CUNY-CL/yoyodyne

Non-autoregressive sequence generation

Open

#262 opened on Oct 25, 2024

View on GitHub
 (0 comments) (0 reactions) (1 assignee)Python (21 forks)auto 404
enhancementgood first issuenew architecture

Repository metrics

Stars
 (35 stars)
PR merge metrics
 (PR metrics pending)

Description

Right now, our dominant paradigm is autoregressive prediction. While this is SOTA ,it doesn't allow experimentation with other sequence-to-sequence tasks. Technically, we can define an arbitrary sequence length >>>> than any input or output and train a model to project an input to this length and then decode over it for non-autoregressive prediction. Example algorithms would be: Listen Attend Spell or CTC.

Contributor guide

Research direction
Review non autoregressive sequence generation papers (e.g., CTC, Listen Attend Spell) and investigate integrating them into the existing sequence to sequence framework. Start by reading the current codebase to understand the autoregressive implementation, then prototype a non autoregressive decoder module.
Tech stack
python
Domain
machine learningai
Issue type
Feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
4
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
Over 1 week
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Fresh
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Mostly clear
Prerequisites
Pythondeep learning basicssequence to sequence models
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
20