publication_date
stringlengths 20
20
| doc_url
stringlengths 33
34
| id
stringlengths 33
34
| authors
stringlengths 5
20.1k
| update_date
stringlengths 20
20
| category_all
stringlengths 5
138
| abstract
stringlengths 123
3.09k
| category_primary
stringclasses 78
values | title
stringlengths 10
211
|
|---|---|---|---|---|---|---|---|---|
2023-03-22T20:26:30Z | http://arxiv.org/pdf/2303.12898v1 | http://arxiv.org/abs/2303.12898v1 | Richard Tarbell, Kim-Kwang Raymond Choo, Glenn Dietrich, Anthony Rios | 2023-03-22T20:26:30Z | cs.CL | Electronic medical records (EMRs) are stored in relational databases. It can
be challenging to access the required information if the user is unfamiliar
with the database schema or general database fundamentals. Hence, researchers
have explored text-to-SQL generation methods that provide healthcare
professionals direct access to EMR data without needing a database expert.
However, currently available datasets have been essentially "solved" with
state-of-the-art models achieving accuracy greater than or near 90%. In this
paper, we show that there is still a long way to go before solving text-to-SQL
generation in the medical domain. To show this, we create new splits of the
existing medical text-to-SQL dataset MIMICSQL that better measure the
generalizability of the resulting models. We evaluate state-of-the-art language
models on our new split showing substantial drops in performance with accuracy
dropping from up to 92% to 28%, thus showing substantial room for improvement.
Moreover, we introduce a novel data augmentation approach to improve the
generalizability of the language models. Overall, this paper is the first step
towards developing more robust text-to-SQL models in the medical
domain.\footnote{The dataset and code will be released upon acceptance. | cs.CL | Towards Understanding the Generalization of Medical Text-to-SQL Models and Datasets |
2023-03-22T20:10:29Z | http://arxiv.org/pdf/2303.12892v2 | http://arxiv.org/abs/2303.12892v2 | Thanh-Dung Le, Philippe Jouvet, Rita Noumeir | 2024-05-25T14:54:20Z | cs.CL, eess.SP | Transformer-based models have shown outstanding results in natural language
processing but face challenges in applications like classifying small-scale
clinical texts, especially with constrained computational resources. This study
presents a customized Mixture of Expert (MoE) Transformer models for
classifying small-scale French clinical texts at CHU Sainte-Justine Hospital.
The MoE-Transformer addresses the dual challenges of effective training with
limited data and low-resource computation suitable for in-house hospital use.
Despite the success of biomedical pre-trained models such as CamemBERT-bio,
DrBERT, and AliBERT, their high computational demands make them impractical for
many clinical settings. Our MoE-Transformer model not only outperforms
DistillBERT, CamemBERT, FlauBERT, and Transformer models on the same dataset
but also achieves impressive results: an accuracy of 87\%, precision of 87\%,
recall of 85\%, and F1-score of 86\%. While the MoE-Transformer does not
surpass the performance of biomedical pre-trained BERT models, it can be
trained at least 190 times faster, offering a viable alternative for settings
with limited data and computational resources. Although the MoE-Transformer
addresses challenges of generalization gaps and sharp minima, demonstrating
some limitations for efficient and accurate clinical text classification, this
model still represents a significant advancement in the field. It is
particularly valuable for classifying small French clinical narratives within
the privacy and constraints of hospital-based computational resources. | cs.CL | Improving Transformer Performance for French Clinical Notes Classification Using Mixture of Experts on a Limited Dataset |
2023-03-22T19:36:07Z | http://arxiv.org/pdf/2304.03169v2 | http://arxiv.org/abs/2304.03169v2 | Rajul Acharya, Ashish Panda, Sunil Kumar Kopparapu | 2023-04-25T11:05:55Z | cs.CL, cs.SD, eess.AS | Speech translation (ST) systems translate speech in one language to text in
another language. End-to-end ST systems (e2e-ST) have gained popularity over
cascade systems because of their enhanced performance due to reduced latency
and computational cost. Though resource intensive, e2e-ST systems have the
inherent ability to retain para and non-linguistic characteristics of the
speech unlike cascade systems. In this paper, we propose to use an e2e
architecture for English-Hindi (en-hi) ST. We use two imperfect machine
translation (MT) services to translate Libri-trans en text into hi text. While
each service gives MT data individually to generate parallel ST data, we
propose a data augmentation strategy of noisy MT data to aid robust ST. The
main contribution of this paper is the proposal of a data augmentation
strategy. We show that this results in better ST (BLEU score) compared to brute
force augmentation of MT data. We observed an absolute improvement of 1.59 BLEU
score with our approach. | cs.CL | Selective Data Augmentation for Robust Speech Translation |
2023-03-22T19:01:25Z | http://arxiv.org/pdf/2303.12869v1 | http://arxiv.org/abs/2303.12869v1 | Jessica López Espejel, Mahaman Sanoussi Yahaya Alassan, Walid Dahhane, El Hassane Ettifouri | 2023-03-22T19:01:25Z | cs.CL | Pretrained transformer-based models have shown high performance in natural
language generation task. However, a new wave of interest has surged: automatic
programming language generation. This task consists of translating natural
language instructions to a programming code. Despite the fact that well-known
pretrained models on language generation have achieved good performance in
learning programming languages, effort is still needed in automatic code
generation. In this paper, we introduce JaCoText, a model based on Transformers
neural network. It aims to generate java source code from natural language
text. JaCoText leverages advantages of both natural language and code
generation models. More specifically, we study some findings from the state of
the art and use them to (1) initialize our model from powerful pretrained
models, (2) explore additional pretraining on our java dataset, (3) carry out
experiments combining the unimodal and bimodal data in the training, and (4)
scale the input and output length during the fine-tuning of the model.
Conducted experiments on CONCODE dataset show that JaCoText achieves new
state-of-the-art results. | cs.CL | JaCoText: A Pretrained Model for Java Code-Text Generation |
2023-03-22T18:49:43Z | http://arxiv.org/pdf/2303.12860v1 | http://arxiv.org/abs/2303.12860v1 | Jeremy R. Cole, Aditi Chaudhary, Bhuwan Dhingra, Partha Talukdar | 2023-03-22T18:49:43Z | cs.CL, cs.AI | Salient Span Masking (SSM) has shown itself to be an effective strategy to
improve closed-book question answering performance. SSM extends general masked
language model pretraining by creating additional unsupervised training
sentences that mask a single entity or date span, thus oversampling factual
information. Despite the success of this paradigm, the span types and sampling
strategies are relatively arbitrary and not widely studied for other tasks.
Thus, we investigate SSM from the perspective of temporal tasks, where learning
a good representation of various temporal expressions is important. To that
end, we introduce Temporal Span Masking (TSM) intermediate training. First, we
find that SSM alone improves the downstream performance on three temporal tasks
by an avg. +5.8 points. Further, we are able to achieve additional improvements
(avg. +0.29 points) by adding the TSM task. These comprise the new best
reported results on the targeted tasks. Our analysis suggests that the
effectiveness of SSM stems from the sentences chosen in the training data
rather than the mask choice: sentences with entities frequently also contain
temporal expressions. Nonetheless, the additional targeted spans of TSM can
still improve performance, especially in a zero-shot context. | cs.CL | Salient Span Masking for Temporal Understanding |
2023-03-22T17:57:47Z | http://arxiv.org/pdf/2303.12788v1 | http://arxiv.org/abs/2303.12788v1 | David Chanin | 2023-03-22T17:57:47Z | cs.CL, cs.AI, cs.LG | While the state-of-the-art for frame semantic parsing has progressed
dramatically in recent years, it is still difficult for end-users to apply
state-of-the-art models in practice. To address this, we present Frame Semantic
Transformer, an open-source Python library which achieves near state-of-the-art
performance on FrameNet 1.7, while focusing on ease-of-use. We use a T5 model
fine-tuned on Propbank and FrameNet exemplars as a base, and improve
performance by using FrameNet lexical units to provide hints to T5 at inference
time. We enhance robustness to real-world data by using textual data
augmentations during training. | cs.CL | Open-source Frame Semantic Parsing |
2023-03-22T17:35:35Z | http://arxiv.org/pdf/2303.12772v1 | http://arxiv.org/abs/2303.12772v1 | Ramisa Anan, Tasnim Sakib Apon, Zeba Tahsin Hossain, Elizabeth Antora Modhu, Sudipta Mondal, MD. Golam Rabiul Alam | 2023-03-22T17:35:35Z | cs.CL, cs.AI | A positive phrase or a sentence with an underlying negative motive is usually
defined as sarcasm that is widely used in today's social media platforms such
as Facebook, Twitter, Reddit, etc. In recent times active users in social media
platforms are increasing dramatically which raises the need for an automated
NLP-based system that can be utilized in various tasks such as determining
market demand, sentiment analysis, threat detection, etc. However, since
sarcasm usually implies the opposite meaning and its detection is frequently a
challenging issue, data meaning extraction through an NLP-based model becomes
more complicated. As a result, there has been a lot of study on sarcasm
detection in English over the past several years, and there's been a noticeable
improvement and yet sarcasm detection in the Bangla language's state remains
the same. In this article, we present a BERT-based system that can achieve
99.60\% while the utilized traditional machine learning algorithms are only
capable of achieving 89.93\%. Additionally, we have employed Local
Interpretable Model-Agnostic Explanations that introduce explainability to our
system. Moreover, we have utilized a newly collected bangla sarcasm dataset,
BanglaSarc that was constructed specifically for the evaluation of this study.
This dataset consists of fresh records of sarcastic and non-sarcastic comments,
the majority of which are acquired from Facebook and YouTube comment sections. | cs.CL | Interpretable Bangla Sarcasm Detection using BERT and Explainable AI |
2023-03-22T17:32:56Z | http://arxiv.org/pdf/2303.12767v1 | http://arxiv.org/abs/2303.12767v1 | Rachith Aiyappa, Jisun An, Haewoon Kwak, Yong-Yeol Ahn | 2023-03-22T17:32:56Z | cs.CL, cs.AI, cs.LG | ChatGPT, the first large language model (LLM) with mass adoption, has
demonstrated remarkable performance in numerous natural language tasks. Despite
its evident usefulness, evaluating ChatGPT's performance in diverse problem
domains remains challenging due to the closed nature of the model and its
continuous updates via Reinforcement Learning from Human Feedback (RLHF). We
highlight the issue of data contamination in ChatGPT evaluations, with a case
study of the task of stance detection. We discuss the challenge of preventing
data contamination and ensuring fair model evaluation in the age of closed and
continuously trained models. | cs.CL | Can we trust the evaluation on ChatGPT? |
2023-03-22T16:51:28Z | http://arxiv.org/pdf/2303.12712v5 | http://arxiv.org/abs/2303.12712v5 | Sébastien Bubeck, Varun Chandrasekaran, Ronen Eldan, Johannes Gehrke, Eric Horvitz, Ece Kamar, Peter Lee, Yin Tat Lee, Yuanzhi Li, Scott Lundberg, Harsha Nori, Hamid Palangi, Marco Tulio Ribeiro, Yi Zhang | 2023-04-13T20:41:31Z | cs.CL, cs.AI | Artificial intelligence (AI) researchers have been developing and refining
large language models (LLMs) that exhibit remarkable capabilities across a
variety of domains and tasks, challenging our understanding of learning and
cognition. The latest model developed by OpenAI, GPT-4, was trained using an
unprecedented scale of compute and data. In this paper, we report on our
investigation of an early version of GPT-4, when it was still in active
development by OpenAI. We contend that (this early version of) GPT-4 is part of
a new cohort of LLMs (along with ChatGPT and Google's PaLM for example) that
exhibit more general intelligence than previous AI models. We discuss the
rising capabilities and implications of these models. We demonstrate that,
beyond its mastery of language, GPT-4 can solve novel and difficult tasks that
span mathematics, coding, vision, medicine, law, psychology and more, without
needing any special prompting. Moreover, in all of these tasks, GPT-4's
performance is strikingly close to human-level performance, and often vastly
surpasses prior models such as ChatGPT. Given the breadth and depth of GPT-4's
capabilities, we believe that it could reasonably be viewed as an early (yet
still incomplete) version of an artificial general intelligence (AGI) system.
In our exploration of GPT-4, we put special emphasis on discovering its
limitations, and we discuss the challenges ahead for advancing towards deeper
and more comprehensive versions of AGI, including the possible need for
pursuing a new paradigm that moves beyond next-word prediction. We conclude
with reflections on societal influences of the recent technological leap and
future research directions. | cs.CL | Sparks of Artificial General Intelligence: Early experiments with GPT-4 |
2023-03-22T15:49:33Z | http://arxiv.org/pdf/2303.12671v2 | http://arxiv.org/abs/2303.12671v2 | Triet Minh Thai, Son T. Luu | 2023-09-03T14:50:34Z | cs.CV, cs.CL | Visual Question Answering (VQA) is a task that requires computers to give
correct answers for the input questions based on the images. This task can be
solved by humans with ease but is a challenge for computers. The
VLSP2022-EVJVQA shared task carries the Visual Question Answering task in the
multilingual domain on a newly released dataset: UIT-EVJVQA, in which the
questions and answers are written in three different languages: English,
Vietnamese and Japanese. We approached the challenge as a sequence-to-sequence
learning task, in which we integrated hints from pre-trained state-of-the-art
VQA models and image features with Convolutional Sequence-to-Sequence network
to generate the desired answers. Our results obtained up to 0.3442 by F1 score
on the public test set, 0.4210 on the private test set, and placed 3rd in the
competition. | cs.CV | Integrating Image Features with Convolutional Sequence-to-sequence Network for Multilingual Visual Question Answering |
2023-03-22T15:44:15Z | http://arxiv.org/pdf/2303.12665v2 | http://arxiv.org/abs/2303.12665v2 | Yue Li, Carolina Scarton | 2024-02-22T15:36:22Z | cs.CL | Considering a conversation thread, rumour stance classification aims to
identify the opinion (e.g. agree or disagree) of replies towards a target
(rumour story). Although the target is expected to be an essential component in
traditional stance classification, we show that rumour stance classification
datasets contain a considerable amount of real-world data whose stance could be
naturally inferred directly from the replies, contributing to the strong
performance of the supervised models without awareness of the target. We find
that current target-aware models underperform in cases where the context of the
target is crucial. Finally, we propose a simple yet effective framework to
enhance reasoning with the targets, achieving state-of-the-art performance on
two benchmark datasets. | cs.CL | Can We Identify Stance Without Target Arguments? A Study for Rumour Stance Classification |
2023-03-22T15:32:40Z | http://arxiv.org/pdf/2303.13463v1 | http://arxiv.org/abs/2303.13463v1 | Wen Cheng, Shichen Dong, Wei Wang | 2023-03-22T15:32:40Z | cs.CL, eess.AS | This paper describes our submission to ICASSP 2023 MUG Challenge Track 4,
Keyphrase Extraction, which aims to extract keyphrases most relevant to the
conference theme from conference materials. We model the challenge as a
single-class Named Entity Recognition task and developed techniques for better
performance on the challenge: For the data preprocessing, we encode the split
keyphrases after word segmentation. In addition, we increase the amount of
input information that the model can accept at one time by fusing multiple
preprocessed sentences into one segment. We replace the loss function with the
multi-class focal loss to address the sparseness of keyphrases. Besides, we
score each appearance of keyphrases and add an extra output layer to fit the
score to rank keyphrases. Exhaustive evaluations are performed to find the best
combination of the word segmentation tool, the pre-trained embedding model, and
the corresponding hyperparameters. With these proposals, we scored 45.04 on the
final test set. | cs.CL | W2KPE: Keyphrase Extraction with Word-Word Relation |
2023-03-22T14:09:20Z | http://arxiv.org/pdf/2303.12582v2 | http://arxiv.org/abs/2303.12582v2 | Chris Chinenye Emezue, Sanchit Gandhi, Lewis Tunstall, Abubakar Abid, Josh Meyer, Quentin Lhoest, Pete Allen, Patrick Von Platen, Douwe Kiela, Yacine Jernite, Julien Chaumond, Merve Noyan, Omar Sanseviero | 2023-04-04T03:32:24Z | cs.CL | The advancement of speech technologies has been remarkable, yet its
integration with African languages remains limited due to the scarcity of
African speech corpora. To address this issue, we present AfroDigits, a
minimalist, community-driven dataset of spoken digits for African languages,
currently covering 38 African languages. As a demonstration of the practical
applications of AfroDigits, we conduct audio digit classification experiments
on six African languages [Igbo (ibo), Yoruba (yor), Rundi (run), Oshiwambo
(kua), Shona (sna), and Oromo (gax)] using the Wav2Vec2.0-Large and XLS-R
models. Our experiments reveal a useful insight on the effect of mixing African
speech corpora during finetuning. AfroDigits is the first published audio digit
dataset for African languages and we believe it will, among other things, pave
the way for Afro-centric speech applications such as the recognition of
telephone numbers, and street numbers. We release the dataset and platform
publicly at https://huggingface.co/datasets/chrisjay/crowd-speech-africa and
https://huggingface.co/spaces/chrisjay/afro-speech respectively. | cs.CL | AfroDigits: A Community-Driven Spoken Digit Dataset for African Languages |
2023-03-22T13:54:46Z | http://arxiv.org/pdf/2303.12570v3 | http://arxiv.org/abs/2303.12570v3 | Fengji Zhang, Bei Chen, Yue Zhang, Jacky Keung, Jin Liu, Daoguang Zan, Yi Mao, Jian-Guang Lou, Weizhu Chen | 2023-10-20T15:21:51Z | cs.CL, cs.AI, cs.PL, cs.SE | The task of repository-level code completion is to continue writing the
unfinished code based on a broader context of the repository. While for
automated code completion tools, it is difficult to utilize the useful
information scattered in different files. We propose RepoCoder, a simple,
generic, and effective framework to address the challenge. It streamlines the
repository-level code completion process by incorporating a similarity-based
retriever and a pre-trained code language model in an iterative
retrieval-generation pipeline. RepoCoder makes effective utilization of
repository-level information for code completion and has the ability to
generate code at various levels of granularity. Moreover, we propose a new
benchmark RepoEval, which consists of the latest and high-quality real-world
repositories covering line, API invocation, and function body completion
scenarios. Experimental results indicate that RepoCoder significantly improves
the In-File completion baseline by over 10% in all settings and consistently
outperforms the vanilla retrieval-augmented code completion approach.
Furthermore, we validate the effectiveness of RepoCoder through comprehensive
analysis, providing valuable insights for future research. Our source code and
benchmark are publicly available:
https://github.com/microsoft/CodeT/tree/main/RepoCoder | cs.CL | RepoCoder: Repository-Level Code Completion Through Iterative Retrieval and Generation |
2023-03-22T13:46:16Z | http://arxiv.org/pdf/2303.13466v2 | http://arxiv.org/abs/2303.13466v2 | Sonish Sivarajkumar, Fengyi Gao, Parker E. Denny, Bayan M. Aldhahwani, Shyam Visweswaran, Allyn Bove, Yanshan Wang | 2024-03-15T17:55:52Z | cs.CL, cs.AI | Post-stroke patient rehabilitation requires precise, personalized treatment
plans. Natural Language Processing (NLP) offers potential to extract valuable
exercise information from clinical notes, aiding in the development of more
effective rehabilitation strategies. Objective: This study aims to develop and
evaluate a variety of NLP algorithms to extract and categorize physical
rehabilitation exercise information from the clinical notes of post-stroke
patients treated at the University of Pittsburgh Medical Center. A cohort of
13,605 patients diagnosed with stroke was identified, and their clinical notes
containing rehabilitation therapy notes were retrieved. A comprehensive
clinical ontology was created to represent various aspects of physical
rehabilitation exercises. State-of-the-art NLP algorithms were then developed
and compared, including rule-based, machine learning-based algorithms, and
large language model (LLM)-based algorithms (ChatGPT). Analysis was conducted
on a dataset comprising 23,724 notes with detailed demographic and clinical
characteristics. The rule-based NLP algorithm demonstrated superior performance
in most areas, particularly in detecting the 'Right Side' location with an F1
score of 0.975, outperforming Gradient Boosting by 0.063. Gradient Boosting
excelled in 'Lower Extremity' location detection (F1 score: 0.978), surpassing
rule-based NLP by 0.023. It also showed notable performance in 'Passive Range
of Motion' with an F1 score of 0.970, a 0.032 improvement over rule-based NLP.
The rule-based algorithm efficiently handled 'Duration', 'Sets', and 'Reps'
with F1 scores up to 0.65. LLM-based NLP, particularly ChatGPT with few-shot
prompts, achieved high recall but generally lower precision and F1 scores.
However, it notably excelled in 'Backward Plane' motion detection, achieving an
F1 score of 0.846, surpassing the rule-based algorithm's 0.720. | cs.CL | Mining Clinical Notes for Physical Rehabilitation Exercise Information: Natural Language Processing Algorithm Development and Validation Study |
2023-03-22T13:03:10Z | http://arxiv.org/pdf/2303.12528v4 | http://arxiv.org/abs/2303.12528v4 | Kabir Ahuja, Harshita Diddee, Rishav Hada, Millicent Ochieng, Krithika Ramesh, Prachi Jain, Akshay Nambi, Tanuja Ganu, Sameer Segal, Maxamed Axmed, Kalika Bali, Sunayana Sitaram | 2023-10-22T22:19:13Z | cs.CL | Generative AI models have shown impressive performance on many Natural
Language Processing tasks such as language understanding, reasoning, and
language generation. An important question being asked by the AI community
today is about the capabilities and limits of these models, and it is clear
that evaluating generative AI is very challenging. Most studies on generative
LLMs have been restricted to English and it is unclear how capable these models
are at understanding and generating text in other languages. We present the
first comprehensive benchmarking of generative LLMs - MEGA, which evaluates
models on standard NLP benchmarks, covering 16 NLP datasets across 70
typologically diverse languages. We compare the performance of generative LLMs
including Chat-GPT and GPT-4 to State of the Art (SOTA) non-autoregressive
models on these tasks to determine how well generative models perform compared
to the previous generation of LLMs. We present a thorough analysis of the
performance of models across languages and tasks and discuss challenges in
improving the performance of generative LLMs on low-resource languages. We
create a framework for evaluating generative LLMs in the multilingual setting
and provide directions for future progress in the field. | cs.CL | MEGA: Multilingual Evaluation of Generative AI |
2023-03-22T09:29:22Z | http://arxiv.org/pdf/2303.13465v3 | http://arxiv.org/abs/2303.13465v3 | Itsugun Cho, Ryota Takahashi, Yusaku Yanase, Hiroaki Saito | 2023-05-15T08:04:18Z | cs.CL, cs.AI | Traditionally, approximate dynamic programming is employed in dialogue
generation with greedy policy improvement through action sampling, as the
natural language action space is vast. However, this practice is inefficient
for reinforcement learning (RL) due to the sparsity of eligible responses with
high action values, which leads to weak improvement sustained by random
sampling. This paper presents theoretical analysis and experiments that reveal
the performance of the dialogue policy is positively correlated with the
sampling size. To overcome this limitation, we introduce a novel
dual-granularity Q-function that explores the most promising response category
to intervene in the sampling process. Our approach extracts actions based on a
grained hierarchy, thereby achieving the optimum with fewer policy iterations.
Additionally, we use offline RL and learn from multiple reward functions
designed to capture emotional nuances in human interactions. Empirical studies
demonstrate that our algorithm outperforms baselines across automatic metrics
and human evaluations. Further testing reveals that our algorithm exhibits both
explainability and controllability and generates responses with higher expected
rewards. | cs.CL | Deep RL with Hierarchical Action Exploration for Dialogue Generation |
2023-03-22T07:34:33Z | http://arxiv.org/pdf/2303.12816v3 | http://arxiv.org/abs/2303.12816v3 | Borui Cai, Yong Xiang, Longxiang Gao, Di Wu, He Zhang, Jiong Jin, Tom Luan | 2023-11-13T05:07:41Z | cs.LG, cs.AI, cs.CL | Knowledge graph embedding (KGE) that maps entities and relations into vector
representations is essential for downstream applications. Conventional KGE
methods require high-dimensional representations to learn the complex structure
of knowledge graph, but lead to oversized model parameters. Recent advances
reduce parameters by low-dimensional entity representations, while developing
techniques (e.g., knowledge distillation or reinvented representation forms) to
compensate for reduced dimension. However, such operations introduce
complicated computations and model designs that may not benefit large knowledge
graphs. To seek a simple strategy to improve the parameter efficiency of
conventional KGE models, we take inspiration from that deeper neural networks
require exponentially fewer parameters to achieve expressiveness comparable to
wider networks for compositional structures. We view all entity representations
as a single-layer embedding network, and conventional KGE methods that adopt
high-dimensional entity representations equal widening the embedding network to
gain expressiveness. To achieve parameter efficiency, we instead propose a
deeper embedding network for entity representations, i.e., a narrow entity
embedding layer plus a multi-layer dimension lifting network (LiftNet).
Experiments on three public datasets show that by integrating LiftNet, four
conventional KGE methods with 16-dimensional representations achieve comparable
link prediction accuracy as original models that adopt 512-dimensional
representations, saving 68.4% to 96.9% parameters. | cs.LG | From Wide to Deep: Dimension Lifting Network for Parameter-efficient Knowledge Graph Embedding |
2023-03-22T05:35:29Z | http://arxiv.org/pdf/2303.12320v2 | http://arxiv.org/abs/2303.12320v2 | Dhaval Taunk, Lakshya Khanna, Pavan Kandru, Vasudeva Varma, Charu Sharma, Makarand Tapaswi | 2023-04-18T05:11:47Z | cs.CL | Commonsense question-answering (QA) methods combine the power of pre-trained
Language Models (LM) with the reasoning provided by Knowledge Graphs (KG). A
typical approach collects nodes relevant to the QA pair from a KG to form a
Working Graph (WG) followed by reasoning using Graph Neural Networks(GNNs).
This faces two major challenges: (i) it is difficult to capture all the
information from the QA in the WG, and (ii) the WG contains some irrelevant
nodes from the KG. To address these, we propose GrapeQA with two simple
improvements on the WG: (i) Prominent Entities for Graph Augmentation
identifies relevant text chunks from the QA pair and augments the WG with
corresponding latent representations from the LM, and (ii) Context-Aware Node
Pruning removes nodes that are less relevant to the QA pair. We evaluate our
results on OpenBookQA, CommonsenseQA and MedQA-USMLE and see that GrapeQA shows
consistent improvements over its LM + KG predecessor (QA-GNN in particular) and
large improvements on OpenBookQA. | cs.CL | GrapeQA: GRaph Augmentation and Pruning to Enhance Question-Answering |
2023-03-22T05:04:21Z | http://arxiv.org/pdf/2303.12314v4 | http://arxiv.org/abs/2303.12314v4 | Kaihang Pan, Juncheng Li, Hongye Song, Jun Lin, Xiaozhong Liu, Siliang Tang | 2023-10-23T12:43:35Z | cs.CL, cs.LG | Prompt tuning is a parameter-efficient method, which learns soft prompts and
conditions frozen language models to perform specific downstream tasks. Though
effective, prompt tuning under few-shot settings on the one hand heavily relies
on a good initialization of soft prompts. On the other hand, it can easily
overfit to few-shot training samples, thereby undermining generalizability.
Existing works leverage pre-training or supervised meta-learning to initialize
soft prompts but they fail to data-efficiently generalize to unseen downstream
tasks. To address the above problems, this paper proposes a novel
Self-sUpervised meta-Prompt learning framework with MEta-gradient
Regularization for few-shot generalization (SUPMER). SUPMER leverages
self-supervised meta-learning with a diverse set of well-designed meta-training
tasks to learn a universal prompt initialization for efficient adaptation using
only unlabeled data. Additionally, it jointly meta-learns a gradient
regularization function to transform raw gradients into a domain-generalizable
direction, thus alleviating the problem of overfitting. Extensive experiments
show that SUPMER achieves better performance for different few-shot downstream
tasks, and also exhibits a stronger domain generalization ability. The code for
SUPMER will be available at https://github.com/beepkh/SUPMER. | cs.CL | Self-supervised Meta-Prompt Learning with Meta-Gradient Regularization for Few-shot Generalization |
2023-03-22T04:52:43Z | http://arxiv.org/pdf/2303.12308v2 | http://arxiv.org/abs/2303.12308v2 | Dhaval Taunk, Shivprasad Sagare, Anupam Patil, Shivansh Subramanian, Manish Gupta, Vasudeva Varma | 2023-04-18T09:38:59Z | cs.CL | Lack of encyclopedic text contributors, especially on Wikipedia, makes
automated text generation for low resource (LR) languages a critical problem.
Existing work on Wikipedia text generation has focused on English only where
English reference articles are summarized to generate English Wikipedia pages.
But, for low-resource languages, the scarcity of reference articles makes
monolingual summarization ineffective in solving this problem. Hence, in this
work, we propose XWikiGen, which is the task of cross-lingual multi-document
summarization of text from multiple reference articles, written in various
languages, to generate Wikipedia-style text. Accordingly, we contribute a
benchmark dataset, XWikiRef, spanning ~69K Wikipedia articles covering five
domains and eight languages. We harness this dataset to train a two-stage
system where the input is a set of citations and a section title and the output
is a section-specific LR summary. The proposed system is based on a novel idea
of neural unsupervised extractive summarization to coarsely identify salient
information followed by a neural abstractive model to generate the
section-specific text. Extensive experiments show that multi-domain training is
better than the multi-lingual setup on average. | cs.CL | XWikiGen: Cross-lingual Summarization for Encyclopedic Text Generation in Low Resource Languages |
2023-03-22T04:11:35Z | http://arxiv.org/pdf/2303.12300v1 | http://arxiv.org/abs/2303.12300v1 | Zeyu Ren, Nurmement Yolwas, Huiru Wang, Wushour Slamu | 2023-03-22T04:11:35Z | cs.SD, cs.CL, eess.AS | In recent years, End-to-End speech recognition technology based on deep
learning has developed rapidly. Due to the lack of Turkish speech data, the
performance of Turkish speech recognition system is poor. Firstly, this paper
studies a series of speech recognition tuning technologies. The results show
that the performance of the model is the best when the data enhancement
technology combining speed perturbation with noise addition is adopted and the
beam search width is set to 16. Secondly, to maximize the use of effective
feature information and improve the accuracy of feature extraction, this paper
proposes a new feature extractor LSPC. LSPC and LiGRU network are combined to
form a shared encoder structure, and model compression is realized. The results
show that the performance of LSPC is better than MSPC and VGGnet when only
using Fbank features, and the WER is improved by 1.01% and 2.53% respectively.
Finally, based on the above two points, a new multi-feature fusion network is
proposed as the main structure of the encoder. The results show that the WER of
the proposed feature fusion network based on LSPC is improved by 0.82% and
1.94% again compared with the single feature (Fbank feature and Spectrogram
feature) extraction using LSPC. Our model achieves performance comparable to
that of advanced End-to-End models. | cs.SD | Exploring Turkish Speech Recognition via Hybrid CTC/Attention Architecture and Multi-feature Fusion Network |
2023-03-22T03:58:38Z | http://arxiv.org/pdf/2303.12294v2 | http://arxiv.org/abs/2303.12294v2 | Xiaomeng Ma, Lingyu Gao | 2023-05-13T21:05:09Z | cs.CL | Neural network models have been proposed to explain the grapheme-phoneme
mapping process in humans for many alphabet languages. These models not only
successfully learned the correspondence of the letter strings and their
pronunciation, but also captured human behavior in nonce word naming tasks. How
would the neural models perform for a non-alphabet language (e.g., Chinese)
unknown character task? How well would the model capture human behavior? In
this study, we first collect human speakers' answers on unknown character
naming tasks and then evaluate a set of transformer models by comparing their
performances with human behaviors on an unknown Chinese character naming task.
We found that the models and humans behaved very similarly, that they had
similar accuracy distribution for each character, and had a substantial overlap
in answers. In addition, the models' answers are highly correlated with humans'
answers. These results suggested that the transformer models can well capture
human's character naming behavior. | cs.CL | Evaluating Transformer Models and Human Behaviors on Chinese Character Naming |
2023-03-21T22:43:41Z | http://arxiv.org/pdf/2303.12220v1 | http://arxiv.org/abs/2303.12220v1 | Kira Droganova, Daniel Zeman | 2023-03-21T22:43:41Z | cs.CL | This paper analyzes multiple deep-syntactic frameworks with the goal of
creating a proposal for a set of universal semantic role labels. The proposal
examines various theoretic linguistic perspectives and focuses on Meaning-Text
Theory and Functional Generative Description frameworks.
For the purpose of this research, data from four languages is used -- Spanish
and Catalan (Taule et al., 2011), Czech (Hajic et al., 2017), and English
(Hajic et al., 2012). This proposal is oriented towards Universal Dependencies
(de Marneffe et al., 2021) with a further intention of applying the universal
semantic role labels to the UD data. | cs.CL | A Unified Taxonomy of Deep Syntactic Relations |
2023-03-21T21:49:39Z | http://arxiv.org/pdf/2303.12208v1 | http://arxiv.org/abs/2303.12208v1 | Sungwoong Kim, Daejin Jo, Donghoon Lee, Jongmin Kim | 2023-03-21T21:49:39Z | cs.CV, cs.CL, cs.LG | While generative modeling on multimodal image-text data has been actively
developed with large-scale paired datasets, there have been limited attempts to
generate both image and text data by a single model rather than a generation of
one fixed modality conditioned on the other modality. In this paper, we explore
a unified generative vision-and-language (VL) model that can produce both
images and text sequences. Especially, we propose a generative VL transformer
based on the non-autoregressive mask prediction, named MAGVLT, and compare it
with an autoregressive generative VL transformer (ARGVLT). In comparison to
ARGVLT, the proposed MAGVLT enables bidirectional context encoding, fast
decoding by parallel token predictions in an iterative refinement, and extended
editing capabilities such as image and text infilling. For rigorous training of
our MAGVLT with image-text pairs from scratch, we combine the image-to-text,
text-to-image, and joint image-and-text mask prediction tasks. Moreover, we
devise two additional tasks based on the step-unrolled mask prediction and the
selective prediction on the mixture of two image-text pairs. Experimental
results on various downstream generation tasks of VL benchmarks show that our
MAGVLT outperforms ARGVLT by a large margin even with significant inference
speedup. Particularly, MAGVLT achieves competitive results on both zero-shot
image-to-text and text-to-image generation tasks from MS-COCO by one
moderate-sized model (fewer than 500M parameters) even without the use of
monomodal data and networks. | cs.CV | MAGVLT: Masked Generative Vision-and-Language Transformer |
2023-03-21T21:38:44Z | http://arxiv.org/pdf/2303.13549v1 | http://arxiv.org/abs/2303.13549v1 | Levi Corallo, Aparna S. Varde | 2023-03-21T21:38:44Z | cs.CV, cs.AI, cs.CL, cs.LG, eess.IV, I.2.1; I.2.7; J.5; K.3.1 | The Berber, or Amazigh language family is a low-resource North African
vernacular language spoken by the indigenous Berber ethnic group. It has its
own unique alphabet called Tifinagh used across Berber communities in Morocco,
Algeria, and others. The Afroasiatic language Berber is spoken by 14 million
people, yet lacks adequate representation in education, research, web
applications etc. For instance, there is no option of translation to or from
Amazigh / Berber on Google Translate, which hosts over 100 languages today.
Consequently, we do not find specialized educational apps, L2 (2nd language
learner) acquisition, automated language translation, and remote-access
facilities enabled in Berber. Motivated by this background, we propose a
supervised approach called DaToBS for Detection and Transcription of Berber
Signs. The DaToBS approach entails the automatic recognition and transcription
of Tifinagh characters from signs in photographs of natural environments. This
is achieved by self-creating a corpus of 1862 pre-processed character images;
curating the corpus with human-guided annotation; and feeding it into an OCR
model via the deployment of CNN for deep learning based on computer vision
models. We deploy computer vision modeling (rather than language models)
because there are pictorial symbols in this alphabet, this deployment being a
novel aspect of our work. The DaToBS experimentation and analyses yield over 92
percent accuracy in our research. To the best of our knowledge, ours is among
the first few works in the automated transcription of Berber signs from
roadside images with deep learning, yielding high accuracy. This can pave the
way for developing pedagogical applications in the Berber language, thereby
addressing an important goal of outreach to underrepresented communities via AI
in education. | cs.CV | Optical Character Recognition and Transcription of Berber Signs from Images in a Low-Resource Language Amazigh |
2023-03-21T18:48:11Z | http://arxiv.org/pdf/2303.12135v4 | http://arxiv.org/abs/2303.12135v4 | Xin Jin, Yuchen Wang | 2023-07-19T05:30:31Z | cs.CL | The growth of pending legal cases in populous countries, such as India, has
become a major issue. Developing effective techniques to process and understand
legal documents is extremely useful in resolving this problem. In this paper,
we present our systems for SemEval-2023 Task 6: understanding legal texts (Modi
et al., 2023). Specifically, we first develop the Legal-BERT-HSLN model that
considers the comprehensive context information in both intra- and
inter-sentence levels to predict rhetorical roles (subtask A) and then train a
Legal-LUKE model, which is legal-contextualized and entity-aware, to recognize
legal entities (subtask B). Our evaluations demonstrate that our designed
models are more accurate than baselines, e.g., with an up to 15.0% better F1
score in subtask B. We achieved notable performance in the task leaderboard,
e.g., 0.834 micro F1 score, and ranked No.5 out of 27 teams in subtask A. | cs.CL | Understand Legal Documents with Contextualized Large Language Models |
2023-03-21T18:45:09Z | http://arxiv.org/pdf/2303.12132v1 | http://arxiv.org/abs/2303.12132v1 | Andrei Kucharavy, Zachary Schillaci, Loïc Maréchal, Maxime Würsch, Ljiljana Dolamic, Remi Sabonnadiere, Dimitri Percia David, Alain Mermoud, Vincent Lenders | 2023-03-21T18:45:09Z | cs.CL, cs.CR, cs.LG, I.2.7; I.2.1; K.6.5; K.4.2; J.7 | Generative Language Models gained significant attention in late 2022 / early
2023, notably with the introduction of models refined to act consistently with
users' expectations of interactions with AI (conversational models). Arguably
the focal point of public attention has been such a refinement of the GPT3
model -- the ChatGPT and its subsequent integration with auxiliary
capabilities, including search as part of Microsoft Bing. Despite extensive
prior research invested in their development, their performance and
applicability to a range of daily tasks remained unclear and niche. However,
their wider utilization without a requirement for technical expertise, made in
large part possible through conversational fine-tuning, revealed the extent of
their true capabilities in a real-world environment. This has garnered both
public excitement for their potential applications and concerns about their
capabilities and potential malicious uses. This review aims to provide a brief
overview of the history, state of the art, and implications of Generative
Language Models in terms of their principles, abilities, limitations, and
future prospects -- especially in the context of cyber-defense, with a focus on
the Swiss operational environment. | cs.CL | Fundamentals of Generative Large Language Models and Perspectives in Cyber-Defense |
2023-03-21T18:03:14Z | http://arxiv.org/pdf/2303.12112v3 | http://arxiv.org/abs/2303.12112v3 | Sara Sarto, Manuele Barraco, Marcella Cornia, Lorenzo Baraldi, Rita Cucchiara | 2023-07-20T08:16:09Z | cs.CV, cs.AI, cs.CL, cs.MM | The CLIP model has been recently proven to be very effective for a variety of
cross-modal tasks, including the evaluation of captions generated from
vision-and-language architectures. In this paper, we propose a new recipe for a
contrastive-based evaluation metric for image captioning, namely
Positive-Augmented Contrastive learning Score (PAC-S), that in a novel way
unifies the learning of a contrastive visual-semantic space with the addition
of generated images and text on curated data. Experiments spanning several
datasets demonstrate that our new metric achieves the highest correlation with
human judgments on both images and videos, outperforming existing
reference-based metrics like CIDEr and SPICE and reference-free metrics like
CLIP-Score. Finally, we test the system-level correlation of the proposed
metric when considering popular image captioning approaches, and assess the
impact of employing different cross-modal features. Our source code and trained
models are publicly available at: https://github.com/aimagelab/pacscore. | cs.CV | Positive-Augmented Contrastive Learning for Image and Video Captioning Evaluation |
2023-03-21T17:51:23Z | http://arxiv.org/pdf/2303.12060v3 | http://arxiv.org/abs/2303.12060v3 | Jingyang Lin, Hang Hua, Ming Chen, Yikang Li, Jenhao Hsiao, Chiuman Ho, Jiebo Luo | 2024-04-23T17:55:37Z | cs.CV, cs.CL | Video summarization aims to distill the most important information from a
source video to produce either an abridged clip or a textual narrative.
Traditionally, different methods have been proposed depending on whether the
output is a video or text, thus ignoring the correlation between the two
semantically related tasks of visual summarization and textual summarization.
We propose a new joint video and text summarization task. The goal is to
generate both a shortened video clip along with the corresponding textual
summary from a long video, collectively referred to as a cross-modal summary.
The generated shortened video clip and text narratives should be semantically
well aligned. To this end, we first build a large-scale human-annotated dataset
-- VideoXum (X refers to different modalities). The dataset is reannotated
based on ActivityNet. After we filter out the videos that do not meet the
length requirements, 14,001 long videos remain in our new dataset. Each video
in our reannotated dataset has human-annotated video summaries and the
corresponding narrative summaries. We then design a novel end-to-end model --
VTSUM-BILP to address the challenges of our proposed task. Moreover, we propose
a new metric called VT-CLIPScore to help evaluate the semantic consistency of
cross-modality summary. The proposed model achieves promising performance on
this new task and establishes a benchmark for future research. | cs.CV | VideoXum: Cross-modal Visual and Textural Summarization of Videos |
2023-03-21T17:48:00Z | http://arxiv.org/pdf/2303.12057v4 | http://arxiv.org/abs/2303.12057v4 | Patrick Y. Wu, Jonathan Nagler, Joshua A. Tucker, Solomon Messing | 2023-09-26T21:24:13Z | cs.CY, cs.CL | Existing approaches to estimating politicians' latent positions along
specific dimensions often fail when relevant data is limited. We leverage the
embedded knowledge in generative large language models (LLMs) to address this
challenge and measure lawmakers' positions along specific political or policy
dimensions. We prompt an instruction/dialogue-tuned LLM to pairwise compare
lawmakers and then scale the resulting graph using the Bradley-Terry model. We
estimate novel measures of U.S. senators' positions on liberal-conservative
ideology, gun control, and abortion. Our liberal-conservative scale, used to
validate LLM-driven scaling, strongly correlates with existing measures and
offsets interpretive gaps, suggesting LLMs synthesize relevant data from
internet and digitized media rather than memorizing existing measures. Our gun
control and abortion measures -- the first of their kind -- differ from the
liberal-conservative scale in face-valid ways and predict interest group
ratings and legislator votes better than ideology alone. Our findings suggest
LLMs hold promise for solving complex social science measurement problems. | cs.CY | Large Language Models Can Be Used to Estimate the Latent Positions of Politicians |
2023-03-21T17:32:01Z | http://arxiv.org/pdf/2304.12155v1 | http://arxiv.org/abs/2304.12155v1 | Chris Emezue, Hellina Nigatu, Cynthia Thinwa, Helper Zhou, Shamsuddeen Muhammad, Lerato Louis, Idris Abdulmumin, Samuel Oyerinde, Benjamin Ajibade, Olanrewaju Samuel, Oviawe Joshua, Emeka Onwuegbuzia, Handel Emezue, Ifeoluwatayo A. Ige, Atnafu Lambebo Tonja, Chiamaka Chukwuneke, Bonaventure F. P. Dossou, Naome A. Etori, Mbonu Chinedu Emmanuel, Oreen Yousuf, Kaosarat Aina, Davis David | 2023-03-21T17:32:01Z | cs.CL, cs.LG | Stopwords are fundamental in Natural Language Processing (NLP) techniques for
information retrieval. One of the common tasks in preprocessing of text data is
the removal of stopwords. Currently, while high-resource languages like English
benefit from the availability of several stopwords, low-resource languages,
such as those found in the African continent, have none that are standardized
and available for use in NLP packages. Stopwords in the context of African
languages are understudied and can reveal information about the crossover
between languages. The \textit{African Stopwords} project aims to study and
curate stopwords for African languages. In this paper, we present our current
progress on ten African languages as well as future plans for the project. | cs.CL | The African Stopwords project: curating stopwords for African languages |
2023-03-21T17:30:40Z | http://arxiv.org/pdf/2303.12513v2 | http://arxiv.org/abs/2303.12513v2 | Morris Alper, Michael Fiman, Hadar Averbuch-Elor | 2023-11-02T10:38:49Z | cs.CV, cs.CL, cs.LG | Most humans use visual imagination to understand and reason about language,
but models such as BERT reason about language using knowledge acquired during
text-only pretraining. In this work, we investigate whether vision-and-language
pretraining can improve performance on text-only tasks that involve implicit
visual reasoning, focusing primarily on zero-shot probing methods. We propose a
suite of visual language understanding (VLU) tasks for probing the visual
reasoning abilities of text encoder models, as well as various non-visual
natural language understanding (NLU) tasks for comparison. We also contribute a
novel zero-shot knowledge probing method, Stroop probing, for applying models
such as CLIP to text-only tasks without needing a prediction head such as the
masked language modelling head of models like BERT. We show that SOTA
multimodally trained text encoders outperform unimodally trained text encoders
on the VLU tasks while being underperformed by them on the NLU tasks, lending
new context to previously mixed results regarding the NLU capabilities of
multimodal models. We conclude that exposure to images during pretraining
affords inherent visual reasoning knowledge that is reflected in language-only
tasks that require implicit visual reasoning. Our findings bear importance in
the broader context of multimodal learning, providing principled guidelines for
the choice of text encoders used in such contexts. | cs.CV | Is BERT Blind? Exploring the Effect of Vision-and-Language Pretraining on Visual Language Understanding |
2023-03-21T17:14:04Z | http://arxiv.org/pdf/2303.12029v1 | http://arxiv.org/abs/2303.12029v1 | Hong Zhang, Haewoon Kwak, Wei Gao, Jisun An | 2023-03-21T17:14:04Z | cs.SI, cs.CL | People who share similar opinions towards controversial topics could form an
echo chamber and may share similar political views toward other topics as well.
The existence of such connections, which we call connected behavior, gives
researchers a unique opportunity to predict how one would behave for a future
event given their past behaviors. In this work, we propose a framework to
conduct connected behavior analysis. Neural stance detection models are trained
on Twitter data collected on three seemingly independent topics, i.e., wearing
a mask, racial equality, and Trump, to detect people's stance, which we
consider as their online behavior in each topic-related event. Our results
reveal a strong connection between the stances toward the three topical events
and demonstrate the power of past behaviors in predicting one's future
behavior. | cs.SI | Wearing Masks Implies Refuting Trump?: Towards Target-specific User Stance Prediction across Events in COVID-19 and US Election 2020 |
2023-03-21T17:04:44Z | http://arxiv.org/pdf/2303.12024v3 | http://arxiv.org/abs/2303.12024v3 | Anirudh S Sundar, Larry Heck | 2023-05-31T00:44:56Z | cs.CL, cs.AI | Optimizing accuracy and performance while eliminating hallucinations of
open-domain conversational large language models (LLMs) is an open research
challenge. A particularly promising direction is to augment and ground LLMs
with information from structured sources. This paper introduces Conversational
Tables (cTBLS), a three-step architecture to retrieve and generate dialogue
responses grounded on retrieved tabular information. cTBLS uses Transformer
encoder embeddings for Dense Table Retrieval and obtains up to 125% relative
improvement over the retriever in the previous state-of-the-art system on the
HyrbiDialogue dataset. cTBLS then uses a shared process between encoder and
decoder models to perform a coarse+fine tabular knowledge (e.g., cell) ranking
combined with a GPT-3.5 LLM response generator to yield a 2x relative
improvement in ROUGE scores. Finally, human evaluators prefer cTBLs +80% of the
time (coherency, fluency) and judge informativeness to be 4x better than the
previous state-of-the-art. | cs.CL | cTBLS: Augmenting Large Language Models with Conversational Tables |
2023-03-21T16:56:05Z | http://arxiv.org/pdf/2303.12023v2 | http://arxiv.org/abs/2303.12023v2 | Zonglin Yang, Xinya Du, Rui Mao, Jinjie Ni, Erik Cambria | 2024-02-16T14:30:33Z | cs.CL, cs.AI | Logical reasoning is central to human cognition and intelligence. It includes
deductive, inductive, and abductive reasoning. Past research of logical
reasoning within AI uses formal language as knowledge representation and
symbolic reasoners. However, reasoning with formal language has proved
challenging (e.g., brittleness and knowledge-acquisition bottleneck). This
paper provides a comprehensive overview on a new paradigm of logical reasoning,
which uses natural language as knowledge representation and pretrained language
models as reasoners, including philosophical definition and categorization of
logical reasoning, advantages of the new paradigm, benchmarks and methods,
challenges of the new paradigm, possible future directions, and relation to
related NLP fields. This new paradigm is promising since it not only alleviates
many challenges of formal representation but also has advantages over
end-to-end neural methods. This survey focus on transformer-based LLMs
explicitly working on deductive, inductive, and abductive reasoning over
English representation. | cs.CL | Logical Reasoning over Natural Language as Knowledge Representation: A Survey |
2023-03-21T14:35:07Z | http://arxiv.org/pdf/2303.13367v2 | http://arxiv.org/abs/2303.13367v2 | Brady Lund, Ting Wang, Nishith Reddy Mannuru, Bing Nie, Somipam Shimray, Ziang Wang | 2023-03-31T17:56:28Z | cs.CL, cs.CY | This paper discusses OpenAIs ChatGPT, a generative pre-trained transformer,
which uses natural language processing to fulfill text-based user requests
(i.e., a chatbot). The history and principles behind ChatGPT and similar models
are discussed. This technology is then discussed in relation to its potential
impact on academia and scholarly research and publishing. ChatGPT is seen as a
potential model for the automated preparation of essays and other types of
scholarly manuscripts. Potential ethical issues that could arise with the
emergence of large language models like GPT-3, the underlying technology behind
ChatGPT, and its usage by academics and researchers, are discussed and situated
within the context of broader advancements in artificial intelligence, machine
learning, and natural language processing for research and scholarly
publishing. | cs.CL | ChatGPT and a New Academic Reality: Artificial Intelligence-Written Research Papers and the Ethics of the Large Language Models in Scholarly Publishing |
2023-03-21T12:55:54Z | http://arxiv.org/pdf/2303.11812v1 | http://arxiv.org/abs/2303.11812v1 | Tongquan Zhou, Siyi Cao, Siruo Zhou, Yao Zhang, Aijing He | 2023-03-21T12:55:54Z | cs.CL | ChatGPT is a publicly available chatbot that can quickly generate texts on
given topics, but it is unknown whether the chatbot is really superior to human
writers in all aspects of writing and whether its writing quality can be
prominently improved on the basis of updating commands. Consequently, this
study compared the writing performance on a narrative topic by ChatGPT and
Chinese intermediate English (CIE) learners so as to reveal the chatbot's
advantage and disadvantage in writing. The data were analyzed in terms of five
discourse components using Coh-Metrix (a special instrument for analyzing
language discourses), and the results revealed that ChatGPT performed better
than human writers in narrativity, word concreteness, and referential cohesion,
but worse in syntactic simplicity and deep cohesion in its initial version.
After more revision commands were updated, while the resulting version was
facilitated in syntactic simplicity, yet it is still lagged far behind CIE
learners' writing in deep cohesion. In addition, the correlation analysis of
the discourse components suggests that narrativity was correlated with
referential cohesion in both ChatGPT and human writers, but the correlations
varied within each group. | cs.CL | Chinese Intermediate English Learners outdid ChatGPT in deep cohesion: Evidence from English narrative writing |
2023-03-21T11:17:37Z | http://arxiv.org/pdf/2303.11750v1 | http://arxiv.org/abs/2303.11750v1 | Lei Lin, Shuangtao Li, Xiaodong Shi | 2023-03-21T11:17:37Z | cs.CL | Simultaneous machine translation, which aims at a real-time translation, is
useful in many live scenarios but very challenging due to the trade-off between
accuracy and latency. To achieve the balance for both, the model needs to wait
for appropriate streaming text (READ policy) and then generates its translation
(WRITE policy). However, WRITE policies of previous work either are specific to
the method itself due to the end-to-end training or suffer from the input
mismatch between training and decoding for the non-end-to-end training.
Therefore, it is essential to learn a generic and better WRITE policy for
simultaneous machine translation. Inspired by strategies utilized by human
interpreters and "wait" policies, we propose a novel adaptive prefix-to-prefix
training policy called LEAPT, which allows our machine translation model to
learn how to translate source sentence prefixes and make use of the future
context. Experiments show that our proposed methods greatly outperform
competitive baselines and achieve promising results. | cs.CL | LEAPT: Learning Adaptive Prefix-to-prefix Translation For Simultaneous Machine Translation |
2023-03-21T10:01:49Z | http://arxiv.org/pdf/2303.11708v2 | http://arxiv.org/abs/2303.11708v2 | Gabriel Skantze, A. Seza Doğruöz | 2023-07-28T09:22:02Z | cs.CL | There is a surge in interest in the development of open-domain chatbots,
driven by the recent advancements of large language models. The "openness" of
the dialogue is expected to be maximized by providing minimal information to
the users about the common ground they can expect, including the presumed joint
activity. However, evidence suggests that the effect is the opposite. Asking
users to "just chat about anything" results in a very narrow form of dialogue,
which we refer to as the "open-domain paradox". In this position paper, we
explain this paradox through the theory of common ground as the basis for
human-like communication. Furthermore, we question the assumptions behind
open-domain chatbots and identify paths forward for enabling common ground in
human-computer dialogue. | cs.CL | The Open-domain Paradox for Chatbots: Common Ground as the Basis for Human-like Dialogue |
2023-03-21T08:39:56Z | http://arxiv.org/pdf/2303.13631v2 | http://arxiv.org/abs/2303.13631v2 | Ping-Rui Tsai, Yen-Ting Chou, Nathan-Christopher Wang, Hui-Ling Chen, Hong-Yue Huang, Zih-Jia Luo, Tzay-Ming Hong | 2024-01-02T09:35:33Z | cs.SD, cs.AI, cs.CL, eess.AS | Music, enchanting and poetic, permeates every corner of human civilization.
Although music is not unfamiliar to people, our understanding of its essence
remains limited, and there is still no universally accepted scientific
description. This is primarily due to music being regarded as a product of both
reason and emotion, making it difficult to define. In this article, we focus on
the fundamental elements of music and construct an evolutionary network from
the perspective of music as a natural language, aligning with the statistical
characteristics of texts. Through this approach, we aim to comprehend the
structural differences in music across different periods, enabling a more
scientific exploration of music. Relying on the advantages of structuralism, we
can concentrate on the relationships and order between the physical elements of
music, rather than getting entangled in the blurred boundaries of science and
philosophy. The scientific framework we present not only conforms to past
conclusions in music, but also serves as a bridge that connects music to
natural language processing and knowledge graphs. | cs.SD | In-depth analysis of music structure as a text network |
2023-03-21T08:08:04Z | http://arxiv.org/pdf/2303.11660v1 | http://arxiv.org/abs/2303.11660v1 | Ming Shen, Jie Ma, Shuai Wang, Yogarshi Vyas, Kalpit Dixit, Miguel Ballesteros, Yassine Benajiba | 2023-03-21T08:08:04Z | cs.CL | Opinion summarization provides an important solution for summarizing opinions
expressed among a large number of reviews. However, generating aspect-specific
and general summaries is challenging due to the lack of annotated data. In this
work, we propose two simple yet effective unsupervised approaches to generate
both aspect-specific and general opinion summaries by training on synthetic
datasets constructed with aspect-related review contents. Our first approach,
Seed Words Based Leave-One-Out (SW-LOO), identifies aspect-related portions of
reviews simply by exact-matching aspect seed words and outperforms existing
methods by 3.4 ROUGE-L points on SPACE and 0.5 ROUGE-1 point on OPOSUM+ for
aspect-specific opinion summarization. Our second approach, Natural Language
Inference Based Leave-One-Out (NLI-LOO) identifies aspect-related sentences
utilizing an NLI model in a more general setting without using seed words and
outperforms existing approaches by 1.2 ROUGE-L points on SPACE for
aspect-specific opinion summarization and remains competitive on other metrics. | cs.CL | Simple Yet Effective Synthetic Dataset Construction for Unsupervised Opinion Summarization |
2023-03-21T07:46:57Z | http://arxiv.org/pdf/2303.11648v1 | http://arxiv.org/abs/2303.11648v1 | Gustavo Penha, Enrico Palumbo, Maryam Aziz, Alice Wang, Hugues Bouchard | 2023-03-21T07:46:57Z | cs.IR, cs.CL, cs.LG | An important goal of online platforms is to enable content discovery, i.e.
allow users to find a catalog entity they were not familiar with. A
pre-requisite to discover an entity, e.g. a book, with a search engine is that
the entity is retrievable, i.e. there are queries for which the system will
surface such entity in the top results. However, machine-learned search engines
have a high retrievability bias, where the majority of the queries return the
same entities. This happens partly due to the predominance of narrow intent
queries, where users create queries using the title of an already known entity,
e.g. in book search 'harry potter'. The amount of broad queries where users
want to discover new entities, e.g. in music search 'chill lyrical electronica
with an atmospheric feeling to it', and have a higher tolerance to what they
might find, is small in comparison. We focus here on two factors that have a
negative impact on the retrievability of the entities (I) the training data
used for dense retrieval models and (II) the distribution of narrow and broad
intent queries issued in the system. We propose CtrlQGen, a method that
generates queries for a chosen underlying intent-narrow or broad. We can use
CtrlQGen to improve factor (I) by generating training data for dense retrieval
models comprised of diverse synthetic queries. CtrlQGen can also be used to
deal with factor (II) by suggesting queries with broader intents to users. Our
results on datasets from the domains of music, podcasts, and books reveal that
we can significantly decrease the retrievability bias of a dense retrieval
model when using CtrlQGen. First, by using the generated queries as training
data for dense models we make 9% of the entities retrievable (go from zero to
non-zero retrievability). Second, by suggesting broader queries to users, we
can make 12% of the entities retrievable in the best case. | cs.IR | Improving Content Retrievability in Search with Controllable Query Generation |
2023-03-21T07:25:36Z | http://arxiv.org/pdf/2303.13373v1 | http://arxiv.org/abs/2303.13373v1 | Eduardo C. Garrido-Merchán, Cristina González-Barthe, María Coronado Vaca | 2023-03-21T07:25:36Z | cs.CL, cs.AI | In recent years there has been a growing demand from financial agents,
especially from particular and institutional investors, for companies to report
on climate-related financial risks. A vast amount of information, in text
format, can be expected to be disclosed in the short term by firms in order to
identify these types of risks in their financial and non financial reports,
particularly in response to the growing regulation that is being passed on the
matter. To this end, this paper applies state-of-the-art NLP techniques to
achieve the detection of climate change in text corpora. We use transfer
learning to fine-tune two transformer models, BERT and ClimateBert -a recently
published DistillRoBERTa-based model that has been specifically tailored for
climate text classification-. These two algorithms are based on the transformer
architecture which enables learning the contextual relationships between words
in a text. We carry out the fine-tuning process of both models on the novel
Clima-Text database, consisting of data collected from Wikipedia, 10K Files
Reports and web-based claims. Our text classification model obtained from the
ClimateBert fine-tuning process on ClimaText, outperforms the models created
with BERT and the current state-of-the-art transformer in this particular
problem. Our study is the first one to implement on the ClimaText database the
recently published ClimateBert algorithm. Based on our results, it can be said
that ClimateBert fine-tuned on ClimaText is an outstanding tool within the NLP
pre-trained transformer models that may and should be used by investors,
institutional agents and companies themselves to monitor the disclosure of
climate risk in financial reports. In addition, our transfer learning
methodology is cheap in computational terms, thus allowing any organization to
perform it. | cs.CL | Fine-tuning ClimateBert transformer with ClimaText for the disclosure analysis of climate-related financial risks |
2023-03-21T06:41:50Z | http://arxiv.org/pdf/2303.11621v1 | http://arxiv.org/abs/2303.11621v1 | Yiwei Li, Shaoxiong Feng, Bin Sun, Kan Li | 2023-03-21T06:41:50Z | cs.CL, cs.AI | With the development of deep learning, advanced dialogue generation methods
usually require a greater amount of computational resources. One promising
approach to obtaining a high-performance and lightweight model is knowledge
distillation, which relies heavily on the pre-trained powerful teacher.
Collaborative learning, also known as online knowledge distillation, is an
effective way to conduct one-stage group distillation in the absence of a
well-trained large teacher model. However, previous work has a severe branch
homogeneity problem due to the same training objective and the independent
identical training sets. To alleviate this problem, we consider the dialogue
attributes in the training of network branches. Each branch learns the
attribute-related features based on the selected subset. Furthermore, we
propose a dual group-based knowledge distillation method, consisting of
positive distillation and negative distillation, to further diversify the
features of different branches in a steadily and interpretable way. The
proposed approach significantly improves branch heterogeneity and outperforms
state-of-the-art collaborative learning methods on two widely used open-domain
dialogue datasets. | cs.CL | Heterogeneous-Branch Collaborative Learning for Dialogue Generation |
2023-03-21T06:00:39Z | http://arxiv.org/pdf/2303.11607v1 | http://arxiv.org/abs/2303.11607v1 | Siddique Latif, Aun Zaidi, Heriberto Cuayahuitl, Fahad Shamshad, Moazzam Shoukat, Junaid Qadir | 2023-03-21T06:00:39Z | cs.CL, cs.SD, eess.AS | The remarkable success of transformers in the field of natural language
processing has sparked the interest of the speech-processing community, leading
to an exploration of their potential for modeling long-range dependencies
within speech sequences. Recently, transformers have gained prominence across
various speech-related domains, including automatic speech recognition, speech
synthesis, speech translation, speech para-linguistics, speech enhancement,
spoken dialogue systems, and numerous multimodal applications. In this paper,
we present a comprehensive survey that aims to bridge research studies from
diverse subfields within speech technology. By consolidating findings from
across the speech technology landscape, we provide a valuable resource for
researchers interested in harnessing the power of transformers to advance the
field. We identify the challenges encountered by transformers in speech
processing while also offering insights into potential solutions to address
these issues. | cs.CL | Transformers in Speech Processing: A Survey |
2023-03-21T04:47:45Z | http://arxiv.org/pdf/2303.11593v4 | http://arxiv.org/abs/2303.11593v4 | Yasuhiro Yoshikai, Tadahaya Mizuno, Shumpei Nemoto, Hiroyuki Kusuhara | 2024-01-14T00:18:44Z | cs.LG, cs.CL, physics.chem-ph, q-bio.BM, J.2; I.2.7 | Recent years have seen rapid development of descriptor generation based on
representation learning of extremely diverse molecules, especially those that
apply natural language processing (NLP) models to SMILES, a literal
representation of molecular structure. However, little research has been done
on how these models understand chemical structure. To address this black box,
we investigated the relationship between the learning progress of SMILES and
chemical structure using a representative NLP model, the Transformer. We show
that while the Transformer learns partial structures of molecules quickly, it
requires extended training to understand overall structures. Consistently, the
accuracy of molecular property predictions using descriptors generated from
models at different learning steps was similar from the beginning to the end of
training. Furthermore, we found that the Transformer requires particularly long
training to learn chirality and sometimes stagnates with low performance due to
misunderstanding of enantiomers. These findings are expected to deepen the
understanding of NLP models in chemistry. | cs.LG | Difficulty in chirality recognition for Transformer architectures learning chemical structures from string |
2023-03-21T01:06:37Z | http://arxiv.org/pdf/2303.11525v3 | http://arxiv.org/abs/2303.11525v3 | Vithursan Thangarasa, Shreyas Saxena, Abhay Gupta, Sean Lie | 2024-03-05T22:12:38Z | cs.LG, cs.CL, cs.CV | Recent research has focused on weight sparsity in neural network training to
reduce FLOPs, aiming for improved efficiency (test accuracy w.r.t training
FLOPs). However, sparse weight training often sacrifices accuracy, requiring
extended training schedules to attain the accuracy of dense models. In
contrast, our approach, Sparse Iso-FLOP Transformations (Sparse-IFT), uses
sparsity to improve accuracy while maintaining dense model FLOPs. Using a
single hyperparameter (i.e., sparsity level), Sparse-IFTs efficiently replace
dense layers, expanding the search space for optimal sparse masks. In addition,
dynamic sparse training with Sparse-IFT models effectively navigates this
larger sparse mask-weight space, which is evidenced by a spectral analysis
using Ramanujan graph properties. Our study reveals a robust correlation among
mask topology, weights, and final performance. Notably, without adjusting
hyperparameters, replacing dense layers with Sparse-IFT yields significant
improvements, such as a +3.5% boost for ResNet-18 on ImageNet and +0.9% for
GPT-3 Small on the Open LLM leaderboard. To our knowledge, this is the first
work to demonstrate the use of sparsity for improving the accuracy of dense
models through a simple-to-use set of sparse transformations. Code is available
at: https://github.com/CerebrasResearch/Sparse-IFT. | cs.LG | Sparse-IFT: Sparse Iso-FLOP Transformations for Maximizing Training Efficiency |
2023-03-20T23:54:26Z | http://arxiv.org/pdf/2303.11504v2 | http://arxiv.org/abs/2303.11504v2 | Tyler A. Chang, Benjamin K. Bergen | 2023-08-26T00:33:23Z | cs.CL | Transformer language models have received widespread public attention, yet
their generated text is often surprising even to NLP researchers. In this
survey, we discuss over 250 recent studies of English language model behavior
before task-specific fine-tuning. Language models possess basic capabilities in
syntax, semantics, pragmatics, world knowledge, and reasoning, but these
capabilities are sensitive to specific inputs and surface features. Despite
dramatic increases in generated text quality as models scale to hundreds of
billions of parameters, the models are still prone to unfactual responses,
commonsense errors, memorized text, and social biases. Many of these weaknesses
can be framed as over-generalizations or under-generalizations of learned
patterns in text. We synthesize recent results to highlight what is currently
known about large language model capabilities, thus providing a resource for
applied work and for research in adjacent fields that use language models. | cs.CL | Language Model Behavior: A Comprehensive Survey |
2023-03-20T21:14:06Z | http://arxiv.org/pdf/2303.11455v1 | http://arxiv.org/abs/2303.11455v1 | Kevin Jesse, Toufique Ahmed, Premkumar T. Devanbu, Emily Morgan | 2023-03-20T21:14:06Z | cs.SE, cs.CL, cs.LG | With the advent of powerful neural language models, AI-based systems to
assist developers in coding tasks are becoming widely available; Copilot is one
such system. Copilot uses Codex, a large language model (LLM), to complete code
conditioned on a preceding "prompt". Codex, however, is trained on public
GitHub repositories, viz., on code that may include bugs and vulnerabilities.
Previous studies [1], [2] show Codex reproduces vulnerabilities seen in
training. In this study, we examine how prone Codex is to generate an
interesting bug category, single statement bugs, commonly referred to as
simple, stupid bugs or SStuBs in the MSR community. We find that Codex and
similar LLMs do help avoid some SStuBs, but do produce known, verbatim SStuBs
as much as 2x as likely than known, verbatim correct code. We explore the
consequences of the Codex generated SStuBs and propose avoidance strategies
that suggest the possibility of reducing the production of known, verbatim
SStubs, and increase the possibility of producing known, verbatim fixes. | cs.SE | Large Language Models and Simple, Stupid Bugs |
2023-03-20T20:28:26Z | http://arxiv.org/pdf/2303.11436v2 | http://arxiv.org/abs/2303.11436v2 | Sifatkaur Dhingra, Manmeet Singh, Vaisakh SB, Neetiraj Malviya, Sukhpal Singh Gill | 2023-04-12T15:46:20Z | cs.CL, cs.AI | Cognitive psychology delves on understanding perception, attention, memory,
language, problem-solving, decision-making, and reasoning. Large language
models (LLMs) are emerging as potent tools increasingly capable of performing
human-level tasks. The recent development in the form of GPT-4 and its
demonstrated success in tasks complex to humans exam and complex problems has
led to an increased confidence in the LLMs to become perfect instruments of
intelligence. Although GPT-4 report has shown performance on some cognitive
psychology tasks, a comprehensive assessment of GPT-4, via the existing
well-established datasets is required. In this study, we focus on the
evaluation of GPT-4's performance on a set of cognitive psychology datasets
such as CommonsenseQA, SuperGLUE, MATH and HANS. In doing so, we understand how
GPT-4 processes and integrates cognitive psychology with contextual
information, providing insight into the underlying cognitive processes that
enable its ability to generate the responses. We show that GPT-4 exhibits a
high level of accuracy in cognitive psychology tasks relative to the prior
state-of-the-art models. Our results strengthen the already available
assessments and confidence on GPT-4's cognitive psychology abilities. It has
significant potential to revolutionize the field of AI, by enabling machines to
bridge the gap between human and machine reasoning. | cs.CL | Mind meets machine: Unravelling GPT-4's cognitive psychology |
2023-03-20T19:20:34Z | http://arxiv.org/pdf/2303.11403v4 | http://arxiv.org/abs/2303.11403v4 | Mustafa Shukor, Corentin Dancette, Matthieu Cord | 2023-10-27T16:38:40Z | cs.CV, cs.CL, cs.LG | Large Language Models (LLMs) have so far impressed the world, with
unprecedented capabilities that emerge in models at large scales. On the vision
side, transformer models (i.e., ViT) are following the same trend, achieving
the best performance on challenging benchmarks. With the abundance of such
unimodal models, a natural question arises; do we need also to follow this
trend to tackle multimodal tasks? In this work, we propose to rather direct
effort to efficient adaptations of existing models, and propose to augment
Language Models with perception. Existing approaches for adapting pretrained
models for vision-language tasks still rely on several key components that
hinder their efficiency. In particular, they still train a large number of
parameters, rely on large multimodal pretraining, use encoders (e.g., CLIP)
trained on huge image-text datasets, and add significant inference overhead. In
addition, most of these approaches have focused on Zero-Shot and In Context
Learning, with little to no effort on direct finetuning. We investigate the
minimal computational effort needed to adapt unimodal models for multimodal
tasks and propose a new challenging setup, alongside different approaches, that
efficiently adapts unimodal pretrained models. We show that by freezing more
than 99% of total parameters, training only one linear projection layer, and
prepending only one trainable token, our approach (dubbed eP-ALM) significantly
outperforms other baselines on VQA and Captioning across Image, Video, and
Audio modalities, following the proposed setup. The code is available here:
https://github.com/mshukor/eP-ALM. | cs.CV | eP-ALM: Efficient Perceptual Augmentation of Language Models |
2023-03-20T18:31:47Z | http://arxiv.org/pdf/2303.11381v1 | http://arxiv.org/abs/2303.11381v1 | Zhengyuan Yang, Linjie Li, Jianfeng Wang, Kevin Lin, Ehsan Azarnasab, Faisal Ahmed, Zicheng Liu, Ce Liu, Michael Zeng, Lijuan Wang | 2023-03-20T18:31:47Z | cs.CV, cs.CL, cs.LG | We propose MM-REACT, a system paradigm that integrates ChatGPT with a pool of
vision experts to achieve multimodal reasoning and action. In this paper, we
define and explore a comprehensive list of advanced vision tasks that are
intriguing to solve, but may exceed the capabilities of existing vision and
vision-language models. To achieve such advanced visual intelligence, MM-REACT
introduces a textual prompt design that can represent text descriptions,
textualized spatial coordinates, and aligned file names for dense visual
signals such as images and videos. MM-REACT's prompt design allows language
models to accept, associate, and process multimodal information, thereby
facilitating the synergetic combination of ChatGPT and various vision experts.
Zero-shot experiments demonstrate MM-REACT's effectiveness in addressing the
specified capabilities of interests and its wide application in different
scenarios that require advanced visual understanding. Furthermore, we discuss
and compare MM-REACT's system paradigm with an alternative approach that
extends language models for multimodal scenarios through joint finetuning.
Code, demo, video, and visualization are available at
https://multimodal-react.github.io/ | cs.CV | MM-REACT: Prompting ChatGPT for Multimodal Reasoning and Action |
2023-03-20T18:08:50Z | http://arxiv.org/pdf/2303.11366v4 | http://arxiv.org/abs/2303.11366v4 | Noah Shinn, Federico Cassano, Edward Berman, Ashwin Gopinath, Karthik Narasimhan, Shunyu Yao | 2023-10-10T05:21:45Z | cs.AI, cs.CL, cs.LG | Large language models (LLMs) have been increasingly used to interact with
external environments (e.g., games, compilers, APIs) as goal-driven agents.
However, it remains challenging for these language agents to quickly and
efficiently learn from trial-and-error as traditional reinforcement learning
methods require extensive training samples and expensive model fine-tuning. We
propose Reflexion, a novel framework to reinforce language agents not by
updating weights, but instead through linguistic feedback. Concretely,
Reflexion agents verbally reflect on task feedback signals, then maintain their
own reflective text in an episodic memory buffer to induce better
decision-making in subsequent trials. Reflexion is flexible enough to
incorporate various types (scalar values or free-form language) and sources
(external or internally simulated) of feedback signals, and obtains significant
improvements over a baseline agent across diverse tasks (sequential
decision-making, coding, language reasoning). For example, Reflexion achieves a
91% pass@1 accuracy on the HumanEval coding benchmark, surpassing the previous
state-of-the-art GPT-4 that achieves 80%. We also conduct ablation and analysis
studies using different feedback signals, feedback incorporation methods, and
agent types, and provide insights into how they affect performance. | cs.AI | Reflexion: Language Agents with Verbal Reinforcement Learning |
2023-03-20T17:59:59Z | http://arxiv.org/pdf/2303.11331v2 | http://arxiv.org/abs/2303.11331v2 | Yuxin Fang, Quan Sun, Xinggang Wang, Tiejun Huang, Xinlong Wang, Yue Cao | 2023-03-22T14:10:37Z | cs.CV, cs.CL | We launch EVA-02, a next-generation Transformer-based visual representation
pre-trained to reconstruct strong and robust language-aligned vision features
via masked image modeling. With an updated plain Transformer architecture as
well as extensive pre-training from an open & accessible giant CLIP vision
encoder, EVA-02 demonstrates superior performance compared to prior
state-of-the-art approaches across various representative vision tasks, while
utilizing significantly fewer parameters and compute budgets. Notably, using
exclusively publicly accessible training data, EVA-02 with only 304M parameters
achieves a phenomenal 90.0 fine-tuning top-1 accuracy on ImageNet-1K val set.
Additionally, our EVA-02-CLIP can reach up to 80.4 zero-shot top-1 on
ImageNet-1K, outperforming the previous largest & best open-sourced CLIP with
only ~1/6 parameters and ~1/6 image-text training data. We offer four EVA-02
variants in various model sizes, ranging from 6M to 304M parameters, all with
impressive performance. To facilitate open access and open research, we release
the complete suite of EVA-02 to the community at
https://github.com/baaivision/EVA/tree/master/EVA-02. | cs.CV | EVA-02: A Visual Representation for Neon Genesis |
2023-03-20T17:54:58Z | http://arxiv.org/pdf/2303.11315v2 | http://arxiv.org/abs/2303.11315v2 | Wenxuan Zhou, Sheng Zhang, Hoifung Poon, Muhao Chen | 2023-10-23T03:25:13Z | cs.CL | Large language models (LLMs) encode parametric knowledge about world facts
and have shown remarkable performance in knowledge-driven NLP tasks. However,
their reliance on parametric knowledge may cause them to overlook contextual
cues, leading to incorrect predictions in context-sensitive NLP tasks (e.g.,
knowledge acquisition tasks). In this paper, we seek to assess and enhance
LLMs' contextual faithfulness in two aspects: knowledge conflict and prediction
with abstention. We demonstrate that LLMs' faithfulness can be significantly
improved using carefully designed prompting strategies. In particular, we
identify opinion-based prompts and counterfactual demonstrations as the most
effective methods. Opinion-based prompts reframe the context as a narrator's
statement and inquire about the narrator's opinions, while counterfactual
demonstrations use instances containing false facts to improve faithfulness in
knowledge conflict situations. Neither technique requires additional training.
We conduct experiments on three datasets of two standard NLP tasks, machine
reading comprehension and relation extraction, and the results demonstrate
significant improvement in faithfulness to contexts. Code and data are released
at https://github.com/wzhouad/context-faithful-llm. | cs.CL | Context-faithful Prompting for Large Language Models |
2023-03-20T17:05:13Z | http://arxiv.org/pdf/2303.13314v1 | http://arxiv.org/abs/2303.13314v1 | Shaina Raza, Syed Raza Bashir | 2023-03-20T17:05:13Z | cs.CL, cs.AI | Infectious diseases are a significant public health concern globally, and
extracting relevant information from scientific literature can facilitate the
development of effective prevention and treatment strategies. However, the
large amount of clinical data available presents a challenge for information
extraction. To address this challenge, this study proposes a natural language
processing (NLP) framework that uses a pre-trained transformer model fine-tuned
on task-specific data to extract key information related to infectious diseases
from free-text clinical data. The proposed framework includes three components:
a data layer for preparing datasets from clinical texts, a foundation model
layer for entity extraction, and an assessment layer for performance analysis.
The results of the evaluation indicate that the proposed method outperforms
standard methods, and leveraging prior knowledge through the pre-trained
transformer model makes it useful for investigating other infectious diseases
in the future. | cs.CL | Leveraging Foundation Models for Clinical Text Analysis |
2023-03-20T16:53:36Z | http://arxiv.org/pdf/2303.13475v2 | http://arxiv.org/abs/2303.13475v2 | Sohom Ghosh, Ankush Chopra, Sudip Kumar Naskar | 2023-08-12T23:51:53Z | cs.CL | Over the years, there has been a paradigm shift in how users access financial
services. With the advancement of digitalization more users have been
preferring the online mode of performing financial activities. This has led to
the generation of a huge volume of financial content. Most investors prefer to
go through these contents before making decisions. Every industry has terms
that are specific to the domain it operates in. Banking and Financial Services
are not an exception to this. In order to fully comprehend these contents, one
needs to have a thorough understanding of the financial terms. Getting a basic
idea about a term becomes easy when it is explained with the help of the broad
category to which it belongs. This broad category is referred to as hypernym.
For example, "bond" is a hypernym of the financial term "alternative
debenture". In this paper, we propose a system capable of extracting and
ranking hypernyms for a given financial term. The system has been trained with
financial text corpora obtained from various sources like DBpedia [4],
Investopedia, Financial Industry Business Ontology (FIBO), prospectus and so
on. Embeddings of these terms have been extracted using FinBERT [3], FinISH [1]
and fine-tuned using SentenceBERT [54]. A novel approach has been used to
augment the training set with negative samples. It uses the hierarchy present
in FIBO. Finally, we benchmark the system performance with that of the existing
ones. We establish that it performs better than the existing ones and is also
scalable. | cs.CL | Learning Semantic Text Similarity to rank Hypernyms of Financial Terms |
2023-03-20T16:18:38Z | http://arxiv.org/pdf/2303.13375v2 | http://arxiv.org/abs/2303.13375v2 | Harsha Nori, Nicholas King, Scott Mayer McKinney, Dean Carignan, Eric Horvitz | 2023-04-12T16:48:39Z | cs.CL, cs.AI | Large language models (LLMs) have demonstrated remarkable capabilities in
natural language understanding and generation across various domains, including
medicine. We present a comprehensive evaluation of GPT-4, a state-of-the-art
LLM, on medical competency examinations and benchmark datasets. GPT-4 is a
general-purpose model that is not specialized for medical problems through
training or engineered to solve clinical tasks. Our analysis covers two sets of
official practice materials for the USMLE, a three-step examination program
used to assess clinical competency and grant licensure in the United States. We
also evaluate performance on the MultiMedQA suite of benchmark datasets. Beyond
measuring model performance, experiments were conducted to investigate the
influence of test questions containing both text and images on model
performance, probe for memorization of content during training, and study
probability calibration, which is of critical importance in high-stakes
applications like medicine. Our results show that GPT-4, without any
specialized prompt crafting, exceeds the passing score on USMLE by over 20
points and outperforms earlier general-purpose models (GPT-3.5) as well as
models specifically fine-tuned on medical knowledge (Med-PaLM, a prompt-tuned
version of Flan-PaLM 540B). In addition, GPT-4 is significantly better
calibrated than GPT-3.5, demonstrating a much-improved ability to predict the
likelihood that its answers are correct. We also explore the behavior of the
model qualitatively through a case study that shows the ability of GPT-4 to
explain medical reasoning, personalize explanations to students, and
interactively craft new counterfactual scenarios around a medical case.
Implications of the findings are discussed for potential uses of GPT-4 in
medical education, assessment, and clinical practice, with appropriate
attention to challenges of accuracy and safety. | cs.CL | Capabilities of GPT-4 on Medical Challenge Problems |
2023-03-20T15:22:11Z | http://arxiv.org/pdf/2303.11192v1 | http://arxiv.org/abs/2303.11192v1 | Vilém Zouhar, Sunit Bhattacharya, Ondřej Bojar | 2023-03-20T15:22:11Z | cs.CL | The Shannon game has long been used as a thought experiment in linguistics
and NLP, asking participants to guess the next letter in a sentence based on
its preceding context. We extend the game by introducing an optional extra
modality in the form of image information. To investigate the impact of
multimodal information in this game, we use human participants and a language
model (LM, GPT-2). We show that the addition of image information improves both
self-reported confidence and accuracy for both humans and LM. Certain word
classes, such as nouns and determiners, benefit more from the additional
modality information. The priming effect in both humans and the LM becomes more
apparent as the context size (extra modality information + sentence context)
increases. These findings highlight the potential of multimodal information in
improving language understanding and modeling. | cs.CL | Multimodal Shannon Game with Images |
2023-03-20T15:10:45Z | http://arxiv.org/pdf/2303.11184v1 | http://arxiv.org/abs/2303.11184v1 | Jiaqing Yuan, Munindar P. Singh | 2023-03-20T15:10:45Z | cs.CL | Conversations among online users sometimes derail, i.e., break down into
personal attacks. Such derailment has a negative impact on the healthy growth
of cyberspace communities. The ability to predict whether ongoing conversations
are likely to derail could provide valuable real-time insight to interlocutors
and moderators. Prior approaches predict conversation derailment
retrospectively without the ability to forestall the derailment proactively.
Some works attempt to make dynamic prediction as the conversation develops, but
fail to incorporate multisource information, such as conversation structure and
distance to derailment.
We propose a hierarchical transformer-based framework that combines
utterance-level and conversation-level information to capture fine-grained
contextual semantics. We propose a domain-adaptive pretraining objective to
integrate conversational structure information and a multitask learning scheme
to leverage the distance from each utterance to derailment. An evaluation of
our framework on two conversation derailment datasets yields improvement over
F1 score for the prediction of derailment. These results demonstrate the
effectiveness of incorporating multisource information. | cs.CL | Conversation Modeling to Predict Derailment |
2023-03-20T14:19:58Z | http://arxiv.org/pdf/2303.11141v2 | http://arxiv.org/abs/2303.11141v2 | Hongbo Wang, Weimin Xiong, Yifan Song, Dawei Zhu, Yu Xia, Sujian Li | 2023-03-21T09:03:14Z | cs.CL | Joint entity and relation extraction (JERE) is one of the most important
tasks in information extraction. However, most existing works focus on
sentence-level coarse-grained JERE, which have limitations in real-world
scenarios. In this paper, we construct a large-scale document-level
fine-grained JERE dataset DocRED-FE, which improves DocRED with Fine-Grained
Entity Type. Specifically, we redesign a hierarchical entity type schema
including 11 coarse-grained types and 119 fine-grained types, and then
re-annotate DocRED manually according to this schema. Through comprehensive
experiments we find that: (1) DocRED-FE is challenging to existing JERE models;
(2) Our fine-grained entity types promote relation classification. We make
DocRED-FE with instruction and the code for our baselines publicly available at
https://github.com/PKU-TANGENT/DOCRED-FE. | cs.CL | DocRED-FE: A Document-Level Fine-Grained Entity And Relation Extraction Dataset |
2023-03-20T14:07:13Z | http://arxiv.org/pdf/2303.11131v1 | http://arxiv.org/abs/2303.11131v1 | Maryam Fazel-Zarandi, Wei-Ning Hsu | 2023-03-20T14:07:13Z | cs.CL, cs.LG, cs.SD, eess.AS | Self-supervised learning leverages unlabeled data effectively, improving
label efficiency and generalization to domains without labeled data. While
recent work has studied generalization to more acoustic/linguistic domains,
languages, and modalities, these investigations are limited to single-source
speech with one primary speaker in the recording. This paper presents Cocktail
HuBERT, a self-supervised learning framework that generalizes to mixture speech
using a masked pseudo source separation objective. This objective encourages
the model to identify the number of sources, separate and understand the
context, and infer the content of masked regions represented as discovered
units. Cocktail HuBERT outperforms state-of-the-art results with 69% lower WER
on multi-speaker ASR, 31% lower DER on diarization, and is competitive on
single- and multi-speaker tasks from SUPERB. | cs.CL | Cocktail HuBERT: Generalized Self-Supervised Pre-training for Mixture and Single-Source Speech |
2023-03-20T13:58:35Z | http://arxiv.org/pdf/2303.11117v5 | http://arxiv.org/abs/2303.11117v5 | Yingjian Liu, Jiang Li, Xiaoping Wang, Zhigang Zeng | 2024-03-20T11:35:04Z | cs.CL | Emotion Recognition in Conversation (ERC) has attracted growing attention in
recent years as a result of the advancement and implementation of
human-computer interface technologies. In this paper, we propose an emotional
inertia and contagion-driven dependency modeling approach (EmotionIC) for ERC
task. Our EmotionIC consists of three main components, i.e., Identity Masked
Multi-Head Attention (IMMHA), Dialogue-based Gated Recurrent Unit (DiaGRU), and
Skip-chain Conditional Random Field (SkipCRF). Compared to previous ERC models,
EmotionIC can model a conversation more thoroughly at both the
feature-extraction and classification levels. The proposed model attempts to
integrate the advantages of attention- and recurrence-based methods at the
feature-extraction level. Specifically, IMMHA is applied to capture
identity-based global contextual dependencies, while DiaGRU is utilized to
extract speaker- and temporal-aware local contextual information. At the
classification level, SkipCRF can explicitly mine complex emotional flows from
higher-order neighboring utterances in the conversation. Experimental results
show that our method can significantly outperform the state-of-the-art models
on four benchmark datasets. The ablation studies confirm that our modules can
effectively model emotional inertia and contagion. | cs.CL | EmotionIC: emotional inertia and contagion-driven dependency modeling for emotion recognition in conversation |
2023-03-20T13:14:59Z | http://arxiv.org/pdf/2303.11082v1 | http://arxiv.org/abs/2303.11082v1 | Blerta Veseli, Sneha Singhania, Simon Razniewski, Gerhard Weikum | 2023-03-20T13:14:59Z | cs.CL, cs.AI | Structured knowledge bases (KBs) are a foundation of many intelligent
applications, yet are notoriously incomplete. Language models (LMs) have
recently been proposed for unsupervised knowledge base completion (KBC), yet,
despite encouraging initial results, questions regarding their suitability
remain open. Existing evaluations often fall short because they only evaluate
on popular subjects, or sample already existing facts from KBs. In this work,
we introduce a novel, more challenging benchmark dataset, and a methodology
tailored for a realistic assessment of the KBC potential of LMs. For automated
assessment, we curate a dataset called WD-KNOWN, which provides an unbiased
random sample of Wikidata, containing over 3.9 million facts. In a second step,
we perform a human evaluation on predictions that are not yet in the KB, as
only this provides real insights into the added value over existing KBs. Our
key finding is that biases in dataset conception of previous benchmarks lead to
a systematic overestimate of LM performance for KBC. However, our results also
reveal strong areas of LMs. We could, for example, perform a significant
completion of Wikidata on the relations nativeLanguage, by a factor of ~21
(from 260k to 5.8M) at 82% precision, usedLanguage, by a factor of ~2.1 (from
2.1M to 6.6M) at 82% precision, and citizenOf by a factor of ~0.3 (from 4.2M to
5.3M) at 90% precision. Moreover, we find that LMs possess surprisingly strong
generalization capabilities: even on relations where most facts were not
directly observed in LM training, prediction quality can be high. | cs.CL | Evaluating Language Models for Knowledge Base Completion |
2023-03-20T11:34:37Z | http://arxiv.org/pdf/2303.11032v2 | http://arxiv.org/abs/2303.11032v2 | Zhengliang Liu, Yue Huang, Xiaowei Yu, Lu Zhang, Zihao Wu, Chao Cao, Haixing Dai, Lin Zhao, Yiwei Li, Peng Shu, Fang Zeng, Lichao Sun, Wei Liu, Dinggang Shen, Quanzheng Li, Tianming Liu, Dajiang Zhu, Xiang Li | 2023-12-21T16:13:05Z | cs.CL, cs.CY | The digitization of healthcare has facilitated the sharing and re-using of
medical data but has also raised concerns about confidentiality and privacy.
HIPAA (Health Insurance Portability and Accountability Act) mandates removing
re-identifying information before the dissemination of medical records. Thus,
effective and efficient solutions for de-identifying medical data, especially
those in free-text forms, are highly needed. While various computer-assisted
de-identification methods, including both rule-based and learning-based, have
been developed and used in prior practice, such solutions still lack
generalizability or need to be fine-tuned according to different scenarios,
significantly imposing restrictions in wider use. The advancement of large
language models (LLM), such as ChatGPT and GPT-4, have shown great potential in
processing text data in the medical domain with zero-shot in-context learning,
especially in the task of privacy protection, as these models can identify
confidential information by their powerful named entity recognition (NER)
capability. In this work, we developed a novel GPT4-enabled de-identification
framework (``DeID-GPT") to automatically identify and remove the identifying
information. Compared to existing commonly used medical text data
de-identification methods, our developed DeID-GPT showed the highest accuracy
and remarkable reliability in masking private information from the unstructured
medical text while preserving the original structure and meaning of the text.
This study is one of the earliest to utilize ChatGPT and GPT-4 for medical text
data processing and de-identification, which provides insights for further
research and solution development on the use of LLMs such as ChatGPT/GPT-4 in
healthcare. Codes and benchmarking data information are available at
https://github.com/yhydhx/ChatGPT-API. | cs.CL | DeID-GPT: Zero-shot Medical Text De-Identification by GPT-4 |
2023-03-20T10:33:06Z | http://arxiv.org/pdf/2303.11005v1 | http://arxiv.org/abs/2303.11005v1 | Li Yi | 2023-03-20T10:33:06Z | cs.CL, cs.AI, cs.IR, cs.LG | Generating lyrics and poems is one of the essential downstream tasks in the
Natural Language Processing (NLP) field. Current methods have performed well in
some lyrics generation scenarios but need further improvements in tasks
requiring fine control. We propose a novel method for generating ancient
Chinese lyrics (Song Ci), a type of ancient lyrics that involves precise
control of song structure. The system is equipped with a phrase retriever and a
phrase connector. Based on an input prompt, the phrase retriever picks phrases
from a database to construct a phrase pool. The phrase connector then selects a
series of phrases from the phrase pool that minimizes a multi-term loss
function that considers rhyme, song structure, and fluency. Experimental
results show that our method can generate high-quality ancient Chinese lyrics
while performing well on topic and song structure control. We also expect our
approach to be generalized to other lyrics-generating tasks. | cs.CL | Controllable Ancient Chinese Lyrics Generation Based on Phrase Prototype Retrieving |
2023-03-20T09:52:52Z | http://arxiv.org/pdf/2303.10974v2 | http://arxiv.org/abs/2303.10974v2 | Andrei Chertkov, Olga Tsymboi, Mikhail Pautov, Ivan Oseledets | 2023-05-23T19:19:54Z | cs.CL, cs.AI | Neural networks are deployed widely in natural language processing tasks on
the industrial scale, and perhaps the most often they are used as compounds of
automatic machine translation systems. In this work, we present a simple
approach to fool state-of-the-art machine translation tools in the task of
translation from Russian to English and vice versa. Using a novel black-box
gradient-free tensor-based optimizer, we show that many online translation
tools, such as Google, DeepL, and Yandex, may both produce wrong or offensive
translations for nonsensical adversarial input queries and refuse to translate
seemingly benign input phrases. This vulnerability may interfere with
understanding a new language and simply worsen the user's experience while
using machine translation systems, and, hence, additional improvements of these
tools are required to establish better translation. | cs.CL | Translate your gibberish: black-box adversarial attack on machine translation systems |
2023-03-20T09:41:28Z | http://arxiv.org/pdf/2303.10966v2 | http://arxiv.org/abs/2303.10966v2 | Rongxiang Weng, Qiang Wang, Wensen Cheng, Changfeng Zhu, Min Zhang | 2023-09-19T07:37:31Z | cs.CL | Neural machine translation (NMT) has achieved remarkable success in producing
high-quality translations. However, current NMT systems suffer from a lack of
reliability, as their outputs that are often affected by lexical or syntactic
changes in inputs, resulting in large variations in quality. This limitation
hinders the practicality and trustworthiness of NMT. A contributing factor to
this problem is that NMT models trained with the one-to-one paradigm struggle
to handle the source diversity phenomenon, where inputs with the same meaning
can be expressed differently. In this work, we treat this problem as a bilevel
optimization problem and present a consistency-aware meta-learning (CAML)
framework derived from the model-agnostic meta-learning (MAML) algorithm to
address it. Specifically, the NMT model with CAML (named CoNMT) first learns a
consistent meta representation of semantically equivalent sentences in the
outer loop. Subsequently, a mapping from the meta representation to the output
sentence is learned in the inner loop, allowing the NMT model to translate
semantically equivalent sentences to the same target sentence. We conduct
experiments on the NIST Chinese to English task, three WMT translation tasks,
and the TED M2O task. The results demonstrate that CoNMT effectively improves
overall translation quality and reliably handles diverse inputs. | cs.CL | Towards Reliable Neural Machine Translation with Consistency-Aware Meta-Learning |
2023-03-20T09:13:27Z | http://arxiv.org/pdf/2303.10949v1 | http://arxiv.org/abs/2303.10949v1 | Haibin Yu, Yuxuan Hu, Yao Qian, Ma Jin, Linquan Liu, Shujie Liu, Yu Shi, Yanmin Qian, Edward Lin, Michael Zeng | 2023-03-20T09:13:27Z | eess.AS, cs.CL, cs.SD | Code-switching speech refers to a means of expression by mixing two or more
languages within a single utterance. Automatic Speech Recognition (ASR) with
End-to-End (E2E) modeling for such speech can be a challenging task due to the
lack of data. In this study, we investigate text generation and injection for
improving the performance of an industry commonly-used streaming model,
Transformer-Transducer (T-T), in Mandarin-English code-switching speech
recognition. We first propose a strategy to generate code-switching text data
and then investigate injecting generated text into T-T model explicitly by
Text-To-Speech (TTS) conversion or implicitly by tying speech and text latent
spaces. Experimental results on the T-T model trained with a dataset containing
1,800 hours of real Mandarin-English code-switched speech show that our
approaches to inject generated code-switching text significantly boost the
performance of T-T models, i.e., 16% relative Token-based Error Rate (TER)
reduction averaged on three evaluation sets, and the approach of tying speech
and text latent spaces is superior to that of TTS conversion on the evaluation
set which contains more homogeneous data with the training set. | eess.AS | Code-Switching Text Generation and Injection in Mandarin-English ASR |
2023-03-20T08:54:40Z | http://arxiv.org/pdf/2303.10942v1 | http://arxiv.org/abs/2303.10942v1 | Bolaji Yusuf, Aditya Gourav, Ankur Gandhe, Ivan Bulyko | 2023-03-20T08:54:40Z | cs.CL, cs.SD, eess.AS | End-to-end speech recognition models are improved by incorporating external
text sources, typically by fusion with an external language model. Such
language models have to be retrained whenever the corpus of interest changes.
Furthermore, since they store the entire corpus in their parameters, rare words
can be challenging to recall. In this work, we propose augmenting a
transducer-based ASR model with a retrieval language model, which directly
retrieves from an external text corpus plausible completions for a partial ASR
hypothesis. These completions are then integrated into subsequent predictions
by an adapter, which is trained once, so that the corpus of interest can be
switched without incurring the computational overhead of retraining. Our
experiments show that the proposed model significantly improves the performance
of a transducer baseline on a pair of question-answering datasets. Further, it
outperforms shallow fusion on recognition of named entities by about 7
relative; when the two are combined, the relative improvement increases to 13%. | cs.CL | On-the-fly Text Retrieval for End-to-End ASR Adaptation |
2023-03-20T07:09:26Z | http://arxiv.org/pdf/2303.10912v1 | http://arxiv.org/abs/2303.10912v1 | Fan Cui, Liyong Guo, Quandong Wang, Peng Gao, Yujun Wang | 2023-03-20T07:09:26Z | cs.SD, cs.CL, cs.LG, eess.AS | In this paper, we investigate representation learning for low-resource
keyword spotting (KWS). The main challenges of KWS are limited labeled data and
limited available device resources. To address those challenges, we explore
representation learning for KWS by self-supervised contrastive learning and
self-training with pretrained model. First, local-global contrastive siamese
networks (LGCSiam) are designed to learn similar utterance-level
representations for similar audio samplers by proposed local-global contrastive
loss without requiring ground-truth. Second, a self-supervised pretrained
Wav2Vec 2.0 model is applied as a constraint module (WVC) to force the KWS
model to learn frame-level acoustic representations. By the LGCSiam and WVC
modules, the proposed small-footprint KWS model can be pretrained with
unlabeled data. Experiments on speech commands dataset show that the
self-training WVC module and the self-supervised LGCSiam module significantly
improve accuracy, especially in the case of training on a small labeled
dataset. | cs.SD | Exploring Representation Learning for Small-Footprint Keyword Spotting |
2023-03-20T06:34:22Z | http://arxiv.org/pdf/2303.10897v1 | http://arxiv.org/abs/2303.10897v1 | Fan Cui, Liyong Guo, Lang He, Jiyao Liu, ErCheng Pei, Yujun Wang, Dongmei Jiang | 2023-03-20T06:34:22Z | cs.SD, cs.CL, eess.AS, q-bio.NC | Electroencephalography (EEG) plays a vital role in detecting how brain
responses to different stimulus. In this paper, we propose a novel Shallow-Deep
Attention-based Network (SDANet) to classify the correct auditory stimulus
evoking the EEG signal. It adopts the Attention-based Correlation Module (ACM)
to discover the connection between auditory speech and EEG from global aspect,
and the Shallow-Deep Similarity Classification Module (SDSCM) to decide the
classification result via the embeddings learned from the shallow and deep
layers. Moreover, various training strategies and data augmentation are used to
boost the model robustness. Experiments are conducted on the dataset provided
by Auditory EEG challenge (ICASSP Signal Processing Grand Challenge 2023).
Results show that the proposed model has a significant gain over the baseline
on the match-mismatch track. | cs.SD | Relate auditory speech to EEG by shallow-deep attention-based network |
2023-03-20T06:20:03Z | http://arxiv.org/pdf/2303.10893v2 | http://arxiv.org/abs/2303.10893v2 | Xinnian Liang, Zefan Zhou, Hui Huang, Shuangzhi Wu, Tong Xiao, Muyun Yang, Zhoujun Li, Chao Bian | 2023-03-22T03:20:27Z | cs.CL | Pretrained language models (PLMs) have shown marvelous improvements across
various NLP tasks. Most Chinese PLMs simply treat an input text as a sequence
of characters, and completely ignore word information. Although Whole Word
Masking can alleviate this, the semantics in words is still not well
represented. In this paper, we revisit the segmentation granularity of Chinese
PLMs. We propose a mixed-granularity Chinese BERT (MigBERT) by considering both
characters and words. To achieve this, we design objective functions for
learning both character and word-level representations. We conduct extensive
experiments on various Chinese NLP tasks to evaluate existing PLMs as well as
the proposed MigBERT. Experimental results show that MigBERT achieves new SOTA
performance on all these tasks. Further analysis demonstrates that words are
semantically richer than characters. More interestingly, we show that MigBERT
also works with Japanese. Our code and model have been released
here~\footnote{https://github.com/xnliang98/MigBERT}. | cs.CL | Character, Word, or Both? Revisiting the Segmentation Granularity for Chinese Pre-trained Language Models |
2023-03-20T06:19:49Z | http://arxiv.org/pdf/2303.11945v1 | http://arxiv.org/abs/2303.11945v1 | Hongyan Ran, Caiyan Jia | 2023-03-20T06:19:49Z | cs.SI, cs.AI, cs.CL, cs.LG | Massive rumors usually appear along with breaking news or trending topics,
seriously hindering the truth. Existing rumor detection methods are mostly
focused on the same domain, and thus have poor performance in cross-domain
scenarios due to domain shift. In this work, we propose an end-to-end
instance-wise and prototype-wise contrastive learning model with a
cross-attention mechanism for cross-domain rumor detection. The model not only
performs cross-domain feature alignment but also enforces target samples to
align with the corresponding prototypes of a given source domain. Since target
labels in a target domain are unavailable, we use a clustering-based approach
with carefully initialized centers by a batch of source domain samples to
produce pseudo labels. Moreover, we use a cross-attention mechanism on a pair
of source data and target data with the same labels to learn domain-invariant
representations. Because the samples in a domain pair tend to express similar
semantic patterns, especially on the people's attitudes (e.g., supporting or
denying) towards the same category of rumors, the discrepancy between a pair of
the source domain and target domain will be decreased. We conduct experiments
on four groups of cross-domain datasets and show that our proposed model
achieves state-of-the-art performance. | cs.SI | Unsupervised Cross-Domain Rumor Detection with Contrastive Learning and Cross-Attention |
2023-03-20T06:13:03Z | http://arxiv.org/pdf/2303.10888v1 | http://arxiv.org/abs/2303.10888v1 | Chanjun Park, Hyeonseok Moon, Seolhwa Lee, Jaehyung Seo, Sugyeong Eo, Heuiseok Lim | 2023-03-20T06:13:03Z | cs.CL, cs.AI | Leaderboard systems allow researchers to objectively evaluate Natural
Language Processing (NLP) models and are typically used to identify models that
exhibit superior performance on a given task in a predetermined setting.
However, we argue that evaluation on a given test dataset is just one of many
performance indications of the model. In this paper, we claim leaderboard
competitions should also aim to identify models that exhibit the best
performance in a real-world setting. We highlight three issues with current
leaderboard systems: (1) the use of a single, static test set, (2) discrepancy
between testing and real-world application (3) the tendency for
leaderboard-centric competition to be biased towards the test set. As a
solution, we propose a new paradigm of leaderboard systems that addresses these
issues of current leaderboard system. Through this study, we hope to induce a
paradigm shift towards more real -world-centric leaderboard competitions. | cs.CL | Self-Improving-Leaderboard(SIL): A Call for Real-World Centric Natural Language Processing Leaderboards |
2023-03-20T05:11:22Z | http://arxiv.org/pdf/2303.10870v1 | http://arxiv.org/abs/2303.10870v1 | Ying Mo, Hongyin Tang, Jiahao Liu, Qifan Wang, Zenglin Xu, Jingang Wang, Wei Wu, Zhoujun Li | 2023-03-20T05:11:22Z | cs.CL | Named entity recognition (NER) is an important research problem in natural
language processing. There are three types of NER tasks, including flat, nested
and discontinuous entity recognition. Most previous sequential labeling models
are task-specific, while recent years have witnessed the rising of generative
models due to the advantage of unifying all NER tasks into the seq2seq model
framework. Although achieving promising performance, our pilot studies
demonstrate that existing generative models are ineffective at detecting entity
boundaries and estimating entity types. This paper proposes a multi-task
Transformer, which incorporates an entity boundary detection task into the
named entity recognition task. More concretely, we achieve entity boundary
detection by classifying the relations between tokens within the sentence. To
improve the accuracy of entity-type mapping during decoding, we adopt an
external knowledge base to calculate the prior entity-type distributions and
then incorporate the information into the model via the self and
cross-attention mechanisms. We perform experiments on an extensive set of NER
benchmarks, including two flat, three nested, and three discontinuous NER
datasets. Experimental results show that our approach considerably improves the
generative NER model's performance. | cs.CL | Multi-task Transformer with Relation-attention and Type-attention for Named Entity Recognition |
2023-03-20T05:07:41Z | http://arxiv.org/pdf/2303.10868v3 | http://arxiv.org/abs/2303.10868v3 | Ruochen Zhao, Hailin Chen, Weishi Wang, Fangkai Jiao, Xuan Long Do, Chengwei Qin, Bosheng Ding, Xiaobao Guo, Minzhi Li, Xingxuan Li, Shafiq Joty | 2023-12-01T02:58:09Z | cs.CL | As Large Language Models (LLMs) become popular, there emerged an important
trend of using multimodality to augment the LLMs' generation ability, which
enables LLMs to better interact with the world. However, there lacks a unified
perception of at which stage and how to incorporate different modalities. In
this survey, we review methods that assist and augment generative models by
retrieving multimodal knowledge, whose formats range from images, codes,
tables, graphs, to audio. Such methods offer a promising solution to important
concerns such as factuality, reasoning, interpretability, and robustness. By
providing an in-depth review, this survey is expected to provide scholars with
a deeper understanding of the methods' applications and encourage them to adapt
existing techniques to the fast-growing field of LLMs. | cs.CL | Retrieving Multimodal Information for Augmented Generation: A Survey |
2023-03-20T03:39:27Z | http://arxiv.org/pdf/2303.10845v1 | http://arxiv.org/abs/2303.10845v1 | Xiaozhe Ren, Pingyi Zhou, Xinfan Meng, Xinjing Huang, Yadao Wang, Weichao Wang, Pengfei Li, Xiaoda Zhang, Alexander Podolskiy, Grigory Arshinov, Andrey Bout, Irina Piontkovskaya, Jiansheng Wei, Xin Jiang, Teng Su, Qun Liu, Jun Yao | 2023-03-20T03:39:27Z | cs.CL | The scaling of large language models has greatly improved natural language
understanding, generation, and reasoning. In this work, we develop a system
that trained a trillion-parameter language model on a cluster of Ascend 910 AI
processors and MindSpore framework, and present the language model with 1.085T
parameters named PanGu-{\Sigma}. With parameter inherent from PanGu-{\alpha},
we extend the dense Transformer model to sparse one with Random Routed Experts
(RRE), and efficiently train the model over 329B tokens by using Expert
Computation and Storage Separation(ECSS). This resulted in a 6.3x increase in
training throughput through heterogeneous computing. Our experimental findings
show that PanGu-{\Sigma} provides state-of-the-art performance in zero-shot
learning of various Chinese NLP downstream tasks. Moreover, it demonstrates
strong abilities when fine-tuned in application data of open-domain dialogue,
question answering, machine translation and code generation. | cs.CL | PanGu-Σ: Towards Trillion Parameter Language Model with Sparse Heterogeneous Computing |
2023-03-20T01:57:52Z | http://arxiv.org/pdf/2304.02016v1 | http://arxiv.org/abs/2304.02016v1 | David Noever, Samantha Elizabeth Miller Noever | 2023-03-20T01:57:52Z | cs.CL, cs.CV, cs.LG | The AI community has embraced multi-sensory or multi-modal approaches to
advance this generation of AI models to resemble expected intelligent
understanding. Combining language and imagery represents a familiar method for
specific tasks like image captioning or generation from descriptions. This
paper compares these monolithic approaches to a lightweight and specialized
method based on employing image models to label objects, then serially
submitting this resulting object list to a large language model (LLM). This use
of multiple Application Programming Interfaces (APIs) enables better than 95%
mean average precision for correct object lists, which serve as input to the
latest Open AI text generator (GPT-4). To demonstrate the API as a modular
alternative, we solve the problem of a user taking a picture of ingredients
available in a refrigerator, then generating novel recipe cards tailored to
complex constraints on cost, preparation time, dietary restrictions, portion
sizes, and multiple meal plans. The research concludes that monolithic
multimodal models currently lack the coherent memory to maintain context and
format for this task and that until recently, the language models like GPT-2/3
struggled to format similar problems without degenerating into repetitive or
non-sensical combinations of ingredients. For the first time, an AI chef or
cook seems not only possible but offers some enhanced capabilities to augment
human recipe libraries in pragmatic ways. The work generates a 100-page recipe
book featuring the thirty top ingredients using over 2000 refrigerator images
as initializing lists. | cs.CL | The Multimodal And Modular Ai Chef: Complex Recipe Generation From Imagery |
2023-03-19T23:41:04Z | http://arxiv.org/pdf/2303.10794v2 | http://arxiv.org/abs/2303.10794v2 | Shenghan Zhang, Haoxuan Li, Ruixiang Tang, Sirui Ding, Laila Rasmy, Degui Zhi, Na Zou, Xia Hu | 2023-04-26T20:40:43Z | cs.LG, cs.CL, cs.MM, q-bio.QM | Detailed phenotype information is fundamental to accurate diagnosis and risk
estimation of diseases. As a rich source of phenotype information, electronic
health records (EHRs) promise to empower diagnostic variant interpretation.
However, how to accurately and efficiently extract phenotypes from the
heterogeneous EHR data remains a challenge. In this work, we present PheME, an
Ensemble framework using Multi-modality data of structured EHRs and
unstructured clinical notes for accurate Phenotype prediction. Firstly, we
employ multiple deep neural networks to learn reliable representations from the
sparse structured EHR data and redundant clinical notes. A multi-modal model
then aligns multi-modal features onto the same latent space to predict
phenotypes. Secondly, we leverage ensemble learning to combine outputs from
single-modal models and multi-modal models to improve phenotype predictions. We
choose seven diseases to evaluate the phenotyping performance of the proposed
framework. Experimental results show that using multi-modal data significantly
improves phenotype prediction in all diseases, the proposed ensemble learning
framework can further boost the performance. | cs.LG | PheME: A deep ensemble framework for improving phenotype prediction from multi-modal data |
2023-03-19T22:15:05Z | http://arxiv.org/pdf/2303.10782v1 | http://arxiv.org/abs/2303.10782v1 | Abhilash Pal, Stephan Huber, Cyrine Chaabani, Alessandro Manzotti, Oscar Koller | 2023-03-19T22:15:05Z | cs.CL, cs.CV | Sign language detection, identifying if someone is signing or not, is
becoming crucially important for its applications in remote conferencing
software and for selecting useful sign data for training sign language
recognition or translation tasks. We argue that the current benchmark data sets
for sign language detection estimate overly positive results that do not
generalize well due to signer overlap between train and test partitions. We
quantify this with a detailed analysis of the effect of signer overlap on
current sign detection benchmark data sets. Comparing accuracy with and without
overlap on the DGS corpus and Signing in the Wild, we observed a relative
decrease in accuracy of 4.17% and 6.27%, respectively. Furthermore, we propose
new data set partitions that are free of overlap and allow for more realistic
performance assessment. We hope this work will contribute to improving the
accuracy and generalization of sign language detection systems. | cs.CL | On the Importance of Signer Overlap for Sign Language Detection |
2023-03-19T16:07:42Z | http://arxiv.org/pdf/2303.10699v1 | http://arxiv.org/abs/2303.10699v1 | Weizhe Lin, Zhilin Wang, Bill Byrne | 2023-03-19T16:07:42Z | cs.CL, cs.CV | The widely used Fact-based Visual Question Answering (FVQA) dataset contains
visually-grounded questions that require information retrieval using common
sense knowledge graphs to answer. It has been observed that the original
dataset is highly imbalanced and concentrated on a small portion of its
associated knowledge graph. We introduce FVQA 2.0 which contains adversarial
variants of test questions to address this imbalance. We show that systems
trained with the original FVQA train sets can be vulnerable to adversarial
samples and we demonstrate an augmentation scheme to reduce this vulnerability
without human annotations. | cs.CL | FVQA 2.0: Introducing Adversarial Samples into Fact-based Visual Question Answering |
2023-03-19T13:47:56Z | http://arxiv.org/pdf/2303.10659v2 | http://arxiv.org/abs/2303.10659v2 | Yuhang Jiang, Ramakanth Kavuluru | 2023-03-22T15:27:08Z | cs.CL | As COVID-19 ravages the world, social media analytics could augment
traditional surveys in assessing how the pandemic evolves and capturing
consumer chatter that could help healthcare agencies in addressing it. This
typically involves mining disclosure events that mention testing positive for
the disease or discussions surrounding perceptions and beliefs in preventative
or treatment options. The 2020 shared task on COVID-19 event extraction
(conducted as part of the W-NUT workshop during the EMNLP conference)
introduced a new Twitter dataset for benchmarking event extraction from
COVID-19 tweets. In this paper, we cast the problem of event extraction as
extractive question answering using recent advances in continuous prompting in
language models. On the shared task test dataset, our approach leads to over 5%
absolute micro-averaged F1-score improvement over prior best results, across
all COVID-19 event slots. Our ablation study shows that continuous prompts have
a major impact on the eventual performance. | cs.CL | COVID-19 event extraction from Twitter via extractive question answering with continuous prompts |
2023-03-19T09:24:48Z | http://arxiv.org/pdf/2303.10612v1 | http://arxiv.org/abs/2303.10612v1 | H. A. Z. Sameen Shahgir, Khondker Salman Sayeed | 2023-03-19T09:24:48Z | cs.CL, cs.LG | This paper presents a method for detecting grammatical errors in Bangla using
a Text-to-Text Transfer Transformer (T5) Language Model, using the small
variant of BanglaT5, fine-tuned on a corpus of 9385 sentences where errors were
bracketed by the dedicated demarcation symbol. The T5 model was primarily
designed for translation and is not specifically designed for this task, so
extensive post-processing was necessary to adapt it to the task of error
detection. Our experiments show that the T5 model can achieve low Levenshtein
Distance in detecting grammatical errors in Bangla, but post-processing is
essential to achieve optimal performance. The final average Levenshtein
Distance after post-processing the output of the fine-tuned model was 1.0394 on
a test set of 5000 sentences. This paper also presents a detailed analysis of
the errors detected by the model and discusses the challenges of adapting a
translation model for grammar. Our approach can be extended to other languages,
demonstrating the potential of T5 models for detecting grammatical errors in a
wide range of languages. | cs.CL | Bangla Grammatical Error Detection Using T5 Transformer Model |
2023-03-19T08:57:39Z | http://arxiv.org/pdf/2303.10606v1 | http://arxiv.org/abs/2303.10606v1 | Mehrdad Rafiepour, Javad Salimi Sartakhti | 2023-03-19T08:57:39Z | cs.CL | Intent-detection and slot-filling are the two main tasks in natural language
understanding. In this study, we propose CTRAN, a novel encoder-decoder
CNN-Transformer-based architecture for intent-detection and slot-filling. In
the encoder, we use BERT, followed by several convolutional layers, and
rearrange the output using window feature sequence. We use stacked Transformer
encoders after the window feature sequence. For the intent-detection decoder,
we utilize self-attention followed by a linear layer. In the slot-filling
decoder, we introduce the aligned Transformer decoder, which utilizes a zero
diagonal mask, aligning output tags with input tokens. We apply our network on
ATIS and SNIPS, and surpass the current state-of-the-art in slot-filling on
both datasets. Furthermore, we incorporate the language model as word
embeddings, and show that this strategy yields a better result when compared to
the language model as an encoder. | cs.CL | CTRAN: CNN-Transformer-based Network for Natural Language Understanding |
2023-03-19T06:09:52Z | http://arxiv.org/pdf/2303.10583v2 | http://arxiv.org/abs/2303.10583v2 | Qihao Zhu, Jianxi Luo | 2023-05-13T05:47:16Z | cs.HC, cs.CL | In the early stages of the design process, designers explore opportunities by
discovering unmet needs and developing innovative concepts as potential
solutions. From a human-centered design perspective, designers must develop
empathy with people to truly understand their needs. However, developing
empathy is a complex and subjective process that relies heavily on the
designer's empathic capability. Therefore, the development of empathic
understanding is intuitive, and the discovery of underlying needs is often
serendipitous. This paper aims to provide insights from artificial intelligence
research to indicate the future direction of AI-driven human-centered design,
taking into account the essential role of empathy. Specifically, we conduct an
interdisciplinary investigation of research areas such as data-driven user
studies, empathic understanding development, and artificial empathy. Based on
this foundation, we discuss the role that artificial empathy can play in
human-centered design and propose an artificial empathy framework for
human-centered design. Building on the mechanisms behind empathy and insights
from empathic design research, the framework aims to break down the rather
complex and subjective concept of empathy into components and modules that can
potentially be modeled computationally. Furthermore, we discuss the expected
benefits of developing such systems and identify current research gaps to
encourage future research efforts. | cs.HC | Toward Artificial Empathy for Human-Centered Design: A Framework |
2023-03-19T05:22:12Z | http://arxiv.org/pdf/2303.10573v1 | http://arxiv.org/abs/2303.10573v1 | Vaibhav Garg, Jiaqing Yuan, Rujie Xi, Munindar P. Singh | 2023-03-19T05:22:12Z | cs.CL | Survivors of sexual harassment frequently share their experiences on social
media, revealing their feelings and emotions and seeking advice. We observed
that on Reddit, survivors regularly share long posts that describe a
combination of (i) a sexual harassment incident, (ii) its effect on the
survivor, including their feelings and emotions, and (iii) the advice being
sought. We term such posts MeToo posts, even though they may not be so tagged
and may appear in diverse subreddits. A prospective helper (such as a counselor
or even a casual reader) must understand a survivor's needs from such posts.
But long posts can be time-consuming to read and respond to.
Accordingly, we address the problem of extracting key information from a long
MeToo post. We develop a natural language-based model to identify sentences
from a post that describe any of the above three categories.
On ten-fold cross-validation of a dataset, our model achieves a macro F1
score of 0.82.
In addition, we contribute MeThree, a dataset comprising 8,947 labeled
sentences extracted from Reddit posts. We apply the LIWC-22 toolkit on MeThree
to understand how different language patterns in sentences of the three
categories can reveal differences in emotional tone, authenticity, and other
aspects. | cs.CL | Extracting Incidents, Effects, and Requested Advice from MeToo Posts |
2023-03-19T04:39:36Z | http://arxiv.org/pdf/2303.12808v1 | http://arxiv.org/abs/2303.12808v1 | Vaibhav Garg, Ganning Xu, Munindar P. Singh | 2023-03-19T04:39:36Z | cs.CL, cs.AI | In India, people identify with a particular group based on certain attributes
such as religion. The same religious groups are often provoked against each
other. Previous studies show the role of provocation in increasing tensions
between India's two prominent religious groups: Hindus and Muslims. With the
advent of the Internet, such provocation also surfaced on social media
platforms such as WhatsApp.
By leveraging an existing dataset of Indian WhatsApp posts, we identified
three categories of provoking sentences against Indian Muslims. Further, we
labeled 7,000 sentences for three provocation categories and called this
dataset PACO. We leveraged PACO to train a model that can identify provoking
sentences from a WhatsApp post. Our best model is fine-tuned RoBERTa and
achieved a 0.851 average AUC score over five-fold cross-validation.
Automatically identifying provoking sentences could stop provoking text from
reaching out to the masses, and can prevent possible discrimination or violence
against the target religious group.
Further, we studied the provocative speech through a pragmatic lens, by
identifying the dialog acts and impoliteness super-strategies used against the
religious group. | cs.CL | PACO: Provocation Involving Action, Culture, and Oppression |
2023-03-19T04:05:10Z | http://arxiv.org/pdf/2303.10560v1 | http://arxiv.org/abs/2303.10560v1 | Brandon Siyuan Loh, Raj Kumar Gupta, Ajay Vishwanath, Andrew Ortony, Yinping Yang | 2023-03-19T04:05:10Z | cs.CL | The COVID-19 pandemic has claimed millions of lives worldwide and elicited
heightened emotions. This study examines the expression of various emotions
pertaining to COVID-19 in the United States and India as manifested in over 54
million tweets, covering the fifteen-month period from February 2020 through
April 2021, a period which includes the beginnings of the huge and disastrous
increase in COVID-19 cases that started to ravage India in March 2021.
Employing pre-trained emotion analysis and topic modeling algorithms, four
distinct types of emotions (fear, anger, happiness, and sadness) and their
time- and location-associated variations were examined. Results revealed
significant country differences and temporal changes in the relative
proportions of fear, anger, and happiness, with fear declining and anger and
happiness fluctuating in 2020 until new situations over the first four months
of 2021 reversed the trends. Detected differences are discussed briefly in
terms of the latent topics revealed and through the lens of appraisal theories
of emotions, and the implications of the findings are discussed. | cs.CL | How People Respond to the COVID-19 Pandemic on Twitter: A Comparative Analysis of Emotional Expressions from US and India |
2023-03-19T01:40:31Z | http://arxiv.org/pdf/2303.10527v1 | http://arxiv.org/abs/2303.10527v1 | Yoav Goldberg | 2023-03-19T01:40:31Z | cs.CL | It is an established assumption that pattern-based models are good at
precision, while learning based models are better at recall. But is that really
the case? I argue that there are two kinds of recall: d-recall, reflecting
diversity, and e-recall, reflecting exhaustiveness. I demonstrate through
experiments that while neural methods are indeed significantly better at
d-recall, it is sometimes the case that pattern-based methods are still
substantially better at e-recall. Ideal methods should aim for both kinds, and
this ideal should in turn be reflected in our evaluations. | cs.CL | Two Kinds of Recall |
2023-03-18T22:36:25Z | http://arxiv.org/pdf/2303.10512v2 | http://arxiv.org/abs/2303.10512v2 | Qingru Zhang, Minshuo Chen, Alexander Bukharin, Nikos Karampatziakis, Pengcheng He, Yu Cheng, Weizhu Chen, Tuo Zhao | 2023-12-20T20:56:14Z | cs.CL, cs.LG | Fine-tuning large pre-trained language models on downstream tasks has become
an important paradigm in NLP. However, common practice fine-tunes all of the
parameters in a pre-trained model, which becomes prohibitive when a large
number of downstream tasks are present. Therefore, many fine-tuning methods are
proposed to learn incremental updates of pre-trained weights in a parameter
efficient way, e.g., low-rank increments. These methods often evenly distribute
the budget of incremental updates across all pre-trained weight matrices, and
overlook the varying importance of different weight parameters. As a
consequence, the fine-tuning performance is suboptimal. To bridge this gap, we
propose AdaLoRA, which adaptively allocates the parameter budget among weight
matrices according to their importance score. In particular, AdaLoRA
parameterizes the incremental updates in the form of singular value
decomposition. Such a novel approach allows us to effectively prune the
singular values of unimportant updates, which is essentially to reduce their
parameter budget but circumvent intensive exact SVD computations. We conduct
extensive experiments with several pre-trained models on natural language
processing, question answering, and natural language generation to validate the
effectiveness of AdaLoRA. Results demonstrate that AdaLoRA manifests notable
improvement over baselines, especially in the low budget settings. Our code is
publicly available at https://github.com/QingruZhang/AdaLoRA . | cs.CL | AdaLoRA: Adaptive Budget Allocation for Parameter-Efficient Fine-Tuning |
2023-03-18T22:19:09Z | http://arxiv.org/pdf/2303.10510v2 | http://arxiv.org/abs/2303.10510v2 | Yanan Jia | 2023-09-27T17:32:30Z | cs.CL, cs.LG, cs.SD, eess.AS | As human-machine voice interfaces provide easy access to increasingly
intelligent machines, many state-of-the-art automatic speech recognition (ASR)
systems are proposed. However, commercial ASR systems usually have poor
performance on domain-specific speech especially under low-resource settings.
The author works with pre-trained DeepSpeech2 and Wav2Vec2 acoustic models to
develop benefit-specific ASR systems. The domain-specific data are collected
using proposed semi-supervised learning annotation with little human
intervention. The best performance comes from a fine-tuned Wav2Vec2-Large-LV60
acoustic model with an external KenLM, which surpasses the Google and AWS ASR
systems on benefit-specific speech. The viability of using error prone ASR
transcriptions as part of spoken language understanding (SLU) is also
investigated. Results of a benefit-specific natural language understanding
(NLU) task show that the domain-specific fine-tuned ASR system can outperform
the commercial ASR systems even when its transcriptions have higher word error
rate (WER), and the results between fine-tuned ASR and human transcriptions are
similar. | cs.CL | A Deep Learning System for Domain-specific Speech Recognition |
2023-03-18T19:17:47Z | http://arxiv.org/pdf/2303.10475v8 | http://arxiv.org/abs/2303.10475v8 | Renze Lou, Kai Zhang, Wenpeng Yin | 2024-05-25T03:21:21Z | cs.CL | Task semantics can be expressed by a set of input-output examples or a piece
of textual instruction. Conventional machine learning approaches for natural
language processing (NLP) mainly rely on the availability of large-scale sets
of task-specific examples. Two issues arise: first, collecting task-specific
labeled examples does not apply to scenarios where tasks may be too complicated
or costly to annotate, or the system is required to handle a new task
immediately; second, this is not user-friendly since end-users are probably
more willing to provide task description rather than a set of examples before
using the system. Therefore, the community is paying increasing interest in a
new supervision-seeking paradigm for NLP: learning to follow task instructions,
i.e., instruction following. Despite its impressive progress, there are some
common issues that the community struggles with. This survey paper tries to
summarize and provide insights to the current research on instruction
following, particularly, by answering the following questions: (i) What is task
instruction, and what instruction types exist? (ii) How to model instructions?
(iii) What are popular instruction following datasets and evaluation metrics?
(iv) What factors influence and explain the instructions' performance? (v) What
challenges remain in instruction following? To our knowledge, this is the first
comprehensive survey about instruction following. | cs.CL | Large Language Model Instruction Following: A Survey of Progresses and Challenges |
2023-03-18T17:56:01Z | http://arxiv.org/pdf/2303.10464v2 | http://arxiv.org/abs/2303.10464v2 | Vithursan Thangarasa, Abhay Gupta, William Marshall, Tianda Li, Kevin Leong, Dennis DeCoste, Sean Lie, Shreyas Saxena | 2023-07-29T19:56:50Z | cs.LG, cs.CL | The pre-training and fine-tuning paradigm has contributed to a number of
breakthroughs in Natural Language Processing (NLP). Instead of directly
training on a downstream task, language models are first pre-trained on large
datasets with cross-domain knowledge (e.g., Pile, MassiveText, etc.) and then
fine-tuned on task-specific data (e.g., natural language generation, text
summarization, etc.). Scaling the model and dataset size has helped improve the
performance of LLMs, but unfortunately, this also lead to highly prohibitive
computational costs. Pre-training LLMs often require orders of magnitude more
FLOPs than fine-tuning and the model capacity often remains the same between
the two phases. To achieve training efficiency w.r.t training FLOPs, we propose
to decouple the model capacity between the two phases and introduce Sparse
Pre-training and Dense Fine-tuning (SPDF). In this work, we show the benefits
of using unstructured weight sparsity to train only a subset of weights during
pre-training (Sparse Pre-training) and then recover the representational
capacity by allowing the zeroed weights to learn (Dense Fine-tuning). We
demonstrate that we can induce up to 75% sparsity into a 1.3B parameter GPT-3
XL model resulting in a 2.5x reduction in pre-training FLOPs, without a
significant loss in accuracy on the downstream tasks relative to the dense
baseline. By rigorously evaluating multiple downstream tasks, we also establish
a relationship between sparsity, task complexity and dataset size. Our work
presents a promising direction to train large GPT models at a fraction of the
training FLOPs using weight sparsity, while retaining the benefits of
pre-trained textual representations for downstream tasks. | cs.LG | SPDF: Sparse Pre-training and Dense Fine-tuning for Large Language Models |
2023-03-18T17:36:21Z | http://arxiv.org/pdf/2303.13365v1 | http://arxiv.org/abs/2303.13365v1 | Shekoufeh Kolahdouz-Rahimi, Kevin Lano, Chenghua Lin | 2023-03-18T17:36:21Z | cs.CL, cs.AI, cs.LG, cs.SE | Improvement of software development methodologies attracts developers to
automatic Requirement Formalisation (RF) in the Requirement Engineering (RE)
field. The potential advantages by applying Natural Language Processing (NLP)
and Machine Learning (ML) in reducing the ambiguity and incompleteness of
requirement written in natural languages is reported in different studies. The
goal of this paper is to survey and classify existing work on NLP and ML for
RF, identifying challenges in this domain and providing promising future
research directions. To achieve this, we conducted a systematic literature
review to outline the current state-of-the-art of NLP and ML techniques in RF
by selecting 257 papers from common used libraries. The search result is
filtered by defining inclusion and exclusion criteria and 47 relevant studies
between 2012 and 2022 are selected. We found that heuristic NLP approaches are
the most common NLP techniques used for automatic RF, primary operating on
structured and semi-structured data. This study also revealed that Deep
Learning (DL) technique are not widely used, instead classical ML techniques
are predominant in the surveyed studies. More importantly, we identified the
difficulty of comparing the performance of different approaches due to the lack
of standard benchmark cases for RF. | cs.CL | Requirement Formalisation using Natural Language Processing and Machine Learning: A Systematic Review |
2023-03-18T15:55:49Z | http://arxiv.org/pdf/2303.10443v1 | http://arxiv.org/abs/2303.10443v1 | Jiexin Ding, Bowen Zhao, Yuqi Huang, Yuntao Wang, Yuanchun Shi | 2023-03-18T15:55:49Z | cs.HC, cs.CL | Automatic unknown word detection techniques can enable new applications for
assisting English as a Second Language (ESL) learners, thus improving their
reading experiences. However, most modern unknown word detection methods
require dedicated eye-tracking devices with high precision that are not easily
accessible to end-users. In this work, we propose GazeReader, an unknown word
detection method only using a webcam. GazeReader tracks the learner's gaze and
then applies a transformer-based machine learning model that encodes the text
information to locate the unknown word. We applied knowledge enhancement
including term frequency, part of speech, and named entity recognition to
improve the performance. The user study indicates that the accuracy and
F1-score of our method were 98.09% and 75.73%, respectively. Lastly, we
explored the design scope for ESL reading and discussed the findings. | cs.HC | GazeReader: Detecting Unknown Word Using Webcam for English as a Second Language (ESL) Learners |
2023-03-18T15:39:23Z | http://arxiv.org/pdf/2303.10439v2 | http://arxiv.org/abs/2303.10439v2 | Yaohou Fan, Chetan Arora, Christoph Treude | 2023-06-12T12:36:28Z | cs.SE, cs.CL | Stop words, which are considered non-predictive, are often eliminated in
natural language processing tasks. However, the definition of uninformative
vocabulary is vague, so most algorithms use general knowledge-based stop lists
to remove stop words. There is an ongoing debate among academics about the
usefulness of stop word elimination, especially in domain-specific settings. In
this work, we investigate the usefulness of stop word removal in a software
engineering context. To do this, we replicate and experiment with three
software engineering research tools from related work. Additionally, we
construct a corpus of software engineering domain-related text from 10,000
Stack Overflow questions and identify 200 domain-specific stop words using
traditional information-theoretic methods. Our results show that the use of
domain-specific stop words significantly improved the performance of research
tools compared to the use of a general stop list and that 17 out of 19
evaluation measures showed better performance.
Online appendix: https://zenodo.org/record/7865748 | cs.SE | Stop Words for Processing Software Engineering Documents: Do they Matter? |
2023-03-18T14:54:57Z | http://arxiv.org/pdf/2303.10430v1 | http://arxiv.org/abs/2303.10430v1 | Yiran Ye, Thai Le, Dongwon Lee | 2023-03-18T14:54:57Z | cs.LG, cs.AI, cs.CL, cs.CY | Online texts with toxic content are a threat in social media that might cause
cyber harassment. Although many platforms applied measures, such as machine
learning-based hate-speech detection systems, to diminish their effect, those
toxic content publishers can still evade the system by modifying the spelling
of toxic words. Those modified words are also known as human-written text
perturbations. Many research works developed certain techniques to generate
adversarial samples to help the machine learning models obtain the ability to
recognize those perturbations. However, there is still a gap between those
machine-generated perturbations and human-written perturbations. In this paper,
we introduce a benchmark test set containing human-written perturbations online
for toxic speech detection models. We also recruited a group of workers to
evaluate the quality of this test set and dropped low-quality samples.
Meanwhile, to check if our perturbation can be normalized to its clean version,
we applied spell corrector algorithms on this dataset. Finally, we test this
data on state-of-the-art language models, such as BERT and RoBERTa, and black
box APIs, such as perspective API, to demonstrate the adversarial attack with
real human-written perturbations is still effective. | cs.LG | NoisyHate: Benchmarking Content Moderation Machine Learning Models with Human-Written Perturbations Online |
2023-03-18T14:02:04Z | http://arxiv.org/pdf/2303.10420v2 | http://arxiv.org/abs/2303.10420v2 | Junjie Ye, Xuanting Chen, Nuo Xu, Can Zu, Zekai Shao, Shichun Liu, Yuhan Cui, Zeyang Zhou, Chao Gong, Yang Shen, Jie Zhou, Siming Chen, Tao Gui, Qi Zhang, Xuanjing Huang | 2023-12-23T12:53:02Z | cs.CL | GPT series models, such as GPT-3, CodeX, InstructGPT, ChatGPT, and so on,
have gained considerable attention due to their exceptional natural language
processing capabilities. However, despite the abundance of research on the
difference in capabilities between GPT series models and fine-tuned models,
there has been limited attention given to the evolution of GPT series models'
capabilities over time. To conduct a comprehensive analysis of the capabilities
of GPT series models, we select six representative models, comprising two GPT-3
series models (i.e., davinci and text-davinci-001) and four GPT-3.5 series
models (i.e., code-davinci-002, text-davinci-002, text-davinci-003, and
gpt-3.5-turbo). We evaluate their performance on nine natural language
understanding (NLU) tasks using 21 datasets. In particular, we compare the
performance and robustness of different models for each task under zero-shot
and few-shot scenarios. Our extensive experiments reveal that the overall
ability of GPT series models on NLU tasks does not increase gradually as the
models evolve, especially with the introduction of the RLHF training strategy.
While this strategy enhances the models' ability to generate human-like
responses, it also compromises their ability to solve some tasks. Furthermore,
our findings indicate that there is still room for improvement in areas such as
model robustness. | cs.CL | A Comprehensive Capability Analysis of GPT-3 and GPT-3.5 Series Models |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.