SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 2005.04560 · ACL · 2020

Posterior Control of Blackbox Generation

Alexander Rush, Li, Lisa Xiang

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 6 functions out of this paper's own repositories and ran 4 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
FranxYao/Gumbel-CRF — 4 of 6
FunctionStatusWhere it lives
Attention Ran FranxYao/Gumbel-CRF/src/modeling/latent_temp_crf_ar.py
pointer only (licence: NONE) · get_code("29319ce69524bccb")
LSTMDecoder Ran FranxYao/Gumbel-CRF/src/modeling/latent_temp_crf_ar.py
pointer only (licence: NONE) · get_code("04cfea23995c7759")
LSTMEncoder Ran FranxYao/Gumbel-CRF/src/modeling/latent_temp_crf_ar.py
pointer only (licence: NONE) · get_code("8b3d28ed401c09b5")
attention Ran FranxYao/Gumbel-CRF/src/modeling/latent_temp_crf_ar.py
pointer only (licence: NONE) · get_code("5dc10625fad08447")
LatentTemplateCRFAR Not yet run FranxYao/Gumbel-CRF/src/modeling/latent_temp_crf_ar.py
pointer only (licence: NONE) · get_code("23c4c07ee6a8b02b")
LinearChainCRF Not yet run FranxYao/Gumbel-CRF/src/modeling/latent_temp_crf_ar.py
pointer only (licence: NONE) · get_code("6e9286da02c5bc13")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

Text generation often requires high-precision output that obeys task-specific rules. This fine-grained control is difficult to enforce with off-the-shelf deep learning models. In this work, we consider augmenting neural generation models with discrete control states learned through a structured latent-variable approach. Under this formulation, task-specific knowledge can be encoded through a range of rich, posterior constraints that are effectively trained into the model. This approach allows users to ground internal model decisions based on prior knowledge, without sacrificing the representational power of neural generative models. Experiments consider applications of this approach for text generation. We find that this method improves over standard benchmarks, while also providing fine-grained control.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("2005.04560")
get_code_for_paper("2005.04560")
have("2005.04560")

Connect an agent — have() is free.