New Model Releases

A
Showing 211–240 of 287
  • arXiv cs.LG (Machine Learning) · EN Infrastructure & Hardware
    The Structure Gap: When More Data Cannot Replace Structured Reasoning
    Neural Network Retrieval-Augmented Generation (RAG)
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.LG (Machine Learning) · EN New Model Releases
    Orthogonal and Parametrized Multi-Attribute Steering of LLMs
    Inference
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • Publickey · JA Infrastructure & Hardware
    Oracle CloudがGoogle TPU導入を発表、2027年に提供を開始。Oracle AI World 2026
    Oracle Cloud to add Google TPUs, with service launching in 2027
    Google Machine Learning Reinforcement Learning
    Oracle announced it will bring Google's AI-specific TPU chips to Oracle Cloud Infrastructure, with Google TPU services becoming available from 2027. The announcement was made at Oracle AI World 2026.
    Read original (Publickey) ↗
  • arXiv cs.LG (Machine Learning) · EN Training & Fine-tuning
    A note on data-efficient finetuning of vision foundation models
    Fine-tuning Machine Learning Reinforcement Learning
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.AI (Artificial Intelligence) · EN New Model Releases
    Strategyproof Deep Reinforcement Learning for Online Resource Allocation
    AI Agents Neural Network Reinforcement Learning
    Read original (arXiv cs.AI (Artificial Intelligence)) ↗
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    semanticfa: An R Package for Response-Free Semantic Factor Analysis of Psychometric Scales
    Embeddings Neural Network Software Engineering
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.AI (Artificial Intelligence) · EN New Model Releases
    From Interaction to Capability: Task Advancement as State and Supervision for GUI Agents
    AI Agents Reinforcement Learning
    Read original (arXiv cs.AI (Artificial Intelligence)) ↗
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    What Carries Forward? Pretraining Disclosure Across Language Model Releases
    Llama Reinforcement Learning
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.AI (Artificial Intelligence) · EN New Model Releases
    Concordat: Typed Sessions for LLM Agents
    AI Agents Claude GPT Reinforcement Learning
    Read original (arXiv cs.AI (Artificial Intelligence)) ↗
  • arXiv cs.AI (Artificial Intelligence) · EN Developer Tools
    M4Bench: Evaluating Procedural Specification for Clinical EHR Derivation Agents
    AI Agents Claude GPT Neural Network Reinforcement Learning
    Read original (arXiv cs.AI (Artificial Intelligence)) ↗
  • arXiv cs.LG (Machine Learning) · EN New Model Releases
    Interpretable Meta-Policies for Neural Decision Making
    AI Agents Meta Neural Network Reinforcement Learning
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.LG (Machine Learning) · EN New Model Releases
    Graph Representation Learning via Adaptive Expectile Aggregation
    Neural Network Reinforcement Learning
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    Representing the Unseen: Rethinking Visual Imagination for Spatial Reasoning
    Neural Network Reinforcement Learning
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.AI (Artificial Intelligence) · EN Developer Tools
    FeatureArc: Benchmarking Coding Agents on Evolving Features
    AI Agents Neural Network Reinforcement Learning
    Read original (arXiv cs.AI (Artificial Intelligence)) ↗
  • arXiv cs.LG (Machine Learning) · EN New Model Releases
    Bayes-Adaptive MDPs for Rollout Allocation to Enhance GRPO Training
    Retrieval-Augmented Generation (RAG)
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • Publickey · JA New Model Releases
    JavaScriptランタイムの「Bun」がPrettier互換で300倍以上高速なフォーマッタに。Prettier設定ファイルもそのまま利用可能
    Bun v1.4.4 adds a Prettier-compatible formatter 300x+ faster
    Anthropic
    Bun, the JavaScript runtime developed by Anthropic, shipped a Prettier-compatible formatter in v1.4.4, released October 22, 2026. It reportedly runs over 300 times faster than Prettier and reads existing Prettier config files as-is.
    Read original (Publickey) ↗
  • arXiv cs.LG (Machine Learning) · EN Inference & Efficiency
    Learning Quality-Controllable Robot Policies with Iterative Hindsight Relabeling
    Inference Reinforcement Learning
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    LinMix: Data Mixing as Linear Domain Attribution
    Neural Network
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.LG (Machine Learning) · EN New Model Releases
    Channel Selection for Mixture-of-Experts Decoding Needs No Calibration Data
    Mixture of Experts (MoE) Neural Network Retrieval-Augmented Generation (RAG)
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.LG (Machine Learning) · EN Multimodal
    PDRVLA: A Vision-Language-Action Model with Physical Dynamics Representations
    Computer Vision Neural Network Reinforcement Learning
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.LG (Machine Learning) · EN Infrastructure & Hardware
    Agent-Conditioned Routing: Diverse Reasoning Paths Within One Frozen Model
    Fine-tuning GPT Mixture of Experts (MoE)
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.LG (Machine Learning) · EN New Model Releases
    The Geometry of Manifold Learning: Singular foliations via Deep Learning models
    Algorithms & Theory Deep Learning Neural Network
    Read original (arXiv cs.LG (Machine Learning)) ↗
  • arXiv cs.CL (Computation and Language) · EN Inference & Efficiency
    EmbeddingGemma 2: Unified and Efficient Multimodal Representations
    Embeddings Quantization
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • Sakana AI Blog (ja) · JA Industry Adoption
    Sakana API Platform、エンタープライズ向けにアーリーアクセスプログラムを開始
    Sakana AI opens early access to its enterprise Sakana API Platform
    Neural Network OpenAI
    Sakana AI announced Sakana API Platform for enterprises and opened an early access program. It offers the company's LLMs via API with admin controls to manage members and usage at the organization and workspace level. The API is OpenAI-compatible, so existing apps only need to switch endpoints. Sakana Fugu and Fugu Ultra come first, with Namazu and Fugu Max to follow.
    Read original (Sakana AI Blog (ja)) ↗
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    UCVE: Graph Structured Retrieval for Compliance Validation
    Computer Vision Retrieval-Augmented Generation (RAG) Reinforcement Learning
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.CL (Computation and Language) · EN Inference & Efficiency
    Watermark Basins: Black-Box Detection of Distilled LLMs
    Fine-tuning
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    Uncovering Shortcut Amplification in Video RLVR through the Lens of Perceptual Uncertainty Collapse
    Reinforcement Learning Software Engineering
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.CL (Computation and Language) · EN Inference & Efficiency
    Self-Regularized Flow Language Models
    Inference Retrieval-Augmented Generation (RAG) Reinforcement Learning
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.CL (Computation and Language) · EN New Model Releases
    Text Generation in Low-Resource Languages Using Human-Curated Vocabularies
    Retrieval-Augmented Generation (RAG) Reinforcement Learning Speech Processing
    Read original (arXiv cs.CL (Computation and Language)) ↗
  • arXiv cs.CL (Computation and Language) · EN Training & Fine-tuning
    Hierarchical Low-Rank Reparameterization for Parameter-Efficient
    Fine-tuning Llama Retrieval-Augmented Generation (RAG)
    Read original (arXiv cs.CL (Computation and Language)) ↗