Skip to content

Popular Terms

Community-favorited, most-read and recently-added AI terms.

⭐ Community Favorites

Most upvoted terms

Not enough votes yet.

🔥 Most Read

All-time view leaders

  1. 1🧮

    Aggregate Table

    2,431

    A table structure that stores summarized results derived from detailed data in order to accelerate analytical queries.

  2. 2➡️

    Autoregressive Decoding

    2,337

    A generation mode in which the model produces output token by token using previous outputs as context.

  3. 3🏥

    Domain-Adaptive Fine-Tuning

    2,333

    An approach that adapts a model to the terminology and usage style of specific domains such as law, healthcare, or finance.

  4. 4📈

    Mean Average Precision

    2,272

    A core evaluation metric that summarizes object detection performance across classes and threshold settings.

  5. 5📍

    Embedding

    2,260

    A learned dense vector representation that carries the meaning of a word, document, image, or another entity.

  6. 6➡️

    Vector

    2,253

    A quantity with direction and magnitude, and one of the most fundamental representations in linear algebra.

  7. 7⚡

    Speculative Decoding

    2,241

    A decoding approach that speeds up generation by validating proposals from a smaller fast model with a larger model.

  8. 8🧠

    Mixture of Experts

    2,239

    An approach in which only relevant expert subnetworks are activated for each input to achieve scale and efficiency.

  9. 9📉

    Post-Training Quantization

    2,232

    A quantization approach that reduces a pretrained model to lower-bit precision to gain memory and speed benefits.

  10. 10🛑

    Abstention

    2,219

    The ability of a model to avoid fabricating certainty and instead decline or express uncertainty when it is not confident.

  11. 11⚖️

    AIC

    2,206

    An information criterion that supports model selection by balancing model fit against model complexity.

  12. 12🔮

    Action Anticipation

    2,206

    A task that attempts to predict a future action from a partially observed video stream before it fully unfolds.

📈 Trending

Highest views-per-day in last 60 days

No new terms in last 60 days.

🆕 Recently Added

Newest published terms

  1. 1🖼️

    Text-to-Image Generation

    01.04.2026

    A generative modeling approach that synthesizes new images from natural language prompts.

  2. 2🧠

    Multimodal Transformer

    01.04.2026

    A model design that processes different data types such as text, images, audio, or video within a shared attention architecture.

  3. 3📏

    Uncertainty Calibration

    01.04.2026

    A quality approach aimed at making model confidence better aligned with actual correctness.

  4. 4🛑

    Abstention

    01.04.2026

    The ability of a model to avoid fabricating certainty and instead decline or express uncertainty when it is not confident.

  5. 5📌

    Factuality

    01.04.2026

    A quality dimension describing how well generated content aligns with real-world facts, source data, or verifiable truth.

  6. 6🌫️

    Hallucination

    01.04.2026

    The phenomenon in which a model generates fluent but unsupported or incorrect content.

  7. 7🏆

    Reward Model

    01.04.2026

    An auxiliary model that estimates how preferable generated outputs are and provides signals for alignment.

  8. 8📜

    Constitutional AI

    01.04.2026

    An alignment approach that tries to guide model behavior through explicit principle sets and normative rules.

  9. 9⚖️

    Direct Preference Optimization

    01.04.2026

    A simpler alignment approach that learns directly from preference pairs.

  10. 10🎛️

    Reinforcement Learning from Human Feedback

    01.04.2026

    An alignment approach that uses reward signals to make model outputs more consistent with human preferences.

  11. 11✅

    Verification Loop

    01.04.2026

    A workflow pattern that attempts to validate model output through additional checks, source review, or second-stage verification.

  12. 12🔗

    Citation Grounding

    01.04.2026

    An approach that improves trust by explicitly showing the source passages supporting the generated answer.