Papers by Kilian Weinberger
Long-term Control for Dialogue Generation: Methods and Evaluation (2022.naacl-main)
Copied to clipboard
| Challenge: | Current approaches for controlling dialogue response generation focus on high-level attributes like style, sentiment, or topic. |
| Approach: | They propose a method that allows for more fine-grained control of dialogue response generation . they propose utterances that encourage the generation of control words in the future . |
| Outcome: | The proposed method outperforms state-of-the-art constrained generation baselines on task-oriented dialogue datasets and shows that it is more fine-grained than previous methods. |
Diffusion Guided Language Modeling (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing guidance methods for text generation are prone to decoding errors and degrade performance. |
| Approach: | They propose a model that steers an auto-regressive language model to generate text with desired properties. |
| Outcome: | The proposed model outperforms existing guidance methods on a wide range of benchmark data sets. |
Re-evaluating the Need for Visual Signals in Unsupervised Grammar Induction (2024.findings-naacl)
Copied to clipboard
Boyi Li, Rodolfo Corona, Karttikeya Mangalam, Catherine Chen, Daniel Flaherty, Serge Belongie, Kilian Weinberger, Jitendra Malik, Trevor Darrell, Dan Klein
| Challenge: | Recent studies show multimodal inputs can improve grammar induction, but weak textual baselines are needed for training. |
| Approach: | They use a fixed grammar family to compare multimodal grammar induction methods . they find multimodal inputs can improve grammar induction by grounding textual inputs to the visual world . |
| Outcome: | The proposed model outperforms weaker baselines on four benchmark datasets. |