Today for AI

EXPLORE / 06

Terms index

Specialized AI and software-development terms with concise explanations

RECORDS

Entity records

3378 records · 52/85

  1. 2041

    Code Sandbox

    An isolated runtime environment used to safely execute generated code to prevent malicious operations.

    Website pending
  2. 2042

    Code Interpreter

    A tool component allowing LLMs to generate and execute code to solve complex problems.

    Website pending
  3. 2043

    Model Hub

    Online platform for storing, versioning, and distributing pre-trained models

    Website pending
  4. 2044

    Multi-process Architecture

    Design running tasks in separate processes to enhance stability and isolation

    Website pending
  5. 2045

    Inference Server

    Software component that receives requests and runs models to generate outputs

    Website pending
  6. 2046

    Guardrail Model

    Auxiliary models designed to detect potential risks and safety violations in inputs or outputs

    Website pending
  7. 2047

    Predefined Policies

    Pre-set rules or constraints used to guide model behavioral boundaries.

    Website pending
  8. 2048

    Vocabulary

    Set of all possible input tokens defined by the model

    Website pending
  9. 2049

    Chat Completion

    A large model text generation interface paradigm based on turn-based conversations.

    Website pending
  10. 2050

    Reasoning LLM

    Large language models capable of logical deduction and multi-step planning

    Website pending
  11. 2051

    Hybrid Architecture

    Model design combining multiple neural components or attention mechanisms to balance efficiency and performance

    Website pending
  12. 2052

    Inference Standard

    Unified technical standards regulating model input/output formats and interaction protocols.

    Website pending
  13. 2053

    Video Diffusion Model

    Deep learning architecture that generates video frames by progressively denoising from random noise.

    Website pending
  14. 2054

    Differential Attention

    A technique that suppresses noise and enhances signals by computing the difference between two sets of attention scores.

    Website pending
  15. 2055

    GQA

    An attention mechanism where multiple query heads share a single set of key-value heads, balancing between MHA and MQA.

    Website pending
  16. 2056

    Asset Operations

    The process of monitoring, maintaining, and managing physical or digital assets

    Website pending
  17. 2057

    Agent Benchmark

    Standardized test sets used to evaluate AI agent performance on specific tasks

    Website pending
  18. 2058

    Open-source LLM

    Large language models with public weights allowing free community modification and use

    Website pending
  19. 2059

    On-policy Collection

    Action sequences generated by the current policy used to update model parameters

    Website pending
  20. 2060

    Dialect Processing

    NLU and generation techniques for specific regional language variants

    Website pending
  21. 2061

    On-policy

    Training method using data generated by the current policy being optimized

    Website pending
  22. 2062

    Text-to-Image Model

    Deep learning models that generate images corresponding to textual descriptions.

    Website pending
  23. 2063

    Visual Debugging

    Monitoring program execution status and data flow intuitively via graphical interfaces

    Website pending
  24. 2064

    Middleware

    Software layer between applications and backend services handling request transformation or logic.

    Website pending
  25. 2065

    Ablation Study

    Method evaluating component contributions by removing or modifying them in a model.

    Website pending
  26. 2066

    Stateful Environment

    A system environment that retains historical interaction records where the current state influences subsequent feedback.

    Website pending
  27. 2067

    UI Localization

    The process of adapting software interfaces to specific linguistic and cultural contexts.

    Website pending
  28. 2068

    ONNX Runtime

    Microsoft's cross-platform machine learning model accelerator

    Website pending
  29. 2069

    Partial Observability

    A scenario where an agent cannot fully access all environmental state information and must make decisions based on incomplete data.

    Website pending
  30. 2070

    One-shot Prompting

    A technique providing a single example or instruction to guide a model in completing a specific task.

    Website pending
  31. 2071

    Agent Skill

    Structured capability modules enabling AI agents to perform specific complex tasks

    Website pending
  32. 2072

    VRAM

    GPU memory determining model loading and computation limits

    Website pending
  33. 2073

    VRAM Optimization

    Techniques to reduce video memory usage during deep learning training

    Website pending
  34. 2074

    LLM Fine-tuning

    Further training a pre-trained LLM on specific data to adapt it for downstream tasks

    Website pending
  35. 2075

    GGML

    A lightweight C library for tensor operations in machine learning

    Website pending
  36. 2076

    On-Device Optimization

    Techniques like quantization and pruning to enable efficient local execution of large models

    Website pending
  37. 2077

    Asynchronous Training

    A training mode that decouples data generation from model updates to overlap computation wait times.

    Website pending
  38. 2078

    DeepSpeed

    Microsoft's open-source library for large model training optimization and distributed parallelism.

    Website pending
  39. 2079

    FlashAttention

    An IO-aware efficient attention algorithm that significantly reduces memory usage.

    Website pending
  40. 2080

    Sequence Parallelism

    A distributed strategy that splits input sequences across multiple devices for parallel computation.

    Website pending