Papers by Zineng Tang
Continuous Language Generative Flow (2021.acl-long)
Copied to clipboard
| Challenge: | Recent years have witnessed various types of generative models for natural language generation (NLG), especially RNNs or transformers. |
| Approach: | They propose a flow-based language generation model that adapts flow-derived generative models to language generation via continuous input embeddings, adapted affine coupling structures, and a novel architecture for autoregressive text generation. |
| Outcome: | The proposed model improves on QG and NMT and improves performance over baselines on SQuAD and TVQA and NML16. |
Dense-Caption Matching and Frame-Selection Gating for Temporal Localization in VideoQA (2020.acl-main)
Copied to clipboard
| Challenge: | Recent years have witnessed a paradigm shift in the way we get our information, and a lot of it. |
| Approach: | They propose a video question answering model which integrates multi-modal input sources and finds temporally relevant information to answer questions. |
| Outcome: | The proposed model outperforms the state-of-the-art on a TVQA dataset. |
Grounding Language in Multi-Perspective Referential Communication (2024.emnlp-main)
Copied to clipboard
| Challenge: | Using a dataset of 2,970 human-written referring expressions, we find that the performance of automated models in both reference generation and comprehension lags behind that of pairs of human agents. |
| Approach: | They propose a task and dataset for referring expression generation and comprehension in multi-agent embodied environments where two agents must take into account one another's visual perspective to produce and understand references to objects in a scene. |
| Outcome: | The proposed model outperforms the strongest proprietary model and improves communicative success from 58.9 to 69.3% when trained with a listener. |
DeCEMBERT: Learning from Noisy Instructional Videos via Dense Captions and Entropy Minimization (2021.naacl-main)
Copied to clipboard
| Challenge: | Existing methods to train models on unlabeled web videos are noisy and temporally misaligned . authors propose a method that adds captions and constrained attention loss to improve performance . |
| Approach: | They propose a method that adds captions from video frames as auxiliary text input to provide visual cues for learning better video and language associations. |
| Outcome: | The proposed method outperforms state-of-the-art methods on video-and-language tasks . it adds captions and constrained attention loss to improve model performance . |