Papers with GPT2
Transformer Feed-Forward Layers Build Predictions by Promoting Concepts in the Vocabulary Space (2022.emnlp-main)
Copied to clipboard
| Challenge: | Fig. 1 shows how feed-forward network (FFN) layers are utilized to build LMs. |
| Approach: | They reverse-engineer the operation of feed-forward network layers to find out how they work . they show that each update can be decomposed to sub-updates corresponding to single parameter vectors . |
| Outcome: | The proposed model reduces the toxicity of GPT2 by almost 50% and improves computation efficiency with a simple early exit rule, saving 20% of computation on average. |
Predict the Next Word: <Humans exhibit uncertainty in this task and language models _____> (2024.eacl-short)
Copied to clipboard
| Challenge: | Language models (LMs) are statistical models trained to assign probability to human-generated text. |
| Approach: | They evaluate language models' ability to reproduce variability that humans exhibit in the ‘next word prediction’ task. |
| Outcome: | The language models are trained to assign probability to human-generated text . they exhibit low calibration to human uncertainty, and advise against it . |
FastSeq: Make Sequence Generation Faster (2021.acl-demo)
Copied to clipboard
Yu Yan, Fei Hu, Jiusheng Chen, Nikhil Bhendawade, Ting Ye, Yeyun Gong, Nan Duan, Desheng Cui, Bingyu Chi, Ruofei Zhang
| Challenge: | Transformer-based models have made tremendous impact in natural language generation, but inference speed is still a bottleneck due to large model size and intensive computing involved in auto-regressive decoding process. |
| Approach: | They propose an attention cache optimization, an efficient algorithm for detecting repeated n-grams, and an asynchronous generation pipeline with parallel I/O to accelerate sequence generation without loss of accuracy. |
| Outcome: | The proposed framework can accelerate the sequence generation by 4x to 9x with a simple one-line code change for a set of widely used and diverse models. |
Exploring Semantics in Pretrained Language Model Attention (2024.starsem-1)
Copied to clipboard
| Challenge: | Abstract Meaning Representations (AMRs) encode the semantics of sentences in the form of graphs. |
| Approach: | They propose to use attention heads of two LMs to detect semantic relations encoded in AMRs. |
| Outcome: | The proposed models detect semantic relations without fine tuning, using both unsupervised and supervised learning techniques. |
Plausibility Processing in Transformer Language Models: Focusing on the Role of Attention Heads in GPT (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Using attention heads, we can explore how Transformer language models process semantic knowledge, especially regarding the plausibility of noun-verb relations. |
| Approach: | They propose to investigate how Transformer language models process semantic knowledge, especially regarding the plausibility of noun-verb relations. |
| Outcome: | The proposed model exhibits a higher degree of similarity with humans in plausibility processing compared to other Transformer language models. |
Language Models that Seek for Knowledge: Modular Search & Generation for Dialogue and Prompt Completion (2022.findings-emnlp)
Copied to clipboard
| Challenge: | Language models (LMs) have been shown to generate more factual responses by employing modularity in combination with retrieval. |
| Approach: | They extend the recent approach of Adolphs et al. (2021) to include internet search as a module. |
| Outcome: | The proposed method outperforms the state-of-the-art model BlenderBot 2 on open-domain knowledge-grounded conversations for the same number of parameters. |
QRelScore: Better Evaluating Generated Questions with Deeper Understanding of Context-aware Relevance (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing metrics for assessing question generation fail to take into account the input context of generation. |
| Approach: | They propose a context-aware Relevance evaluation metric for Question Generation that takes into account the context of question generation into account. |
| Outcome: | The proposed metric achieves higher correlation with human judgments while being much more robust to adversarial samples. |
A Generative Language Model for Few-shot Aspect-Based Sentiment Analysis (2022.findings-naacl)
Copied to clipboard
| Challenge: | Pre-trained language models are often used to achieve state-of-the-art results . eval paper shows that generative language model can handle joint and multi-task settings . |
| Approach: | They propose to reformulate extraction and prediction tasks into a sequence generation task . they propose a generative language model with unidirectional attention that learns to accomplish the tasks via language generation . |
| Outcome: | The proposed model outperforms the state-of-the-art in few-shot and full-shot settings. |
Rethinking Why Intermediate-Task Fine-Tuning Works (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Pretrained language models can be fine-tuned on intermediate labeled-data tasks before fine- tuning the models on the target task of interest. |
| Approach: | They conduct extensive experiments to study the impact of different factors on STILT . they find that the improvement from an intermediate task could be orthogonal to it containing reasoning or other complex skills. |
| Outcome: | The proposed method improves the performance of pretrained language models on various target tasks. |
Spelling convention sensitivity in neural language models (2023.findings-eacl)
Copied to clipboard
| Challenge: | Various long-distance dependencies have been investigated using neural language models. |
| Approach: | They examine whether large neural language models learn the long-distance dependency of British versus American spelling conventions . a large T5 language model does internalize consistency, but only with respect to observed lexical items . |
| Outcome: | The proposed model internalizes consistency with the training corpora, but only with respect to observed lexical items. |
Simulated Chats for Building Dialog Systems: Learning to Generate Conversations from Instructions (2021.findings-emnlp)
Copied to clipboard
| Challenge: | Popular dialog datasets such as MultiWOZ are created by providing crowd workers with instructions that describe the task to be accomplished. |
| Approach: | They propose a data creation strategy that uses a pre-trained language model to simulate the interaction between crowd workers by creating a user bot and an agent bot. |
| Outcome: | The proposed data creation strategy improves on two publicly available datasets using a pre-trained language model and a smaller percentage of actual crowd-generated conversations and their corresponding instructions. |
Beyond [CLS] through Ranking by Generation (2020.emnlp-main)
Copied to clipboard
| Challenge: | Recent work on generative ranking models for Information Retrieval has focused on discriminative methods that learn a similarity function to compare questions and candidates answers. |
| Approach: | They propose to use a language model to train a ranking function that model the semantic similarity of documents and queries instead of discriminative ranking functions. |
| Outcome: | The proposed approaches are as effective as state-of-the-art discriminative models for the answer selection task and show unlikelihood losses are reduced for IR. |
Probing Pre-trained Auto-regressive Language Models for Named Entity Typing and Recognition (2022.lrec-1)
Copied to clipboard
| Challenge: | Existing studies have focused on auto-regressive models for generalization in named entity (NE) typing (NET) and recognition (NER) . however, little has been done in this direction for auto-Regressive LMs despite their popularity and potential to express a wide variety of NLP tasks in the same unified format. |
| Approach: | They propose to probe auto-regressive LMs for NET and NER generalization by resorting to meta-learning to assess the model's memorization of NEs. |
| Outcome: | The proposed model performs well on NET and NER generalization tasks, while relying more on NE than contextual cues in few-shot NER. |
Manipulating the Perceived Personality Traits of Language Models (2023.findings-emnlp)
Copied to clipboard
| Challenge: | Psychology research has long explored aspects of human personality like extroversion, agreeableness and emotional stability, three of the personality traits that make up the ‘Big Five’. |
| Approach: | They propose to use text generated from large language models to evaluate perceived personality traits and to frame them as tools for controlling personas in dialog systems. |
| Outcome: | The proposed models predict personality traits in different contexts and can be manipulated in a predictable way. |
Facts2Story: Controlling Text Generation by Key Facts (2020.coling-main)
Copied to clipboard
| Challenge: | Existing methods for story generation struggle with staying coherent for long periods of time. |
| Approach: | They propose a controlled generation task which expands a sequence of facts into a longer narrative. |
| Outcome: | The proposed model produces competitive fluency while adhering to the requested facts. |
Bot-Adversarial Dialogue for Safe Conversational Agents (2021.naacl-main)
Copied to clipboard
| Challenge: | a new method for evaluating chatbot safety is proposed to mimic human-generated data . a bot-adversarial dialogue model learns undesirable features from this data, a study finds . |
| Approach: | They propose a human-and-model-in-the-loop framework for evaluating toxicity of chatbots . they propose two methods for safe conversational agents by either training on data or ”baking-in” safety to the generative model itself. |
| Outcome: | The proposed methods are safer than existing models while maintaining usability metrics, the authors say . they show that the proposed methods can be used to make safer models with human-model interactions . |
Enhancing Language Generation with Effective Checkpoints of Pre-trained Language Model (2021.findings-acl)
Copied to clipboard
| Challenge: | Existing methods to exploit PrLMs for NLG tasks do not get as much performance gain as in the NLU task. |
| Approach: | They propose a method to integrate public checkpoints of PrLMs for the most convenience. |
| Outcome: | The proposed method significantly improves the quality of the language generation tasks on 6 different kinds of PrLMs. |
Fingerprinting Fine-tuned Language Models in the Wild (2021.findings-acl)
Copied to clipboard
| Challenge: | Existing fingerprinting methods to fingerprint language models are limited to attributing organic text . however, fine-tuned LMs can generate long, coherent, and grammatically valid synthetic text. |
| Approach: | They conduct extensive experiments to demonstrate the limitations of existing fingerprinting approaches. |
| Outcome: | The proposed fingerprinting methods are limited to attributing synthetic text generated by 10 pre-trained LMs. |
Structural and Functional Decomposition for Personality Image Captioning in a Communication Game (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Personality image captioning (PIC) aims to describe an image with a natural language caption given a personality trait. |
| Approach: | They propose to use a communication game between a speaker and a listener to generate captions for PIC. |
| Outcome: | The proposed model achieves state-of-the-art performance for personal image captioning (PIC) the proposed model is based on a communication game between a speaker and a listener . |
Mitigating the Learning Bias towards Repetition by Self-Contrastive Training for Open-Ended Generation (2023.findings-acl)
Copied to clipboard
| Challenge: | Existing language models generate repetitive texts with greedy decoding or beam search. |
| Approach: | They propose a self-contrastive training technique to penalize the output of a premature checkpoint of the same model when it incorrectly predicts repetition. |
| Outcome: | The proposed training mitigates repetition while maintaining fluency while minimizing the overestimation of token-level repetition probabilities. |
Towards Robust Personalized Dialogue Generation via Order-Insensitive Representation Regularization (2023.findings-acl)
Copied to clipboard
| Challenge: | Recent studies have focused on developing persona consistent dialogue models . order sensitivity affects the quality and consistency of generated response . |
| Approach: | They propose a model-agnostic framework to improve persona consistent dialogue response generation by concatenating persona texts and dialogue history as a single input sequence. |
| Outcome: | The proposed framework outperforms existing models on the Persona-Chat dataset and shows that it is more robust under different persona orders and more consistent with the persona profile. |
Hit the Nail on the Head: Parameter-Efficient Multi-task Tuning via Human Language Intervention (2024.findings-emnlp)
Copied to clipboard
| Challenge: | Recent studies show that PEFT on small pre-trained language models improves multitasking capabilities. |
| Approach: | They propose a multi-task learning framework that enables transfer of prior knowledge across tasks . they attach task descriptions to input samples and map them to task embeddings . |
| Outcome: | The proposed method improves performance on a T5 model and in decoder-only models . |
Characterizing Mechanisms for Factual Recall in Language Models (2023.emnlp-main)
Copied to clipboard
| Challenge: | Language Models often integrate facts they memorized with new information that appears in a given context, causing competition within the model. |
| Approach: | They investigate distributional and mechanistic determinants of LM behavior in a dataset that queries for knowledge of world capitals . they use head attribution to identify individual attention heads that either promote the memorized answer or the in-context answer in the logits . |
| Outcome: | The proposed method can increase the rate of generating the in-context answer to 88% of the time by scaling up or down the value vector of individual attention heads at runtime. |
Authorship Attribution for Neural Text Generation (2020.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in deep learning have enabled the generation of realistic artifacts . however, the qualities of texts generated by these models are better, often confusing classifiers if they are not real. |
| Approach: | They propose to use neural network-based language models to generate realistic texts . they investigate the authorship attribution problem in three versions of a text . |
| Outcome: | The proposed models generate texts that are difficult to distinguish from human-written ones . the results show that most generators still generate texts significantly different from human ones compared to other models . |
PyMT5: multi-mode translation of natural language and Python code with transformers (2020.emnlp-main)
Copied to clipboard
| Challenge: | Using Python method text-to-text transfer transformers, developers can easily model source code and natural language. |
| Approach: | They propose a Python method text-to-text transfer transformer that can translate between all pairs of Python method feature combinations. |
| Outcome: | The proposed model outperforms similar-sized auto-regressive language models on a large-scale parallel corpus of 26 million methods and 7.7 million method-docstring pairs on the CodeSearchNet test set. |
Focus Your Attention (with Adaptive IIR Filters) (2023.emnlp-main)
Copied to clipboard
| Challenge: | Existing methods to deal with long-range data processing are implicit convolutions and regularized parameterization. |
| Approach: | They propose a new layer where dynamic (i.e., input-dependent) IIR filters are used to process the input sequence prior to applying conventional attention. |
| Outcome: | The proposed layer performs on-par with state-of-the-art networks with a fraction of their parameters and time complexity that is sub-quadratic with input size. |
A Curious Case of Searching for the Correlation between Training Data and Adversarial Robustness of Transformer Textual Models (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing studies show that fine-tuned textual transformer models are vulnerable to adversarial text perturbations. |
| Approach: | They extract 13 different features representing a wide range of input fine-tuning corpora properties and use them to predict adversarial robustness of the fine- tuned models. |
| Outcome: | The proposed framework can be used as an additional tool for robustness evaluation since it saves 30x-193x runtime compared to the traditional technique and can be easily used under adversarial training. |
Token Prediction as Implicit Classification to Identify LLM-Generated Text (2023.emnlp-main)
Copied to clipboard
| Challenge: | a novel approach for identifying large language models (LLMs) involved in text generation is proposed . instead of adding an additional classification layer, we reframe the classification task as a next-token prediction task . |
| Approach: | They propose a novel approach for identifying large language models involved in text generation . instead of adding an additional classification layer, they reframe the task as a next-token prediction task . |
| Outcome: | The proposed method performs exceptionally well in the text classification task . it can distinguish distinctive writing styles among various LLMs even without an explicit classifier. |
InstructEd: Soft-Instruction Tuning for Model Editing with Hops (2024.findings-acl)
Copied to clipboard
| Challenge: | Existing methods for model editing are limited due to excessive memorization and knowledge conflict issues. |
| Approach: | They propose to insert soft instructions into the attention module to facilitate interactions between instructions and questions and to understand and utilize new facts. |
| Outcome: | The proposed method achieves 10% improvement in one-hop (multi-hop) model editing on three datasets with LLaMAs and GPT2 . |
Composable Text Controls in Latent Space with ODEs (2023.emnlp-main)
Copied to clipboard
Guangyi Liu, Zeyu Feng, Yuan Gao, Zichao Yang, Xiaodan Liang, Junwei Bao, Xiaodong He, Shuguang Cui, Zhen Li, Zhiting Hu
| Challenge: | Existing approaches to composable text operations often require plug-and-play . a single LM can perform arbitrary text operation composition in the latent space . |
| Approach: | They propose an efficient approach for composable text operations in the latent space of text . they connect pretrained LMs to the laten space and adapt them to the space . |
| Outcome: | The proposed approach improves on existing methods in the latent space of text. |