AdaRNN

Closed weights Beihang University 13K parameters June 2014

No estimate

No hardware requirements for this model

The weights for this model have not been published, so it cannot be downloaded or run on your own hardware at any size. It is reachable only through its provider, and no graphics card changes that.

On record

Full specification

Everything on record for this model. Most of it describes how it was trained rather than how it runs — useful context for judging how much work went into it, and how it compares with models built at a different scale.

Origin

Who built this model, where, and when it was published.

Organisation
Beihang University
Organisation type
Academia
Country
China
Published
1 June 2014
Authors
Li Dong, Furu Wei, Chuanqi Tan, Duyu Tang, M. Zhou, Ke Xu

What it does

The problem areas the model was built for. A model can carry several of each.

Domain
Language
Task
Sentiment classification

Size

How large the model is and how much data it was trained on. Parameters are the figure that decides whether it fits on a given graphics card.

Parameters
13K

D=25 "For recursive neural models, the dimension of word vector is set to 25, and f = tanh is used as the nonlinearity function. We employ 10 composition matrices in AdaRNN." Composition matrices: "W ∈ R D×2D is the composition matrix, and b is the bias vector." C=10 "We employ 10 composition matrices in AdaRNN." Combination matrix: "S ∈ R C×(2D+|e|) is the matrix used to determine which composition function we use, vl , vr are the left and right child vectors, and e are external feature vector.…

Training data
6,248 tokens

"Training data consists of 6,248 tweets,"

How it is classified

Labels the source dataset applies when tracking notable models, and how confident it is in the entry.

Why it is tracked
Highly cited
Record confidence
Confident

Sources

Where this record came from and when it was last checked.

Reference
Adaptive Recursive Neural Network for Target-dependent Twitter Sentiment Classification
Last updated
28 November 2025

What the numbers mean

About this model

AdaRNN was published by Beihang University, in China, in June 2014. The organisation is categorised as academia.

It works in Language, and is recorded as doing sentiment classification.

Its weights were never published, so it can only be reached through its provider. No graphics card changes that.

What went into building it

Around 6,248 tokens went into training it.

Its inclusion criterion is highly cited.

Answers

AdaRNN — common questions

01

When was AdaRNN released?

AdaRNN was published in June 2014. Capability per parameter has improved considerably since, so a newer model of the same size is often the better use of the same hardware.

02

What is AdaRNN used for?

AdaRNN works in Language, and is recorded as handling sentiment classification. A model can carry several of each, so these are the areas it was built for rather than a limit on what it will attempt.

03

What GPU do I need to run AdaRNN?

None. AdaRNN is a closed model — its weights were never published, so it cannot be downloaded or run on your own hardware at any price. It is reachable only through its provider.

04

Is AdaRNN open source?

The licensing for AdaRNN was never recorded in our source data. We treat unstated licensing as closed, because an unrecorded licence is not one to rely on.

05

How many parameters does AdaRNN have?

AdaRNN has 13K parameters. D=25 "For recursive neural models, the dimension of word vector is set to 25, and f = tanh is used as the nonlinearity function. We employ 10 composition matrices in AdaRNN." Composition matrices: "W ∈ R D×2D is the composition matrix, and b is the bias vector." C=10 "We employ 10 composition matrices in AdaRNN." Combination matrix: "S ∈ R C×(2D+|e|) is the matrix used to determine which composition function we use, vl , vr are the left and right child vectors, and e are external feature vector. In this work, e is a one-hot binary feature vector which indicates what the dependency type is." |e| > 4: (see Figure 2) Weights: 10 * 25 * 50 + 10 * (50+4) =13040 (ignoring embedding to the 25 dimension embedding space). That figure is the total, and it is what decides how much memory the model needs — roughly half a gigabyte per billion at the compression most people use.

06

Who created AdaRNN?

AdaRNN was published by Beihang University, based in China, categorised as academia.

Source

Original publication

Record last updated 28 November 2025

The other direction

Looking at it from the other side?

This page starts from the model. If you already own a card and want to know everything it will run, start from the hardware instead.