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
|
|---|---|---|---|---|---|---|---|---|
2023-05-19T04:46:04Z | http://arxiv.org/pdf/2305.11426v3 | http://arxiv.org/abs/2305.11426v3 | Satyapriya Krishna, Jiaqi Ma, Dylan Slack, Asma Ghandeharioun, Sameer Singh, Himabindu Lakkaraju | 2023-12-07T14:27:29Z | cs.CL, cs.AI | Large Language Models (LLMs) have demonstrated remarkable capabilities in
performing complex tasks. Moreover, recent research has shown that
incorporating human-annotated rationales (e.g., Chain-of-Thought prompting)
during in-context learning can significantly enhance the performance of these
models, particularly on tasks that require reasoning capabilities. However,
incorporating such rationales poses challenges in terms of scalability as this
requires a high degree of human involvement. In this work, we present a novel
framework, Amplifying Model Performance by Leveraging In-Context Learning with
Post Hoc Explanations (AMPLIFY), which addresses the aforementioned challenges
by automating the process of rationale generation. To this end, we leverage
post hoc explanation methods which output attribution scores (explanations)
capturing the influence of each of the input features on model predictions.
More specifically, we construct automated natural language rationales that
embed insights from post hoc explanations to provide corrective signals to
LLMs. Extensive experimentation with real-world datasets demonstrates that our
framework, AMPLIFY, leads to prediction accuracy improvements of about 10-25%
over a wide range of tasks, including those where prior approaches which rely
on human-annotated rationales such as Chain-of-Thought prompting fall short.
Our work makes one of the first attempts at highlighting the potential of post
hoc explanations as valuable tools for enhancing the effectiveness of LLMs.
Furthermore, we conduct additional empirical analyses and ablation studies to
demonstrate the impact of each of the components of AMPLIFY, which, in turn,
leads to critical insights for refining in-context learning. | cs.CL | Post Hoc Explanations of Language Models Can Improve Language Models |
2023-05-19T03:48:16Z | http://arxiv.org/pdf/2305.11411v1 | http://arxiv.org/abs/2305.11411v1 | Dong Zhang, Rong Ye, Tom Ko, Mingxuan Wang, Yaqian Zhou | 2023-05-19T03:48:16Z | cs.CL, cs.SD, eess.AS | How can speech-to-text translation (ST) perform as well as machine
translation (MT)? The key point is to bridge the modality gap between speech
and text so that useful MT techniques can be applied to ST. Recently, the
approach of representing speech with unsupervised discrete units yields a new
way to ease the modality problem. This motivates us to propose Discrete Unit
Back-translation (DUB) to answer two questions: (1) Is it better to represent
speech with discrete units than with continuous features in direct ST? (2) How
much benefit can useful MT techniques bring to ST? With DUB, the
back-translation technique can successfully be applied on direct ST and obtains
an average boost of 5.5 BLEU on MuST-C En-De/Fr/Es. In the low-resource
language scenario, our method achieves comparable performance to existing
methods that rely on large-scale external data. Code and models are available
at https://github.com/0nutation/DUB. | cs.CL | DUB: Discrete Unit Back-translation for Speech Translation |
2023-05-19T03:31:42Z | http://arxiv.org/pdf/2305.11408v2 | http://arxiv.org/abs/2305.11408v2 | Sara Papi, Marco Turchi, Matteo Negri | 2023-07-20T00:58:30Z | cs.CL, cs.LG, cs.SD, eess.AS | Attention is the core mechanism of today's most used architectures for
natural language processing and has been analyzed from many perspectives,
including its effectiveness for machine translation-related tasks. Among these
studies, attention resulted to be a useful source of information to get
insights about word alignment also when the input text is substituted with
audio segments, as in the case of the speech translation (ST) task. In this
paper, we propose AlignAtt, a novel policy for simultaneous ST (SimulST) that
exploits the attention information to generate source-target alignments that
guide the model during inference. Through experiments on the 8 language pairs
of MuST-C v1.0, we show that AlignAtt outperforms previous state-of-the-art
SimulST policies applied to offline-trained models with gains in terms of BLEU
of 2 points and latency reductions ranging from 0.5s to 0.8s across the 8
languages. | cs.CL | AlignAtt: Using Attention-based Audio-Translation Alignments as a Guide for Simultaneous Speech Translation |
2023-05-19T02:51:25Z | http://arxiv.org/pdf/2305.11398v1 | http://arxiv.org/abs/2305.11398v1 | Yelena Mejova, Lydia Manikonda | 2023-05-19T02:51:25Z | cs.SI, cs.CL | Unprecedented lockdowns at the start of the COVID-19 pandemic have
drastically changed the routines of millions of people, potentially impacting
important health-related behaviors. In this study, we use YouTube videos
embedded in tweets about diet, exercise and fitness posted before and during
COVID-19 to investigate the influence of the pandemic lockdowns on diet and
nutrition. In particular, we examine the nutritional profile of the foods
mentioned in the transcript, description and title of each video in terms of
six macronutrients (protein, energy, fat, sodium, sugar, and saturated fat).
These macronutrient values were further linked to demographics to assess if
there are specific effects on those potentially having insufficient access to
healthy sources of food. Interrupted time series analysis revealed a
considerable shift in the aggregated macronutrient scores before and during
COVID-19. In particular, whereas areas with lower incomes showed decrease in
energy, fat, and saturated fat, those with higher percentage of African
Americans showed an elevation in sodium. Word2Vec word similarities and odds
ratio analysis suggested a shift from popular diets and lifestyle bloggers
before the lockdowns to the interest in a variety of healthy foods, communal
sharing of quick and easy recipes, as well as a new emphasis on comfort foods.
To the best of our knowledge, this work is novel in terms of linking attention
signals in tweets, content of videos, their nutrients profile, and aggregate
demographics of the users. The insights made possible by this combination of
resources are important for monitoring the secondary health effects of social
distancing, and informing social programs designed to alleviate these effects. | cs.SI | Comfort Foods and Community Connectedness: Investigating Diet Change during COVID-19 Using YouTube Videos on Twitter |
2023-05-19T02:42:35Z | http://arxiv.org/pdf/2305.11392v1 | http://arxiv.org/abs/2305.11392v1 | Mingliang Zhai, Yulin Li, Xiameng Qin, Chen Yi, Qunyi Xie, Chengquan Zhang, Kun Yao, Yuwei Wu, Yunde Jia | 2023-05-19T02:42:35Z | cs.CV, cs.CL | Transformers achieve promising performance in document understanding because
of their high effectiveness and still suffer from quadratic computational
complexity dependency on the sequence length. General efficient transformers
are challenging to be directly adapted to model document. They are unable to
handle the layout representation in documents, e.g. word, line and paragraph,
on different granularity levels and seem hard to achieve a good trade-off
between efficiency and performance. To tackle the concerns, we propose
Fast-StrucTexT, an efficient multi-modal framework based on the StrucTexT
algorithm with an hourglass transformer architecture, for visual document
understanding. Specifically, we design a modality-guided dynamic token merging
block to make the model learn multi-granularity representation and prunes
redundant tokens. Additionally, we present a multi-modal interaction module
called Symmetry Cross Attention (SCA) to consider multi-modal fusion and
efficiently guide the token mergence. The SCA allows one modality input as
query to calculate cross attention with another modality in a dual phase.
Extensive experiments on FUNSD, SROIE, and CORD datasets demonstrate that our
model achieves the state-of-the-art performance and almost 1.9X faster
inference time than the state-of-the-art methods. | cs.CV | Fast-StrucTexT: An Efficient Hourglass Transformer with Modality-guided Dynamic Token Merge for Document Understanding |
2023-05-19T02:09:52Z | http://arxiv.org/pdf/2306.05552v1 | http://arxiv.org/abs/2306.05552v1 | Anaelia Ovalle, Mehrab Beikzadeh, Parshan Teimouri, Kai-Wei Chang, Majid Sarrafzadeh | 2023-05-19T02:09:52Z | cs.CL, I.2; I.7 | Large language models have been useful in expanding mental health care
delivery. ChatGPT, in particular, has gained popularity for its ability to
generate human-like dialogue. However, data-sensitive domains -- including but
not limited to healthcare -- face challenges in using ChatGPT due to privacy
and data-ownership concerns. To enable its utilization, we propose a text
ambiguation framework that preserves user privacy. We ground this in the task
of addressing stress prompted by user-provided texts to demonstrate the
viability and helpfulness of privacy-preserved generations. Our results suggest
that chatGPT recommendations are still able to be moderately helpful and
relevant, even when the original user text is not provided. | cs.CL | ChatGPT for Us: Preserving Data Privacy in ChatGPT via Dialogue Text Ambiguation to Expand Mental Health Care Delivery |
2023-05-19T01:59:20Z | http://arxiv.org/pdf/2305.13330v3 | http://arxiv.org/abs/2305.13330v3 | Tatiana Likhomanenko, Loren Lugosch, Ronan Collobert | 2024-02-16T16:20:18Z | eess.AS, cs.CL, cs.LG, cs.SD | Recent work has shown that it is possible to train an $\textit{unsupervised}$
automatic speech recognition (ASR) system using only unpaired audio and text.
Existing unsupervised ASR methods assume that no labeled data can be used for
training. We argue that even if one does not have any labeled audio for a given
language, there is $\textit{always}$ labeled data available for other
languages. We show that it is possible to use character-level acoustic models
(AMs) from other languages to bootstrap an $\textit{unsupervised}$ AM in a new
language. Here, "unsupervised" means no labeled audio is available for the
$\textit{target}$ language. Our approach is based on two key ingredients: (i)
generating pseudo-labels (PLs) of the $\textit{target}$ language using some
$\textit{other}$ language AM and (ii) constraining these PLs with a
$\textit{target language model}$. Our approach is effective on Common Voice:
e.g. transfer of English AM to Swahili achieves 18% WER. It also outperforms
character-based wav2vec-U 2.0 by 15% absolute WER on LJSpeech with 800h of
labeled German data instead of 60k hours of unlabeled English data. | eess.AS | Unsupervised ASR via Cross-Lingual Pseudo-Labeling |
2023-05-19T01:27:29Z | http://arxiv.org/pdf/2305.11374v1 | http://arxiv.org/abs/2305.11374v1 | Dhara Yu, Noah D. Goodman, Jesse Mu | 2023-05-19T01:27:29Z | cs.CL | Humans teach others about the world through language and demonstration. When
might one of these modalities be more effective than the other? In this work,
we study the factors that modulate the effectiveness of language vs.
demonstration using multi-agent systems to model human communication.
Specifically, we train neural network agents to teach via language or
demonstration in a grounded communication task, manipulating 1) the inherent
difficulty of the task and 2) the competence of the teacher. We find that
teaching by demonstration is more effective in the simplest settings, but
language is more effective as task difficulty increases, due to its ability to
generalize more effectively to unseen scenarios. Overall, these results provide
converging evidence for a tradeoff between language and demonstration as
teaching modalities in humans, and make the novel predictions that
demonstration may be optimal for easy tasks, while language enables
generalization in more challenging settings. | cs.CL | Characterizing tradeoffs between teaching via language and demonstrations in multi-agent systems |
2023-05-19T01:04:16Z | http://arxiv.org/pdf/2305.11366v2 | http://arxiv.org/abs/2305.11366v2 | Zifeng Wang, Cao Xiao, Jimeng Sun | 2023-10-08T03:44:35Z | cs.CL | Clinical trials are critical for drug development. Constructing the
appropriate eligibility criteria (i.e., the inclusion/exclusion criteria for
patient recruitment) is essential for the trial's success. Proper design of
clinical trial protocols should consider similar precedent trials and their
eligibility criteria to ensure sufficient patient coverage. In this paper, we
present a method named AutoTrial to aid the design of clinical eligibility
criteria using language models. It allows (1) controllable generation under
instructions via a hybrid of discrete and neural prompting, (2) scalable
knowledge incorporation via in-context learning, and (3) explicit reasoning
chains to provide rationales for understanding the outputs. Experiments on over
70K clinical trials verify that AutoTrial generates high-quality criteria texts
that are fluent and coherent and with high accuracy in capturing the relevant
clinical concepts to the target trial. It is noteworthy that our method, with a
much smaller parameter size, gains around 60% winning rate against the GPT-3.5
baselines via human evaluations. | cs.CL | AutoTrial: Prompting Language Models for Clinical Trial Design |
2023-05-19T00:53:45Z | http://arxiv.org/pdf/2305.11364v2 | http://arxiv.org/abs/2305.11364v2 | Emily Reif, Minsuk Kahng, Savvas Petridis | 2023-09-27T22:08:13Z | cs.CL, cs.AI | Large language models (LLMs) can be used to generate smaller, more refined
datasets via few-shot prompting for benchmarking, fine-tuning or other use
cases. However, understanding and evaluating these datasets is difficult, and
the failure modes of LLM-generated data are still not well understood.
Specifically, the data can be repetitive in surprising ways, not only
semantically but also syntactically and lexically. We present LinguisticLens, a
novel inter-active visualization tool for making sense of and analyzing
syntactic diversity of LLM-generated datasets. LinguisticLens clusters text
along syntactic, lexical, and semantic axes. It supports hierarchical
visualization of a text dataset, allowing users to quickly scan for an overview
and inspect individual examples. The live demo is available at
shorturl.at/zHOUV. | cs.CL | Visualizing Linguistic Diversity of Text Datasets Synthesized by Large Language Models |
2023-05-19T00:14:10Z | http://arxiv.org/pdf/2305.11355v1 | http://arxiv.org/abs/2305.11355v1 | Jacob Eisenstein, Vinodkumar Prabhakaran, Clara Rivera, Dorottya Demszky, Devyani Sharma | 2023-05-19T00:14:10Z | cs.CL | We introduce a new dataset of conversational speech representing English from
India, Nigeria, and the United States. The Multi-Dialect Dataset of Dialogues
(MD3) strikes a new balance between open-ended conversational speech and
task-oriented dialogue by prompting participants to perform a series of short
information-sharing tasks. This facilitates quantitative cross-dialectal
comparison, while avoiding the imposition of a restrictive task structure that
might inhibit the expression of dialect features. Preliminary analysis of the
dataset reveals significant differences in syntax and in the use of discourse
markers. The dataset, which will be made publicly available with the
publication of this paper, includes more than 20 hours of audio and more than
200,000 orthographically-transcribed tokens. | cs.CL | MD3: The Multi-Dialect Dataset of Dialogues |
2023-05-18T23:58:53Z | http://arxiv.org/pdf/2305.11351v2 | http://arxiv.org/abs/2305.11351v2 | Zhifeng Kong, Kamalika Chaudhuri | 2024-02-20T22:35:32Z | cs.LG, cs.CL, cs.CV | Deep generative models are known to produce undesirable samples such as
harmful content. Traditional mitigation methods include re-training from
scratch, filtering, or editing; however, these are either computationally
expensive or can be circumvented by third parties. In this paper, we take a
different approach and study how to post-edit an already-trained conditional
generative model so that it redacts certain conditionals that will, with high
probability, lead to undesirable content. This is done by distilling the
conditioning network in the models, giving a solution that is effective,
efficient, controllable, and universal for a class of deep generative models.
We conduct experiments on redacting prompts in text-to-image models and
redacting voices in text-to-speech models. Our method is computationally light,
leads to better redaction quality and robustness than baseline methods while
still retaining high generation quality. | cs.LG | Data Redaction from Conditional Generative Models |
2023-05-18T23:49:31Z | http://arxiv.org/pdf/2305.11349v1 | http://arxiv.org/abs/2305.11349v1 | Amila Silva, Ling Luo, Shanika Karunasekera, Christopher Leckie | 2023-05-18T23:49:31Z | cs.LG, cs.CL | The emergence of social media as one of the main platforms for people to
access news has enabled the wide dissemination of fake news. This has motivated
numerous studies on automating fake news detection. Although there have been
limited attempts at unsupervised fake news detection, their performance suffers
due to not exploiting the knowledge from various modalities related to news
records and due to the presence of various latent biases in the existing news
datasets. To address these limitations, this work proposes an effective
framework for unsupervised fake news detection, which first embeds the
knowledge available in four modalities in news records and then proposes a
novel noise-robust self-supervised learning technique to identify the veracity
of news records from the multi-modal embeddings. Also, we propose a novel
technique to construct news datasets minimizing the latent biases in existing
news datasets. Following the proposed approach for dataset construction, we
produce a Large-scale Unlabelled News Dataset consisting 419,351 news articles
related to COVID-19, acronymed as LUND-COVID. We trained the proposed
unsupervised framework using LUND-COVID to exploit the potential of large
datasets, and evaluate it using a set of existing labelled datasets. Our
results show that the proposed unsupervised framework largely outperforms
existing unsupervised baselines for different tasks such as multi-modal fake
news detection, fake news early detection and few-shot fake news detection,
while yielding notable improvements for unseen domains during training. | cs.LG | Unsupervised Domain-agnostic Fake News Detection using Multi-modal Weak Signals |
2023-05-18T23:47:00Z | http://arxiv.org/pdf/2305.11348v2 | http://arxiv.org/abs/2305.11348v2 | Yuxin Xiao, Shulammite Lim, Tom Joseph Pollard, Marzyeh Ghassemi | 2024-01-03T04:00:15Z | cs.LG, cs.CL, cs.CR, cs.CY | Data sharing is crucial for open science and reproducible research, but the
legal sharing of clinical data requires the removal of protected health
information from electronic health records. This process, known as
de-identification, is often achieved through the use of machine learning
algorithms by many commercial and open-source systems. While these systems have
shown compelling results on average, the variation in their performance across
different demographic groups has not been thoroughly examined. In this work, we
investigate the bias of de-identification systems on names in clinical notes
via a large-scale empirical analysis. To achieve this, we create 16 name sets
that vary along four demographic dimensions: gender, race, name popularity, and
the decade of popularity. We insert these names into 100 manually curated
clinical templates and evaluate the performance of nine public and private
de-identification methods. Our findings reveal that there are statistically
significant performance gaps along a majority of the demographic dimensions in
most methods. We further illustrate that de-identification quality is affected
by polysemy in names, gender context, and clinical note characteristics. To
mitigate the identified gaps, we propose a simple and method-agnostic solution
by fine-tuning de-identification methods with clinical context and diverse
names. Overall, it is imperative to address the bias in existing methods
immediately so that downstream stakeholders can build high-quality systems to
serve all demographic parties fairly. | cs.LG | In the Name of Fairness: Assessing the Bias in Clinical Record De-identification |
2023-05-18T22:47:06Z | http://arxiv.org/pdf/2305.11334v1 | http://arxiv.org/abs/2305.11334v1 | Giorgi Kokaia, Pratyush Sinha, Yutong Jiang, Nozha Boujemaa | 2023-05-18T22:47:06Z | cs.CL, cs.AI | We introduce two novel methods, Tree-Search and Self-contextualizing QA,
designed to enhance the performance of large language models (LLMs) in
question-answering tasks. Tree-Search is a sampling technique specifically
created to extract diverse information from an LLM for a given prompt.
Self-contextualizing QA leverages Tree-Search to enable the model to create its
own context using a wide range of information relevant to the prompt, evaluate
it explicitly and return a open book answer to the initial prompt . We
demonstrate that the quality of generated answers improves according to various
metrics, including accuracy, informativeness, coherence, and consistency, as
evaluated by GPT3.5(text-davinci-003). Furthermore, we show that our methods
result in increased robustness and that performance is positively correlated
with tree size, benefiting both answer quality and robustness. Finally, we
discuss other promising applications of Tree-Search, highlighting its potential
to enhance a broad range of tasks beyond question-answering.
\noindent We also discuss several areas for future work, including refining
the Tree-Search and Self-Contextualizing QA methods, improving the coherence of
the generated context, and investigating the impact of bootstrapping on model
robustness | cs.CL | Writing your own book: A method for going from closed to open book QA to improve robustness and performance of smaller LLMs |
2023-05-18T22:23:40Z | http://arxiv.org/pdf/2305.11326v2 | http://arxiv.org/abs/2305.11326v2 | Marcos Gomez, Jordi Cabot, Robert Clarisó | 2023-05-24T11:55:16Z | cs.CL | Tabular data is the most common format to publish and exchange structured
data online. A clear example is the growing number of open data portals
published by all types of public administrations. However, exploitation of
these data sources is currently limited to technical people able to
programmatically manipulate and digest such data. As an alternative, we propose
the use of chatbots to offer a conversational interface to facilitate the
exploration of tabular data sources. With our approach, any regular citizen can
benefit and leverage them. Moreover, our chatbots are not manually created:
instead, they are automatically generated from the data source itself thanks to
the instantiation of a configurable collection of conversation patterns. | cs.CL | Towards the Automatic Generation of Conversational Interfaces to Facilitate the Exploration of Tabular Data |
2023-05-18T21:53:58Z | http://arxiv.org/pdf/2305.11317v2 | http://arxiv.org/abs/2305.11317v2 | Wanrong Zhu, Xinyi Wang, Yujie Lu, Tsu-Jui Fu, Xin Eric Wang, Miguel Eckstein, William Yang Wang | 2023-10-28T04:13:44Z | cs.CL | The field of text-to-image (T2I) generation has garnered significant
attention both within the research community and among everyday users. Despite
the advancements of T2I models, a common issue encountered by users is the need
for repetitive editing of input prompts in order to receive a satisfactory
image, which is time-consuming and labor-intensive. Given the demonstrated text
generation power of large-scale language models, such as GPT-k, we investigate
the potential of utilizing such models to improve the prompt editing process
for T2I generation. We conduct a series of experiments to compare the common
edits made by humans and GPT-k, evaluate the performance of GPT-k in prompting
T2I, and examine factors that may influence this process. We found that GPT-k
models focus more on inserting modifiers while humans tend to replace words and
phrases, which includes changes to the subject matter. Experimental results
show that GPT-k are more effective in adjusting modifiers rather than
predicting spontaneous changes in the primary subject matters. Adopting the
edit suggested by GPT-k models may reduce the percentage of remaining edits by
20-30%. | cs.CL | Collaborative Generative AI: Integrating GPT-k for Efficient Editing in Text-to-Image Generation |
2023-05-18T21:52:48Z | http://arxiv.org/pdf/2305.11315v1 | http://arxiv.org/abs/2305.11315v1 | Zeyu Zhang, Steven Bethard | 2023-05-18T21:52:48Z | cs.CL | Geocoding is the task of converting location mentions in text into structured
data that encodes the geospatial semantics. We propose a new architecture for
geocoding, GeoNorm. GeoNorm first uses information retrieval techniques to
generate a list of candidate entries from the geospatial ontology. Then it
reranks the candidate entries using a transformer-based neural network that
incorporates information from the ontology such as the entry's population. This
generate-and-rerank process is applied twice: first to resolve the less
ambiguous countries, states, and counties, and second to resolve the remaining
location mentions, using the identified countries, states, and counties as
context. Our proposed toponym resolution framework achieves state-of-the-art
performance on multiple datasets. Code and models are available at
\url{https://github.com/clulab/geonorm}. | cs.CL | Improving Toponym Resolution with Better Candidate Generation, Transformer-based Reranking, and Two-Stage Resolution |
2023-05-18T19:42:04Z | http://arxiv.org/pdf/2305.11271v1 | http://arxiv.org/abs/2305.11271v1 | Cristian-Paul Bara, Ziqiao Ma, Yingzhuo Yu, Julie Shah, Joyce Chai | 2023-05-18T19:42:04Z | cs.AI, cs.CL, cs.CV, cs.LG | Collaborative tasks often begin with partial task knowledge and incomplete
initial plans from each partner. To complete these tasks, agents need to engage
in situated communication with their partners and coordinate their partial
plans towards a complete plan to achieve a joint task goal. While such
collaboration seems effortless in a human-human team, it is highly challenging
for human-AI collaboration. To address this limitation, this paper takes a step
towards collaborative plan acquisition, where humans and agents strive to learn
and communicate with each other to acquire a complete plan for joint tasks.
Specifically, we formulate a novel problem for agents to predict the missing
task knowledge for themselves and for their partners based on rich perceptual
and dialogue history. We extend a situated dialogue benchmark for symmetric
collaborative tasks in a 3D blocks world and investigate computational
strategies for plan acquisition. Our empirical results suggest that predicting
the partner's missing knowledge is a more viable approach than predicting one's
own. We show that explicit modeling of the partner's dialogue moves and mental
states produces improved and more stable results than without. These results
provide insight for future AI agents that can predict what knowledge their
partner is missing and, therefore, can proactively communicate such information
to help their partner acquire such missing knowledge toward a common
understanding of joint tasks. | cs.AI | Towards Collaborative Plan Acquisition through Theory of Mind Modeling in Situated Dialogue |
2023-05-18T18:58:30Z | http://arxiv.org/pdf/2305.11262v1 | http://arxiv.org/abs/2305.11262v1 | Jiaxu Zhao, Meng Fang, Zijing Shi, Yitong Li, Ling Chen, Mykola Pechenizkiy | 2023-05-18T18:58:30Z | cs.CL | \textit{\textbf{\textcolor{red}{Warning}:} This paper contains content that
may be offensive or upsetting.} Pretrained conversational agents have been
exposed to safety issues, exhibiting a range of stereotypical human biases such
as gender bias. However, there are still limited bias categories in current
research, and most of them only focus on English. In this paper, we introduce a
new Chinese dataset, CHBias, for bias evaluation and mitigation of Chinese
conversational language models. Apart from those previous well-explored bias
categories, CHBias includes under-explored bias categories, such as ageism and
appearance biases, which received less attention. We evaluate two popular
pretrained Chinese conversational models, CDial-GPT and EVA2.0, using CHBias.
Furthermore, to mitigate different biases, we apply several debiasing methods
to the Chinese pretrained models. Experimental results show that these Chinese
pretrained models are potentially risky for generating texts that contain
social biases, and debiasing methods using the proposed dataset can make
response generation less biased while preserving the models' conversational
capabilities. | cs.CL | CHBias: Bias Evaluation and Mitigation of Chinese Conversational Language Models |
2023-05-18T18:38:32Z | http://arxiv.org/pdf/2305.11255v4 | http://arxiv.org/abs/2305.11255v4 | Hao Fei, Bobo Li, Qian Liu, Lidong Bing, Fei Li, Tat-Seng Chua | 2023-06-09T01:27:58Z | cs.CL | While sentiment analysis systems try to determine the sentiment polarities of
given targets based on the key opinion expressions in input texts, in implicit
sentiment analysis (ISA) the opinion cues come in an implicit and obscure
manner. Thus detecting implicit sentiment requires the common-sense and
multi-hop reasoning ability to infer the latent intent of opinion. Inspired by
the recent chain-of-thought (CoT) idea, in this work we introduce a Three-hop
Reasoning (THOR) CoT framework to mimic the human-like reasoning process for
ISA. We design a three-step prompting principle for THOR to step-by-step induce
the implicit aspect, opinion, and finally the sentiment polarity. Our
THOR+Flan-T5 (11B) pushes the state-of-the-art (SoTA) by over 6% F1 on
supervised setup. More strikingly, THOR+GPT3 (175B) boosts the SoTA by over 50%
F1 on zero-shot setting. Our code is open at
https://github.com/scofield7419/THOR-ISA. | cs.CL | Reasoning Implicit Sentiment with Chain-of-Thought Prompting |
2023-05-18T18:32:03Z | http://arxiv.org/pdf/2305.11251v1 | http://arxiv.org/abs/2305.11251v1 | Oleg Sobchuk, Artjoms Šeļa | 2023-05-18T18:32:03Z | cs.CL | What are the best methods of capturing thematic similarity between literary
texts? Knowing the answer to this question would be useful for automatic
clustering of book genres, or any other thematic grouping. This paper compares
a variety of algorithms for unsupervised learning of thematic similarities
between texts, which we call "computational thematics". These algorithms belong
to three steps of analysis: text preprocessing, extraction of text features,
and measuring distances between the lists of features. Each of these steps
includes a variety of options. We test all the possible combinations of these
options: every combination of algorithms is given a task to cluster a corpus of
books belonging to four pre-tagged genres of fiction. This clustering is then
validated against the "ground truth" genre labels. Such comparison of
algorithms allows us to learn the best and the worst combinations for
computational thematic analysis. To illustrate the sharp difference between the
best and the worst methods, we then cluster 5000 random novels from the
HathiTrust corpus of fiction. | cs.CL | Computational thematics: Comparing algorithms for clustering the genres of literary fiction |
2023-05-18T18:15:53Z | http://arxiv.org/pdf/2305.11244v2 | http://arxiv.org/abs/2305.11244v2 | Srijith Radhakrishnan, Chao-Han Huck Yang, Sumeer Ahmad Khan, Narsis A. Kiani, David Gomez-Cabrero, Jesper N. Tegner | 2023-10-03T19:45:25Z | cs.CL, cs.AI, cs.LG, cs.NE, eess.AS | In this work, we explore Parameter-Efficient-Learning (PEL) techniques to
repurpose a General-Purpose-Speech (GSM) model for Arabic dialect
identification (ADI). Specifically, we investigate different setups to
incorporate trainable features into a multi-layer encoder-decoder GSM
formulation under frozen pre-trained settings. Our architecture includes
residual adapter and model reprogramming (input-prompting). We design a
token-level label mapping to condition the GSM for Arabic Dialect
Identification (ADI). This is challenging due to the high variation in
vocabulary and pronunciation among the numerous regional dialects. We achieve
new state-of-the-art accuracy on the ADI-17 dataset by vanilla fine-tuning. We
further reduce the training budgets with the PEL method, which performs within
1.86% accuracy to fine-tuning using only 2.5% of (extra) network trainable
parameters. Our study demonstrates how to identify Arabic dialects using a
small dataset and limited computation with open source code and pre-trained
models. | cs.CL | A Parameter-Efficient Learning Approach to Arabic Dialect Identification with Pre-Trained General-Purpose Speech Model |
2023-05-18T18:15:43Z | http://arxiv.org/pdf/2305.11243v2 | http://arxiv.org/abs/2305.11243v2 | Eliza Kosoy, Emily Rose Reagan, Leslie Lai, Alison Gopnik, Danielle Krettek Cobb | 2023-11-07T20:20:47Z | cs.CL, cs.AI | Developmental psychologists have spent decades devising experiments to test
the intelligence and knowledge of infants and children, tracing the origin of
crucial concepts and capacities. Moreover, experimental techniques in
developmental psychology have been carefully designed to discriminate the
cognitive capacities that underlie particular behaviors. We propose that using
classical experiments from child development is a particularly effective way to
probe the computational abilities of AI models, in general, and LLMs in
particular. First, the methodological techniques of developmental psychology,
such as the use of novel stimuli to control for past experience or control
conditions to determine whether children are using simple associations, can be
equally helpful for assessing the capacities of LLMs. In parallel, testing LLMs
in this way can tell us whether the information that is encoded in text is
sufficient to enable particular responses, or whether those responses depend on
other kinds of information, such as information from exploration of the
physical world. In this work we adapt classical developmental experiments to
evaluate the capabilities of LaMDA, a large language model from Google. We
propose a novel LLM Response Score (LRS) metric which can be used to evaluate
other language models, such as GPT. We find that LaMDA generates appropriate
responses that are similar to those of children in experiments involving social
understanding, perhaps providing evidence that knowledge of these domains is
discovered through language. On the other hand, LaMDA's responses in early
object and action understanding, theory of mind, and especially causal
reasoning tasks are very different from those of young children, perhaps
showing that these domains require more real-world, self-initiated exploration
and cannot simply be learned from patterns in language input. | cs.CL | Comparing Machines and Children: Using Developmental Psychology Experiments to Assess the Strengths and Weaknesses of LaMDA Responses |
2023-05-18T18:15:07Z | http://arxiv.org/pdf/2305.11242v1 | http://arxiv.org/abs/2305.11242v1 | Sharon Levy, Neha Anna John, Ling Liu, Yogarshi Vyas, Jie Ma, Yoshinari Fujinuma, Miguel Ballesteros, Vittorio Castelli, Dan Roth | 2023-05-18T18:15:07Z | cs.CL | Studies in bias and fairness in natural language processing have primarily
examined social biases within a single language and/or across few attributes
(e.g. gender, race). However, biases can manifest differently across various
languages for individual attributes. As a result, it is critical to examine
biases within each language and attribute. Of equal importance is to study how
these biases compare across languages and how the biases are affected when
training a model on multilingual data versus monolingual data. We present a
bias analysis across Italian, Chinese, English, Hebrew, and Spanish on the
downstream sentiment analysis task to observe whether specific demographics are
viewed more positively. We study bias similarities and differences across these
languages and investigate the impact of multilingual vs. monolingual training
data. We adapt existing sentiment bias templates in English to Italian,
Chinese, Hebrew, and Spanish for four attributes: race, religion, nationality,
and gender. Our results reveal similarities in bias expression such as
favoritism of groups that are dominant in each language's culture (e.g.
majority religions and nationalities). Additionally, we find an increased
variation in predictions across protected groups, indicating bias
amplification, after multilingual finetuning in comparison to multilingual
pretraining. | cs.CL | Comparing Biases and the Impact of Multilingual Training across Multiple Languages |
2023-05-18T18:00:44Z | http://arxiv.org/pdf/2305.11231v1 | http://arxiv.org/abs/2305.11231v1 | Mohammad Khosravani, Amine Trabelsi | 2023-05-18T18:00:44Z | cs.CL | Unsupervised summarization is a powerful technique that enables training
summarizing models without requiring labeled datasets. This survey covers
different recent techniques and models used for unsupervised summarization. We
cover extractive, abstractive, and hybrid models and strategies used to achieve
unsupervised summarization. While the main focus of this survey is on recent
research, we also cover some of the important previous research. We
additionally introduce a taxonomy, classifying different research based on
their approach to unsupervised training. Finally, we discuss the current
approaches and mention some datasets and evaluation methods. | cs.CL | Recent Trends in Unsupervised Summarization |
2023-05-18T17:59:06Z | http://arxiv.org/pdf/2305.11172v1 | http://arxiv.org/abs/2305.11172v1 | Peng Wang, Shijie Wang, Junyang Lin, Shuai Bai, Xiaohuan Zhou, Jingren Zhou, Xinggang Wang, Chang Zhou | 2023-05-18T17:59:06Z | cs.CV, cs.CL, cs.SD, eess.AS | In this work, we explore a scalable way for building a general representation
model toward unlimited modalities. We release ONE-PEACE, a highly extensible
model with 4B parameters that can seamlessly align and integrate
representations across vision, audio, and language modalities. The architecture
of ONE-PEACE comprises modality adapters, shared self-attention layers, and
modality FFNs. This design allows for the easy extension of new modalities by
adding adapters and FFNs, while also enabling multi-modal fusion through
self-attention layers. To pretrain ONE-PEACE, we develop two modality-agnostic
pretraining tasks, cross-modal aligning contrast and intra-modal denoising
contrast, which align the semantic space of different modalities and capture
fine-grained details within modalities concurrently. With the scaling-friendly
architecture and pretraining tasks, ONE-PEACE has the potential to expand to
unlimited modalities. Without using any vision or language pretrained model for
initialization, ONE-PEACE achieves leading results on a wide range of uni-modal
and multi-modal tasks, including image classification (ImageNet), semantic
segmentation (ADE20K), audio-text retrieval (AudioCaps, Clotho), audio
classification (ESC-50, FSD50K, VGGSound), audio question answering (AVQA),
image-text retrieval (MSCOCO, Flickr30K), and visual grounding (RefCOCO/+/g).
Code is available at https://github.com/OFA-Sys/ONE-PEACE. | cs.CV | ONE-PEACE: Exploring One General Representation Model Toward Unlimited Modalities |
2023-05-18T17:58:35Z | http://arxiv.org/pdf/2305.11171v3 | http://arxiv.org/abs/2305.11171v3 | Zorik Gekhman, Jonathan Herzig, Roee Aharoni, Chen Elkind, Idan Szpektor | 2023-10-18T19:16:18Z | cs.CL | Factual consistency evaluation is often conducted using Natural Language
Inference (NLI) models, yet these models exhibit limited success in evaluating
summaries. Previous work improved such models with synthetic training data.
However, the data is typically based on perturbed human-written summaries,
which often differ in their characteristics from real model-generated summaries
and have limited coverage of possible factual errors. Alternatively, large
language models (LLMs) have recently shown promising results in directly
evaluating generative tasks, but are too computationally expensive for
practical use. Motivated by these limitations, we introduce TrueTeacher, a
method for generating synthetic data by annotating diverse model-generated
summaries using a LLM. Unlike prior work, TrueTeacher does not rely on
human-written summaries, and is multilingual by nature. Experiments on the TRUE
benchmark show that a student model trained using our data, substantially
outperforms both the state-of-the-art model with similar capacity, and the LLM
teacher. In a systematic study, we compare TrueTeacher to existing synthetic
data generation methods and demonstrate its superiority and robustness to
domain-shift. We also show that our method generalizes to multilingual
scenarios. Lastly, we release our large scale synthetic dataset (1.4M
examples), generated using TrueTeacher, and a checkpoint trained on this data. | cs.CL | TrueTeacher: Learning Factual Consistency Evaluation with Large Language Models |
2023-05-18T17:58:31Z | http://arxiv.org/pdf/2305.11170v1 | http://arxiv.org/abs/2305.11170v1 | Wangchunshu Zhou, Yuchen Eleanor Jiang, Ryan Cotterell, Mrinmaya Sachan | 2023-05-18T17:58:31Z | cs.CL, cs.AI, cs.LG | The primary way of building AI applications is shifting from training
specialist models to prompting generalist models. A common practice for
prompting generalist models, often referred to as in-context learning, is to
append a few examples (demonstrations) to the prompt to help the model better
understand the task. While effective, in-context learning can be inefficient
because it makes the input prompt much longer, consuming valuable space in the
context window and leading to larger computational costs. In this paper, we
propose DynaICL, a recipe for efficient prompting with black-box generalist
models that dynamically allocate in-context examples according to the input
complexity and the computational budget. To achieve this, we train a meta
controller that predicts the number of in-context examples suitable for the
generalist model to make a good prediction based on the performance-efficiency
trade-off for a specific input. We then dynamically allocate the number of
demonstrations for an input according to predictions from the meta controller
and the given computation budget. Experimental results show that dynamic
example allocation helps achieve a better performance-efficiency trade-off in
two practical settings where computational resources or the required
performance is constrained. Specifically, DynaICL saves up to 46% token budget
compared to the common practice that allocates the same number of in-context
examples to each input. We also find that a meta controller trained on a
certain backbone model and tasks can successfully generalize to unseen models
and tasks. | cs.CL | Efficient Prompting via Dynamic In-Context Learning |
2023-05-18T17:58:08Z | http://arxiv.org/pdf/2305.11169v2 | http://arxiv.org/abs/2305.11169v2 | Charles Jin, Martin Rinard | 2023-05-24T11:52:13Z | cs.LG, cs.AI, cs.CL, cs.PL | We present evidence that language models can learn meaning despite being
trained only to perform next token prediction on text, specifically a corpus of
programs. Each program is preceded by a specification in the form of (textual)
input-output examples. Working with programs enables us to precisely define
concepts relevant to meaning in language (e.g., correctness and semantics),
making program synthesis well-suited as an intermediate testbed for
characterizing the presence (or absence) of meaning in language models.
We first train a Transformer model on the corpus of programs, then probe the
trained model's hidden states as it completes a program given a specification.
Despite providing no inductive bias toward learning the semantics of the
language, we find that a linear probe is able to extract abstractions of both
current and future program states from the model states. Moreover, there is a
strong, statistically significant correlation between the accuracy of the probe
and the model's ability to generate a program that implements the
specification. To evaluate whether the semantics are represented in the model
states rather than learned by the probe, we design a novel experimental
procedure that intervenes on the semantics of the language while preserving the
lexicon and syntax. We also demonstrate that the model learns to generate
correct programs that are, on average, shorter than those in the training set,
which is evidence that language model outputs may differ from the training
distribution in semantically meaningful ways. In summary, this paper does not
propose any new techniques for training language models, but develops an
experimental framework for and provides insights into the acquisition and
representation of (formal) meaning in language models. | cs.LG | Evidence of Meaning in Language Models Trained on Programs |
2023-05-18T17:48:03Z | http://arxiv.org/pdf/2305.11159v1 | http://arxiv.org/abs/2305.11159v1 | Kai Zhang, Bernal Jiménez Gutiérrez, Yu Su | 2023-05-18T17:48:03Z | cs.CL | Recent work has shown that fine-tuning large language models (LLMs) on
large-scale instruction-following datasets substantially improves their
performance on a wide range of NLP tasks, especially in the zero-shot setting.
However, even advanced instruction-tuned LLMs still fail to outperform small
LMs on relation extraction (RE), a fundamental information extraction task. We
hypothesize that instruction-tuning has been unable to elicit strong RE
capabilities in LLMs due to RE's low incidence in instruction-tuning datasets,
making up less than 1% of all tasks (Wang et al., 2022). To address this
limitation, we propose QA4RE, a framework that aligns RE with question
answering (QA), a predominant task in instruction-tuning datasets.
Comprehensive zero-shot RE experiments over four datasets with two series of
instruction-tuned LLMs (six LLMs in total) demonstrate that our QA4RE framework
consistently improves LLM performance, strongly verifying our hypothesis and
enabling LLMs to outperform strong zero-shot baselines by a large margin.
Additionally, we provide thorough experiments and discussions to show the
robustness, few-shot effectiveness, and strong transferability of our QA4RE
framework. This work illustrates a promising way of adapting LLMs to
challenging and underrepresented tasks by aligning these tasks with more common
instruction-tuning tasks like QA. | cs.CL | Aligning Instruction Tasks Unlocks Large Language Models as Zero-Shot Relation Extractors |
2023-05-18T17:45:22Z | http://arxiv.org/pdf/2305.11206v1 | http://arxiv.org/abs/2305.11206v1 | Chunting Zhou, Pengfei Liu, Puxin Xu, Srini Iyer, Jiao Sun, Yuning Mao, Xuezhe Ma, Avia Efrat, Ping Yu, Lili Yu, Susan Zhang, Gargi Ghosh, Mike Lewis, Luke Zettlemoyer, Omer Levy | 2023-05-18T17:45:22Z | cs.CL, cs.AI, cs.LG | Large language models are trained in two stages: (1) unsupervised pretraining
from raw text, to learn general-purpose representations, and (2) large scale
instruction tuning and reinforcement learning, to better align to end tasks and
user preferences. We measure the relative importance of these two stages by
training LIMA, a 65B parameter LLaMa language model fine-tuned with the
standard supervised loss on only 1,000 carefully curated prompts and responses,
without any reinforcement learning or human preference modeling. LIMA
demonstrates remarkably strong performance, learning to follow specific
response formats from only a handful of examples in the training data,
including complex queries that range from planning trip itineraries to
speculating about alternate history. Moreover, the model tends to generalize
well to unseen tasks that did not appear in the training data. In a controlled
human study, responses from LIMA are either equivalent or strictly preferred to
GPT-4 in 43% of cases; this statistic is as high as 58% when compared to Bard
and 65% versus DaVinci003, which was trained with human feedback. Taken
together, these results strongly suggest that almost all knowledge in large
language models is learned during pretraining, and only limited instruction
tuning data is necessary to teach models to produce high quality output. | cs.CL | LIMA: Less Is More for Alignment |
2023-05-18T17:36:41Z | http://arxiv.org/pdf/2305.11142v1 | http://arxiv.org/abs/2305.11142v1 | Yuchen Eleanor Jiang, Tianyu Liu, Shuming Ma, Dongdong Zhang, Mrinmaya Sachan, Ryan Cotterell | 2023-05-18T17:36:41Z | cs.CL | Several recent papers claim human parity at sentence-level Machine
Translation (MT), especially in high-resource languages. Thus, in response, the
MT community has, in part, shifted its focus to document-level translation.
Translating documents requires a deeper understanding of the structure and
meaning of text, which is often captured by various kinds of discourse
phenomena such as consistency, coherence, and cohesion. However, this renders
conventional sentence-level MT evaluation benchmarks inadequate for evaluating
the performance of context-aware MT systems. This paper presents a new dataset
with rich discourse annotations, built upon the large-scale parallel corpus BWB
introduced in Jiang et al. (2022). The new BWB annotation introduces four extra
evaluation aspects, i.e., entity, terminology, coreference, and quotation,
covering 15,095 entity mentions in both languages. Using these annotations, we
systematically investigate the similarities and differences between the
discourse structures of source and target languages, and the challenges they
pose to MT. We discover that MT outputs differ fundamentally from human
translations in terms of their latent discourse structures. This gives us a new
perspective on the challenges and opportunities in document-level MT. We make
our resource publicly available to spur future research in document-level MT
and the generalization to other language translation tasks. | cs.CL | Discourse Centric Evaluation of Machine Translation with a Densely Annotated Parallel Corpus |
2023-05-18T17:35:28Z | http://arxiv.org/pdf/2305.11140v1 | http://arxiv.org/abs/2305.11140v1 | Chantal Amrhein, Florian Schottmann, Rico Sennrich, Samuel Läubli | 2023-05-18T17:35:28Z | cs.CL, I.2.7 | Natural language generation models reproduce and often amplify the biases
present in their training data. Previous research explored using
sequence-to-sequence rewriting models to transform biased model outputs (or
original texts) into more gender-fair language by creating pseudo training data
through linguistic rules. However, this approach is not practical for languages
with more complex morphology than English. We hypothesise that creating
training data in the reverse direction, i.e. starting from gender-fair text, is
easier for morphologically complex languages and show that it matches the
performance of state-of-the-art rewriting models for English. To eliminate the
rule-based nature of data creation, we instead propose using machine
translation models to create gender-biased text from real gender-fair text via
round-trip translation. Our approach allows us to train a rewriting model for
German without the need for elaborate handcrafted rules. The outputs of this
model increased gender-fairness as shown in a human evaluation study. | cs.CL | Exploiting Biased Models to De-bias Text: A Gender-Fair Rewriting Model |
2023-05-18T17:22:53Z | http://arxiv.org/pdf/2305.11129v2 | http://arxiv.org/abs/2305.11129v2 | David Uthus, Santiago Ontañón, Joshua Ainslie, Mandy Guo | 2023-10-26T22:43:26Z | cs.CL | We present our work on developing a multilingual, efficient text-to-text
transformer that is suitable for handling long inputs. This model, called
mLongT5, builds upon the architecture of LongT5, while leveraging the
multilingual datasets used for pretraining mT5 and the pretraining tasks of
UL2. We evaluate this model on a variety of multilingual summarization and
question-answering tasks, and the results show stronger performance for mLongT5
when compared to existing multilingual models such as mBART or M-BERT. | cs.CL | mLongT5: A Multilingual and Efficient Text-To-Text Transformer for Longer Sequences |
2023-05-18T16:57:57Z | http://arxiv.org/pdf/2305.11116v1 | http://arxiv.org/abs/2305.11116v1 | Yujie Lu, Xianjun Yang, Xiujun Li, Xin Eric Wang, William Yang Wang | 2023-05-18T16:57:57Z | cs.CV, cs.CL | Existing automatic evaluation on text-to-image synthesis can only provide an
image-text matching score, without considering the object-level
compositionality, which results in poor correlation with human judgments. In
this work, we propose LLMScore, a new framework that offers evaluation scores
with multi-granularity compositionality. LLMScore leverages the large language
models (LLMs) to evaluate text-to-image models. Initially, it transforms the
image into image-level and object-level visual descriptions. Then an evaluation
instruction is fed into the LLMs to measure the alignment between the
synthesized image and the text, ultimately generating a score accompanied by a
rationale. Our substantial analysis reveals the highest correlation of LLMScore
with human judgments on a wide range of datasets (Attribute Binding Contrast,
Concept Conjunction, MSCOCO, DrawBench, PaintSkills). Notably, our LLMScore
achieves Kendall's tau correlation with human evaluations that is 58.8% and
31.2% higher than the commonly-used text-image matching metrics CLIP and BLIP,
respectively. | cs.CV | LLMScore: Unveiling the Power of Large Language Models in Text-to-Image Synthesis Evaluation |
2023-05-18T16:34:18Z | http://arxiv.org/pdf/2305.11096v4 | http://arxiv.org/abs/2305.11096v4 | Jinhui Ye, Wenxiang Jiao, Xing Wang, Zhaopeng Tu, Hui Xiong | 2024-06-04T08:27:40Z | cs.CL | End-to-end sign language translation (SLT) aims to convert sign language
videos into spoken language texts directly without intermediate
representations. It has been a challenging task due to the modality gap between
sign videos and texts and the data scarcity of labeled data. Due to these
challenges, the input and output distributions of end-to-end sign language
translation (i.e., video-to-text) are less effective compared to the
gloss-to-text approach (i.e., text-to-text). To tackle these challenges, we
propose a novel Cross-modality Data Augmentation (XmDA) framework to transfer
the powerful gloss-to-text translation capabilities to end-to-end sign language
translation (i.e. video-to-text) by exploiting pseudo gloss-text pairs from the
sign gloss translation model. Specifically, XmDA consists of two key
components, namely, cross-modality mix-up and cross-modality knowledge
distillation. The former explicitly encourages the alignment between sign video
features and gloss embeddings to bridge the modality gap. The latter utilizes
the generation knowledge from gloss-to-text teacher models to guide the spoken
language text generation. Experimental results on two widely used SLT datasets,
i.e., PHOENIX-2014T and CSL-Daily, demonstrate that the proposed XmDA framework
significantly and consistently outperforms the baseline models. Extensive
analyses confirm our claim that XmDA enhances spoken language text generation
by reducing the representation distance between videos and texts, as well as
improving the processing of low-frequency words and long sentences. | cs.CL | Cross-modality Data Augmentation for End-to-End Sign Language Translation |
2023-05-18T16:32:58Z | http://arxiv.org/pdf/2305.11095v3 | http://arxiv.org/abs/2305.11095v3 | Puyuan Peng, Brian Yan, Shinji Watanabe, David Harwath | 2023-08-16T00:57:34Z | eess.AS, cs.AI, cs.CL, cs.LG, cs.SD | We investigate the emergent abilities of the recently proposed web-scale
speech model Whisper, by adapting it to unseen tasks with prompt engineering.
We selected three tasks: audio-visual speech recognition (AVSR), code-switched
speech recognition (CS-ASR), and speech translation (ST) on unseen language
pairs. We design task-specific prompts, by either leveraging another
large-scale model, or simply manipulating the special tokens in the default
prompts. Experiments show that compared to the default prompts, our proposed
prompts improve performance by 10% to 45% on the three zero-shot tasks, and
even outperform SotA supervised models on some datasets. In addition, our
experiments reveal many interesting properties of Whisper, including its
robustness to prompts, bias on accents, and the multilingual understanding in
its latent space. Code is available at
https://github.com/jasonppy/PromptingWhisper | eess.AS | Prompting the Hidden Talent of Web-Scale Speech Models for Zero-Shot Task Generalization |
2023-05-18T16:08:11Z | http://arxiv.org/pdf/2305.11080v1 | http://arxiv.org/abs/2305.11080v1 | Abhipsa Basu, R. Venkatesh Babu, Danish Pruthi | 2023-05-18T16:08:11Z | cs.CV, cs.CL | Recent progress in generative models has resulted in models that produce both
realistic as well as relevant images for most textual inputs. These models are
being used to generate millions of images everyday, and hold the potential to
drastically impact areas such as generative art, digital marketing and data
augmentation. Given their outsized impact, it is important to ensure that the
generated content reflects the artifacts and surroundings across the globe,
rather than over-representing certain parts of the world. In this paper, we
measure the geographical representativeness of common nouns (e.g., a house)
generated through DALL.E 2 and Stable Diffusion models using a crowdsourced
study comprising 540 participants across 27 countries. For deliberately
underspecified inputs without country names, the generated images most reflect
the surroundings of the United States followed by India, and the top
generations rarely reflect surroundings from all other countries (average score
less than 3 out of 5). Specifying the country names in the input increases the
representativeness by 1.44 points on average for DALL.E 2 and 0.75 for Stable
Diffusion, however, the overall scores for many countries still remain low,
highlighting the need for future models to be more geographically inclusive.
Lastly, we examine the feasibility of quantifying the geographical
representativeness of generated images without conducting user studies. | cs.CV | Inspecting the Geographical Representativeness of Images from Text-to-Image Models |
2023-05-18T16:00:48Z | http://arxiv.org/pdf/2305.11073v1 | http://arxiv.org/abs/2305.11073v1 | Yifan Peng, Kwangyoun Kim, Felix Wu, Brian Yan, Siddhant Arora, William Chen, Jiyang Tang, Suwon Shon, Prashant Sridhar, Shinji Watanabe | 2023-05-18T16:00:48Z | cs.CL, cs.SD, eess.AS | Conformer, a convolution-augmented Transformer variant, has become the de
facto encoder architecture for speech processing due to its superior
performance in various tasks, including automatic speech recognition (ASR),
speech translation (ST) and spoken language understanding (SLU). Recently, a
new encoder called E-Branchformer has outperformed Conformer in the LibriSpeech
ASR benchmark, making it promising for more general speech applications. This
work compares E-Branchformer and Conformer through extensive experiments using
different types of end-to-end sequence-to-sequence models. Results demonstrate
that E-Branchformer achieves comparable or better performance than Conformer in
almost all evaluation sets across 15 ASR, 2 ST, and 3 SLU benchmarks, while
being more stable during training. We will release our training configurations
and pre-trained models for reproducibility, which can benefit the speech
community. | cs.CL | A Comparative Study on E-Branchformer vs Conformer in Speech Recognition, Translation, and Understanding Tasks |
2023-05-18T15:59:36Z | http://arxiv.org/pdf/2305.11072v1 | http://arxiv.org/abs/2305.11072v1 | Heng-Jui Chang, Alexander H. Liu, James Glass | 2023-05-18T15:59:36Z | cs.CL, eess.AS | Self-supervised speech representation models have succeeded in various tasks,
but improving them for content-related problems using unlabeled data is
challenging. We propose speaker-invariant clustering (Spin), a novel
self-supervised learning method that clusters speech representations and
performs swapped prediction between the original and speaker-perturbed
utterances. Spin disentangles speaker information and preserves content
representations with just 45 minutes of fine-tuning on a single GPU. Spin
improves pre-trained networks and outperforms prior methods in speech
recognition and acoustic unit discovery. | cs.CL | Self-supervised Fine-tuning for Improved Content Representations by Speaker-invariant Clustering |
2023-05-18T15:43:09Z | http://arxiv.org/pdf/2305.11052v1 | http://arxiv.org/abs/2305.11052v1 | Shicheng Xu, Liang Pang, Huawei Shen, Xueqi Cheng | 2023-05-18T15:43:09Z | cs.IR, cs.CL | Dense retrieval has shown promise in the first-stage retrieval process when
trained on in-domain labeled datasets. However, previous studies have found
that dense retrieval is hard to generalize to unseen domains due to its weak
modeling of domain-invariant and interpretable feature (i.e., matching signal
between two texts, which is the essence of information retrieval). In this
paper, we propose a novel method to improve the generalization of dense
retrieval via capturing matching signal called BERM. Fully fine-grained
expression and query-oriented saliency are two properties of the matching
signal. Thus, in BERM, a single passage is segmented into multiple units and
two unit-level requirements are proposed for representation as the constraint
in training to obtain the effective matching signal. One is semantic unit
balance and the other is essential matching unit extractability. Unit-level
view and balanced semantics make representation express the text in a
fine-grained manner. Essential matching unit extractability makes passage
representation sensitive to the given query to extract the pure matching
information from the passage containing complex context. Experiments on BEIR
show that our method can be effectively combined with different dense retrieval
training methods (vanilla, hard negatives mining and knowledge distillation) to
improve its generalization ability without any additional inference overhead
and target domain data. | cs.IR | BERM: Training the Balanced and Extractable Representation for Matching to Improve Generalization Ability of Dense Retrieval |
2023-05-18T15:31:34Z | http://arxiv.org/pdf/2305.11038v3 | http://arxiv.org/abs/2305.11038v3 | Jiawei Chen, Yaojie Lu, Hongyu Lin, Jie Lou, Wei Jia, Dai Dai, Hua Wu, Boxi Cao, Xianpei Han, Le Sun | 2023-05-26T05:44:00Z | cs.CL | Named entity recognition in real-world applications suffers from the
diversity of entity types, the emergence of new entity types, and the lack of
high-quality annotations. To address the above problems, this paper proposes an
in-context learning-based NER approach, which can effectively inject in-context
NER ability into PLMs and recognize entities of novel types on-the-fly using
only a few demonstrative instances. Specifically, we model PLMs as a
meta-function $\mathcal{ \lambda_ {\text{instruction, demonstrations, text}}.
M}$, and a new entity extractor can be implicitly constructed by applying new
instruction and demonstrations to PLMs, i.e., $\mathcal{ (\lambda . M)
}$(instruction, demonstrations) $\to$ $\mathcal{F}$ where $\mathcal{F}$ will be
a new entity extractor, i.e., $\mathcal{F}$: text $\to$ entities. To inject the
above in-context NER ability into PLMs, we propose a meta-function pre-training
algorithm, which pre-trains PLMs by comparing the (instruction,
demonstration)-initialized extractor with a surrogate golden extractor.
Experimental results on 4 few-shot NER datasets show that our method can
effectively inject in-context NER ability into PLMs and significantly
outperforms the PLMs+fine-tuning counterparts. | cs.CL | Learning In-context Learning for Named Entity Recognition |
2023-05-18T15:22:00Z | http://arxiv.org/pdf/2305.11034v1 | http://arxiv.org/abs/2305.11034v1 | Samuel Mensah, Kai Sun, Nikolaos Aletras | 2023-05-18T15:22:00Z | cs.CL | State-of-the-art target-oriented opinion word extraction (TOWE) models
typically use BERT-based text encoders that operate on the word level, along
with graph convolutional networks (GCNs) that incorporate syntactic information
extracted from syntax trees. These methods achieve limited gains with GCNs and
have difficulty using BERT wordpieces. Meanwhile, BERT wordpieces are known to
be effective at representing rare words or words with insufficient context
information. To address this issue, this work trades syntax trees for BERT
wordpieces by entirely removing the GCN component from the methods'
architectures. To enhance TOWE performance, we tackle the issue of aspect
representation loss during encoding. Instead of solely utilizing a sentence as
the input, we use a sentence-aspect pair. Our relatively simple approach
achieves state-of-the-art results on benchmark datasets and should serve as a
strong baseline for further research. | cs.CL | Trading Syntax Trees for Wordpieces: Target-oriented Opinion Words Extraction with Wordpieces and Aspect Enhancement |
2023-05-18T15:15:56Z | http://arxiv.org/pdf/2305.11029v2 | http://arxiv.org/abs/2305.11029v2 | Qi Sun, Kun Huang, Xiaocui Yang, Pengfei Hong, Kun Zhang, Soujanya Poria | 2023-05-26T08:23:43Z | cs.CL, cs.AI | Document-level relation extraction (DocRE) aims to infer complex semantic
relations among entities in a document. Distant supervision (DS) is able to
generate massive auto-labeled data, which can improve DocRE performance. Recent
works leverage pseudo labels generated by the pre-denoising model to reduce
noise in DS data. However, unreliable pseudo labels bring new noise, e.g.,
adding false pseudo labels and losing correct DS labels. Therefore, how to
select effective pseudo labels to denoise DS data is still a challenge in
document-level distant relation extraction. To tackle this issue, we introduce
uncertainty estimation technology to determine whether pseudo labels can be
trusted. In this work, we propose a Document-level distant Relation Extraction
framework with Uncertainty Guided label denoising, UGDRE. Specifically, we
propose a novel instance-level uncertainty estimation method, which measures
the reliability of the pseudo labels with overlapping relations. By further
considering the long-tail problem, we design dynamic uncertainty thresholds for
different types of relations to filter high-uncertainty pseudo labels. We
conduct experiments on two public datasets. Our framework outperforms strong
baselines by 1.91 F1 and 2.28 Ign F1 on the RE-DocRED dataset. | cs.CL | Uncertainty Guided Label Denoising for Document-level Distant Relation Extraction |
2023-05-18T15:04:52Z | http://arxiv.org/pdf/2305.11023v1 | http://arxiv.org/abs/2305.11023v1 | Harshil Shah, Arthur Wilcke, Marius Cobzarenco, Cristi Cobzarenco, Edward Challis, David Barber | 2023-05-18T15:04:52Z | cs.CL | Natural language understanding includes the tasks of intent detection
(identifying a user's objectives) and slot filling (extracting the entities
relevant to those objectives). Prior slot filling methods assume that each
intent type cannot occur more than once within a message, however this is often
not a valid assumption for real-world settings. In this work, we generalize
slot filling by removing the constraint of unique intents in a message. We cast
this as a JSON generation task and approach it using a language model. We
create a pre-training dataset by combining DBpedia and existing slot filling
datasets that we convert for JSON generation. We also generate an in-domain
dataset using GPT-3. We train T5 models for this task (with and without
exemplars in the prompt) and find that both training datasets improve
performance, and that the model is able to generalize to intent types not seen
during training. | cs.CL | Generalized Multiple Intent Conditioned Slot Filling |
2023-05-18T14:49:19Z | http://arxiv.org/pdf/2305.11016v1 | http://arxiv.org/abs/2305.11016v1 | Elisa Bassignana, Filip Ginter, Sampo Pyysalo, Rob van der Goot, Barbara Plank | 2023-05-18T14:49:19Z | cs.CL | Relation Extraction (RE) remains a challenging task, especially when
considering realistic out-of-domain evaluations. One of the main reasons for
this is the limited training size of current RE datasets: obtaining
high-quality (manually annotated) data is extremely expensive and cannot
realistically be repeated for each new domain. An intermediate training step on
data from related tasks has shown to be beneficial across many NLP
tasks.However, this setup still requires supplementary annotated data, which is
often not available. In this paper, we investigate intermediate pre-training
specifically for RE. We exploit the affinity between syntactic structure and
semantic RE, and identify the syntactic relations which are closely related to
RE by being on the shortest dependency path between two entities. We then take
advantage of the high accuracy of current syntactic parsers in order to
automatically obtain large amounts of low-cost pre-training data. By
pre-training our RE model on the relevant syntactic relations, we are able to
outperform the baseline in five out of six cross-domain setups, without any
additional annotated data. | cs.CL | Silver Syntax Pre-training for Cross-Domain Relation Extraction |
2023-05-18T14:45:09Z | http://arxiv.org/pdf/2305.11013v1 | http://arxiv.org/abs/2305.11013v1 | Zhifu Gao, Zerui Li, Jiaming Wang, Haoneng Luo, Xian Shi, Mengzhe Chen, Yabin Li, Lingyun Zuo, Zhihao Du, Zhangyu Xiao, Shiliang Zhang | 2023-05-18T14:45:09Z | cs.SD, cs.CL, eess.AS | This paper introduces FunASR, an open-source speech recognition toolkit
designed to bridge the gap between academic research and industrial
applications. FunASR offers models trained on large-scale industrial corpora
and the ability to deploy them in applications. The toolkit's flagship model,
Paraformer, is a non-autoregressive end-to-end speech recognition model that
has been trained on a manually annotated Mandarin speech recognition dataset
that contains 60,000 hours of speech. To improve the performance of Paraformer,
we have added timestamp prediction and hotword customization capabilities to
the standard Paraformer backbone. In addition, to facilitate model deployment,
we have open-sourced a voice activity detection model based on the Feedforward
Sequential Memory Network (FSMN-VAD) and a text post-processing punctuation
model based on the controllable time-delay Transformer (CT-Transformer), both
of which were trained on industrial corpora. These functional modules provide a
solid foundation for building high-precision long audio speech recognition
services. Compared to other models trained on open datasets, Paraformer
demonstrates superior performance. | cs.SD | FunASR: A Fundamental End-to-End Speech Recognition Toolkit |
2023-05-18T14:34:58Z | http://arxiv.org/pdf/2305.11004v4 | http://arxiv.org/abs/2305.11004v4 | Wei Xue, Yongliang Shen, Wenqi Ren, Jietian Guo, Shiliang Pu, Weiming Lu | 2024-06-18T04:20:35Z | cs.CL | Taxonomy completion, enriching existing taxonomies by inserting new concepts
as parents or attaching them as children, has gained significant interest.
Previous approaches embed concepts as vectors in Euclidean space, which makes
it difficult to model asymmetric relations in taxonomy. In addition, they
introduce pseudo-leaves to convert attachment cases into insertion cases,
leading to an incorrect bias in network learning dominated by numerous
pseudo-leaves. Addressing these, our framework, TaxBox, leverages box
containment and center closeness to design two specialized geometric scorers
within the box embedding space. These scorers are tailored for insertion and
attachment operations and can effectively capture intrinsic relationships
between concepts by optimizing on a granular box constraint loss. We employ a
dynamic ranking loss mechanism to balance the scores from these scorers,
allowing adaptive adjustments of insertion and attachment scores. Experiments
on four real-world datasets show that TaxBox significantly outperforms previous
methods, yielding substantial improvements over prior methods in real-world
datasets, with average performance boosts of 6.7%, 34.9%, and 51.4% in MRR,
Hit@1, and Prec@1, respectively. | cs.CL | Insert or Attach: Taxonomy Completion via Box Embedding |
2023-05-18T14:23:25Z | http://arxiv.org/pdf/2305.11000v2 | http://arxiv.org/abs/2305.11000v2 | Dong Zhang, Shimin Li, Xin Zhang, Jun Zhan, Pengyu Wang, Yaqian Zhou, Xipeng Qiu | 2023-05-19T14:41:16Z | cs.CL | Multi-modal large language models are regarded as a crucial step towards
Artificial General Intelligence (AGI) and have garnered significant interest
with the emergence of ChatGPT. However, current speech-language models
typically adopt the cascade paradigm, preventing inter-modal knowledge
transfer. In this paper, we propose SpeechGPT, a large language model with
intrinsic cross-modal conversational abilities, capable of perceiving and
generating multi-model content. With discrete speech representations, we first
construct SpeechInstruct, a large-scale cross-modal speech instruction dataset.
Additionally, we employ a three-stage training strategy that includes
modality-adaptation pre-training, cross-modal instruction fine-tuning, and
chain-of-modality instruction fine-tuning. The experimental results demonstrate
that SpeechGPT has an impressive capacity to follow multi-modal human
instructions and highlight the potential of handling multiple modalities with
one model. Demos are shown in https://0nutation.github.io/SpeechGPT.github.io/. | cs.CL | SpeechGPT: Empowering Large Language Models with Intrinsic Cross-Modal Conversational Abilities |
2023-05-18T14:20:32Z | http://arxiv.org/pdf/2305.10998v2 | http://arxiv.org/abs/2305.10998v2 | Junyi Li, Tianyi Tang, Wayne Xin Zhao, Jingyuan Wang, Jian-Yun Nie, Ji-Rong Wen | 2023-05-24T09:35:39Z | cs.CL | Large language models (LLMs) encode a large amount of world knowledge.
However, as such knowledge is frozen at the time of model training, the models
become static and limited by the training data at that time. In order to
further improve the capacity of LLMs for knowledge-intensive tasks, we consider
augmenting LLMs with the large-scale web using search engine. Unlike previous
augmentation sources (e.g., Wikipedia data dump), the web provides broader,
more comprehensive and constantly updated information. In this paper, we
present a web-augmented LLM UNIWEB, which is trained over 16
knowledge-intensive tasks in a unified text-to-text format. Instead of simply
using the retrieved contents from web, our approach has made two major
improvements. Firstly, we propose an adaptive search engine assisted learning
method that can self-evaluate the confidence level of LLM's predictions, and
adaptively determine when to refer to the web for more data, which can avoid
useless or noisy augmentation from web. Secondly, we design a pretraining task,
i.e., continual knowledge learning, based on salient spans prediction, to
reduce the discrepancy between the encoded and retrieved knowledge. Experiments
on a wide range of knowledge-intensive tasks show that our model significantly
outperforms previous retrieval-augmented methods. | cs.CL | The Web Can Be Your Oyster for Improving Large Language Models |
2023-05-18T14:11:57Z | http://arxiv.org/pdf/2305.10992v1 | http://arxiv.org/abs/2305.10992v1 | Atsuki Yamaguchi, Hiroaki Ozaki, Terufumi Morishita, Gaku Morio, Yasuhiro Sogawa | 2023-05-18T14:11:57Z | cs.CL, cs.AI | Masked language modeling (MLM) is a widely used self-supervised pretraining
objective, where a model needs to predict an original token that is replaced
with a mask given contexts. Although simpler and computationally efficient
pretraining objectives, e.g., predicting the first character of a masked token,
have recently shown comparable results to MLM, no objectives with a masking
scheme actually outperform it in downstream tasks. Motivated by the assumption
that their lack of complexity plays a vital role in the degradation, we
validate whether more complex masked objectives can achieve better results and
investigate how much complexity they should have to perform comparably to MLM.
Our results using GLUE, SQuAD, and Universal Dependencies benchmarks
demonstrate that more complicated objectives tend to show better downstream
results with at least half of the MLM complexity needed to perform comparably
to MLM. Finally, we discuss how we should pretrain a model using a masked
objective from the task complexity perspective. | cs.CL | How does the task complexity of masked pretraining objectives affect downstream performance? |
2023-05-18T14:09:52Z | http://arxiv.org/pdf/2305.10991v1 | http://arxiv.org/abs/2305.10991v1 | Luca Herranz-Celotti, Ermal Rrapaj | 2023-05-18T14:09:52Z | cs.CL, cs.AI | The softmax attention mechanism has emerged as a noteworthy development in
the field of Artificial Intelligence research, building on the successes of
Transformer-based architectures. However, their ever increasing sizes
necessitate ever increasing computational memory, that limits their usage. We
propose KgV, a sigmoid gating mechanism that, in conjunction with softmax
attention, significantly boosts performance without increasing architecture
size. To amend the size requirements, we leverage Tensor Chains to identify and
prune the excess parameters. We find that such excess resides primarily within
the embedding layer, and not in the output linear layer. To further improve
embedding and significantly reduce parameters, we introduce H-SoftPOS, a
hierarchical embedding layer which simultaneously enhances performance.
Remarkably, on the WMT14 English-German validation set, our approach yields a
threefold reduction in perplexity, surpassing the current state-of-the-art,
while reducing parameter counts also by a factor of 3. When we further reduce
the number of parameters up to sevenfold, we can still achieve a 21\% decrease
in perplexity with respect to the baseline Transformer. To understand
generalization capabilities, we conduct experiments on the 7 language pairs of
the WMT17 dataset. Our method outperforms existing techniques in terms of test
loss while simultaneously halving the number of parameters. Moreover, we
observe a 70 times reduction in variance with respect to the prior
state-of-the-art. In conclusion, our proposed method yields significant
improvements in performance and much lower memory cost. We call the resulting
architecture Anthe. | cs.CL | Less is More! A slim architecture for optimal language translation |
2023-05-18T14:01:33Z | http://arxiv.org/pdf/2305.10985v1 | http://arxiv.org/abs/2305.10985v1 | Elisa Bassignana, Filip Ginter, Sampo Pyysalo, Rob van der Goot, Barbara Plank | 2023-05-18T14:01:33Z | cs.CL | Most research in Relation Extraction (RE) involves the English language,
mainly due to the lack of multi-lingual resources. We propose Multi-CrossRE,
the broadest multi-lingual dataset for RE, including 26 languages in addition
to English, and covering six text domains. Multi-CrossRE is a machine
translated version of CrossRE (Bassignana and Plank, 2022), with a sub-portion
including more than 200 sentences in seven diverse languages checked by native
speakers. We run a baseline model over the 26 new datasets and--as sanity
check--over the 26 back-translations to English. Results on the back-translated
data are consistent with the ones on the original English CrossRE, indicating
high quality of the translation and the resulting dataset. | cs.CL | Multi-CrossRE A Multi-Lingual Multi-Domain Dataset for Relation Extraction |
2023-05-18T13:38:36Z | http://arxiv.org/pdf/2305.10971v2 | http://arxiv.org/abs/2305.10971v2 | Iyanuoluwa Shode, David Ifeoluwa Adelani, Jing Peng, Anna Feldman | 2023-08-22T07:25:43Z | cs.CL | Africa has over 2000 indigenous languages but they are under-represented in
NLP research due to lack of datasets. In recent years, there have been progress
in developing labeled corpora for African languages. However, they are often
available in a single domain and may not generalize to other domains. In this
paper, we focus on the task of sentiment classification for cross domain
adaptation. We create a new dataset, NollySenti - based on the Nollywood movie
reviews for five languages widely spoken in Nigeria (English, Hausa, Igbo,
Nigerian-Pidgin, and Yoruba. We provide an extensive empirical evaluation using
classical machine learning methods and pre-trained language models. Leveraging
transfer learning, we compare the performance of cross-domain adaptation from
Twitter domain, and cross-lingual adaptation from English language. Our
evaluation shows that transfer from English in the same target domain leads to
more than 5% improvement in accuracy compared to transfer from Twitter in the
same language. To further mitigate the domain difference, we leverage machine
translation (MT) from English to other Nigerian languages, which leads to a
further improvement of 7% over cross-lingual evaluation. While MT to
low-resource languages are often of low quality, through human evaluation, we
show that most of the translated sentences preserve the sentiment of the
original English reviews. | cs.CL | NollySenti: Leveraging Transfer Learning and Machine Translation for Nigerian Movie Sentiment Classification |
2023-05-18T13:20:38Z | http://arxiv.org/pdf/2305.10951v2 | http://arxiv.org/abs/2305.10951v2 | Martijn Bartelds, Nay San, Bradley McDonnell, Dan Jurafsky, Martijn Wieling | 2023-05-19T02:15:41Z | cs.CL, eess.AS | The performance of automatic speech recognition (ASR) systems has advanced
substantially in recent years, particularly for languages for which a large
amount of transcribed speech is available. Unfortunately, for low-resource
languages, such as minority languages, regional languages or dialects, ASR
performance generally remains much lower. In this study, we investigate whether
data augmentation techniques could help improve low-resource ASR performance,
focusing on four typologically diverse minority languages or language variants
(West Germanic: Gronings, West-Frisian; Malayo-Polynesian: Besemah, Nasal). For
all four languages, we examine the use of self-training, where an ASR system
trained with the available human-transcribed data is used to generate
transcriptions, which are then combined with the original data to train a new
ASR system. For Gronings, for which there was a pre-existing text-to-speech
(TTS) system available, we also examined the use of TTS to generate ASR
training data from text-only sources. We find that using a self-training
approach consistently yields improved performance (a relative WER reduction up
to 20.5% compared to using an ASR system trained on 24 minutes of manually
transcribed speech). The performance gain from TTS augmentation for Gronings
was even stronger (up to 25.5% relative reduction in WER compared to a system
based on 24 minutes of manually transcribed speech). In sum, our results show
the benefit of using self-training or (if possible) TTS-generated data as an
efficient solution to overcome the limitations of data availability for
resource-scarce languages in order to improve ASR performance. | cs.CL | Making More of Little Data: Improving Low-Resource Automatic Speech Recognition Using Data Augmentation |
2023-05-18T12:43:31Z | http://arxiv.org/pdf/2305.10930v3 | http://arxiv.org/abs/2305.10930v3 | Liang Chen, Shuming Ma, Dongdong Zhang, Furu Wei, Baobao Chang | 2023-06-02T02:58:03Z | cs.CL, cs.AI | While multilingual neural machine translation has achieved great success, it
suffers from the off-target issue, where the translation is in the wrong
language. This problem is more pronounced on zero-shot translation tasks. In
this work, we find that failing in encoding discriminative target language
signal will lead to off-target and a closer lexical distance (i.e.,
KL-divergence) between two languages' vocabularies is related with a higher
off-target rate. We also find that solely isolating the vocab of different
languages in the decoder can alleviate the problem. Motivated by the findings,
we propose Language Aware Vocabulary Sharing (LAVS), a simple and effective
algorithm to construct the multilingual vocabulary, that greatly alleviates the
off-target problem of the translation model by increasing the KL-divergence
between languages. We conduct experiments on a multilingual machine translation
benchmark in 11 languages. Experiments show that the off-target rate for 90
translation tasks is reduced from 29\% to 8\%, while the overall BLEU score is
improved by an average of 1.9 points without extra training cost or sacrificing
the supervised directions' performance. We release the code at
https://github.com/PKUnlp-icler/Off-Target-MNMT for reproduction. | cs.CL | On the Off-Target Problem of Zero-Shot Multilingual Neural Machine Translation |
2023-05-18T12:40:41Z | http://arxiv.org/pdf/2305.10928v1 | http://arxiv.org/abs/2305.10928v1 | Nadav Borenstein, Natalia da Silva Perez, Isabelle Augenstein | 2023-05-18T12:40:41Z | cs.CL, cs.LG | NLP methods can aid historians in analyzing textual materials in greater
volumes than manually feasible. Developing such methods poses substantial
challenges though. First, acquiring large, annotated historical datasets is
difficult, as only domain experts can reliably label them. Second, most
available off-the-shelf NLP models are trained on modern language texts,
rendering them significantly less effective when applied to historical corpora.
This is particularly problematic for less well studied tasks, and for languages
other than English. This paper addresses these challenges while focusing on the
under-explored task of event extraction from a novel domain of historical
texts. We introduce a new multilingual dataset in English, French, and Dutch
composed of newspaper ads from the early modern colonial period reporting on
enslaved people who liberated themselves from enslavement. We find that: 1)
even with scarce annotated data, it is possible to achieve surprisingly good
results by formulating the problem as an extractive QA task and leveraging
existing datasets and models for modern languages; and 2) cross-lingual
low-resource learning for historical languages is highly challenging, and
machine translation of the historical datasets to the considered target
languages is, in practice, often the best-performing solution. | cs.CL | Multilingual Event Extraction from Historical Newspaper Adverts |
2023-05-18T12:39:25Z | http://arxiv.org/pdf/2305.10927v3 | http://arxiv.org/abs/2305.10927v3 | Yingxiu Zhao, Bowen Yu, Haiyang Yu, Bowen Li, Jinyang Li, Chao Wang, Fei Huang, Yongbin Li, Nevin L. Zhang | 2023-11-05T15:26:49Z | cs.CL | The goal of document-grounded dialogue (DocGD) is to generate a response by
grounding the evidence in a supporting document in accordance with the dialogue
context. This process involves four variables that are causally connected.
Recently, task-specific pre-training has greatly boosted performances on many
downstream tasks. Existing DocGD methods, however, continue to rely on general
pre-trained language models without a specifically tailored pre-training
approach that explicitly captures the causal relationships. To tackle this
issue, we are the first to present a causally-complete dataset construction
strategy for building million-level DocGD pre-training corpora. To better
capture causality, we further propose a causally-perturbed pre-training
strategy, which introduces causal perturbations on the variables and optimizes
the overall causal effect. Experiments on three benchmark datasets demonstrate
that our causal pre-training achieves considerable and consistent improvements
under fully-supervised, low-resource, few-shot, and zero-shot settings. | cs.CL | Causal Document-Grounded Dialogue Pre-training |
2023-05-18T12:37:01Z | http://arxiv.org/pdf/2305.10923v1 | http://arxiv.org/abs/2305.10923v1 | Chuan Meng, Negar Arabzadeh, Mohammad Aliannejadi, Maarten de Rijke | 2023-05-18T12:37:01Z | cs.IR, cs.CL, cs.LG, H.3.3 | Query performance prediction (QPP) is a core task in information retrieval.
The QPP task is to predict the retrieval quality of a search system for a query
without relevance judgments. Research has shown the effectiveness and
usefulness of QPP for ad-hoc search. Recent years have witnessed considerable
progress in conversational search (CS). Effective QPP could help a CS system to
decide an appropriate action to be taken at the next turn. Despite its
potential, QPP for CS has been little studied. We address this research gap by
reproducing and studying the effectiveness of existing QPP methods in the
context of CS. While the task of passage retrieval remains the same in the two
settings, a user query in CS depends on the conversational history, introducing
novel QPP challenges. In particular, we seek to explore to what extent findings
from QPP methods for ad-hoc search generalize to three CS settings: (i)
estimating the retrieval quality of different query rewriting-based retrieval
methods, (ii) estimating the retrieval quality of a conversational dense
retrieval method, and (iii) estimating the retrieval quality for top ranks vs.
deeper-ranked lists. Our findings can be summarized as follows: (i) supervised
QPP methods distinctly outperform unsupervised counterparts only when a
large-scale training set is available; (ii) point-wise supervised QPP methods
outperform their list-wise counterparts in most cases; and (iii) retrieval
score-based unsupervised QPP methods show high effectiveness in assessing the
conversational dense retrieval method, ConvDR. | cs.IR | Query Performance Prediction: From Ad-hoc to Conversational Search |
2023-05-18T12:31:45Z | http://arxiv.org/pdf/2305.10920v1 | http://arxiv.org/abs/2305.10920v1 | Ryokan Ri, Ryo Ueda, Jason Naradowsky | 2023-05-18T12:31:45Z | cs.CL, cs.AI | To develop computational agents that better communicate using their own
emergent language, we endow the agents with an ability to focus their attention
on particular concepts in the environment. Humans often understand an object or
scene as a composite of concepts and those concepts are further mapped onto
words. We implement this intuition as cross-modal attention mechanisms in
Speaker and Listener agents in a referential game and show attention leads to
more compositional and interpretable emergent language. We also demonstrate how
attention aids in understanding the learned communication protocol by
investigating the attention weights associated with each message symbol and the
alignment of attention weights between Speaker and Listener agents. Overall,
our results suggest that attention is a promising mechanism for developing more
human-like emergent language. | cs.CL | Emergent Communication with Attention |
2023-05-18T12:25:07Z | http://arxiv.org/pdf/2305.10913v2 | http://arxiv.org/abs/2305.10913v2 | Davide Rigoni, Luca Parolari, Luciano Serafini, Alessandro Sperduti, Lamberto Ballan | 2023-09-26T09:29:26Z | cs.CV, cs.AI, cs.CL, cs.LG, cs.MM | Using only image-sentence pairs, weakly-supervised visual-textual grounding
aims to learn region-phrase correspondences of the respective entity mentions.
Compared to the supervised approach, learning is more difficult since bounding
boxes and textual phrases correspondences are unavailable. In light of this, we
propose the Semantic Prior Refinement Model (SPRM), whose predictions are
obtained by combining the output of two main modules. The first untrained
module aims to return a rough alignment between textual phrases and bounding
boxes. The second trained module is composed of two sub-components that refine
the rough alignment to improve the accuracy of the final phrase-bounding box
alignments. The model is trained to maximize the multimodal similarity between
an image and a sentence, while minimizing the multimodal similarity of the same
sentence and a new unrelated image, carefully selected to help the most during
training. Our approach shows state-of-the-art results on two popular datasets,
Flickr30k Entities and ReferIt, shining especially on ReferIt with a 9.6%
absolute improvement. Moreover, thanks to the untrained component, it reaches
competitive performances just using a small fraction of training examples. | cs.CV | Weakly-Supervised Visual-Textual Grounding with Semantic Prior Refinement |
2023-05-18T12:10:06Z | http://arxiv.org/pdf/2305.10907v1 | http://arxiv.org/abs/2305.10907v1 | Xinze Li, Yixin Cao, Muhao Chen, Aixin Sun | 2023-05-18T12:10:06Z | cs.CL | Goal-oriented Script Generation is a new task of generating a list of steps
that can fulfill the given goal. In this paper, we propose to extend the task
from the perspective of cognitive theory. Instead of a simple flat structure,
the steps are typically organized hierarchically - Human often decompose a
complex task into subgoals, where each subgoal can be further decomposed into
steps. To establish the benchmark, we contribute a new dataset, propose several
baseline methods, and set up evaluation metrics. Both automatic and human
evaluation verify the high-quality of dataset, as well as the effectiveness of
incorporating subgoals into hierarchical script generation. Furthermore, We
also design and evaluate the model to discover subgoal, and find that it is a
bit more difficult to decompose the goals than summarizing from segmented
steps. | cs.CL | Take a Break in the Middle: Investigating Subgoals towards Hierarchical Script Generation |
2023-05-18T11:41:56Z | http://arxiv.org/pdf/2305.10892v2 | http://arxiv.org/abs/2305.10892v2 | Marco Rovera | 2024-02-18T19:29:44Z | cs.CL | This paper introduces EventNet-ITA, a large, multi-domain corpus annotated
full-text with event frames for Italian. Moreover, we present and thoroughly
evaluate an efficient multi-label sequence labeling approach for Frame Parsing.
Covering a wide range of individual, social and historical phenomena, with more
than 53,000 annotated sentences and over 200 modeled frames, EventNet-ITA
constitutes the first systematic attempt to provide the Italian language with a
publicly available resource for Frame Parsing of events, useful for a broad
spectrum of research and application tasks. Our approach achieves a promising
0.9 strict F1-score for frame classification and 0.72 for frame element
classification, on top of minimizing computational requirements. The annotated
corpus and the frame parsing model are released under open license. | cs.CL | EventNet-ITA: Italian Frame Parsing for Events |
2023-05-18T10:38:06Z | http://arxiv.org/pdf/2305.10866v1 | http://arxiv.org/abs/2305.10866v1 | Wei Xiang, Chao Liang, Bang Wang | 2023-05-18T10:38:06Z | cs.CL | Implicit Discourse Relation Recognition (IDRR) aims at classifying the
relation sense between two arguments without an explicit connective. Recently,
the ConnPrompt~\cite{Wei.X:et.al:2022:COLING} has leveraged the powerful prompt
learning for IDRR based on the fusion of multi-prompt decisions from three
different yet much similar connective prediction templates. Instead of
multi-prompt ensembling, we propose to design auxiliary tasks with enlightened
prompt learning for the IDRR task. Although an auxiliary task is not used to
directly output final prediction, we argue that during the joint training some
of its learned features can be useful to boost the main task. In light of such
motivations, we propose a task enlightenment prompt learning model, called
TEPrompt, to fuse learned features from three related tasks for IDRR. In
particular, the TEPrompt contains three tasks, viz., Discourse Relation
Recognition (DRR), Sense Semantics Classification (SSC) and Annotated
Connective Prediction (ACP), each with a unique prompt template and an answer
space. In the training phase, we jointly train three prompt learning tasks with
shared argument representation. In the testing phase, we only take the DRR
output with fused features as the final IDRR decision. Experiments with the
same conditions have shown that the proposed TEPrompt outperforms the
ConnPrompt. This can be attributed to the promoted decision features and
language models benefited from joint-training of auxiliary tasks. | cs.CL | TEPrompt: Task Enlightenment Prompt Learning for Implicit Discourse Relation Recognition |
2023-05-18T10:37:54Z | http://arxiv.org/pdf/2305.10865v2 | http://arxiv.org/abs/2305.10865v2 | Wenhao Li, Dan Qiao, Baoxiang Wang, Xiangfeng Wang, Bo Jin, Hongyuan Zha | 2023-09-30T08:27:28Z | cs.LG, cs.AI, cs.CL, cs.MA | The difficulty of appropriately assigning credit is particularly heightened
in cooperative MARL with sparse reward, due to the concurrent time and
structural scales involved. Automatic subgoal generation (ASG) has recently
emerged as a viable MARL approach inspired by utilizing subgoals in
intrinsically motivated reinforcement learning. However, end-to-end learning of
complex task planning from sparse rewards without prior knowledge, undoubtedly
requires massive training samples. Moreover, the diversity-promoting nature of
existing ASG methods can lead to the "over-representation" of subgoals,
generating numerous spurious subgoals of limited relevance to the actual task
reward and thus decreasing the sample efficiency of the algorithm. To address
this problem and inspired by the disentangled representation learning, we
propose a novel "disentangled" decision-making method, Semantically Aligned
task decomposition in MARL (SAMA), that prompts pretrained language models with
chain-of-thought that can suggest potential goals, provide suitable goal
decomposition and subgoal allocation as well as self-reflection-based
replanning. Additionally, SAMA incorporates language-grounded RL to train each
agent's subgoal-conditioned policy. SAMA demonstrates considerable advantages
in sample efficiency compared to state-of-the-art ASG methods, as evidenced by
its performance on two challenging sparse-reward tasks, Overcooked and MiniRTS. | cs.LG | Semantically Aligned Task Decomposition in Multi-Agent Reinforcement Learning |
2023-05-18T10:07:50Z | http://arxiv.org/pdf/2305.10848v1 | http://arxiv.org/abs/2305.10848v1 | Dmitriy Kalugin-Balashov | 2023-05-18T10:07:50Z | cs.CL | In this paper, we unveil a groundbreaking method to amplify full-text search
lemmatization, utilizing the OpenCorpora dataset and a bespoke paradigm
retrieval algorithm. Our primary aim is to streamline the extraction of a
word's primary form or lemma - a crucial factor in full-text search.
Additionally, we propose a compact dictionary storage strategy, significantly
boosting the speed and precision of lemma retrieval. | cs.CL | Advancing Full-Text Search Lemmatization Techniques with Paradigm Retrieval from OpenCorpora |
2023-05-18T10:03:25Z | http://arxiv.org/pdf/2305.10847v6 | http://arxiv.org/abs/2305.10847v6 | Ning Lu, Shengcai Liu, Rui He, Qi Wang, Yew-Soon Ong, Ke Tang | 2024-05-15T08:00:09Z | cs.CL, cs.AI | Large language models (LLMs) have shown remarkable performance in various
tasks and have been extensively utilized by the public. However, the increasing
concerns regarding the misuse of LLMs, such as plagiarism and spamming, have
led to the development of multiple detectors, including fine-tuned classifiers
and statistical methods. In this study, we equip LLMs with prompts, rather than
relying on an external paraphraser, to evaluate the vulnerability of these
detectors. We propose a novel Substitution-based In-Context example
Optimization method (SICO) to automatically construct prompts for evading the
detectors. SICO is cost-efficient as it requires only 40 human-written examples
and a limited number of LLM inferences to generate a prompt. Moreover, once a
task-specific prompt has been constructed, it can be universally used against a
wide range of detectors. Extensive experiments across three real-world tasks
demonstrate that SICO significantly outperforms the paraphraser baselines and
enables GPT-3.5 to successfully evade six detectors, decreasing their AUC by
0.5 on average. Furthermore, a comprehensive human evaluation show that the
SICO-generated text achieves human-level readability and task completion rates,
while preserving high imperceptibility. Finally, we propose an ensemble
approach to enhance the robustness of detectors against SICO attack. The code
is publicly available at https://github.com/ColinLu50/Evade-GPT-Detector. | cs.CL | Large Language Models can be Guided to Evade AI-Generated Text Detection |
2023-05-18T09:58:19Z | http://arxiv.org/pdf/2305.10845v1 | http://arxiv.org/abs/2305.10845v1 | Patrick Kahardipraja, Brielen Madureira, David Schlangen | 2023-05-18T09:58:19Z | cs.CL | Language is by its very nature incremental in how it is produced and
processed. This property can be exploited by NLP systems to produce fast
responses, which has been shown to be beneficial for real-time interactive
applications. Recent neural network-based approaches for incremental processing
mainly use RNNs or Transformers. RNNs are fast but monotonic (cannot correct
earlier output, which can be necessary in incremental processing).
Transformers, on the other hand, consume whole sequences, and hence are by
nature non-incremental. A restart-incremental interface that repeatedly passes
longer input prefixes can be used to obtain partial outputs, while providing
the ability to revise. However, this method becomes costly as the sentence
grows longer. In this work, we propose the Two-pass model for AdaPtIve Revision
(TAPIR) and introduce a method to obtain an incremental supervision signal for
learning an adaptive revision policy. Experimental results on sequence
labelling show that our model has better incremental performance and faster
inference speed compared to restart-incremental Transformers, while showing
little degradation on full sequences. | cs.CL | TAPIR: Learning Adaptive Revision for Incremental Natural Language Understanding with a Two-Pass Model |
2023-05-18T09:50:47Z | http://arxiv.org/pdf/2305.10839v1 | http://arxiv.org/abs/2305.10839v1 | Chong-En Lin, Kuan-Yu Chen | 2023-05-18T09:50:47Z | cs.CL, cs.SD, eess.AS | Non-autoregressive automatic speech recognition (ASR) has become a mainstream
of ASR modeling because of its fast decoding speed and satisfactory result. To
further boost the performance, relaxing the conditional independence assumption
and cascading large-scaled pre-trained models are two active research
directions. In addition to these strategies, we propose a lexical-aware
non-autoregressive Transformer-based (LA-NAT) ASR framework, which consists of
an acoustic encoder, a speech-text shared encoder, and a speech-text shared
decoder. The acoustic encoder is used to process the input speech features as
usual, and the speech-text shared encoder and decoder are designed to train
speech and text data simultaneously. By doing so, LA-NAT aims to make the ASR
model aware of lexical information, so the resulting model is expected to
achieve better results by leveraging the learned linguistic knowledge. A series
of experiments are conducted on the AISHELL-1, CSJ, and TEDLIUM 2 datasets.
According to the experiments, the proposed LA-NAT can provide superior results
than other recently proposed non-autoregressive ASR models. In addition, LA-NAT
is a relatively compact model than most non-autoregressive ASR models, and it
is about 58 times faster than the classic autoregressive model. | cs.CL | A Lexical-aware Non-autoregressive Transformer-based ASR Model |
2023-05-18T09:24:53Z | http://arxiv.org/pdf/2305.10835v1 | http://arxiv.org/abs/2305.10835v1 | Daniil Gavrilov, Nikita Balagansky | 2023-05-18T09:24:53Z | cs.LG, cs.CL | In this paper, we propose Ahead-of-Time (AoT) P-Tuning, a novel
parameter-efficient fine-tuning method for pre-trained Language Models (LMs)
that adds input-dependent bias before each Transformer layer. We evaluate AoT
P-Tuning on GLUE and SuperGLUE benchmarking datasets using RoBERTa and DeBERTa
models, showing that it outperforms BitFit and is comparable or better than
other baseline methods for efficient fine-tuning. Additionally, we assess the
inference overhead of AoT P-Tuning and demonstrate that it introduces
negligible overhead compared to established baseline methods. Our method
enables multi-task inference with a single backbone LM, making it a practical
solution for real-world applications. | cs.LG | Ahead-of-Time P-Tuning |
2023-05-18T09:23:05Z | http://arxiv.org/pdf/2305.10834v1 | http://arxiv.org/abs/2305.10834v1 | Scott Rettberg, Talan Memmott, Jill Walker Rettberg, Jason Nelson, Patrick Lichty | 2023-05-18T09:23:05Z | cs.AI, cs.CL, cs.HC, cs.MM, J.5 | During 2022, both transformer-based AI text generation sys-tems such as GPT-3
and AI text-to-image generation systems such as DALL-E 2 and Stable Diffusion
made exponential leaps forward and are unquestionably altering the fields of
digital art and electronic literature. In this panel a group of electronic
literature authors and theorists consider new oppor-tunities for human
creativity presented by these systems and present new works have produced
during the past year that specifically address these systems as environments
for literary expressions that are translated through iterative interlocutive
processes into visual representations. The premise that binds these
presentations is that these systems and the works gener-ated must be considered
from a literary perspective, as they originate in human writing. In works
ranging from a visual memoir of the personal experience of a health crisis, to
interac-tive web comics, to architectures based on abstract poetic language, to
political satire, four artists explore the capabili-ties of these writing
environments for new genres of literary artist practice, while a digital
culture theorist considers the origins and effects of the particular training
datasets of human language and images on which these new hybrid forms are
based. | cs.AI | AIwriting: Relations Between Image Generation and Digital Writing |
2023-05-18T09:22:29Z | http://arxiv.org/pdf/2305.10833v3 | http://arxiv.org/abs/2305.10833v3 | Amal Haddad Haddad, Damith Premasiri, Tharindu Ranasinghe, Ruslan Mitkov | 2023-06-01T13:59:23Z | cs.CL | The domain of Botany is rich with metaphorical terms. Those terms play an
important role in the description and identification of flowers and plants.
However, the identification of such terms in discourse is an arduous task. This
leads in some cases to committing errors during translation processes and
lexicographic tasks. The process is even more challenging when it comes to
machine translation, both in the cases of single-word terms and multi-word
terms. One of the recent concerns of Natural Language Processing (NLP)
applications and Machine Translation (MT) technologies is the automatic
identification of metaphor-based words in discourse through Deep Learning (DL).
In this study, we seek to fill this gap through the use of thirteen popular
transformer based models, as well as ChatGPT, and we show that discriminative
models perform better than GPT-3.5 model with our best performer reporting
92.2349% F1 score in metaphoric flower and plant names identification task. | cs.CL | Deep Learning Methods for Extracting Metaphorical Names of Flowers and Plants |
2023-05-18T08:57:17Z | http://arxiv.org/pdf/2305.10819v2 | http://arxiv.org/abs/2305.10819v2 | Jingheng Ye, Yinghui Li, Qingyu Zhou, Yangning Li, Shirong Ma, Hai-Tao Zheng, Ying Shen | 2023-10-17T04:56:57Z | cs.CL | Evaluating the performance of Grammatical Error Correction (GEC) systems is a
challenging task due to its subjectivity. Designing an evaluation metric that
is as objective as possible is crucial to the development of GEC task. However,
mainstream evaluation metrics, i.e., reference-based metrics, introduce bias
into the multi-reference evaluation by extracting edits without considering the
presence of multiple references. To overcome this issue, we propose Chunk-LEvel
Multi-reference Evaluation (CLEME), designed to evaluate GEC systems in the
multi-reference evaluation setting. CLEME builds chunk sequences with
consistent boundaries for the source, the hypothesis and references, thus
eliminating the bias caused by inconsistent edit boundaries. Furthermore, we
observe the consistent boundary could also act as the boundary of grammatical
errors, based on which the F$_{0.5}$ score is then computed following the
correction independence assumption. We conduct experiments on six English
reference sets based on the CoNLL-2014 shared task. Extensive experiments and
detailed analyses demonstrate the correctness of our discovery and the
effectiveness of CLEME. Further analysis reveals that CLEME is robust to
evaluate GEC systems across reference sets with varying numbers of references
and annotation style. | cs.CL | CLEME: Debiasing Multi-reference Evaluation for Grammatical Error Correction |
2023-05-18T08:56:05Z | http://arxiv.org/pdf/2305.10818v4 | http://arxiv.org/abs/2305.10818v4 | Sofia Maria Lo Cicero Vaina, Nikita Balagansky, Daniil Gavrilov | 2024-02-12T09:34:39Z | cs.LG, cs.CL | Diffusion Language models (DLMs) are a promising avenue for text generation
due to their practical properties on tractable controllable generation. They
also have the advantage of not having to predict text autoregressively.
However, despite these notable features, DLMs have not yet reached the
performance levels of their autoregressive counterparts. One of the ways to
reduce the performance gap between these two types of language models is to
speed up the generation of DLMs. Therefore, we propose a novel methodology to
address this issue in this work. It enables the execution of more generation
steps within a given time frame, leading to higher-quality outputs.
Specifically, our methods estimate DLMs completeness of text generation and
allow adaptive halting of the generation process. We evaluate our methods on
Plaid, SSD, and CDCD DLMs and create a cohesive perspective on their generation
workflows. Finally, we confirm that our methods allow halting these models and
decrease the generation time by $10$-$40$\% without a drop in the quality of
model samples. | cs.LG | Diffusion Language Models Generation Can Be Halted Early |
2023-05-18T08:00:09Z | http://arxiv.org/pdf/2305.10788v1 | http://arxiv.org/abs/2305.10788v1 | Hang Shao, Wei Wang, Bei Liu, Xun Gong, Haoyu Wang, Yanmin Qian | 2023-05-18T08:00:09Z | cs.SD, cs.CL, eess.AS | Due to the rapid development of computing hardware resources and the dramatic
growth of data, pre-trained models in speech recognition, such as Whisper, have
significantly improved the performance of speech recognition tasks. However,
these models usually have a high computational overhead, making it difficult to
execute effectively on resource-constrained devices. To speed up inference and
reduce model size while maintaining performance, we propose a novel guided
knowledge distillation and quantization for large pre-trained model Whisper.
The student model selects distillation and quantization layers based on
quantization loss and distillation loss, respectively. We compressed
$\text{Whisper}_\text{small}$ to $\text{Whisper}_\text{base}$ and
$\text{Whisper}_\text{tiny}$ levels, making $\text{Whisper}_\text{small}$
5.18x/10.48x smaller, respectively. Moreover, compared to the original
$\text{Whisper}_\text{base}$ and $\text{Whisper}_\text{tiny}$, there is also a
relative character error rate~(CER) reduction of 11.3% and 14.0% for the new
compressed model respectively. | cs.SD | Whisper-KDQ: A Lightweight Whisper via Guided Knowledge Distillation and Quantization for Efficient ASR |
2023-05-18T07:56:40Z | http://arxiv.org/pdf/2305.10786v2 | http://arxiv.org/abs/2305.10786v2 | Qian Chen, Wen Wang, Qinglin Zhang, Siqi Zheng, Chong Deng, Hai Yu, Jiaqing Liu, Yukun Ma, Chong Zhang | 2023-10-23T06:34:50Z | cs.CL | Prior studies diagnose the anisotropy problem in sentence representations
from pre-trained language models, e.g., BERT, without fine-tuning. Our analysis
reveals that the sentence embeddings from BERT suffer from a bias towards
uninformative words, limiting the performance in semantic textual similarity
(STS) tasks. To address this bias, we propose a simple and efficient
unsupervised approach, Diagonal Attention Pooling (Ditto), which weights words
with model-based importance estimations and computes the weighted average of
word representations from pre-trained models as sentence embeddings. Ditto can
be easily applied to any pre-trained language model as a postprocessing
operation. Compared to prior sentence embedding approaches, Ditto does not add
parameters nor requires any learning. Empirical evaluations demonstrate that
our proposed Ditto can alleviate the anisotropy problem and improve various
pre-trained models on STS tasks. | cs.CL | Ditto: A Simple and Efficient Approach to Improve Sentence Embeddings |
2023-05-18T06:15:45Z | http://arxiv.org/pdf/2305.10736v1 | http://arxiv.org/abs/2305.10736v1 | Chenhe Dong, Yuexiang Xie, Yaliang Li, Ying Shen | 2023-05-18T06:15:45Z | cs.CL, cs.AI | Despite substantial progress in abstractive text summarization to generate
fluent and informative texts, the factual inconsistency in the generated
summaries remains an important yet challenging problem to be solved. In this
paper, we construct causal graphs for abstractive text summarization and
identify the intrinsic causes of the factual inconsistency, i.e., the language
bias and irrelevancy bias, and further propose a debiasing framework, named
CoFactSum, to alleviate the causal effects of these biases by counterfactual
estimation. Specifically, the proposed CoFactSum provides two counterfactual
estimation strategies, i.e., Explicit Counterfactual Masking with an explicit
dynamic masking strategy, and Implicit Counterfactual Training with an implicit
discriminative cross-attention mechanism. Meanwhile, we design a Debiasing
Degree Adjustment mechanism to dynamically adapt the debiasing degree at each
decoding step. Extensive experiments on two widely-used summarization datasets
demonstrate the effectiveness of CoFactSum in enhancing the factual consistency
of generated summaries compared with several baselines. | cs.CL | Counterfactual Debiasing for Generating Factually Consistent Text Summaries |
2023-05-18T06:10:49Z | http://arxiv.org/pdf/2305.10734v2 | http://arxiv.org/abs/2305.10734v2 | Hao Shi, Kazuki Shimada, Masato Hirano, Takashi Shibuya, Yuichiro Koyama, Zhi Zhong, Shusuke Takahashi, Tatsuya Kawahara, Yuki Mitsufuji | 2024-02-28T12:10:19Z | cs.SD, cs.CL, eess.AS | Diffusion-based generative speech enhancement (SE) has recently received
attention, but reverse diffusion remains time-consuming. One solution is to
initialize the reverse diffusion process with enhanced features estimated by a
predictive SE system. However, the pipeline structure currently does not
consider for a combined use of generative and predictive decoders. The
predictive decoder allows us to use the further complementarity between
predictive and diffusion-based generative SE. In this paper, we propose a
unified system that use jointly generative and predictive decoders across two
levels. The encoder encodes both generative and predictive information at the
shared encoding level. At the decoded feature level, we fuse the two decoded
features by generative and predictive decoders. Specifically, the two SE
modules are fused in the initial and final diffusion steps: the initial fusion
initializes the diffusion process with the predictive SE to improve
convergence, and the final fusion combines the two complementary SE outputs to
enhance SE performance. Experiments conducted on the Voice-Bank dataset
demonstrate that incorporating predictive information leads to faster decoding
and higher PESQ scores compared with other score-based diffusion SE (StoRM and
SGMSE+). | cs.SD | Diffusion-Based Speech Enhancement with Joint Generative and Predictive Decoders |
2023-05-18T05:58:27Z | http://arxiv.org/pdf/2305.10731v2 | http://arxiv.org/abs/2305.10731v2 | Jihyung Moon, Dong-Ho Lee, Hyundong Cho, Woojeong Jin, Chan Young Park, Minwoo Kim, Jonathan May, Jay Pujara, Sungjoon Park | 2023-10-08T03:57:31Z | cs.CL | Toxic language, such as hate speech, can deter users from participating in
online communities and enjoying popular platforms. Previous approaches to
detecting toxic language and norm violations have been primarily concerned with
conversations from online forums and social media, such as Reddit and Twitter.
These approaches are less effective when applied to conversations on
live-streaming platforms, such as Twitch and YouTube Live, as each comment is
only visible for a limited time and lacks a thread structure that establishes
its relationship with other comments. In this work, we share the first NLP
study dedicated to detecting norm violations in conversations on live-streaming
platforms. We define norm violation categories in live-stream chats and
annotate 4,583 moderated comments from Twitch. We articulate several facets of
live-stream data that differ from other forums, and demonstrate that existing
models perform poorly in this setting. By conducting a user study, we identify
the informational context humans use in live-stream moderation, and train
models leveraging context to identify norm violations. Our results show that
appropriate contextual information can boost moderation performance by 35\%. | cs.CL | Analyzing Norm Violations in Live-Stream Chat |
2023-05-18T05:17:57Z | http://arxiv.org/pdf/2305.10713v2 | http://arxiv.org/abs/2305.10713v2 | Lingfeng Shen, Weiting Tan, Boyuan Zheng, Daniel Khashabi | 2023-10-23T01:22:44Z | cs.CL, cs.LG | With growing capabilities of large language models, prompting them has become
the dominant way to access them. This has motivated the development of
strategies for automatically selecting effective language prompts. In this
paper, we introduce prompt flatness, a new metric to quantify the expected
utility of a language prompt. This metric is inspired by flatness
regularization in statistical learning that quantifies the robustness of the
model towards its parameter perturbations. We provide theoretical foundations
for this metric and its relationship with other prompt selection metrics,
providing a comprehensive understanding of existing methods. Empirically, we
show that combining prompt flatness with existing metrics improves both
performance and sample efficiency. Our metric outperforms the previous prompt
selection metrics with an average increase of 5% in accuracy and 10% in Pearson
correlation across 6 classification benchmarks. | cs.CL | Flatness-Aware Prompt Selection Improves Accuracy and Sample Efficiency |
2023-05-18T05:01:04Z | http://arxiv.org/pdf/2305.10709v1 | http://arxiv.org/abs/2305.10709v1 | Tingting Wu, Xiao Ding, Minji Tang, Hao Zhang, Bing Qin, Ting Liu | 2023-05-18T05:01:04Z | cs.CL, cs.AI | Large-scale datasets in the real world inevitably involve label noise. Deep
models can gradually overfit noisy labels and thus degrade model
generalization. To mitigate the effects of label noise, learning with noisy
labels (LNL) methods are designed to achieve better generalization performance.
Due to the lack of suitable datasets, previous studies have frequently employed
synthetic label noise to mimic real-world label noise. However, synthetic noise
is not instance-dependent, making this approximation not always effective in
practice. Recent research has proposed benchmarks for learning with real-world
noisy labels. However, the noise sources within may be single or fuzzy, making
benchmarks different from data with heterogeneous label noises in the real
world. To tackle these issues, we contribute NoisywikiHow, the largest NLP
benchmark built with minimal supervision. Specifically, inspired by human
cognition, we explicitly construct multiple sources of label noise to imitate
human errors throughout the annotation, replicating real-world noise, whose
corruption is affected by both ground-truth labels and instances. Moreover, we
provide a variety of noise levels to support controlled experiments on noisy
data, enabling us to evaluate LNL methods systematically and comprehensively.
After that, we conduct extensive multi-dimensional experiments on a broad range
of LNL methods, obtaining new and intriguing findings. | cs.CL | NoisywikiHow: A Benchmark for Learning with Real-world Noisy Labels in Natural Language Processing |
2023-05-18T04:30:09Z | http://arxiv.org/pdf/2305.10703v1 | http://arxiv.org/abs/2305.10703v1 | Yue Yu, Yuchen Zhuang, Rongzhi Zhang, Yu Meng, Jiaming Shen, Chao Zhang | 2023-05-18T04:30:09Z | cs.CL, cs.IR, cs.LG | With the development of large language models (LLMs), zero-shot learning has
attracted much attention for various NLP tasks. Different from prior works that
generate training data with billion-scale natural language generation (NLG)
models, we propose a retrieval-enhanced framework to create training data from
a general-domain unlabeled corpus. To realize this, we first conduct
contrastive pretraining to learn an unsupervised dense retriever for extracting
the most relevant documents using class-descriptive verbalizers. We then
further propose two simple strategies, namely Verbalizer Augmentation with
Demonstrations and Self-consistency Guided Filtering to improve the topic
coverage of the dataset while removing noisy examples. Experiments on nine
datasets demonstrate that REGEN achieves 4.3% gain over the strongest baselines
and saves around 70% of the time compared to baselines using large NLG models.
Besides, REGEN can be naturally integrated with recently proposed large
language models to boost performance. | cs.CL | ReGen: Zero-Shot Text Classification via Training Data Generation with Progressive Dense Retrieval |
2023-05-18T03:58:19Z | http://arxiv.org/pdf/2305.10688v2 | http://arxiv.org/abs/2305.10688v2 | Zequn Liu, Wei Zhang, Yingce Xia, Lijun Wu, Shufang Xie, Tao Qin, Ming Zhang, Tie-Yan Liu | 2023-05-26T04:35:46Z | cs.CL | Generative pre-trained Transformer (GPT) has demonstrates its great success
in natural language processing and related techniques have been adapted into
molecular modeling. Considering that text is the most important record for
scientific discovery, in this paper, we propose MolXPT, a unified language
model of text and molecules pre-trained on SMILES (a sequence representation of
molecules) wrapped by text. Briefly, we detect the molecule names in each
sequence and replace them to the corresponding SMILES. In this way, the SMILES
could leverage the information from surrounding text, and vice versa. The above
wrapped sequences, text sequences from PubMed and SMILES sequences from PubChem
are all fed into a language model for pre-training. Experimental results
demonstrate that MolXPT outperforms strong baselines of molecular property
prediction on MoleculeNet, performs comparably to the best model in
text-molecule translation while using less than half of its parameters, and
enables zero-shot molecular generation without finetuning. | cs.CL | MolXPT: Wrapping Molecules with Text for Generative Pre-training |
2023-05-18T03:57:51Z | http://arxiv.org/pdf/2305.10686v1 | http://arxiv.org/abs/2305.10686v1 | Jinzheng He, Jinglin Liu, Zhenhui Ye, Rongjie Huang, Chenye Cui, Huadai Liu, Zhou Zhao | 2023-05-18T03:57:51Z | cs.SD, cs.CL, eess.AS | We are interested in a challenging task, Realistic-Music-Score based Singing
Voice Synthesis (RMS-SVS). RMS-SVS aims to generate high-quality singing voices
given realistic music scores with different note types (grace, slur, rest,
etc.). Though significant progress has been achieved, recent singing voice
synthesis (SVS) methods are limited to fine-grained music scores, which require
a complicated data collection pipeline with time-consuming manual annotation to
align music notes with phonemes. Furthermore, these manual annotation destroys
the regularity of note durations in music scores, making fine-grained music
scores inconvenient for composing. To tackle these challenges, we propose
RMSSinger, the first RMS-SVS method, which takes realistic music scores as
input, eliminating most of the tedious manual annotation and avoiding the
aforementioned inconvenience. Note that music scores are based on words rather
than phonemes, in RMSSinger, we introduce word-level modeling to avoid the
time-consuming phoneme duration annotation and the complicated phoneme-level
mel-note alignment. Furthermore, we propose the first diffusion-based pitch
modeling method, which ameliorates the naturalness of existing pitch-modeling
methods. To achieve these, we collect a new dataset containing realistic music
scores and singing voices according to these realistic music scores from
professional singers. Extensive experiments on the dataset demonstrate the
effectiveness of our methods. Audio samples are available at
https://rmssinger.github.io/. | cs.SD | RMSSinger: Realistic-Music-Score based Singing Voice Synthesis |
2023-05-18T03:53:59Z | http://arxiv.org/pdf/2305.10683v4 | http://arxiv.org/abs/2305.10683v4 | Zhenhailong Wang, Ansel Blume, Sha Li, Genglin Liu, Jaemin Cho, Zineng Tang, Mohit Bansal, Heng Ji | 2023-10-21T16:34:03Z | cs.CV, cs.CL | Action knowledge involves the understanding of textual, visual, and temporal
aspects of actions. We introduce the Action Dynamics Benchmark (ActionBench)
containing two carefully designed probing tasks: Action Antonym and Video
Reversal, which targets multimodal alignment capabilities and temporal
understanding skills of the model, respectively. Despite recent video-language
models' (VidLM) impressive performance on various benchmark tasks, our
diagnostic tasks reveal their surprising deficiency (near-random performance)
in action knowledge, suggesting that current models rely on object recognition
abilities as a shortcut for action understanding. To remedy this, we propose a
novel framework, Paxion, along with a new Discriminative Video Dynamics
Modeling (DVDM) objective. The Paxion framework utilizes a Knowledge Patcher
network to encode new action knowledge and a Knowledge Fuser component to
integrate the Patcher into frozen VidLMs without compromising their existing
capabilities. Due to limitations of the widely-used Video-Text Contrastive
(VTC) loss for learning action knowledge, we introduce the DVDM objective to
train the Knowledge Patcher. DVDM forces the model to encode the correlation
between the action text and the correct ordering of video frames. Our extensive
analyses show that Paxion and DVDM together effectively fill the gap in action
knowledge understanding (~50% to 80%), while maintaining or improving
performance on a wide spectrum of both object- and action-centric downstream
tasks. The code and data will be made publicly available for research purposes
at https://github.com/MikeWangWZHL/Paxion.git. | cs.CV | Paxion: Patching Action Knowledge in Video-Language Foundation Models |
2023-05-18T03:34:50Z | http://arxiv.org/pdf/2305.10680v2 | http://arxiv.org/abs/2305.10680v2 | Xian Shi, Haoneng Luo, Zhifu Gao, Shiliang Zhang, Zhijie Yan | 2023-05-25T02:26:35Z | cs.SD, cs.CL, eess.AS | Estimating confidence scores for recognition results is a classic task in ASR
field and of vital importance for kinds of downstream tasks and training
strategies. Previous end-to-end~(E2E) based confidence estimation models (CEM)
predict score sequences of equal length with input transcriptions, leading to
unreliable estimation when deletion and insertion errors occur. In this paper
we proposed CIF-Aligned confidence estimation model (CA-CEM) to achieve
accurate and reliable confidence estimation based on novel non-autoregressive
E2E ASR model - Paraformer. CA-CEM utilizes the modeling character of
continuous integrate-and-fire (CIF) mechanism to generate token-synchronous
acoustic embedding, which solves the estimation failure issue above. We measure
the quality of estimation with AUC and RMSE in token level and ECE-U - a
proposed metrics in utterance level. CA-CEM gains 24% and 19% relative
reduction on ECE-U and also better AUC and RMSE on two test sets. Furthermore,
we conduct analysis to explore the potential of CEM for different ASR related
usage. | cs.SD | Accurate and Reliable Confidence Estimation Based on Non-Autoregressive End-to-End Speech Recognition System |
2023-05-18T03:32:54Z | http://arxiv.org/pdf/2305.10679v1 | http://arxiv.org/abs/2305.10679v1 | Xin-Ye Li, Jiang-Tian Xue, Zheng Xie, Ming Li | 2023-05-18T03:32:54Z | cs.AI, cs.CL, cs.SE | Code generation aims to automatically generate source code from high-level
task specifications, which can significantly increase productivity of software
engineering. Recently, approaches based on large language models (LLMs) have
shown remarkable code generation abilities on simple tasks. However, generate
code for more complex tasks, such as competition-level problems, remains
challenging. In this paper, we introduce Brainstorm framework for code
generation. It leverages a brainstorming step that generates and selects
diverse thoughts on the problem to facilitate algorithmic reasoning, where the
thoughts are possible blueprint of solving the problem. We demonstrate that
Brainstorm significantly enhances the ability of LLMs to solve
competition-level programming problems, resulting in a more than 50% increase
in the pass@$k$ metrics for ChatGPT on the CodeContests benchmark, achieving
state-of-the-art performance. Furthermore, our experiments conducted on
LeetCode contests show that our framework boosts the ability of ChatGPT to a
level comparable to that of human programmers. | cs.AI | Think Outside the Code: Brainstorming Boosts Large Language Models in Code Generation |
2023-05-18T02:57:54Z | http://arxiv.org/pdf/2305.10666v3 | http://arxiv.org/abs/2305.10666v3 | Zelin Ying, Chen Li, Yu Dong, Qiuqiang Kong, Qiao Tian, Yuanyuan Huo, Yuxuan Wang | 2024-03-25T10:59:04Z | cs.CL, cs.AI, cs.SD, eess.AS | The front-end is a critical component of English text-to-speech (TTS)
systems, responsible for extracting linguistic features that are essential for
a text-to-speech model to synthesize speech, such as prosodies and phonemes.
The English TTS front-end typically consists of a text normalization (TN)
module, a prosody word prosody phrase (PWPP) module, and a grapheme-to-phoneme
(G2P) module. However, current research on the English TTS front-end focuses
solely on individual modules, neglecting the interdependence between them and
resulting in sub-optimal performance for each module. Therefore, this paper
proposes a unified front-end framework that captures the dependencies among the
English TTS front-end modules. Extensive experiments have demonstrated that the
proposed method achieves state-of-the-art (SOTA) performance in all modules. | cs.CL | A unified front-end framework for English text-to-speech synthesis |
2023-05-18T02:19:05Z | http://arxiv.org/pdf/2305.10652v3 | http://arxiv.org/abs/2305.10652v3 | Peter Ochieng | 2023-09-12T06:47:52Z | cs.SD, cs.CL, eess.AS | The current monaural state of the art tools for speech separation relies on
supervised learning. This means that they must deal with permutation problem,
they are impacted by the mismatch on the number of speakers used in training
and inference. Moreover, their performance heavily relies on the presence of
high-quality labelled data. These problems can be effectively addressed by
employing a fully unsupervised technique for speech separation. In this paper,
we use contrastive learning to establish the representations of frames then use
the learned representations in the downstream deep modularization task.
Concretely, we demonstrate experimentally that in speech separation, different
frames of a speaker can be viewed as augmentations of a given hidden standard
frame of that speaker. The frames of a speaker contain enough prosodic
information overlap which is key in speech separation. Based on this, we
implement a self-supervised learning to learn to minimize the distance between
frames belonging to a given speaker. The learned representations are used in a
downstream deep modularization task to cluster frames based on speaker
identity. Evaluation of the developed technique on WSJ0-2mix and WSJ0-3mix
shows that the technique attains SI-SNRi and SDRi of 20.8 and 21.0 respectively
in WSJ0-2mix. In WSJ0-3mix, it attains SI-SNRi and SDRi of 20.7 and 20.7
respectively in WSJ0-2mix. Its greatest strength being that as the number of
speakers increase, its performance does not degrade significantly. | cs.SD | Speech Separation based on Contrastive Learning and Deep Modularization |
2023-05-18T02:08:33Z | http://arxiv.org/pdf/2305.10649v1 | http://arxiv.org/abs/2305.10649v1 | Xingchen Song, Di Wu, Binbin Zhang, Zhendong Peng, Bo Dang, Fuping Pan, Zhiyong Wu | 2023-05-18T02:08:33Z | cs.SD, cs.CL, eess.AS, I.2.7 | In this paper, we present ZeroPrompt (Figure 1-(a)) and the corresponding
Prompt-and-Refine strategy (Figure 3), two simple but effective
\textbf{training-free} methods to decrease the Token Display Time (TDT) of
streaming ASR models \textbf{without any accuracy loss}. The core idea of
ZeroPrompt is to append zeroed content to each chunk during inference, which
acts like a prompt to encourage the model to predict future tokens even before
they were spoken. We argue that streaming acoustic encoders naturally have the
modeling ability of Masked Language Models and our experiments demonstrate that
ZeroPrompt is engineering cheap and can be applied to streaming acoustic
encoders on any dataset without any accuracy loss. Specifically, compared with
our baseline models, we achieve 350 $\sim$ 700ms reduction on First Token
Display Time (TDT-F) and 100 $\sim$ 400ms reduction on Last Token Display Time
(TDT-L), with theoretically and experimentally equal WER on both Aishell-1 and
Librispeech datasets. | cs.SD | ZeroPrompt: Streaming Acoustic Encoders are Zero-Shot Masked LMs |
2023-05-18T02:04:38Z | http://arxiv.org/pdf/2305.10647v1 | http://arxiv.org/abs/2305.10647v1 | Sreyan Ghosh, Utkarsh Tyagi, Sonal Kumar, Dinesh Manocha | 2023-05-18T02:04:38Z | cs.CL, cs.AI, cs.IR | Biomedical Named Entity Recognition (BioNER) is the fundamental task of
identifying named entities from biomedical text. However, BioNER suffers from
severe data scarcity and lacks high-quality labeled data due to the highly
specialized and expert knowledge required for annotation. Though data
augmentation has shown to be highly effective for low-resource NER in general,
existing data augmentation techniques fail to produce factual and diverse
augmentations for BioNER. In this paper, we present BioAug, a novel data
augmentation framework for low-resource BioNER. BioAug, built on BART, is
trained to solve a novel text reconstruction task based on selective masking
and knowledge augmentation. Post training, we perform conditional generation
and generate diverse augmentations conditioning BioAug on selectively corrupted
text similar to the training stage. We demonstrate the effectiveness of BioAug
on 5 benchmark BioNER datasets and show that BioAug outperforms all our
baselines by a significant margin (1.5%-21.5% absolute improvement) and is able
to generate augmentations that are both more factual and diverse. Code:
https://github.com/Sreyan88/BioAug. | cs.CL | BioAug: Conditional Generation based Data Augmentation for Low-Resource Biomedical NER |
2023-05-18T02:03:55Z | http://arxiv.org/pdf/2305.10645v2 | http://arxiv.org/abs/2305.10645v2 | Peter Ochieng | 2023-05-19T10:47:56Z | cs.CL | This paper looks at the ability of large language models to participate in
educational guided reading. We specifically, evaluate their ability to generate
meaningful questions from the input text, generate diverse questions both in
terms of content coverage and difficulty of the questions and evaluate their
ability to recommend part of the text that a student should re-read based on
the student's responses to the questions. Based on our evaluation of ChatGPT
and Bard, we report that,
1) Large language models are able to generate high quality meaningful
questions that have high correlation with the input text, 2) They generate
diverse question that cover most topics in the input text even though this
ability is significantly degraded as the input text increases, 3)The large
language models are able to generate both low and high cognitive questions even
though they are significantly biased toward low cognitive question, 4) They are
able to effectively summarize responses and extract a portion of text that
should be re-read. | cs.CL | Are Large Language Models Fit For Guided Reading? |
2023-05-18T00:35:38Z | http://arxiv.org/pdf/2305.10626v3 | http://arxiv.org/abs/2305.10626v3 | Jiannan Xiang, Tianhua Tao, Yi Gu, Tianmin Shu, Zirui Wang, Zichao Yang, Zhiting Hu | 2023-10-28T17:55:44Z | cs.CL, cs.AI, cs.LG | While large language models (LMs) have shown remarkable capabilities across
numerous tasks, they often struggle with simple reasoning and planning in
physical environments, such as understanding object permanence or planning
household activities. The limitation arises from the fact that LMs are trained
only on written text and miss essential embodied knowledge and skills. In this
paper, we propose a new paradigm of enhancing LMs by finetuning them with world
models, to gain diverse embodied knowledge while retaining their general
language capabilities. Our approach deploys an embodied agent in a world model,
particularly a simulator of the physical world (VirtualHome), and acquires a
diverse set of embodied experiences through both goal-oriented planning and
random exploration. These experiences are then used to finetune LMs to teach
diverse abilities of reasoning and acting in the physical world, e.g., planning
and completing goals, object permanence and tracking, etc. Moreover, it is
desirable to preserve the generality of LMs during finetuning, which
facilitates generalizing the embodied knowledge across tasks rather than being
tied to specific simulations. We thus further introduce the classical (EWC) for
selective weight updates, combined with low-rank adapters (LoRA) for training
efficiency. Extensive experiments show our approach substantially improves base
LMs on 18 downstream tasks by 64.28% on average. In particular, the small LMs
(1.3B, 6B, and 13B) enhanced by our approach match or even outperform much
larger LMs (e.g., ChatGPT). | cs.CL | Language Models Meet World Models: Embodied Experiences Enhance Language Models |
2023-05-18T00:01:27Z | http://arxiv.org/pdf/2305.10615v2 | http://arxiv.org/abs/2305.10615v2 | Jiatong Shi, Dan Berrebbi, William Chen, Ho-Lam Chung, En-Pei Hu, Wei Ping Huang, Xuankai Chang, Shang-Wen Li, Abdelrahman Mohamed, Hung-yi Lee, Shinji Watanabe | 2023-08-11T17:39:21Z | cs.SD, cs.CL, eess.AS | Speech processing Universal PERformance Benchmark (SUPERB) is a leaderboard
to benchmark the performance of Self-Supervised Learning (SSL) models on
various speech processing tasks. However, SUPERB largely considers English
speech in its evaluation. This paper presents multilingual SUPERB (ML-SUPERB),
covering 143 languages (ranging from high-resource to endangered), and
considering both automatic speech recognition and language identification.
Following the concept of SUPERB, ML-SUPERB utilizes frozen SSL features and
employs a simple framework for multilingual tasks by learning a shallow
downstream model. Similar to the SUPERB benchmark, we find speech SSL models
can significantly improve performance compared to FBANK features. Furthermore,
we find that multilingual models do not always perform better than their
monolingual counterparts. We will release ML-SUPERB as a challenge with
organized datasets and reproducible training scripts for future multilingual
representation research. | cs.SD | ML-SUPERB: Multilingual Speech Universal PERformance Benchmark |
2023-05-17T23:55:32Z | http://arxiv.org/pdf/2305.10614v2 | http://arxiv.org/abs/2305.10614v2 | Byung-Doh Oh, William Schuler | 2023-06-02T22:50:07Z | cs.CL, cs.AI | While there is much recent interest in studying why Transformer-based large
language models make predictions the way they do, the complex computations
performed within each layer have made their behavior somewhat opaque. To
mitigate this opacity, this work presents a linear decomposition of final
hidden states from autoregressive language models based on each initial input
token, which is exact for virtually all contemporary Transformer architectures.
This decomposition allows the definition of probability distributions that
ablate the contribution of specific input tokens, which can be used to analyze
their influence on model probabilities over a sequence of upcoming words with
only one forward pass from the model. Using the change in next-word probability
as a measure of importance, this work first examines which context words make
the biggest contribution to language model predictions. Regression experiments
suggest that Transformer-based language models rely primarily on collocational
associations, followed by linguistic factors such as syntactic dependencies and
coreference relationships in making next-word predictions. Additionally,
analyses using these measures to predict syntactic dependencies and coreferent
mention spans show that collocational association and repetitions of the same
token largely explain the language models' predictions on these tasks. | cs.CL | Token-wise Decomposition of Autoregressive Language Model Hidden States for Analyzing Model Predictions |
2023-05-17T23:50:28Z | http://arxiv.org/pdf/2305.10613v3 | http://arxiv.org/abs/2305.10613v3 | Dong-Ho Lee, Kian Ahrabian, Woojeong Jin, Fred Morstatter, Jay Pujara | 2023-10-20T04:46:44Z | cs.CL | Temporal knowledge graph (TKG) forecasting benchmarks challenge models to
predict future facts using knowledge of past facts. In this paper, we apply
large language models (LLMs) to these benchmarks using in-context learning
(ICL). We investigate whether and to what extent LLMs can be used for TKG
forecasting, especially without any fine-tuning or explicit modules for
capturing structural and temporal information. For our experiments, we present
a framework that converts relevant historical facts into prompts and generates
ranked predictions using token probabilities. Surprisingly, we observe that
LLMs, out-of-the-box, perform on par with state-of-the-art TKG models carefully
designed and trained for TKG forecasting. Our extensive evaluation presents
performances across several models and datasets with different characteristics,
compares alternative heuristics for preparing contextual information, and
contrasts to prominent TKG methods and simple frequency and recency baselines.
We also discover that using numerical indices instead of entity/relation names,
i.e., hiding semantic information, does not significantly affect the
performance ($\pm$0.4\% Hit@1). This shows that prior semantic knowledge is
unnecessary; instead, LLMs can leverage the existing patterns in the context to
achieve such performance. Our analysis also reveals that ICL enables LLMs to
learn irregular patterns from the historical context, going beyond simple
predictions based on common or recent information. | cs.CL | Temporal Knowledge Graph Forecasting Without Knowledge Using In-Context Learning |
2023-05-17T23:41:30Z | http://arxiv.org/pdf/2305.10610v1 | http://arxiv.org/abs/2305.10610v1 | Saeth Wannasuphoprasit, Yi Zhou, Danushka Bollegala | 2023-05-17T23:41:30Z | cs.CL | Cosine similarity between two words, computed using their contextualised
token embeddings obtained from masked language models (MLMs) such as BERT has
shown to underestimate the actual similarity between those words (Zhou et al.,
2022). This similarity underestimation problem is particularly severe for
highly frequent words. Although this problem has been noted in prior work, no
solution has been proposed thus far. We observe that the L2 norm of
contextualised embeddings of a word correlates with its log-frequency in the
pretraining corpus. Consequently, the larger L2 norms associated with the
highly frequent words reduce the cosine similarity values measured between
them, thus underestimating the similarity scores. To solve this issue, we
propose a method to discount the L2 norm of a contextualised word embedding by
the frequency of that word in a corpus when measuring the cosine similarities
between words. We show that the so called stop words behave differently from
the rest of the words, which require special consideration during their
discounting process. Experimental results on a contextualised word similarity
dataset show that our proposed discounting method accurately solves the
similarity underestimation problem. | cs.CL | Solving Cosine Similarity Underestimation between High Frequency Words by L2 Norm Discounting |
2023-05-17T23:16:17Z | http://arxiv.org/pdf/2305.10601v2 | http://arxiv.org/abs/2305.10601v2 | Shunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran, Thomas L. Griffiths, Yuan Cao, Karthik Narasimhan | 2023-12-03T22:50:35Z | cs.CL, cs.AI, cs.LG | Language models are increasingly being deployed for general problem solving
across a wide range of tasks, but are still confined to token-level,
left-to-right decision-making processes during inference. This means they can
fall short in tasks that require exploration, strategic lookahead, or where
initial decisions play a pivotal role. To surmount these challenges, we
introduce a new framework for language model inference, Tree of Thoughts (ToT),
which generalizes over the popular Chain of Thought approach to prompting
language models, and enables exploration over coherent units of text (thoughts)
that serve as intermediate steps toward problem solving. ToT allows LMs to
perform deliberate decision making by considering multiple different reasoning
paths and self-evaluating choices to decide the next course of action, as well
as looking ahead or backtracking when necessary to make global choices. Our
experiments show that ToT significantly enhances language models'
problem-solving abilities on three novel tasks requiring non-trivial planning
or search: Game of 24, Creative Writing, and Mini Crosswords. For instance, in
Game of 24, while GPT-4 with chain-of-thought prompting only solved 4% of
tasks, our method achieved a success rate of 74%. Code repo with all prompts:
https://github.com/princeton-nlp/tree-of-thought-llm. | cs.CL | Tree of Thoughts: Deliberate Problem Solving with Large Language Models |
2023-05-17T22:30:01Z | http://arxiv.org/pdf/2306.05539v1 | http://arxiv.org/abs/2306.05539v1 | Himanshu Gupta, Saurabh Arjun Sawant, Swaroop Mishra, Mutsumi Nakamura, Arindam Mitra, Santosh Mashetty, Chitta Baral | 2023-05-17T22:30:01Z | cs.CL | Instruction tuning of language models has demonstrated the ability to enhance
model generalization to unseen tasks via in-context learning using a few
examples. However, typical supervised learning still requires a plethora of
downstream training data for finetuning. Often in real-world situations, there
is a scarcity of data available for finetuning, falling somewhere between few
shot inference and fully supervised finetuning. In this work, we demonstrate
the sample efficiency of instruction tuned models over various tasks by
estimating the minimal downstream training data required by them to perform
transfer learning and match the performance of state-of-the-art (SOTA)
supervised models. We conduct experiments on 119 tasks from Super Natural
Instructions (SuperNI) in both the single task learning (STL) and multi task
learning (MTL) settings. Our findings reveal that, in the STL setting,
instruction tuned models equipped with 25% of the downstream train data surpass
the SOTA performance on the downstream tasks. In the MTL setting, an
instruction tuned model trained on only 6% of downstream training data achieve
SOTA, while using 100% of the training data results in a 3.69% points
improvement (ROUGE-L 74.68) over the previous SOTA. We conduct an analysis on
T5 vs Tk-Instruct by developing several baselines to demonstrate that
instruction tuning aids in increasing both sample efficiency and transfer
learning. Additionally, we observe a consistent ~4% performance increase in
both settings when pre-finetuning is performed with instructions. Finally, we
conduct a categorical study and find that contrary to previous results, tasks
in the question rewriting and title generation categories suffer from
instruction tuning. | cs.CL | Instruction Tuned Models are Quick Learners |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.