Damai Dai


pdf bib
Why Can GPT Learn In-Context? Language Models Secretly Perform Gradient Descent as Meta-Optimizers
Damai Dai | Yutao Sun | Li Dong | Yaru Hao | Shuming Ma | Zhifang Sui | Furu Wei
Findings of the Association for Computational Linguistics: ACL 2023

Large pretrained language models have shown surprising in-context learning (ICL) ability. With a few demonstration input-label pairs, they can predict the label for an unseen input without parameter updates. Despite the great success in performance, its working mechanism still remains an open question. In this paper, we explain language models as meta-optimizers and understand in-context learning as implicit finetuning. Theoretically, we figure out that Transformer attention has a dual form of gradient descent. On top of it, we understand ICL as follows: GPT first produces meta-gradients according to the demonstration examples, and then these meta-gradients are applied to the original GPT to build an ICL model. We comprehensively compare the behaviors of in-context learning and explicit finetuning on real tasks to provide empirical evidence that supports our understanding. Experimental results show that in-context learning behaves similarly to explicit finetuning from multiple perspectives. Inspired by the dual form between Transformer attention and gradient descent, we design a momentum-based attention by analogy with gradient descent with momentum. The improved performance over vanilla attention further supports our understanding from another perspective, and more importantly, shows the potential to utilize our understanding for future model design. The code is available at https://aka.ms/icl.

pdf bib
Bi-Drop: Enhancing Fine-tuning Generalization via Synchronous sub-net Estimation and Optimization
Shoujie Tong | Heming Xia | Damai Dai | Runxin Xu | Tianyu Liu | Binghuai Lin | Yunbo Cao | Zhifang Sui
Findings of the Association for Computational Linguistics: EMNLP 2023

Pretrained language models have achieved remarkable success in natural language understanding. However, fine-tuning pretrained models on limited training data tends to overfit and thus diminish performance. This paper presents Bi-Drop, a fine-tuning strategy that selectively updates model parameters using gradients from various sub-nets dynamically generated by dropout. The sub-net estimation of Bi-Drop is performed in an in-batch manner, so it overcomes the problem of hysteresis in sub-net updating, which is possessed by previous methods that perform asynchronous sub-net estimation. Also, Bi-Drop needs only one mini-batch to estimate the sub-net so it achieves higher utility of training data. Experiments on the GLUE benchmark demonstrate that Bi-Drop consistently outperforms previous fine-tuning methods. Furthermore, empirical results also show that Bi-Drop exhibits excellent generalization ability and robustness for domain transfer, data imbalance, and low-resource scenarios.

pdf bib
Not All Demonstration Examples are Equally Beneficial: Reweighting Demonstration Examples for In-Context Learning
Zhe Yang | Damai Dai | Peiyi Wang | Zhifang Sui
Findings of the Association for Computational Linguistics: EMNLP 2023

Large Language Models (LLMs) have recently gained the In-Context Learning (ICL) ability with the models scaling up, allowing them to quickly adapt to downstream tasks with only a few demonstration examples prepended in the input sequence. Nonetheless, the current practice of ICL treats all demonstration examples equally, which still warrants improvement, as the quality of examples is usually uneven. In this paper, we investigate how to determine approximately optimal weights for demonstration examples and how to apply them during ICL. To assess the quality of weights in the absence of additional validation data, we design a masked self-prediction (MSP) score that exhibits a strong correlation with the final ICL performance. To expedite the weight-searching process, we discretize the continuous weight space and adopt beam search. With approximately optimal weights obtained, we further propose two strategies to apply them to demonstrations at different model positions. Experimental results on 8 text classification tasks show that our approach outperforms conventional ICL by a large margin. Our code are publicly available at https:github.com/Zhe-Young/WICL.

pdf bib
Label Words are Anchors: An Information Flow Perspective for Understanding In-Context Learning
Lean Wang | Lei Li | Damai Dai | Deli Chen | Hao Zhou | Fandong Meng | Jie Zhou | Xu Sun
Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing

In-context learning (ICL) emerges as a promising capability of large language models (LLMs) by providing them with demonstration examples to perform diverse tasks. However, the underlying mechanism of how LLMs learn from the provided context remains under-explored. In this paper, we investigate the working mechanism of ICL through an information flow lens. Our findings reveal that label words in the demonstration examples function as anchors: (1) semantic information aggregates into label word representations during the shallow computation layers’ processing; (2) the consolidated information in label words serves as a reference for LLMs’ final predictions. Based on these insights, we introduce an anchor re-weighting method to improve ICL performance, a demonstration compression technique to expedite inference, and an analysis framework for diagnosing ICL errors in GPT2-XL. The promising applications of our findings again validate the uncovered ICL working mechanism and pave the way for future studies.

pdf bib
Denoising Bottleneck with Mutual Information Maximization for Video Multimodal Fusion
Shaoxiang Wu | Damai Dai | Ziwei Qin | Tianyu Liu | Binghuai Lin | Yunbo Cao | Zhifang Sui
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

Video multimodal fusion aims to integrate multimodal signals in videos, such as visual, audio and text, to make a complementary prediction with multiple modalities contents. However, unlike other image-text multimodal tasks, video has longer multimodal sequences with more redundancy and noise in both visual and audio modalities. Prior denoising methods like forget gate are coarse in the granularity of noise filtering. They often suppress the redundant and noisy information at the risk of losing critical information. Therefore, we propose a denoising bottleneck fusion (DBF) model for fine-grained video multimodal fusion. On the one hand, we employ a bottleneck mechanism to filter out noise and redundancy with a restrained receptive field. On the other hand, we use a mutual information maximization module to regulate the filter-out module to preserve key information within different modalities. Our DBF model achieves significant improvement over current state-of-the-art baselines on multiple benchmarks covering multimodal sentiment analysis and multimodal summarization tasks. It proves that our model can effectively capture salient features from noisy and redundant video, audio, and text inputs. The code for this paper will be publicly available at https://github.com/WSXRHFG/DBF


pdf bib
StableMoE: Stable Routing Strategy for Mixture of Experts
Damai Dai | Li Dong | Shuming Ma | Bo Zheng | Zhifang Sui | Baobao Chang | Furu Wei
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

The Mixture-of-Experts (MoE) technique can scale up the model size of Transformers with an affordable computational overhead. We point out that existing learning-to-route MoE methods suffer from the routing fluctuation issue, i.e., the target expert of the same input may change along with training, but only one expert will be activated for the input during inference. The routing fluctuation tends to harm sample efficiency because the same input updates different experts but only one is finally used. In this paper, we propose StableMoE with two training stages to address the routing fluctuation problem. In the first training stage, we learn a balanced and cohesive routing strategy and distill it into a lightweight router decoupled from the backbone model. In the second training stage, we utilize the distilled router to determine the token-to-expert assignment and freeze it for a stable routing strategy. We validate our method on language modeling and multilingual machine translation. The results show that StableMoE outperforms existing MoE methods in terms of both convergence speed and performance.

pdf bib
Knowledge Neurons in Pretrained Transformers
Damai Dai | Li Dong | Yaru Hao | Zhifang Sui | Baobao Chang | Furu Wei
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

Large-scale pretrained language models are surprisingly good at recalling factual knowledge presented in the training corpus. In this paper, we present preliminary studies on how factual knowledge is stored in pretrained Transformers by introducing the concept of knowledge neurons. Specifically, we examine the fill-in-the-blank cloze task for BERT. Given a relational fact, we propose a knowledge attribution method to identify the neurons that express the fact. We find that the activation of such knowledge neurons is positively correlated to the expression of their corresponding facts. In our case studies, we attempt to leverage knowledge neurons to edit (such as update, and erase) specific factual knowledge without fine-tuning. Our results shed light on understanding the storage of knowledge within pretrained Transformers.

pdf bib
Hierarchical Curriculum Learning for AMR Parsing
Peiyi Wang | Liang Chen | Tianyu Liu | Damai Dai | Yunbo Cao | Baobao Chang | Zhifang Sui
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)

Abstract Meaning Representation (AMR) parsing aims to translate sentences to semantic representation with a hierarchical structure, and is recently empowered by pretrained sequence-to-sequence models. However, there exists a gap between their flat training objective (i.e., equally treats all output tokens) and the hierarchical AMR structure, which limits the model generalization. To bridge this gap, we propose a Hierarchical Curriculum Learning (HCL) framework with Structure-level (SC) and Instance-level Curricula (IC). SC switches progressively from core to detail AMR semantic elements while IC transits from structure-simple to -complex AMR instances during training. Through these two warming-up processes, HCL reduces the difficulty of learning complex structures, thus the flat model can better adapt to the AMR hierarchy. Extensive experiments on AMR2.0, AMR3.0, structure-complex and out-of-distribution situations verify the effectiveness of HCL.

pdf bib
Calibrating Factual Knowledge in Pretrained Language Models
Qingxiu Dong | Damai Dai | Yifan Song | Jingjing Xu | Zhifang Sui | Lei Li
Findings of the Association for Computational Linguistics: EMNLP 2022

Previous literature has proved that Pretrained Language Models (PLMs) can store factual knowledge. However, we find that facts stored in the PLMs are not always correct. It motivates us to explore a fundamental question: How do we calibrate factual knowledge in PLMs without re-training from scratch? In this work, we propose a simple and lightweight method CaliNet to achieve this goal. To be specific, we first detect whether PLMs can learn the right facts via a contrastive score between right and fake facts. If not, we then use a lightweight method to add and adapt new parameters to specific factual texts. Experiments on the knowledge probing task show the calibration effectiveness and efficiency. In addition, through closed-book question answering, we find that the calibrated PLM possesses knowledge generalization ability after finetuning.Beyond the calibration performance, we further investigate and visualize the knowledge calibration mechanism.


pdf bib
Decompose, Fuse and Generate: A Formation-Informed Method for Chinese Definition Generation
Hua Zheng | Damai Dai | Lei Li | Tianyu Liu | Zhifang Sui | Baobao Chang | Yang Liu
Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies

In this paper, we tackle the task of Definition Generation (DG) in Chinese, which aims at automatically generating a definition for a word. Most existing methods take the source word as an indecomposable semantic unit. However, in parataxis languages like Chinese, word meanings can be composed using the word formation process, where a word (“桃花”, peach-blossom) is formed by formation components (“桃”, peach; “花”, flower) using a formation rule (Modifier-Head). Inspired by this process, we propose to enhance DG with word formation features. We build a formation-informed dataset, and propose a model DeFT, which Decomposes words into formation features, dynamically Fuses different features through a gating mechanism, and generaTes word definitions. Experimental results show that our method is both effective and robust.

pdf bib
基于词信息嵌入的汉语构词结构识别研究(Chinese Word-Formation Prediction based on Representations of Word-Related Features)
Hua Zheng (郑婳) | Yaqi Yan (殷雅琦) | Yue Wang (王悦) | Damai Dai (代达劢) | Yang Liu (刘扬)
Proceedings of the 20th Chinese National Conference on Computational Linguistics


pdf bib
Inductively Representing Out-of-Knowledge-Graph Entities by Optimal Estimation Under Translational Assumptions
Damai Dai | Hua Zheng | Fuli Luo | Pengcheng Yang | Tianyu Liu | Zhifang Sui | Baobao Chang
Proceedings of the 6th Workshop on Representation Learning for NLP (RepL4NLP-2021)

Conventional Knowledge Graph Completion (KGC) assumes that all test entities appear during training. However, in real-world scenarios, Knowledge Graphs (KG) evolve fast with out-of-knowledge-graph (OOKG) entities added frequently, and we need to efficiently represent these entities. Most existing Knowledge Graph Embedding (KGE) methods cannot represent OOKG entities without costly retraining on the whole KG. To enhance efficiency, we propose a simple and effective method that inductively represents OOKG entities by their optimal estimation under translational assumptions. Moreover, given pretrained embeddings of the in-knowledge-graph (IKG) entities, our method even needs no additional learning. Experimental results on two KGC tasks with OOKG entities show that our method outperforms the previous methods by a large margin with higher efficiency.

pdf bib
Leveraging Word-Formation Knowledge for Chinese Word Sense Disambiguation
Hua Zheng | Lei Li | Damai Dai | Deli Chen | Tianyu Liu | Xu Sun | Yang Liu
Findings of the Association for Computational Linguistics: EMNLP 2021

In parataxis languages like Chinese, word meanings are constructed using specific word-formations, which can help to disambiguate word senses. However, such knowledge is rarely explored in previous word sense disambiguation (WSD) methods. In this paper, we propose to leverage word-formation knowledge to enhance Chinese WSD. We first construct a large-scale Chinese lexical sample WSD dataset with word-formations. Then, we propose a model FormBERT to explicitly incorporate word-formations into sense disambiguation. To further enhance generalizability, we design a word-formation predictor module in case word-formation annotations are unavailable. Experimental results show that our method brings substantial performance improvement over strong baselines.


pdf bib
Learning to Control the Fine-grained Sentiment for Story Ending Generation
Fuli Luo | Damai Dai | Pengcheng Yang | Tianyu Liu | Baobao Chang | Zhifang Sui | Xu Sun
Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics

Automatic story ending generation is an interesting and challenging task in natural language generation. Previous studies are mainly limited to generate coherent, reasonable and diversified story endings, and few works focus on controlling the sentiment of story endings. This paper focuses on generating a story ending which meets the given fine-grained sentiment intensity. There are two major challenges to this task. First is the lack of story corpus which has fine-grained sentiment labels. Second is the difficulty of explicitly controlling sentiment intensity when generating endings. Therefore, we propose a generic and novel framework which consists of a sentiment analyzer and a sentimental generator, respectively addressing the two challenges. The sentiment analyzer adopts a series of methods to acquire sentiment intensities of the story dataset. The sentimental generator introduces the sentiment intensity into decoder via a Gaussian Kernel Layer to control the sentiment of the output. To the best of our knowledge, this is the first endeavor to control the fine-grained sentiment for story ending generation without manually annotating sentiment labels. Experiments show that our proposed framework can generate story endings which are not only more coherent and fluent but also able to meet the given sentiment intensity better.