Reinforcement learning on a reward model trained from human summary comparisons produces summaries humans prefer over supervised fine-tuning or human references on TL;DR and transfers to CNN/DM.
ProphetNet: Predicting Future N-gram for Sequence-to-Sequence Pre-training
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
abstract
This paper presents a new sequence-to-sequence pre-training model called ProphetNet, which introduces a novel self-supervised objective named future n-gram prediction and the proposed n-stream self-attention mechanism. Instead of optimizing one-step-ahead prediction in the traditional sequence-to-sequence model, the ProphetNet is optimized by n-step ahead prediction that predicts the next n tokens simultaneously based on previous context tokens at each time step. The future n-gram prediction explicitly encourages the model to plan for the future tokens and prevent overfitting on strong local correlations. We pre-train ProphetNet using a base scale dataset (16GB) and a large-scale dataset (160GB), respectively. Then we conduct experiments on CNN/DailyMail, Gigaword, and SQuAD 1.1 benchmarks for abstractive summarization and question generation tasks. Experimental results show that ProphetNet achieves new state-of-the-art results on all these datasets compared to the models using the same scale pre-training corpus.
representative citing papers
Multi-token prediction accounts for nearly all rollout stability gains on synthetic three-component seismograms, with sharp dependence on context covering the full P-S interval and magnitude-based losses unable to prevent polarity flips.
Prefixing BART encoder input with bucketized CTR and length control tokens lets a single fine-tuned model generate ad headlines with controllable length and higher estimated CTR than prior baselines.
citing papers explorer
-
Learning to summarize from human feedback
Reinforcement learning on a reward model trained from human summary comparisons produces summaries humans prefer over supervised fine-tuning or human references on TL;DR and transfers to CNN/DM.
-
When Do Autoregressive Sequence Models Forecast Physical Wavefields? A Controlled Study on Synthetic Seismograms
Multi-token prediction accounts for nearly all rollout stability gains on synthetic three-component seismograms, with sharp dependence on context covering the full P-S interval and magnitude-based losses unable to prevent polarity flips.
-
COBART: Controlled, Optimized, Bidirectional and Auto-Regressive Transformer for Ad Headline Generation
Prefixing BART encoder input with bucketized CTR and length control tokens lets a single fine-tuned model generate ad headlines with controllable length and higher estimated CTR than prior baselines.