New Model Releases

A
Showing 151–180 of 218
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    RESKILL: Explicit Failure Attribution and Structured Repair for Interactive Language Agents
    AI Agents Deep Learning Retrieval-Augmented Generation (RAG) Reinforcement Learning
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.AI (Artificial Intelligence) · EN New Model Releases
    Circuit-MLLM: Topological Logic-Guided Latent-Space Visual Reasoning for Circuit Schematic Understanding
    GPT Inference Machine Learning Retrieval-Augmented Generation (RAG)
    Read original (arXiv cs.AI (Artificial Intelligence)) ↗
  • arXiv cs.AI (Artificial Intelligence) · EN Developer Tools
    CIDERS: Cloud-Edge LLM Collaborative Learning via Accelerating Personalized Bilevel Optimization
    Embeddings Reinforcement Learning
    Read original (arXiv cs.AI (Artificial Intelligence)) ↗
  • arXiv cs.AI (Artificial Intelligence) · EN New Model Releases
    Kaininja: Extending Native 3D Generators to the Part Level
    Neural Network
    Read original (arXiv cs.AI (Artificial Intelligence)) ↗
  • arXiv cs.LG (Machine Learning) · EN Training & Fine-tuning
    Principal-timestep Restricted Init via Sparse Matrix-decomposition in Flow-matching
    Fine-tuning Neural Network
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.AI (Artificial Intelligence) · EN New Model Releases
    FedLTLib: A Comprehensive Benchmark for Federated Long-Tail Learning
    Algorithms & Theory Deep Learning Meta Retrieval-Augmented Generation (RAG) Reinforcement Learning
    Read original (arXiv cs.AI (Artificial Intelligence)) ↗
  • arXiv cs.LG (Machine Learning) · EN New Model Releases
    Where to Compute and How to Interact: Operator-Readable Adaptation with Gauge-Aware Transport
    Software Engineering
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    Can We Trust the Judges? Validation of Factuality Evaluation Methods via Answer Perturbation
    Meta Retrieval-Augmented Generation (RAG) Software Engineering
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.LG (Machine Learning) · EN New Model Releases
    Bayesian Optimisation Using Product-of-Experts Gaussian Process Models with Uncertainty Calibration
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.LG (Machine Learning) · EN New Model Releases
    The Token Before the Value Is the Key: How Hybrid Architectures Organize Induction Circuits
    Neural Network Reinforcement Learning
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.LG (Machine Learning) · EN New Model Releases
    Specifying Reward Functions for RL Without Environment Sampling
    AI Agents Deep Learning Reinforcement Learning Reinforcement Learning from Human Feedback (RLHF)
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.LG (Machine Learning) · EN New Model Releases
    Strong and Compact Policies for Submodular Markov Decision Processes via LP-Based Submodular Orienteering
    Deep Learning Machine Learning Reinforcement Learning
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.LG (Machine Learning) · EN New Model Releases
    The Misery of Mechanistic Interpretability: A Formal Perspective
    GPT Llama Neural Network
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    To Each Language Its Tokenizer: Modular Tokenizers for Efficient Multilingual LLMs
    Embeddings Inference Meta
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    Authorship attribution and aesthetic evaluation of AI poetry: a case study with Haiku
    Gemini GPT Llama Neural Network
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    SlopShape: Identifying AI-Generated Commercial Web Content
    GPT Neural Network
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    RSIAgent: Autonomous Exploration for Recursive Self-improvement in New Environments
    AI Agents GPT Reinforcement Learning
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.CL (Computation and Language) · EN Inference & Efficiency
    Reason What Matters: Retrieval-Grounded Reasoning for Universal Multimodal Embeddings
    Embeddings Inference Reinforcement Learning
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    Artificial entrepreneurial cognition: Locating and causally steering an opportunity recognition dial inside large language models (LLMs)
    Llama Neural Network
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    CITECHOICE: A Causal Audit of How Document Presentation Redistributes Citation Credit in Agentic Search
    Reinforcement Learning Software Engineering
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    EMR: Self-Evolving Medical Multi-Agent System via Experience Mining and Reuse
    AI Agents Inference Neural Network
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • ITmedia AI+ · JA New Model Releases
    「FDE」「ハーネス」「オープンウェイト」って何？　AI時代に現場で飛び交う10用語をGitHubが整理
    GitHub publishes a guide to 10 AI-era dev terms: FDE, harness, open-weight
    GitHub released a guide explaining ten AI-related terms now spreading quickly across software development, including "FDE," "harness," and "open-weight." The guide aims to give teams a shared vocabulary as AI tooling reshapes everyday engineering work.
    Read original (ITmedia AI+) ↗
  • Simon Willison's Weblog · EN New Model Releases
    commit-rewriter 0.1
    Simon Willison releases commit-rewriter 0.1 to clean up commit messages
    Simon Willison released commit-rewriter 0.1, a small web app for editing and rewriting Git commit messages. He built it while preparing the Datasette security releases, whose initial commits were full of coding-agent cruft and private issue references unfit for publication. Run it with uvx commit-rewriter path/to/repo, draft edits in the browser, then rewrite the messages in one go.
    Read original (Simon Willison's Weblog) ↗
  • Simon Willison's Weblog · EN New Model Releases
    shot-scraper 1.12
    shot-scraper 1.12 adds WebP screenshot output with quality control
    Simon Willison shipped shot-scraper 1.12, adding WebP support to his screenshot automation tool. Pass -o screenshot.webp --quality 80 to get a compressed WebP; omitting --quality yields lossless output. He notes WebP screenshots are almost always much smaller than JPEG or PNG equivalents, and built the feature to generate the screenshot for his new commit-rewriter tool.
    Read original (Simon Willison's Weblog) ↗
  • ITmedia AI+ · JA New Model Releases
    日立HMAX新技術「AIオーケストレーション」、特急注文など部門間調整を自動化
    Hitachi shows AI orchestration automating cross-department plant scheduling
    At Hitachi Social Innovation Forum 2026, Hitachi presented concrete scenarios for its AI orchestration technology, which automates schedule coordination across manufacturing departments. Department-specific AI agents feed a coordinator AI, a planning optimization engine (MLCP) generates feasible plant-wide plans, and humans make the final decision. Part of the HMAX Industry lineup, availability is planned for 2027.
    Read original (ITmedia AI+) ↗
  • ITmedia AI+ · JA New Model Releases
    AI技術を“無料で公開”、なぜ？　米IT業界の文化「オープン化」の裏にしたたかな戦略、日本は追随できるか
    Why AI firms open up their tech: Agentic AI Foundation CTO explains
    ITmedia looks at AI's open turn, from OpenAI's gpt-oss and Google's Gemma to Anthropic donating MCP to a nonprofit. Agentic AI Foundation's CTO explained at an Aug. 25 briefing why firms give technology away and where Japanese firms could win.
    Read original (ITmedia AI+) ↗
  • Publickey · JA Agents & Tool Use
    TailscaleのVPNにAIエージェントを組み込める「Aperture」正式リリース。AIによるTailscaleやノードの操作も可能に
    Tailscale ships Aperture AI gateway GA, adds MCPs for agent-driven network ops
    AI Agents Machine Learning
    Tailscale announced general availability of Aperture, an AI gateway that lets any node in a tailnet reach AI services without distributing API keys. It now bundles cost controls, guardrails, audit logs, MCP/API proxies, and in-gateway token purchasing. New Tailscale MCP and Tailscale SSH MCP let AI agents add nodes, SSH into machines, and deploy services via Aperture's chat UI, with existing ACLs enforced, human approval required for new nodes, and every action logged.
    Read original (Publickey) ↗
  • ITmedia AI+ · JA New Model Releases
    AnthropicのアモデイCEO、AI開発の「ペース調整」訴えるエッセイ公開　アルトマン氏とマスク氏も賛同
    Anthropic's Amodei calls to pace frontier AI; Altman and Musk back the plan
    Anthropic OpenAI
    Anthropic CEO Dario Amodei published an essay proposing a three-stage plan to slow the pace of model capability gains, citing rapid self-improvement and safety risks. He calls for embedded third-party evaluators and coordination among democracies. OpenAI's Sam Altman and Elon Musk endorsed the approach and said they would adopt similar evaluation regimes.
    Read original (ITmedia AI+) ↗
  • arXiv cs.AI (Artificial Intelligence) · EN Inference & Efficiency
    Rethinking Heterogeneous System Disaggregation for Subquadratic Attention
    Inference Retrieval-Augmented Generation (RAG)
    Read original (arXiv cs.AI (Artificial Intelligence)) ↗
  • arXiv cs.AI (Artificial Intelligence) · EN New Model Releases
    CMA-OT: Hierarchical Expert Supervision for Dance-to-Music Generation
    Retrieval-Augmented Generation (RAG) Reinforcement Learning
    Read original (arXiv cs.AI (Artificial Intelligence)) ↗