Papers with graphical models
On the Rejection Criterion for Proxy-based Test-time Alignment (2026.acl-short)
Copied to clipboard
| Challenge: | Recent work suggests that test-time alignment methods rely on a small aligned model as a proxy that guides the generation of a larger base model. |
| Approach: | They propose a rejection criterion based on a conservative confidence bet for test-time alignment methods that use a small aligned model as a proxy to guide the generation of a larger base model. |
| Outcome: | The proposed approach outperforms previous work on several datasets. |
Joint Learning of Representations for Web-tables, Entities and Types using Graph Convolutional Network (2021.eacl-main)
Copied to clipboard
| Challenge: | Existing approaches for table annotation with entities and types capture the syntactic structure of tables using graphical models or learn embeddings of table entries without accounting for the complete syntaktic structure. |
| Approach: | They propose a Graph Convolutional Network that captures the complete structure of tables, knowledge graph and the training annotations and jointly learns embeddings for table elements as well as the entities and types. |
| Outcome: | The proposed model significantly outperforms state-of-the-art methods on 5 benchmark datasets while showing promising performance on downstream table-related applications. |
Neural Sparse Topical Coding (P18-1)
Copied to clipboard
| Challenge: | Topic models with sparsity enhancement are effective at learning discriminative and coherent latent topics of short texts. |
| Approach: | They propose a novel sparsity-enhanced topic model with back propagation that replaces the inference process with the back propagations, making it easy to explore extensions. |
| Outcome: | The proposed model outperforms existing methods on Web Snippet and 20Newsgroups datasets. |
On the Inductive Bias of Masked Language Modeling: From Statistical to Syntactic Dependencies (2021.naacl-main)
Copied to clipboard
| Challenge: | Recent studies suggest that pretrained language models acquire useful inductive biases through masks that implicitly act as cloze reductions for downstream tasks. |
| Approach: | They construct cloze-like masks using task-specific lexicons to explain their results . they show that the majority of performance gains come from generic masks that are not associated with the lexical . |
| Outcome: | The proposed method outperforms a classic method for unsupervised parsing. |
Training for Gibbs Sampling on Conditional Random Fields with Neural Scoring Factors (2020.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in NLP focus on simple approaches to model the output label space . graphical models are often limited to (heuristic) greedy search and its variants . |
| Approach: | They propose an approach for efficiently training and decoding hybrids of graphical and graphical models based on Gibbs sampling. |
| Outcome: | The proposed approach improves on Dutch and Dutch with graphical models . the proposed model improves over a strong baseline on three languages . |