud-ai

For many data professionals, the most daunting part of a new project isn't the complexity of the data or the sophistication of the model, it’s the 'blank slate.' Staring at an empty notebook while cre

Google’s DeepMind just released new 4B and 27B MedGemma models!

Google DeepMind released MedGemma, a collection of medical vision-language foundation models based on Gemma 3 in 4B and 27B parameter sizes, demonstrating advanced medical understanding and reasoning

model-releasesr-ollama

Google’s Gemini 3.1 Flash TTS model offers unparalleled control over AI voices

model-releasessiliconangle

Google LLC’s DeepMind artificial intelligence unit today rolled out a new text-to-speech model called Gemini 3.1 Flash TTS. Unlike its earlier, robotic predecessors, it enables users to direct the voc

Had a great time at PyCon & PyData DE. Highly recommend it. Great open-source, community-focused conference with lots of builders in the Pyt…

model-releasessebastian-raschka--x

Had a great time at PyCon & PyData DE. Highly recommend it. Great open-source, community-focused conference with lots of builders in the Python AI, LLM and agent space. Taking a short family break, my

Happy coding! Opus 4.7 is a significant step up. To get the most out of it, take the time to adjust your workflow to take advantage of Claud…

Happy coding! Opus 4.7 is a significant step up. To get the most out of it, take the time to adjust your workflow to take advantage of Claude running for longer & being more agentic. It feels like a n

Here's Qwen 3.6-35B-A3B v.s. Claude Opus 4.7 for 'Generate an SVG of a flamingo riding a unicycle', in case you thought Qwen might be cheati…

model-releasessimon-willison--x

This post compares the performance of Qwen 3.6-35B-A3B and Claude Opus 4.7 models on a creative task of generating SVG code for a flamingo riding a unicycle, likely demonstrating differences in their

Hessian-Enhanced Token Attribution (HETA): Interpreting Autoregressive LLMs

arXiv:2604.13258v1 Announce Type: new Abstract: Attribution methods seek to explain language model predictions by quantifying the contribution of input tokens to generated outputs. However, most exist

Hi! I'm here with another launch, it just happens to be extremely niche, nerdy, and probably only for a handful of people. In the desktop …

Hi! I'm here with *another launch*, it just happens to be extremely niche, nerdy, and probably only for a handful of people. In the desktop app, Claude Cowork and Code now have a little Bluetooth API

Hierarchical Reinforcement Learning with Runtime Safety Shielding for Power Grid Operation

arXiv:2604.14032v1 Announce Type: cross Abstract: Reinforcement learning has shown promise for automating power-grid operation tasks such as topology control and congestion management. However, its de