Browse › Natural Language Processing › Multimodal Text Prediction › MultiSubs
MultiSubs Benchmark (Multimodal Text Prediction)
Multimodal text prediction is a type of natural language processing that involves predicting the next word or sequence of words in a sentence, given multiple modalities or types of input. In traditional text prediction, the prediction is based solely on the context of the sentence, such as the words that precede the target word. In multimodal text prediction, additional modalities, such as images, audio, or user behavior, are also used to inform the prediction.
For example, in a multimodal text prediction system for captioning images, the system may use both the content of the image and the words that have been typed so far to generate the next word in the caption. The image may provide additional context or information about the content of the caption, while the typed words may provide information about the style or tone of the caption.
Multimodal text prediction can be achieved using a variety of techniques, including deep learning models and statistical models. These models can be trained on large datasets of text and multimodal inputs to learn the relationships between the different types of data and improve the accuracy of the predictions.
Multimodal text prediction has many applications, including chatbots, virtual assistants, and predictive text input for mobile devices. By incorporating additional modalities into the prediction process, multimodal text prediction systems can provide more accurate and useful predictions, improving the overall user experience.
The archive carries no text for this table; the description above is the archive's text for the task Multimodal Text Prediction. archive 2025-07-28
Over time archive 2025-07-28
The chart needs JavaScript; the table below carries every value.
Direction inferred from the metric name, not from the archive: Accuracy (higher is better). Not inferred (points only, no best-so-far line): Word similarity. Points are placed at the row's paper date; 1 of 1 rows carry one.
Results archive 2025-07-28
Archive rows end at the archive snapshot, 2025-07-28: no result published after that date is in this table. Rank is the archive's row order at that snapshot; not re-ranked here. Metric values are the archive's strings. Column headers sort the table in your browser; each row keeps its archive rank.
| Paper | Code | Ran Syntology | Report | ||||||
|---|---|---|---|---|---|---|---|---|---|
| 1 | 9-gram LM with back-off | 30.35 | 0.44 | – | Paper | Code | 2021 | linked, not harvested | report |
All 1 rows shown. 1 link to a paper page on this site; 0 are marked as using additional training data in the archive. No GitHub stars are tracked; "Code" is the first repository the archive lists for the row. The archive carries no row tags, review links or community-submitted rows for this table; none are shown. archive 2025-07-28
Syntology Ran reads "N of M ran · U unverified": of the M code samples Syntology harvested from repositories linked to that row's paper (joined by arXiv id), N executed on a synthesized input and the other U = M−N are unverified (harvested, no recorded run). It counts code from repositories linked to that row's paper, not this result: the row's number was not reproduced and nothing here is a correctness claim. The other cell texts mean no graph line for the row: "linked, not harvested" (the archive links code, Syntology has not harvested it), "no code linked" (no code link in the archive), "not matched" (the row's paper URL matched no paper on this site). 0 rows have a graph line, from 0 distinct papers; 0 rows (0 papers) have at least one sample that ran. Counting each paper once: Syntology ran 0 of 0 samples; 0 unverified. Separately, 0 of those 0 are pointer-only (licence): the site points at that code rather than redistributing it, a licence property recorded for ran and unverified samples alike; each cell's tooltip carries the row's own pointer-only count. Read from the graph 2026-09-24. Per-sample status is on the paper page.
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections