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
|
|---|---|---|---|---|---|---|---|---|
2023-11-24T14:20:12Z | http://arxiv.org/pdf/2311.14505v1 | http://arxiv.org/abs/2311.14505v1 | Victor Bystrov, Viktoriia Naboka-Krell, Anna Staszewska-Bystrova, Peter Winker | 2023-11-24T14:20:12Z | cs.CL | An initial procedure in text-as-data applications is text preprocessing. One
of the typical steps, which can substantially facilitate computations, consists
in removing infrequent words believed to provide limited information about the
corpus. Despite popularity of vocabulary pruning, not many guidelines on how to
implement it are available in the literature. The aim of the paper is to fill
this gap by examining the effects of removing infrequent words for the quality
of topics estimated using Latent Dirichlet Allocation. The analysis is based on
Monte Carlo experiments taking into account different criteria for infrequent
terms removal and various evaluation metrics. The results indicate that pruning
is beneficial and that the share of vocabulary which might be eliminated can be
quite considerable. | cs.CL | Analysing the Impact of Removing Infrequent Words on Topic Quality in LDA Models |
2023-11-24T14:08:31Z | http://arxiv.org/pdf/2311.14495v4 | http://arxiv.org/abs/2311.14495v4 | Shida Wang, Qianxiao Li | 2024-06-05T05:15:16Z | cs.LG, cs.AI, cs.CL, math.DS | In this paper, we investigate the long-term memory learning capabilities of
state-space models (SSMs) from the perspective of parameterization. We prove
that state-space models without any reparameterization exhibit a memory
limitation similar to that of traditional RNNs: the target relationships that
can be stably approximated by state-space models must have an exponential
decaying memory. Our analysis identifies this "curse of memory" as a result of
the recurrent weights converging to a stability boundary, suggesting that a
reparameterization technique can be effective. To this end, we introduce a
class of reparameterization techniques for SSMs that effectively lift its
memory limitations. Besides improving approximation capabilities, we further
illustrate that a principled choice of reparameterization scheme can also
enhance optimization stability. We validate our findings using synthetic
datasets, language models and image classifications. | cs.LG | StableSSM: Alleviating the Curse of Memory in State-space Models through Stable Reparameterization |
2023-11-24T13:48:02Z | http://arxiv.org/pdf/2312.00804v1 | http://arxiv.org/abs/2312.00804v1 | Elke Smith, Nils Reiter, Jan Peters | 2023-11-24T13:48:02Z | cs.CL, cs.LG | Problem gambling is a major public health concern and is associated with
profound psychological distress and economic problems. There are numerous
gambling communities on the internet where users exchange information about
games, gambling tactics, as well as gambling-related problems. Individuals
exhibiting higher levels of problem gambling engage more in such communities.
Online gambling communities may provide insights into problem-gambling
behaviour. Using data scraped from a major German gambling discussion board, we
fine-tuned a large language model, specifically a Bidirectional Encoder
Representations from Transformers (BERT) model, to predict signs of
problem-gambling from forum posts. Training data were generated by manual
annotation and by taking into account diagnostic criteria and gambling-related
cognitive distortions. Using k-fold cross-validation, our models achieved a
precision of 0.95 and F1 score of 0.71, demonstrating that satisfactory
classification performance can be achieved by generating high-quality training
material through manual annotation based on diagnostic criteria. The current
study confirms that a BERT-based model can be reliably used on small data sets
and to detect signatures of problem gambling in online communication data. Such
computational approaches may have potential for the detection of changes in
problem-gambling prevalence among online users. | cs.CL | Automatic detection of problem-gambling signs from online texts using large language models |
2023-11-24T13:47:25Z | http://arxiv.org/pdf/2311.14483v2 | http://arxiv.org/abs/2311.14483v2 | Alessandra Grossi, Francesca Gasparini | 2023-12-14T15:39:04Z | eess.AS, cs.CL, cs.SD | In this paper, SER_AMPEL, a multi-source dataset for speech emotion
recognition (SER) is presented. The peculiarity of the dataset is that it is
collected with the aim of providing a reference for speech emotion recognition
in case of Italian older adults. The dataset is collected following different
protocols, in particular considering acted conversations, extracted from movies
and TV series, and recording natural conversations where the emotions are
elicited by proper questions. The evidence of the need for such a dataset
emerges from the analysis of the state of the art. Preliminary considerations
on the critical issues of SER are reported analyzing the classification results
on a subset of the proposed dataset. | eess.AS | SER_AMPEL: a multi-source dataset for speech emotion recognition of Italian older adults |
2023-11-24T13:41:12Z | http://arxiv.org/pdf/2311.14479v2 | http://arxiv.org/abs/2311.14479v2 | Jasper Dekoninck, Marc Fischer, Luca Beurer-Kellner, Martin Vechev | 2024-03-06T09:36:54Z | cs.CL | As Large Language Models (LLMs) are deployed more widely, customization with
respect to vocabulary, style, and character becomes more important. In this
work, we introduce model arithmetic, a novel inference framework for composing
and biasing LLMs without the need for model (re)training or highly specific
datasets. In addition, the framework allows for more precise control of
generated text than direct prompting and prior controlled text generation (CTG)
techniques. Using model arithmetic, we can express prior CTG techniques as
simple formulas and naturally extend them to new and more effective
formulations. Further, we show that speculative sampling, a technique for
efficient LLM sampling, extends to our setting. This enables highly efficient
text generation with multiple composed models with only marginal overhead over
a single model. Our empirical evaluation demonstrates that model arithmetic
allows fine-grained control of generated text while outperforming
state-of-the-art on the task of toxicity reduction. We release an open source
easy-to-use implementation of our framework at
https://github.com/eth-sri/language-model-arithmetic. | cs.CL | Controlled Text Generation via Language Model Arithmetic |
2023-11-24T13:19:47Z | http://arxiv.org/pdf/2311.14465v2 | http://arxiv.org/abs/2311.14465v2 | Timour Igamberdiev, Doan Nam Long Vu, Felix Künnecke, Zhuo Yu, Jannik Holmer, Ivan Habernal | 2024-04-24T13:31:38Z | cs.CL | Neural machine translation (NMT) is a widely popular text generation task,
yet there is a considerable research gap in the development of
privacy-preserving NMT models, despite significant data privacy concerns for
NMT systems. Differentially private stochastic gradient descent (DP-SGD) is a
popular method for training machine learning models with concrete privacy
guarantees; however, the implementation specifics of training a model with
DP-SGD are not always clarified in existing models, with differing software
libraries used and code bases not always being public, leading to
reproducibility issues. To tackle this, we introduce DP-NMT, an open-source
framework for carrying out research on privacy-preserving NMT with DP-SGD,
bringing together numerous models, datasets, and evaluation metrics in one
systematic software package. Our goal is to provide a platform for researchers
to advance the development of privacy-preserving NMT systems, keeping the
specific details of the DP-SGD algorithm transparent and intuitive to
implement. We run a set of experiments on datasets from both general and
privacy-related domains to demonstrate our framework in use. We make our
framework publicly available and welcome feedback from the community. | cs.CL | DP-NMT: Scalable Differentially-Private Machine Translation |
2023-11-24T13:09:34Z | http://arxiv.org/pdf/2311.14455v4 | http://arxiv.org/abs/2311.14455v4 | Javier Rando, Florian Tramèr | 2024-04-29T12:37:59Z | cs.AI, cs.CL, cs.CR, cs.LG | Reinforcement Learning from Human Feedback (RLHF) is used to align large
language models to produce helpful and harmless responses. Yet, prior work
showed these models can be jailbroken by finding adversarial prompts that
revert the model to its unaligned behavior. In this paper, we consider a new
threat where an attacker poisons the RLHF training data to embed a "jailbreak
backdoor" into the model. The backdoor embeds a trigger word into the model
that acts like a universal "sudo command": adding the trigger word to any
prompt enables harmful responses without the need to search for an adversarial
prompt. Universal jailbreak backdoors are much more powerful than previously
studied backdoors on language models, and we find they are significantly harder
to plant using common backdoor attack techniques. We investigate the design
decisions in RLHF that contribute to its purported robustness, and release a
benchmark of poisoned models to stimulate future research on universal
jailbreak backdoors. | cs.AI | Universal Jailbreak Backdoors from Poisoned Human Feedback |
2023-11-24T10:15:34Z | http://arxiv.org/pdf/2311.14391v2 | http://arxiv.org/abs/2311.14391v2 | Milan Straka | 2023-12-08T02:13:25Z | cs.CL | We present CorPipe, the winning entry to the CRAC 2023 Shared Task on
Multilingual Coreference Resolution. Our system is an improved version of our
earlier multilingual coreference pipeline, and it surpasses other participants
by a large margin of 4.5 percent points. CorPipe first performs mention
detection, followed by coreference linking via an antecedent-maximization
approach on the retrieved spans. Both tasks are trained jointly on all
available corpora using a shared pretrained language model. Our main
improvements comprise inputs larger than 512 subwords and changing the mention
decoding to support ensembling. The source code is available at
https://github.com/ufal/crac2023-corpipe. | cs.CL | ÚFAL CorPipe at CRAC 2023: Larger Context Improves Multilingual Coreference Resolution |
2023-11-24T08:53:52Z | http://arxiv.org/pdf/2311.14353v2 | http://arxiv.org/abs/2311.14353v2 | Yasumasa Kano, Katsuhito Sudoh, Satoshi Nakamura | 2023-11-27T16:55:29Z | cs.CL | Simultaneous translation is a task in which the translation begins before the
end of an input speech segment. Its evaluation should be conducted based on
latency in addition to quality, and for users, the smallest possible amount of
latency is preferable. Most existing metrics measure latency based on the start
timings of partial translations and ignore their duration. This means such
metrics do not penalize the latency caused by long translation output, which
delays the comprehension of users and subsequent translations. In this work, we
propose a novel latency evaluation metric for simultaneous translation called
\emph{Average Token Delay} (ATD) that focuses on the duration of partial
translations. We demonstrate its effectiveness through analyses simulating
user-side latency based on Ear-Voice Span (EVS). In our experiment, ATD had the
highest correlation with EVS among baseline latency metrics under most
conditions. | cs.CL | Average Token Delay: A Duration-aware Latency Metric for Simultaneous Translation |
2023-11-24T07:53:48Z | http://arxiv.org/pdf/2311.14324v1 | http://arxiv.org/abs/2311.14324v1 | Shengyin Sun, Yuxiang Ren, Chen Ma, Xuecang Zhang | 2023-11-24T07:53:48Z | cs.AI, cs.CL, cs.LG | The latest advancements in large language models (LLMs) have revolutionized
the field of natural language processing (NLP). Inspired by the success of LLMs
in NLP tasks, some recent work has begun investigating the potential of
applying LLMs in graph learning tasks. However, most of the existing work
focuses on utilizing LLMs as powerful node feature augmenters, leaving
employing LLMs to enhance graph topological structures an understudied problem.
In this work, we explore how to leverage the information retrieval and text
generation capabilities of LLMs to refine/enhance the topological structure of
text-attributed graphs (TAGs) under the node classification setting. First, we
propose using LLMs to help remove unreliable edges and add reliable ones in the
TAG. Specifically, we first let the LLM output the semantic similarity between
node attributes through delicate prompt designs, and then perform edge deletion
and edge addition based on the similarity. Second, we propose using
pseudo-labels generated by the LLM to improve graph topology, that is, we
introduce the pseudo-label propagation as a regularization to guide the graph
neural network (GNN) in learning proper edge weights. Finally, we incorporate
the two aforementioned LLM-based methods for graph topological refinement into
the process of GNN training, and perform extensive experiments on four
real-world datasets. The experimental results demonstrate the effectiveness of
LLM-based graph topology refinement (achieving a 0.15%--2.47% performance gain
on public benchmarks). | cs.AI | Large Language Models as Topological Structure Enhancers for Text-Attributed Graphs |
2023-11-23T21:54:22Z | http://arxiv.org/pdf/2311.14212v3 | http://arxiv.org/abs/2311.14212v3 | Christoph Kern, Stephanie Eckman, Jacob Beck, Rob Chew, Bolei Ma, Frauke Kreuter | 2024-01-22T15:05:30Z | stat.ML, cs.CL, cs.LG, stat.ME | When training data are collected from human annotators, the design of the
annotation instrument, the instructions given to annotators, the
characteristics of the annotators, and their interactions can impact training
data. This study demonstrates that design choices made when creating an
annotation instrument also impact the models trained on the resulting
annotations. We introduce the term annotation sensitivity to refer to the
impact of annotation data collection methods on the annotations themselves and
on downstream model performance and predictions. We collect annotations of hate
speech and offensive language in five experimental conditions of an annotation
instrument, randomly assigning annotators to conditions. We then fine-tune BERT
models on each of the five resulting datasets and evaluate model performance on
a holdout portion of each condition. We find considerable differences between
the conditions for 1) the share of hate speech/offensive language annotations,
2) model performance, 3) model predictions, and 4) model learning curves. Our
results emphasize the crucial role played by the annotation instrument which
has received little attention in the machine learning literature. We call for
additional research into how and why the instrument impacts the annotations to
inform the development of best practices in instrument design. | stat.ML | Annotation Sensitivity: Training Data Collection Methods Affect Model Performance |
2023-11-23T20:52:44Z | http://arxiv.org/pdf/2311.14199v2 | http://arxiv.org/abs/2311.14199v2 | Chang Liu, Yuanhe Tian, Yan Song | 2024-04-25T09:25:24Z | cs.CV, cs.CL | Radiology report generation (RRG) aims to automatically generate free-text
descriptions from clinical radiographs, e.g., chest X-Ray images. RRG plays an
essential role in promoting clinical automation and presents significant help
to provide practical assistance for inexperienced doctors and alleviate
radiologists' workloads. Therefore, consider these meaningful potentials,
research on RRG is experiencing explosive growth in the past half-decade,
especially with the rapid development of deep learning approaches. Existing
studies perform RRG from the perspective of enhancing different modalities,
provide insights on optimizing the report generation process with elaborated
features from both visual and textual information, and further facilitate RRG
with the cross-modal interactions among them. In this paper, we present a
comprehensive review of deep learning-based RRG from various perspectives.
Specifically, we firstly cover pivotal RRG approaches based on the
task-specific features of radiographs, reports, and the cross-modal relations
between them, and then illustrate the benchmark datasets conventionally used
for this task with evaluation metrics, subsequently analyze the performance of
different approaches and finally offer our summary on the challenges and the
trends in future directions. Overall, the goal of this paper is to serve as a
tool for understanding existing literature and inspiring potential valuable
research in the field of RRG. | cs.CV | A Systematic Review of Deep Learning-based Research on Radiology Report Generation |
2023-11-23T19:20:59Z | http://arxiv.org/pdf/2311.14169v1 | http://arxiv.org/abs/2311.14169v1 | Ramon Pires, Thales Sales Almeida, Hugo Abonizio, Rodrigo Nogueira | 2023-11-23T19:20:59Z | cs.CL, cs.AI, cs.LG | Recent advancements in language models have showcased human-comparable
performance in academic entrance exams. However, existing studies often
overlook questions that require the integration of visual comprehension, thus
compromising the full spectrum and complexity inherent in real-world scenarios.
To address this gap, we present a comprehensive framework to evaluate language
models on entrance exams, which incorporates both textual and visual elements.
We evaluate the two most recent editions of Exame Nacional do Ensino M\'edio
(ENEM), the main standardized entrance examination adopted by Brazilian
universities. Our study not only reaffirms the capabilities of GPT-4 as the
state of the art for handling complex multidisciplinary questions, but also
pioneers in offering a realistic assessment of multimodal language models on
Portuguese examinations. One of the highlights is that text captions
transcribing visual content outperform the direct use of images, suggesting
that the vision model has room for improvement. Yet, despite improvements
afforded by images or captions, mathematical questions remain a challenge for
these state-of-the-art models. The code and data used on experiments are
available at https://github.com/piresramon/gpt-4-enem. | cs.CL | Evaluating GPT-4's Vision Capabilities on Brazilian University Admission Exams |
2023-11-23T17:47:14Z | http://arxiv.org/pdf/2311.14126v1 | http://arxiv.org/abs/2311.14126v1 | Wu Zekun, Sahan Bulathwela, Adriano Soares Koshiyama | 2023-11-23T17:47:14Z | cs.CL, cs.AI, cs.CY, cs.LG | Large Language Models (LLM) have made significant advances in the recent past
becoming more mainstream in Artificial Intelligence (AI) enabled human-facing
applications. However, LLMs often generate stereotypical output inherited from
historical data, amplifying societal biases and raising ethical concerns. This
work introduces i) the Multi-Grain Stereotype Dataset, which includes 52,751
instances of gender, race, profession and religion stereotypic text and ii) a
novel stereotype classifier for English text. We design several experiments to
rigorously test the proposed model trained on the novel dataset. Our
experiments show that training the model in a multi-class setting can
outperform the one-vs-all binary counterpart. Consistent feature importance
signals from different eXplainable AI tools demonstrate that the new model
exploits relevant text features. We utilise the newly created model to assess
the stereotypic behaviour of the popular GPT family of models and observe the
reduction of bias over time. In summary, our work establishes a robust and
practical framework for auditing and evaluating the stereotypic bias in LLM. | cs.CL | Towards Auditing Large Language Models: Improving Text-based Stereotype Detection |
2023-11-23T17:20:36Z | http://arxiv.org/pdf/2311.14115v3 | http://arxiv.org/abs/2311.14115v3 | Vincent Dumoulin, Daniel D. Johnson, Pablo Samuel Castro, Hugo Larochelle, Yann Dauphin | 2024-01-10T16:11:32Z | cs.LG, cs.AI, cs.CL | Learning from human feedback (LHF) -- and in particular learning from
pairwise preferences -- has recently become a crucial ingredient in training
large language models (LLMs), and has been the subject of much research. Most
recent works frame it as a reinforcement learning problem, where a reward
function is learned from pairwise preference data and the LLM is treated as a
policy which is adapted to maximize the rewards, often under additional
regularization constraints. We propose an alternative interpretation which
centers on the generative process for pairwise preferences and treats LHF as a
density estimation problem. We provide theoretical and empirical results
showing that for a family of generative processes defined via preference
behavior distribution equations, training a reward function on pairwise
preferences effectively models an annotator's implicit preference distribution.
Finally, we discuss and present findings on "annotator misspecification" --
failure cases where wrong modeling assumptions are made about annotator
behavior, resulting in poorly-adapted models -- suggesting that approaches that
learn from pairwise human preferences could have trouble learning from a
population of annotators with diverse viewpoints. | cs.LG | A density estimation perspective on learning from pairwise human preferences |
2023-11-23T16:45:56Z | http://arxiv.org/pdf/2311.14096v2 | http://arxiv.org/abs/2311.14096v2 | Yan Tao, Olga Viberg, Ryan S. Baker, Rene F. Kizilcec | 2024-06-26T15:26:44Z | cs.CL, cs.AI | Culture fundamentally shapes people's reasoning, behavior, and communication.
As people increasingly use generative artificial intelligence (AI) to expedite
and automate personal and professional tasks, cultural values embedded in AI
models may bias people's authentic expression and contribute to the dominance
of certain cultures. We conduct a disaggregated evaluation of cultural bias for
five widely used large language models (OpenAI's GPT-4o/4-turbo/4/3.5-turbo/3)
by comparing the models' responses to nationally representative survey data.
All models exhibit cultural values resembling English-speaking and Protestant
European countries. We test cultural prompting as a control strategy to
increase cultural alignment for each country/territory. For recent models
(GPT-4, 4-turbo, 4o), this improves the cultural alignment of the models'
output for 71-81% of countries and territories. We suggest using cultural
prompting and ongoing evaluation to reduce cultural bias in the output of
generative AI. | cs.CL | Cultural Bias and Cultural Alignment of Large Language Models |
2023-11-23T16:26:24Z | http://arxiv.org/pdf/2311.14087v1 | http://arxiv.org/abs/2311.14087v1 | Armand Stricker | 2023-11-23T16:26:24Z | cs.CL | Although general question answering has been well explored in recent years,
temporal question answering is a task which has not received as much focus. Our
work aims to leverage a popular approach used for general question answering,
answer extraction, in order to find answers to temporal questions within a
paragraph. To train our model, we propose a new dataset, inspired by SQuAD,
specifically tailored to provide rich temporal information. We chose to adapt
the corpus WikiWars, which contains several documents on history's greatest
conflicts. Our evaluation shows that a deep learning model trained to perform
pattern matching, often used in general question answering, can be adapted to
temporal question answering, if we accept to ask questions whose answers must
be directly present within a text. | cs.CL | Question Answering in Natural Language: the Special Case of Temporal Expressions |
2023-11-23T16:08:39Z | http://arxiv.org/pdf/2311.14076v1 | http://arxiv.org/abs/2311.14076v1 | Armand Stricker, Patrick Paroubek | 2023-11-23T16:08:39Z | cs.CL | Most existing dialogue corpora and models have been designed to fit into 2
predominant categories : task-oriented dialogues portray functional goals, such
as making a restaurant reservation or booking a plane ticket, while
chit-chat/open-domain dialogues focus on holding a socially engaging talk with
a user. However, humans tend to seamlessly switch between modes and even use
chitchat to enhance task-oriented conversations. To bridge this gap, new
datasets have recently been created, blending both communication modes into
conversation examples. The approaches used tend to rely on adding chit-chat
snippets to pre-existing, human-generated task-oriented datasets. Given the
tendencies observed in humans, we wonder however if the latter do not
\textit{already} hold chit-chat sequences. By using topic modeling and
searching for topics which are most similar to a set of keywords related to
social talk, we explore the training sets of Schema-Guided Dialogues and
MultiWOZ. Our study shows that sequences related to social talk are indeed
naturally present, motivating further research on ways chitchat is combined
into task-oriented dialogues. | cs.CL | Searching for Snippets of Open-Domain Dialogue in Task-Oriented Dialogue Datasets |
2023-11-23T15:50:42Z | http://arxiv.org/pdf/2311.14067v2 | http://arxiv.org/abs/2311.14067v2 | Armand Stricker, Patrick Paroubek | 2024-01-24T20:56:59Z | cs.CL | As a recent development, task-oriented dialogues (TODs) have been enriched
with chitchat in an effort to make dialogues more diverse and engaging. This
enhancement is particularly valuable as TODs are often confined to narrow
domains, making the mitigation of repetitive and predictable responses a
significant challenge. This paper presents a comparative analysis of three
chitchat enhancements, aiming to identify the most effective approach in terms
of diversity. Additionally, we quantify the divergence between the added
chitchat, the original task-oriented language, and chitchat typically found in
chitchat datasets, highlighting the top 20 divergent keywords for each
comparison. Our findings drive a discussion on future enhancements for
augmenting TODs, emphasizing the importance of grounding dialogues beyond the
task to achieve more diverse and natural exchanges. | cs.CL | Enhancing Task-Oriented Dialogues with Chitchat: a Comparative Study Based on Lexical Diversity and Divergence |
2023-11-23T15:42:00Z | http://arxiv.org/pdf/2311.14063v1 | http://arxiv.org/abs/2311.14063v1 | Yasser Abdelaziz Dahou Djilali, Sanath Narayan, Eustache Le Bihan, Haithem Boussaid, Ebtessam Almazrouei, Merouane Debbah | 2023-11-23T15:42:00Z | cs.CV, cs.CL, cs.LG | The Lip Reading Sentences-3 (LRS3) benchmark has primarily been the focus of
intense research in visual speech recognition (VSR) during the last few years.
As a result, there is an increased risk of overfitting to its excessively used
test set, which is only one hour duration. To alleviate this issue, we build a
new VSR test set named WildVSR, by closely following the LRS3 dataset creation
processes. We then evaluate and analyse the extent to which the current VSR
models generalize to the new test data. We evaluate a broad range of publicly
available VSR models and find significant drops in performance on our test set,
compared to their corresponding LRS3 results. Our results suggest that the
increase in word error rates is caused by the models inability to generalize to
slightly harder and in the wild lip sequences than those found in the LRS3 test
set. Our new test benchmark is made public in order to enable future research
towards more robust VSR models. | cs.CV | Do VSR Models Generalize Beyond LRS3? |
2023-11-23T13:13:48Z | http://arxiv.org/pdf/2311.13987v1 | http://arxiv.org/abs/2311.13987v1 | Ondřej Cífka, Constantinos Dimitriou, Cheng-i Wang, Hendrik Schreiber, Luke Miner, Fabian-Robert Stöter | 2023-11-23T13:13:48Z | eess.AS, cs.CL, cs.LG, cs.SD | Current automatic lyrics transcription (ALT) benchmarks focus exclusively on
word content and ignore the finer nuances of written lyrics including
formatting and punctuation, which leads to a potential misalignment with the
creative products of musicians and songwriters as well as listeners'
experiences. For example, line breaks are important in conveying information
about rhythm, emotional emphasis, rhyme, and high-level structure. To address
this issue, we introduce Jam-ALT, a new lyrics transcription benchmark based on
the JamendoLyrics dataset. Our contribution is twofold. Firstly, a complete
revision of the transcripts, geared specifically towards ALT evaluation by
following a newly created annotation guide that unifies the music industry's
guidelines, covering aspects such as punctuation, line breaks, spelling,
background vocals, and non-word sounds. Secondly, a suite of evaluation metrics
designed, unlike the traditional word error rate, to capture such phenomena. We
hope that the proposed benchmark contributes to the ALT task, enabling more
precise and reliable assessments of transcription systems and enhancing the
user experience in lyrics applications such as subtitle renderings for live
captioning or karaoke. | eess.AS | Jam-ALT: A Formatting-Aware Lyrics Transcription Benchmark |
2023-11-23T12:52:37Z | http://arxiv.org/pdf/2311.13982v1 | http://arxiv.org/abs/2311.13982v1 | Shulin Cao, Jiajie Zhang, Jiaxin Shi, Xin Lv, Zijun Yao, Qi Tian, Juanzi Li, Lei Hou | 2023-11-23T12:52:37Z | cs.CL, cs.AI | Large language models (LLMs) are capable of answering knowledge-intensive
complex questions with chain-of-thought (CoT) reasoning. However, they tend to
generate factually incorrect reasoning steps when the required knowledge is not
available or up-to-date in models' parameters. Recent works turn to retrieving
external knowledge to augment CoT reasoning. Despite being promising, these
chain-based methods suffer from: 1) Negative retrieval. Unnecessary or
incorrect retrieval may mislead the reasoning; 2) Limited sight. Lacking the
ability to look backward or forward, a local error in one step will propagate
along the chain.
In this paper, we propose a novel approach: Probabilistic Tree-of-thought
Reasoning (ProbTree). First, LLMs translate a complex question into a query
tree, in which each non-root node denotes a sub-question of its parent node.
Then, probabilistic reasoning is conducted over the tree, by solving questions
from leaf to root considering the confidence of both question decomposing and
answering. During reasoning, for leaf nodes, LLMs choose a more confident
answer from Closed-book QA that employs parametric knowledge and Open-book QA
that employs retrieved external knowledge, thus eliminating the negative
retrieval problem. For non-leaf nodes, with the hierarchical structure, LLMs
have broader sights and are able to globally reason with the information from
child nodes, thus recovering from local errors. The experiments on three
Complex QA datasets under the open-domain setting show that our approach
outperforms SOTA methods significantly, demonstrating the effect of
probabilistic tree-of-thought reasoning. | cs.CL | Probabilistic Tree-of-thought Reasoning for Answering Knowledge-intensive Complex Questions |
2023-11-23T12:19:41Z | http://arxiv.org/pdf/2312.07556v1 | http://arxiv.org/abs/2312.07556v1 | Mengling Hu, Chaochao Chen, Weiming Liu, Xinting Liao, Xiaolin Zheng | 2023-11-23T12:19:41Z | cs.CL, cs.AI, cs.IR | Short text clustering has been popularly studied for its significance in
mining valuable insights from many short texts. In this paper, we focus on the
federated short text clustering (FSTC) problem, i.e., clustering short texts
that are distributed in different clients, which is a realistic problem under
privacy requirements. Compared with the centralized short text clustering
problem that short texts are stored on a central server, the FSTC problem has
not been explored yet. To fill this gap, we propose a Federated Robust Short
Text Clustering (FSTC) framework. FSTC includes two main modules, i.e., robust
short text clustering module and federated cluster center aggregation module.
The robust short text clustering module aims to train an effective short text
clustering model with local data in each client. We innovatively combine
optimal transport to generate pseudo-labels with Gaussian-uniform mixture model
to ensure the reliability of the pseudo-supervised data. The federated cluster
center aggregation module aims to exchange knowledge across clients without
sharing local raw data in an efficient way. The server aggregates the local
cluster centers from different clients and then sends the global centers back
to all clients in each communication round. Our empirical studies on three
short text clustering datasets demonstrate that FSTC significantly outperforms
the federated short text clustering baselines. | cs.CL | Federated Learning for Short Text Clustering |
2023-11-23T12:15:56Z | http://arxiv.org/pdf/2311.13957v1 | http://arxiv.org/abs/2311.13957v1 | Yueqi Zeng, Ziqiang Li, Pengfei Xia, Lei Liu, Bin Li | 2023-11-23T12:15:56Z | cs.CR, cs.CL | With the boom in the natural language processing (NLP) field these years,
backdoor attacks pose immense threats against deep neural network models.
However, previous works hardly consider the effect of the poisoning rate. In
this paper, our main objective is to reduce the number of poisoned samples
while still achieving a satisfactory Attack Success Rate (ASR) in text backdoor
attacks. To accomplish this, we propose an efficient trigger word insertion
strategy in terms of trigger word optimization and poisoned sample selection.
Extensive experiments on different datasets and models demonstrate that our
proposed method can significantly improve attack effectiveness in text
classification tasks. Remarkably, our approach achieves an ASR of over 90% with
only 10 poisoned samples in the dirty-label setting and requires merely 1.5% of
the training data in the clean-label setting. | cs.CR | Efficient Trigger Word Insertion |
2023-11-23T12:04:25Z | http://arxiv.org/pdf/2311.13951v2 | http://arxiv.org/abs/2311.13951v2 | Wentao Ge, Shunian Chen, Guiming Hardy Chen, Zhihong Chen, Junying Chen, Shuo Yan, Chenghao Zhu, Ziyue Lin, Wenya Xie, Xinyi Zhang, Yichen Chai, Xiaoyu Liu, Dingjie Song, Xidong Wang, Anningzhe Gao, Zhiyi Zhang, Jianquan Li, Xiang Wan, Benyou Wang | 2024-04-27T04:32:05Z | cs.CL | Multimodal large language models (MLLMs) (e.g., GPT-4V, LLaVA, and Claude-3)
have broadened the scope of AI applications. Yet, evaluating their performance
presents a significant challenge owing to the inherently subjective nature of
tasks that do not yield clear-cut solutions especially for those open-ended
queries. Existing automatic evaluation methodologies are mainly limited in
evaluating objective queries without considering real-world user experiences,
inadequately addressing the nuances of creative and associative multimodal
tasks. In our paper, we propose a new evaluation paradigm for MLLMs, which is
evaluating MLLMs with \textit{per-sample criteria} using potent MLLM as the
judge. To validate the feasibility and effectiveness of this paradigm, we
design a benchmark, dubbed \textit{MLLM-Bench}, with the evaluation samples
across six critical levels following the revised Bloom's Taxonomy with the
ethical consideration. We benchmark 21 popular MLLMs in a pairwise-comparison
fashion, showing diverse performance across models. Moreover, the validity of
our benchmark manifests itself in reaching 88.02\% agreement with human
evaluation. We contend that the proposed paradigm explores the potential of
MLLMs as effective evaluation tools with the help of per-sample criteria, and
that MLLM-Bench will serve as a catalyst for encouraging the development of
user-centric MLLMs tailored to real-world applications. Our benchmark data,
online leaderboard and submission entry are at https://mllm-bench.llmzoo.com. | cs.CL | MLLM-Bench: Evaluating Multimodal LLMs with Per-sample Criteria |
2023-11-23T11:40:28Z | http://arxiv.org/pdf/2311.13937v1 | http://arxiv.org/abs/2311.13937v1 | Daryna Dementieva, Daniil Moskovskiy, David Dale, Alexander Panchenko | 2023-11-23T11:40:28Z | cs.CL | Text detoxification is the task of transferring the style of text from toxic
to neutral. While here are approaches yielding promising results in monolingual
setup, e.g., (Dale et al., 2021; Hallinan et al., 2022), cross-lingual transfer
for this task remains a challenging open problem (Moskovskiy et al., 2022). In
this work, we present a large-scale study of strategies for cross-lingual text
detoxification -- given a parallel detoxification corpus for one language; the
goal is to transfer detoxification ability to another language for which we do
not have such a corpus. Moreover, we are the first to explore a new task where
text translation and detoxification are performed simultaneously, providing
several strong baselines for this task. Finally, we introduce new automatic
detoxification evaluation metrics with higher correlations with human judgments
than previous benchmarks. We assess the most promising approaches also with
manual markup, determining the answer for the best strategy to transfer the
knowledge of text detoxification between languages. | cs.CL | Exploring Methods for Cross-lingual Text Style Transfer: The Case of Text Detoxification |
2023-11-23T11:14:13Z | http://arxiv.org/pdf/2311.13921v1 | http://arxiv.org/abs/2311.13921v1 | Jiří Bednář, Jakub Náplava, Petra Barančíková, Ondřej Lisický | 2023-11-23T11:14:13Z | cs.CL, cs.IR | This article focuses on the development and evaluation of Small-sized Czech
sentence embedding models. Small models are important components for real-time
industry applications in resource-constrained environments. Given the limited
availability of labeled Czech data, alternative approaches, including
pre-training, knowledge distillation, and unsupervised contrastive fine-tuning,
are investigated. Comprehensive intrinsic and extrinsic analyses are conducted,
showcasing the competitive performance of our models compared to significantly
larger counterparts, with approximately 8 times smaller size and 5 times faster
speed than conventional Base-sized models. To promote cooperation and
reproducibility, both the models and the evaluation pipeline are made publicly
accessible. Ultimately, this article presents practical applications of the
developed sentence embedding models in Seznam.cz, the Czech search engine.
These models have effectively replaced previous counterparts, enhancing the
overall search experience for instance, in organic search, featured snippets,
and image search. This transition has yielded improved performance. | cs.CL | Some Like It Small: Czech Semantic Embedding Models for Industry Applications |
2023-11-23T10:56:14Z | http://arxiv.org/pdf/2311.13910v1 | http://arxiv.org/abs/2311.13910v1 | John Mendonça, Patrícia Pereira, Miguel Menezes, Vera Cabarrão, Ana C. Farinha, Helena Moniz, João Paulo Carvalho, Alon Lavie, Isabel Trancoso | 2023-11-23T10:56:14Z | cs.CL | Task-oriented conversational datasets often lack topic variability and
linguistic diversity. However, with the advent of Large Language Models (LLMs)
pretrained on extensive, multilingual and diverse text data, these limitations
seem overcome. Nevertheless, their generalisability to different languages and
domains in dialogue applications remains uncertain without benchmarking
datasets. This paper presents a holistic annotation approach for emotion and
conversational quality in the context of bilingual customer support
conversations. By performing annotations that take into consideration the
complete instances that compose a conversation, one can form a broader
perspective of the dialogue as a whole. Furthermore, it provides a unique and
valuable resource for the development of text classification models. To this
end, we present benchmarks for Emotion Recognition and Dialogue Quality
Estimation and show that further research is needed to leverage these models in
a production setting. | cs.CL | Dialogue Quality and Emotion Annotations for Customer Support Conversations |
2023-11-23T10:23:51Z | http://arxiv.org/pdf/2311.13892v3 | http://arxiv.org/abs/2311.13892v3 | Bingkang Shi, Xiaodan Zhang, Dehan Kong, Yulei Wu, Zongzhen Liu, Honglei Lyu, Longtao Huang | 2024-01-25T15:36:44Z | cs.CL, cs.AI | The social biases and unwelcome stereotypes revealed by pretrained language
models are becoming obstacles to their application. Compared to numerous
debiasing methods targeting word level, there has been relatively less
attention on biases present at phrase level, limiting the performance of
debiasing in discipline domains. In this paper, we propose an automatic
multi-token debiasing pipeline called \textbf{General Phrase Debiaser}, which
is capable of mitigating phrase-level biases in masked language models.
Specifically, our method consists of a \textit{phrase filter stage} that
generates stereotypical phrases from Wikipedia pages as well as a \textit{model
debias stage} that can debias models at the multi-token level to tackle bias
challenges on phrases. The latter searches for prompts that trigger model's
bias, and then uses them for debiasing. State-of-the-art results on standard
datasets and metrics show that our approach can significantly reduce gender
biases on both career and multiple disciplines, across models with varying
parameter sizes. | cs.CL | General Phrase Debiaser: Debiasing Masked Language Models at a Multi-Token Level |
2023-11-23T09:58:39Z | http://arxiv.org/pdf/2311.13878v1 | http://arxiv.org/abs/2311.13878v1 | Muneeswaran I, Shreya Saxena, Siva Prasad, M V Sai Prakash, Advaith Shankar, Varun V, Vishal Vaddina, Saisubramaniam Gopalakrishnan | 2023-11-23T09:58:39Z | cs.CL, cs.AI | Large Language Models (LLMs) are widely used in critical fields such as
healthcare, education, and finance due to their remarkable proficiency in
various language-related tasks. However, LLMs are prone to generating factually
incorrect responses or "hallucinations," which can lead to a loss of
credibility and trust among users. To address this issue, we propose a
multi-stage framework that generates the rationale first, verifies and refines
incorrect ones, and uses them as supporting references to generate the answer.
The generated rationale enhances the transparency of the answer and our
framework provides insights into how the model arrived at this answer, by using
this rationale and the references to the context. In this paper, we demonstrate
its effectiveness in improving the quality of responses to drug-related
inquiries in the life sciences industry. Our framework improves traditional
Retrieval Augmented Generation (RAG) by enabling OpenAI GPT-3.5-turbo to be
14-25% more faithful and 16-22% more accurate on two datasets. Furthermore,
fine-tuning samples based on our framework improves the accuracy of smaller
open-access LLMs by 33-42% and competes with RAG on commercial models. | cs.CL | Minimizing Factual Inconsistency and Hallucination in Large Language Models |
2023-11-23T08:56:41Z | http://arxiv.org/pdf/2311.13857v1 | http://arxiv.org/abs/2311.13857v1 | Neo Christopher Chung, George Dyer, Lennart Brocki | 2023-11-23T08:56:41Z | cs.CL, cs.AI, cs.HC | The global mental health crisis is looming with a rapid increase in mental
disorders, limited resources, and the social stigma of seeking treatment. As
the field of artificial intelligence (AI) has witnessed significant
advancements in recent years, large language models (LLMs) capable of
understanding and generating human-like text may be used in supporting or
providing psychological counseling. However, the application of LLMs in the
mental health domain raises concerns regarding the accuracy, effectiveness, and
reliability of the information provided. This paper investigates the major
challenges associated with the development of LLMs for psychological
counseling, including model hallucination, interpretability, bias, privacy, and
clinical effectiveness. We explore potential solutions to these challenges that
are practical and applicable to the current paradigm of AI. From our experience
in developing and deploying LLMs for mental health, AI holds a great promise
for improving mental health care, if we can carefully navigate and overcome
pitfalls of LLMs. | cs.CL | Challenges of Large Language Models for Mental Health Counseling |
2023-11-23T08:34:37Z | http://arxiv.org/pdf/2311.13848v1 | http://arxiv.org/abs/2311.13848v1 | Jiahao Li, Quan Wang, Chiwei Zhu, Zhendong Mao, Yongdong Zhang | 2023-11-23T08:34:37Z | cs.CL | The task of Grammatical Error Correction (GEC) aims to automatically correct
grammatical errors in natural texts. Almost all previous works treat annotated
training data equally, but inherent discrepancies in data are neglected. In
this paper, the inherent discrepancies are manifested in two aspects, namely,
accuracy of data annotation and diversity of potential annotations. To this
end, we propose MainGEC, which designs token-level and sentence-level training
weights based on inherent discrepancies in accuracy and potential diversity of
data annotation, respectively, and then conducts mixed-grained weighted
training to improve the training effect for GEC. Empirical evaluation shows
that whether in the Seq2Seq or Seq2Edit manner, MainGEC achieves consistent and
significant performance improvements on two benchmark datasets, demonstrating
the effectiveness and superiority of the mixed-grained weighted training.
Further ablation experiments verify the effectiveness of designed weights of
both granularities in MainGEC. | cs.CL | Grammatical Error Correction via Mixed-Grained Weighted Training |
2023-11-23T07:33:38Z | http://arxiv.org/pdf/2311.13833v1 | http://arxiv.org/abs/2311.13833v1 | Saman Motamed, Danda Pani Paudel, Luc Van Gool | 2023-11-23T07:33:38Z | cs.CV, cs.CL, cs.LG | Diffusion models have revolutionized generative content creation and
text-to-image (T2I) diffusion models in particular have increased the creative
freedom of users by allowing scene synthesis using natural language. T2I models
excel at synthesizing concepts such as nouns, appearances, and styles. To
enable customized content creation based on a few example images of a concept,
methods such as Textual Inversion and DreamBooth invert the desired concept and
enable synthesizing it in new scenes. However, inverting more general concepts
that go beyond object appearance and style (adjectives and verbs) through
natural language, remains a challenge. Two key characteristics of these
concepts contribute to the limitations of current inversion methods. 1)
Adjectives and verbs are entangled with nouns (subject) and can hinder
appearance-based inversion methods, where the subject appearance leaks into the
concept embedding and 2) describing such concepts often extends beyond single
word embeddings (being frozen in ice, walking on a tightrope, etc.) that
current methods do not handle.
In this study, we introduce Lego, a textual inversion method designed to
invert subject entangled concepts from a few example images. Lego disentangles
concepts from their associated subjects using a simple yet effective Subject
Separation step and employs a Context Loss that guides the inversion of
single/multi-embedding concepts. In a thorough user study, Lego-generated
concepts were preferred over 70% of the time when compared to the baseline.
Additionally, visual question answering using a large language model suggested
Lego-generated concepts are better aligned with the text description of the
concept. | cs.CV | Lego: Learning to Disentangle and Invert Concepts Beyond Object Appearance in Text-to-Image Diffusion Models |
2023-11-23T04:42:27Z | http://arxiv.org/pdf/2311.13806v1 | http://arxiv.org/abs/2311.13806v1 | Madelon Hulsebos, Paul Groth, Çağatay Demiralp | 2023-11-23T04:42:27Z | cs.DB, cs.CL, cs.LG | Understanding the semantics of relational tables is instrumental for
automation in data exploration and preparation systems. A key source for
understanding a table is the semantics of its columns. With the rise of deep
learning, learned table representations are now available, which can be applied
for semantic type detection and achieve good performance on benchmarks.
Nevertheless, we observe a gap between this performance and its applicability
in practice. In this paper, we propose AdaTyper to address one of the most
critical deployment challenges: adaptation. AdaTyper uses weak-supervision to
adapt a hybrid type predictor towards new semantic types and shifted data
distributions at inference time, using minimal human feedback. The hybrid type
predictor of AdaTyper combines rule-based methods and a light machine learning
model for semantic column type detection. We evaluate the adaptation
performance of AdaTyper on real-world database tables hand-annotated with
semantic column types through crowdsourcing and find that the f1-score improves
for new and existing types. AdaTyper approaches an average precision of 0.6
after only seeing 5 examples, significantly outperforming existing adaptation
methods based on human-provided regular expressions or dictionaries. | cs.DB | AdaTyper: Adaptive Semantic Column Type Detection |
2023-11-23T03:03:54Z | http://arxiv.org/pdf/2311.13784v1 | http://arxiv.org/abs/2311.13784v1 | Dongjun Jang, Sangah Lee, Sungjoo Byun, Jinwoong Kim, Jean Seo, Minseok Kim, Soyeon Kim, Chaeyoung Oh, Jaeyoon Kim, Hyemi Jo, Hyopil Shin | 2023-11-23T03:03:54Z | cs.CL | This paper presents the DaG LLM (David and Goliath Large Language Model), a
language model specialized for Korean and fine-tuned through Instruction Tuning
across 41 tasks within 13 distinct categories. | cs.CL | DaG LLM ver 1.0: Pioneering Instruction-Tuned Language Modeling for Korean NLP |
2023-11-23T01:06:08Z | http://arxiv.org/pdf/2311.13755v1 | http://arxiv.org/abs/2311.13755v1 | Milad Baghalzadeh Shishehgarkhaneh, Robert C. Moehler, Yihai Fang, Amer A. Hijazi, Hamed Aboutorab | 2023-11-23T01:06:08Z | cs.CL | The construction industry in Australia is characterized by its intricate
supply chains and vulnerability to myriad risks. As such, effective supply
chain risk management (SCRM) becomes imperative. This paper employs different
transformer models, and train for Named Entity Recognition (NER) in the context
of Australian construction SCRM. Utilizing NER, transformer models identify and
classify specific risk-associated entities in news articles, offering a
detailed insight into supply chain vulnerabilities. By analysing news articles
through different transformer models, we can extract relevant entities and
insights related to specific risk taxonomies local (milieu) to the Australian
construction landscape. This research emphasises the potential of NLP-driven
solutions, like transformer models, in revolutionising SCRM for construction in
geo-media specific contexts. | cs.CL | Transformer-based Named Entity Recognition in Construction Supply Chain Risk Management in Australia |
2023-11-22T23:35:13Z | http://arxiv.org/pdf/2311.13735v1 | http://arxiv.org/abs/2311.13735v1 | Zhichao Yang, Sanjit Singh Batra, Joel Stremmel, Eran Halperin | 2023-11-22T23:35:13Z | cs.CL | Recent advances in large language models (LLMs) show potential for clinical
applications, such as clinical decision support and trial recommendations.
However, the GPT-4 LLM predicts an excessive number of ICD codes for medical
coding tasks, leading to high recall but low precision. To tackle this
challenge, we introduce LLM-codex, a two-stage approach to predict ICD codes
that first generates evidence proposals using an LLM and then employs an
LSTM-based verification stage. The LSTM learns from both the LLM's high recall
and human expert's high precision, using a custom loss function. Our model is
the only approach that simultaneously achieves state-of-the-art results in
medical coding accuracy, accuracy on rare codes, and sentence-level evidence
identification to support coding decisions without training on human-annotated
evidence according to experiments on the MIMIC dataset. | cs.CL | Surpassing GPT-4 Medical Coding with a Two-Stage Approach |
2023-11-22T22:52:00Z | http://arxiv.org/pdf/2311.13729v2 | http://arxiv.org/abs/2311.13729v2 | Shashank Gupta, Xuguang Ai, Ramakanth Kavuluru | 2024-03-10T01:11:20Z | cs.CL | End-to-end relation extraction (E2ERE) is an important and realistic
application of natural language processing (NLP) in biomedicine. In this paper,
we aim to compare three prevailing paradigms for E2ERE using a complex dataset
focused on rare diseases involving discontinuous and nested entities. We use
the RareDis information extraction dataset to evaluate three competing
approaches (for E2ERE): NER $\rightarrow$ RE pipelines, joint sequence to
sequence models, and generative pre-trained transformer (GPT) models. We use
comparable state-of-the-art models and best practices for each of these
approaches and conduct error analyses to assess their failure modes. Our
findings reveal that pipeline models are still the best, while
sequence-to-sequence models are not far behind; GPT models with eight times as
many parameters are worse than even sequence-to-sequence models and lose to
pipeline models by over 10 F1 points. Partial matches and discontinuous
entities caused many NER errors contributing to lower overall E2E performances.
We also verify these findings on a second E2ERE dataset for chemical-protein
interactions. Although generative LM-based methods are more suitable for
zero-shot settings, when training data is available, our results show that it
is better to work with more conventional models trained and tailored for E2ERE.
More innovative methods are needed to marry the best of the both worlds from
smaller encoder-decoder pipeline models and the larger GPT models to improve
E2ERE. As of now, we see that well designed pipeline models offer substantial
performance gains at a lower cost and carbon footprint for E2ERE. Our
contribution is also the first to conduct E2ERE for the RareDis dataset. | cs.CL | Comparison of pipeline, sequence-to-sequence, and GPT models for end-to-end relation extraction: experiments with the rare disease use-case |
2023-11-22T21:59:46Z | http://arxiv.org/pdf/2311.13708v4 | http://arxiv.org/abs/2311.13708v4 | Weiwei Li, Xing Liu, Wei Wang, Lu Chen, Sizhe Li, Hui Fan | 2024-01-27T07:32:54Z | cs.CL | To address the challenge of identifying hidden danger in substations from
unstructured text, a novel dynamic analysis method is proposed. We first
extract relevant information from the unstructured text, and then leverages a
flexible distributed search engine built on Elastic-Search to handle the data.
Following this, the hidden Markov model is employed to train the data within
the engine. The Viterbi algorithm is integrated to decipher the hidden state
sequences, facilitating the segmentation and labeling of entities related to
hidden dangers. The final step involves using the Neo4j graph database to
dynamically create a knowledge graph that visualizes hidden dangers in the
substation. The effectiveness of the proposed method is demonstrated through a
case analysis from a specific substation with hidden dangers revealed in the
text records. | cs.CL | Dynamic Fault Analysis in Substations Based on Knowledge Graphs |
2023-11-22T19:45:40Z | http://arxiv.org/pdf/2311.13668v3 | http://arxiv.org/abs/2311.13668v3 | Stephanie L. Hyland, Shruthi Bannur, Kenza Bouzid, Daniel C. Castro, Mercy Ranjit, Anton Schwaighofer, Fernando Pérez-García, Valentina Salvatelli, Shaury Srivastav, Anja Thieme, Noel Codella, Matthew P. Lungren, Maria Teodora Wetscherek, Ozan Oktay, Javier Alvarez-Valle | 2024-04-26T16:29:54Z | cs.CL, cs.AI, cs.CV | We present a radiology-specific multimodal model for the task for generating
radiological reports from chest X-rays (CXRs). Our work builds on the idea that
large language model(s) can be equipped with multimodal capabilities through
alignment with pre-trained vision encoders. On natural images, this has been
shown to allow multimodal models to gain image understanding and description
capabilities. Our proposed model (MAIRA-1) leverages a CXR-specific image
encoder in conjunction with a fine-tuned large language model based on
Vicuna-7B, and text-based data augmentation, to produce reports with
state-of-the-art quality. In particular, MAIRA-1 significantly improves on the
radiologist-aligned RadCliQ metric and across all lexical metrics considered.
Manual review of model outputs demonstrates promising fluency and accuracy of
generated reports while uncovering failure modes not captured by existing
evaluation practices. More information and resources can be found on the
project website: https://aka.ms/maira. | cs.CL | MAIRA-1: A specialised large multimodal model for radiology report generation |
2023-11-22T19:19:37Z | http://arxiv.org/pdf/2311.13657v1 | http://arxiv.org/abs/2311.13657v1 | Nathan Brown, Ashton Williamson, Tahj Anderson, Logan Lawrence | 2023-11-22T19:19:37Z | cs.CL, cs.LG | As pretrained transformer language models continue to achieve
state-of-the-art performance, the Natural Language Processing community has
pushed for advances in model compression and efficient attention mechanisms to
address high computational requirements and limited input sequence length.
Despite these separate efforts, no investigation has been done into the
intersection of these two fields. In this work, we provide an evaluation of
model compression via knowledge distillation on efficient attention
transformers. We provide cost-performance trade-offs for the compression of
state-of-the-art efficient attention architectures and the gains made in
performance in comparison to their full attention counterparts. Furthermore, we
introduce a new long-context Named Entity Recognition dataset, GONERD, to train
and test the performance of NER models on long sequences. We find that
distilled efficient attention transformers can preserve a significant amount of
original model performance, preserving up to 98.6% across short-context tasks
(GLUE, SQUAD, CoNLL-2003), up to 94.6% across long-context
Question-and-Answering tasks (HotpotQA, TriviaQA), and up to 98.8% on
long-context Named Entity Recognition (GONERD), while decreasing inference
times by up to 57.8%. We find that, for most models on most tasks, performing
knowledge distillation is an effective method to yield high-performing
efficient attention models with low costs. | cs.CL | Efficient Transformer Knowledge Distillation: A Performance Review |
2023-11-22T19:04:04Z | http://arxiv.org/pdf/2311.13647v1 | http://arxiv.org/abs/2311.13647v1 | John X. Morris, Wenting Zhao, Justin T. Chiu, Vitaly Shmatikov, Alexander M. Rush | 2023-11-22T19:04:04Z | cs.CL, cs.LG | Language models produce a distribution over the next token; can we use this
information to recover the prompt tokens? We consider the problem of language
model inversion and show that next-token probabilities contain a surprising
amount of information about the preceding text. Often we can recover the text
in cases where it is hidden from the user, motivating a method for recovering
unknown prompts given only the model's current distribution output. We consider
a variety of model access scenarios, and show how even without predictions for
every token in the vocabulary we can recover the probability vector through
search. On Llama-2 7b, our inversion method reconstructs prompts with a BLEU of
$59$ and token-level F1 of $78$ and recovers $27\%$ of prompts exactly. Code
for reproducing all experiments is available at
http://github.com/jxmorris12/vec2text. | cs.CL | Language Model Inversion |
2023-11-22T18:50:47Z | http://arxiv.org/pdf/2311.13628v2 | http://arxiv.org/abs/2311.13628v2 | Thomas P. Zollo, Todd Morrill, Zhun Deng, Jake C. Snell, Toniann Pitassi, Richard Zemel | 2024-03-27T20:20:22Z | cs.LG, cs.AI, cs.CL | The recent explosion in the capabilities of large language models has led to
a wave of interest in how best to prompt a model to perform a given task. While
it may be tempting to simply choose a prompt based on average performance on a
validation set, this can lead to a deployment where unexpectedly poor responses
are generated, especially for the worst-off users. To mitigate this prospect,
we propose Prompt Risk Control, a lightweight framework for selecting a prompt
based on rigorous upper bounds on families of informative risk measures. We
offer methods for producing bounds on a diverse set of metrics, including
quantities that measure worst-case responses and disparities in generation
quality across the population of users. In addition, we extend the underlying
statistical bounding techniques to accommodate the possibility of distribution
shifts in deployment. Experiments on applications such as open-ended chat,
medical question summarization, and code generation highlight how such a
framework can foster responsible deployment by reducing the risk of the worst
outcomes. | cs.LG | Prompt Risk Control: A Rigorous Framework for Responsible Deployment of Large Language Models |
2023-11-22T18:37:27Z | http://arxiv.org/pdf/2311.13581v1 | http://arxiv.org/abs/2311.13581v1 | Giovanni Monea, Armand Joulin, Edouard Grave | 2023-11-22T18:37:27Z | cs.CL | Scaling the size of language models to tens of billions of parameters has led
to impressive performance on a wide range of tasks. At generation, these models
are used auto-regressively, requiring a forward pass for each generated token,
and thus reading the full set of parameters from memory. This memory access
forms the primary bottleneck for generation and it worsens as the model size
increases. Moreover, executing a forward pass for multiple tokens in parallel
often takes nearly the same time as it does for just one token. These two
observations lead to the development of speculative sampling, where a second
smaller model is used to draft a few tokens, that are then validated or
rejected using a single forward pass of the large model. Unfortunately, this
method requires two models that share the same tokenizer and thus limits its
adoption. As an alternative, we propose to use parallel decoding as a way to
draft multiple tokens from a single model with no computational cost, nor the
need for a second model. Our approach only requires an additional input token
that marks the words that will be generated simultaneously. We show promising
performance (up to $30\%$ speed-up) while requiring only as few as $O(d_{emb})$
additional parameters. | cs.CL | PaSS: Parallel Speculative Sampling |
2023-11-22T18:22:56Z | http://arxiv.org/pdf/2311.13565v1 | http://arxiv.org/abs/2311.13565v1 | Inderjeet Nair, Shwetha Somasundaram, Apoorv Saxena, Koustava Goswami | 2023-11-22T18:22:56Z | cs.CL, cs.AI, cs.IR | We address the task of evidence retrieval for long document question
answering, which involves locating relevant paragraphs within a document to
answer a question. We aim to assess the applicability of large language models
(LLMs) in the task of zero-shot long document evidence retrieval, owing to
their unprecedented performance across various NLP tasks. However, currently
the LLMs can consume limited context lengths as input, thus providing document
chunks as inputs might overlook the global context while missing out on
capturing the inter-segment dependencies. Moreover, directly feeding the large
input sets can incur significant computational costs, particularly when
processing the entire document (and potentially incurring monetary expenses
with enterprise APIs like OpenAI's GPT variants). To address these challenges,
we propose a suite of techniques that exploit the discourse structure commonly
found in documents. By utilizing this structure, we create a condensed
representation of the document, enabling a more comprehensive understanding and
analysis of relationships between different parts. We retain $99.6\%$ of the
best zero-shot approach's performance, while processing only $26\%$ of the
total tokens used by the best approach in the information seeking evidence
retrieval setup. We also show how our approach can be combined with
\textit{self-ask} reasoning agent to achieve best zero-shot performance in
complex multi-hop question answering, just $\approx 4\%$ short of zero-shot
performance using gold evidence. | cs.CL | Drilling Down into the Discourse Structure with LLMs for Long Document Question Answering |
2023-11-22T17:14:54Z | http://arxiv.org/pdf/2311.13534v4 | http://arxiv.org/abs/2311.13534v4 | Shitao Xiao, Zheng Liu, Peitian Zhang, Xingrun Xing | 2023-12-08T16:45:22Z | cs.CL, cs.AI, cs.IR | The pre-trained language models are continually fine-tuned to better support
downstream applications. However, this operation may result in significant
performance degeneration on general tasks beyond the targeted domain. To
overcome this problem, we propose LM-Cocktail which enables the fine-tuned
model to stay resilient in general perspectives. Our method is conducted in the
form of model merging, where the fine-tuned language model is merged with the
pre-trained base model or the peer models from other domains through weighted
average. Despite simplicity, LM-Cocktail is surprisingly effective: the
resulted model is able to achieve a strong empirical performance in the whole
scope of general tasks while preserving a superior capacity in its targeted
domain. We conduct comprehensive experiments with LLama and BGE model on
popular benchmarks, including FLAN, MMLU, MTEB, whose results validate the
efficacy of our proposed method. The code and checkpoints are available at
https://github.com/FlagOpen/FlagEmbedding/tree/master/LM_Cocktail. | cs.CL | LM-Cocktail: Resilient Tuning of Language Models via Model Merging |
2023-11-22T17:14:18Z | http://arxiv.org/pdf/2311.17065v2 | http://arxiv.org/abs/2311.17065v2 | Rongxiang Wang, Felix Xiaozhu Lin | 2023-12-04T15:37:57Z | eess.AS, cs.CL, cs.LG | In contemporary speech understanding (SU), a sophisticated pipeline is
employed, encompassing the ingestion of streaming voice input. The pipeline
executes beam search iteratively, invoking a deep neural network to generate
tentative outputs (referred to as hypotheses) in an autoregressive manner.
Periodically, the pipeline assesses attention and Connectionist Temporal
Classification (CTC) scores.
This paper aims to enhance SU performance on edge devices with limited
resources. Adopting a hybrid strategy, our approach focuses on accelerating
on-device execution and offloading inputs surpassing the device's capacity.
While this approach is established, we tackle SU's distinctive challenges
through innovative techniques: (1) Late Contextualization: This involves the
parallel execution of a model's attentive encoder during input ingestion. (2)
Pilot Inference: Addressing temporal load imbalances in the SU pipeline, this
technique aims to mitigate them effectively. (3) Autoregression Offramps:
Decisions regarding offloading are made solely based on hypotheses, presenting
a novel approach.
These techniques are designed to seamlessly integrate with existing speech
models, pipelines, and frameworks, offering flexibility for independent or
combined application. Collectively, they form a hybrid solution for edge SU.
Our prototype, named XYZ, has undergone testing on Arm platforms featuring 6 to
8 cores, demonstrating state-of-the-art accuracy. Notably, it achieves a 2x
reduction in end-to-end latency and a corresponding 2x decrease in offloading
requirements. | eess.AS | Efficient Deep Speech Understanding at the Edge |
2023-11-22T16:12:42Z | http://arxiv.org/pdf/2311.13495v1 | http://arxiv.org/abs/2311.13495v1 | Colleen Farrelly, Yashbir Singh, Quincy A. Hathaway, Gunnar Carlsson, Ashok Choudhary, Rahul Paul, Gianfranco Doretto, Yassine Himeur, Shadi Atalls, Wathiq Mansoor | 2023-11-22T16:12:42Z | cs.CY, cs.CL, cs.LG | Institutional bias can impact patient outcomes, educational attainment, and
legal system navigation. Written records often reflect bias, and once bias is
identified; it is possible to refer individuals for training to reduce bias.
Many machine learning tools exist to explore text data and create predictive
models that can search written records to identify real-time bias. However, few
previous studies investigate large language model embeddings and geometric
models of biased text data to understand geometry's impact on bias modeling
accuracy. To overcome this issue, this study utilizes the RedditBias database
to analyze textual biases. Four transformer models, including BERT and RoBERTa
variants, were explored. Post-embedding, t-SNE allowed two-dimensional
visualization of data. KNN classifiers differentiated bias types, with lower
k-values proving more effective. Findings suggest BERT, particularly mini BERT,
excels in bias classification, while multilingual models lag. The
recommendation emphasizes refining monolingual models and exploring
domain-specific biases. | cs.CY | Current Topological and Machine Learning Applications for Bias Detection in Text |
2023-11-22T15:42:51Z | http://arxiv.org/pdf/2311.13475v1 | http://arxiv.org/abs/2311.13475v1 | Harshita Tyagi, Prashasta Jung, Hyowon Lee | 2023-11-22T15:42:51Z | cs.CL, cs.HC, cs.LG | Formality plays a significant role in language communication, especially in
low-resource languages such as Hindi, Japanese and Korean. These languages
utilise formal and informal expressions to convey messages based on social
contexts and relationships. When a language translation technique is used to
translate from a source language that does not pertain the formality (e.g.
English) to a target language that does, there is a missing information on
formality that could be a challenge in producing an accurate outcome. This
research explores how this issue should be resolved when machine learning
methods are used to translate from English to languages with formality, using
Hindi as the example data. This was done by training a bilingual model in a
formality-controlled setting and comparing its performance with a pre-trained
multilingual model in a similar setting. Since there are not a lot of training
data with ground truth, automated annotation techniques were employed to
increase the data size. The primary modeling approach involved leveraging
transformer models, which have demonstrated effectiveness in various natural
language processing tasks. We evaluate the official formality accuracy(ACC) by
comparing the predicted masked tokens with the ground truth. This metric
provides a quantitative measure of how well the translations align with the
desired outputs. Our study showcases a versatile translation strategy that
considers the nuances of formality in the target language, catering to diverse
language communication needs and scenarios. | cs.CL | Machine Translation to Control Formality Features in the Target Language |
2023-11-22T15:40:57Z | http://arxiv.org/pdf/2311.13472v1 | http://arxiv.org/abs/2311.13472v1 | Nidhi Vakil, Hadi Amiri | 2023-11-22T15:40:57Z | cs.CL, cs.AI | Curriculum learning provides a systematic approach to training. It refines
training progressively, tailors training to task requirements, and improves
generalization through exposure to diverse examples. We present a curriculum
learning approach that builds on existing knowledge about text and graph
complexity formalisms for training with text graph data. The core part of our
approach is a novel data scheduler, which employs "spaced repetition" and
complexity formalisms to guide the training process. We demonstrate the
effectiveness of the proposed approach on several text graph tasks and graph
neural network architectures. The proposed model gains more and uses less data;
consistently prefers text over graph complexity indices throughout training,
while the best curricula derived from text and graph complexity indices are
equally effective; and it learns transferable curricula across GNN models and
datasets. In addition, we find that both node-level (local) and graph-level
(global) graph complexity indices, as well as shallow and traditional text
complexity indices play a crucial role in effective curriculum learning. | cs.CL | Complexity-Guided Curriculum Learning for Text Graphs |
2023-11-22T15:22:04Z | http://arxiv.org/pdf/2311.13455v1 | http://arxiv.org/abs/2311.13455v1 | Yanyi Pu | 2023-11-22T15:22:04Z | cs.AI, cs.CL | This thesis delves into a fortiori arguments in deductive reasoning,
underscoring their relevance in various domains such as law, philosophy, and
artificial intelligence. The research is centred on employing GPT-3.5-turbo to
automate the analysis of these arguments, with a focus on understanding
intricate reasoning processes, generating clear and coherent explanations, and
creating novel arguments. The methodology encompasses a series of tasks
including detailed reasoning, interpretation, and the augmentation of a
fortiori arguments. It involves meticulously identifying these arguments in
diverse contexts, differentiating comparative elements, and categorizing them
based on their logical structure.
Extensive experiments reveals the challenges encountered by GPT-3.5-turbo in
accurately detecting and classifying a fortiori arguments. Nevertheless, the
model demonstrates a performance that rivals specialized models, particularly
in extracting key components and interpreting underlying properties. The
integration of external information into the model's processing significantly
elevates the quality of the generated explanations. Additionally, the model
exhibits a noteworthy capability in augmenting arguments, thus contributing to
the enrichment of the data set.
Despite facing certain limitations, this thesis makes significant
contributions to the fields of artificial intelligence and logical reasoning.
It introduces novel methodologies, establishes a rigorous evaluation framework,
and provides deep insights that set the stage for future advancements in
automated logical reasoning. The findings and methodologies presented herein
not only underscore the potential of AI in complex reasoning tasks but also
highlight areas for future research and development. | cs.AI | Generation of Explanations for Logic Reasoning |
2023-11-22T12:39:28Z | http://arxiv.org/pdf/2311.13350v1 | http://arxiv.org/abs/2311.13350v1 | Shubham Kumar Nigam, Aniket Deroy | 2023-11-22T12:39:28Z | cs.CL, cs.AI, cs.IR, cs.LG | This extended abstract extends the research presented in "ILDC for CJPE:
Indian Legal Documents Corpus for Court Judgment Prediction and Explanation"
\cite{malik-etal-2021-ildc}, focusing on fact-based judgment prediction within
the context of Indian legal documents. We introduce two distinct problem
variations: one based solely on facts, and another combining facts with rulings
from lower courts (RLC). Our research aims to enhance early-phase case outcome
prediction, offering significant benefits to legal professionals and the
general public. The results, however, indicated a performance decline compared
to the original ILDC for CJPE study, even after implementing various weightage
schemes in our DELSumm algorithm. Additionally, using only facts for legal
judgment prediction with different transformer models yielded results inferior
to the state-of-the-art outcomes reported in the "ILDC for CJPE" study. | cs.CL | Fact-based Court Judgment Prediction |
2023-11-22T11:08:38Z | http://arxiv.org/pdf/2311.13314v1 | http://arxiv.org/abs/2311.13314v1 | Xinyan Guan, Yanjiang Liu, Hongyu Lin, Yaojie Lu, Ben He, Xianpei Han, Le Sun | 2023-11-22T11:08:38Z | cs.CL | Incorporating factual knowledge in knowledge graph is regarded as a promising
approach for mitigating the hallucination of large language models (LLMs).
Existing methods usually only use the user's input to query the knowledge
graph, thus failing to address the factual hallucination generated by LLMs
during its reasoning process. To address this problem, this paper proposes
Knowledge Graph-based Retrofitting (KGR), a new framework that incorporates
LLMs with KGs to mitigate factual hallucination during the reasoning process by
retrofitting the initial draft responses of LLMs based on the factual knowledge
stored in KGs. Specifically, KGR leverages LLMs to extract, select, validate,
and retrofit factual statements within the model-generated responses, which
enables an autonomous knowledge verifying and refining procedure without any
additional manual efforts. Experiments show that KGR can significantly improve
the performance of LLMs on factual QA benchmarks especially when involving
complex reasoning processes, which demonstrates the necessity and effectiveness
of KGR in mitigating hallucination and enhancing the reliability of LLMs. | cs.CL | Mitigating Large Language Model Hallucinations via Autonomous Knowledge Graph-based Retrofitting |
2023-11-22T10:55:36Z | http://arxiv.org/pdf/2311.13307v2 | http://arxiv.org/abs/2311.13307v2 | Xiao Song, Jiafan Liu, Yun Li, Wenbin Lei, Ruxin Wang | 2023-12-05T10:25:23Z | cs.CV, cs.CL, cs.MM | Radiology Report Generation (RRG) draws attention as an interaction between
vision and language fields. Previous works inherited the ideology of
vision-to-language generation tasks,aiming to generate paragraphs with high
consistency as reports. However, one unique characteristic of RRG, the
independence between diseases, was neglected, leading to the injection of
disease co-occurrence as a confounder that effects the results through backdoor
path. Unfortunately, this confounder confuses the process of report generation
worse because of the biased RRG data distribution. In this paper, to rethink
this issue thoroughly, we reason about its causes and effects from a novel
perspective of statistics and causality, where the Joint Vision Coupling and
the Conditional Sentence Coherence Coupling are two aspects prone to implicitly
decrease the accuracy of reports. Then, a counterfactual augmentation strategy
that contains the Counterfactual Sample Synthesis and the Counterfactual Report
Reconstruction sub-methods is proposed to break these two aspects of spurious
effects. Experimental results and further analyses on two widely used datasets
justify our reasoning and proposed methods. | cs.CV | Rethinking Radiology Report Generation via Causal Reasoning and Counterfactual Augmentation |
2023-11-22T10:04:29Z | http://arxiv.org/pdf/2311.13281v1 | http://arxiv.org/abs/2311.13281v1 | Nick Goodson, Rongfei Lu | 2023-11-22T10:04:29Z | cs.CY, cs.CL | Large Language Models (LLMs) and chatbots show significant promise in
streamlining the legal intake process. This advancement can greatly reduce the
workload and costs for legal aid organizations, improving availability while
making legal assistance more accessible to a broader audience. However, a key
challenge with current LLMs is their tendency to overconfidently deliver an
immediate 'best guess' to a client's question based on the output distribution
learned over the training data. This approach often overlooks the client's
actual intentions or the specifics of their legal situation. As a result,
clients may not realize the importance of providing essential additional
context or expressing their underlying intentions, which are crucial for their
legal cases. Traditionally, logic based decision trees have been used to
automate intake for specific access to justice issues, such as immigration and
eviction. But those solutions lack scalability. We demonstrate a
proof-of-concept using LLMs to elicit and infer clients' underlying intentions
and specific legal circumstances through free-form, language-based
interactions. We also propose future research directions to use supervised
fine-tuning or offline reinforcement learning to automatically incorporate
intention and context elicitation in chatbots without explicit prompting. | cs.CY | Intention and Context Elicitation with Large Language Models in the Legal Aid Intake Process |
2023-11-22T09:51:53Z | http://arxiv.org/pdf/2311.13274v2 | http://arxiv.org/abs/2311.13274v2 | Daphne van Zandvoort, Laura Wiersema, Tom Huibers, Sandra van Dulmen, Sjaak Brinkkemper | 2024-01-19T10:06:50Z | cs.CL | Customized medical prompts enable Large Language Models (LLM) to effectively
address medical dialogue summarization. The process of medical reporting is
often time-consuming for healthcare professionals. Implementing medical
dialogue summarization techniques presents a viable solution to alleviate this
time constraint by generating automated medical reports. The effectiveness of
LLMs in this process is significantly influenced by the formulation of the
prompt, which plays a crucial role in determining the quality and relevance of
the generated reports. In this research, we used a combination of two distinct
prompting strategies, known as shot prompting and pattern prompting to enhance
the performance of automated medical reporting. The evaluation of the automated
medical reports is carried out using the ROUGE score and a human evaluation
with the help of an expert panel. The two-shot prompting approach in
combination with scope and domain context outperforms other methods and
achieves the highest score when compared to the human reference set by a
general practitioner. However, the automated reports are approximately twice as
long as the human references, due to the addition of both redundant and
relevant statements that are added to the report. | cs.CL | Enhancing Summarization Performance through Transformer-Based Prompt Engineering in Automated Medical Reporting |
2023-11-22T09:51:43Z | http://arxiv.org/pdf/2311.13273v2 | http://arxiv.org/abs/2311.13273v2 | Wouter Faber, Renske Eline Bootsma, Tom Huibers, Sandra van Dulmen, Sjaak Brinkkemper | 2024-01-08T14:19:29Z | cs.CL | Generative Artificial Intelligence (AI) can be used to automatically generate
medical reports based on transcripts of medical consultations. The aim is to
reduce the administrative burden that healthcare professionals face. The
accuracy of the generated reports needs to be established to ensure their
correctness and usefulness. There are several metrics for measuring the
accuracy of AI generated reports, but little work has been done towards the
application of these metrics in medical reporting. A comparative
experimentation of 10 accuracy metrics has been performed on AI generated
medical reports against their corresponding General Practitioner's (GP) medical
reports concerning Otitis consultations. The number of missing, incorrect, and
additional statements of the generated reports have been correlated with the
metric scores. In addition, we introduce and define a Composite Accuracy Score
which produces a single score for comparing the metrics within the field of
automated medical reporting. Findings show that based on the correlation study
and the Composite Accuracy Score, the ROUGE-L and Word Mover's Distance metrics
are the preferred metrics, which is not in line with previous work. These
findings help determine the accuracy of an AI generated medical report, which
aids the development of systems that generate medical reports for GPs to reduce
the administrative burden. | cs.CL | Comparative Experimentation of Accuracy Metrics in Automated Medical Reporting: The Case of Otitis Consultations |
2023-11-22T09:23:34Z | http://arxiv.org/pdf/2311.13258v1 | http://arxiv.org/abs/2311.13258v1 | Yangyi Chen, Xingyao Wang, Manling Li, Derek Hoiem, Heng Ji | 2023-11-22T09:23:34Z | cs.CV, cs.CL, cs.LG | State-of-the-art vision-language models (VLMs) still have limited performance
in structural knowledge extraction, such as relations between objects. In this
work, we present ViStruct, a training framework to learn VLMs for effective
visual structural knowledge extraction. Two novel designs are incorporated.
First, we propose to leverage the inherent structure of programming language to
depict visual structural information. This approach enables explicit and
consistent representation of visual structural information of multiple
granularities, such as concepts, relations, and events, in a well-organized
structured format. Second, we introduce curriculum-based learning for VLMs to
progressively comprehend visual structures, from fundamental visual concepts to
intricate event structures. Our intuition is that lower-level knowledge may
contribute to complex visual structure understanding. Furthermore, we compile
and release a collection of datasets tailored for visual structural knowledge
extraction. We adopt a weakly-supervised approach to directly generate visual
event structures from captions for ViStruct training, capitalizing on abundant
image-caption pairs from the web. In experiments, we evaluate ViStruct on
visual structure prediction tasks, demonstrating its effectiveness in improving
the understanding of visual structures. The code is public at
\url{https://github.com/Yangyi-Chen/vi-struct}. | cs.CV | ViStruct: Visual Structural Knowledge Extraction via Curriculum Guided Code-Vision Representation |
2023-11-22T09:06:11Z | http://arxiv.org/pdf/2311.14741v1 | http://arxiv.org/abs/2311.14741v1 | Oliver Bendel, Karim N'diaye | 2023-11-22T09:06:11Z | cs.CL, cs.AI, cs.RO, I.2; K.3 | Dead, extinct, and endangered languages have been preserved primarily through
audio conservation and the collection and digitization of scripts and have been
promoted through targeted language acquisition efforts. Another possibility
would be to build conversational agents that can master these languages. This
would provide an artificial, active conversational partner which has knowledge
of the vocabulary and grammar, and one learns with it in a different way. The
chatbot @ve, with which one can communicate in Latin, was developed in
2022/2023 based on GPT-3.0. It was additionally equipped with a manually
created knowledge base. After conceptual groundwork, this paper presents the
preparation and implementation of the project. In addition, it summarizes the
test that a Latin expert conducted with the chatbot. A critical discussion
elaborates advantages and disadvantages. @ve could be a new tool for teaching
Latin in a memorable and entertaining way through dialogue. However, the
present implementation is still too prone to glitches for stand-alone use -
i.e., without the accompaniment of a teacher. The use of GPT-4 could be a
solution as well as the extension of the knowledge base. In conclusion, it can
be argued that conversational agents are an innovative approach to promoting
and preserving languages. | cs.CL | @ve: A Chatbot for Latin |
2023-11-22T09:04:57Z | http://arxiv.org/pdf/2311.13246v2 | http://arxiv.org/abs/2311.13246v2 | Yilun Liu, Shimin Tao, Xiaofeng Zhao, Ming Zhu, Wenbing Ma, Junhao Zhu, Chang Su, Yutai Hou, Miao Zhang, Min Zhang, Hongxia Ma, Li Zhang, Hao Yang, Yanfei Jiang | 2024-03-21T03:50:32Z | cs.CL | Instruction tuning is crucial for enabling Language Learning Models (LLMs) in
responding to human instructions. The quality of instruction pairs used for
tuning greatly affects the performance of LLMs. However, the manual creation of
high-quality instruction datasets is costly, leading to the adoption of
automatic generation of instruction pairs by LLMs as a popular alternative. To
ensure the high quality of LLM-generated instruction datasets, several
approaches have been proposed. Nevertheless, existing methods either compromise
dataset integrity by filtering a large proportion of samples, or are unsuitable
for industrial applications. In this paper, instead of discarding low-quality
samples, we propose CoachLM, a novel approach to enhance the quality of
instruction datasets through automatic revisions on samples in the dataset.
CoachLM is trained from the samples revised by human experts and significantly
increases the proportion of high-quality samples in the dataset from 17.7% to
78.9%. The effectiveness of CoachLM is further assessed on various real-world
instruction test sets. The results show that CoachLM improves the
instruction-following capabilities of the instruction-tuned LLM by an average
of 29.9%, which even surpasses larger LLMs with nearly twice the number of
parameters. Furthermore, CoachLM is successfully deployed in a data management
system for LLMs at Huawei, resulting in an efficiency improvement of up to 20%
in the cleaning of 40k real-world instruction pairs. We release various assets
of CoachLM, including the training data, code and test set
(https://github.com/lunyiliu/CoachLM). | cs.CL | CoachLM: Automatic Instruction Revisions Improve the Data Quality in LLM Instruction Tuning |
2023-11-22T08:58:25Z | http://arxiv.org/pdf/2311.14740v1 | http://arxiv.org/abs/2311.14740v1 | Bohan Chen, Andrea L. Bertozzi | 2023-11-22T08:58:25Z | cs.CL | Traditional methods of linking large language models (LLMs) to knowledge
bases via the semantic similarity search often fall short of capturing complex
relational dynamics. To address these limitations, we introduce AutoKG, a
lightweight and efficient approach for automated knowledge graph (KG)
construction. For a given knowledge base consisting of text blocks, AutoKG
first extracts keywords using a LLM and then evaluates the relationship weight
between each pair of keywords using graph Laplace learning. We employ a hybrid
search scheme combining vector similarity and graph-based associations to
enrich LLM responses. Preliminary experiments demonstrate that AutoKG offers a
more comprehensive and interconnected knowledge retrieval mechanism compared to
the semantic similarity search, thereby enhancing the capabilities of LLMs in
generating more insightful and relevant outputs. | cs.CL | AutoKG: Efficient Automated Knowledge Graph Generation for Language Models |
2023-11-22T08:57:55Z | http://arxiv.org/pdf/2311.13240v1 | http://arxiv.org/abs/2311.13240v1 | Chiwei Zhu, Benfeng Xu, Quan Wang, Yongdong Zhang, Zhendong Mao | 2023-11-22T08:57:55Z | cs.CL | As large language models attract increasing attention and find widespread
application, concurrent challenges of reliability also arise at the same time.
Confidence calibration, an effective analysis method for gauging the
reliability of deep models, serves as a crucial tool for assessing and
improving their reliability. However, such investigation has been comparatively
underexplored. In this work, we conduct a systematic examination of the
calibration of aligned language models throughout the entire construction
process, including pretraining and alignment training. At each stage, we
investigate how different training settings, such as parameter scales and
training data, affect model calibration. To thoroughly assess model
calibration, we evaluate models on three most concerned aspects: generation,
factuality and understanding. Our work sheds light on whether popular LLMs are
well-calibrated and how the training process influences model calibration. | cs.CL | On the Calibration of Large Language Models and Alignment |
2023-11-22T08:39:17Z | http://arxiv.org/pdf/2311.13230v1 | http://arxiv.org/abs/2311.13230v1 | Tianhang Zhang, Lin Qiu, Qipeng Guo, Cheng Deng, Yue Zhang, Zheng Zhang, Chenghu Zhou, Xinbing Wang, Luoyi Fu | 2023-11-22T08:39:17Z | cs.CL, cs.AI | Large Language Models (LLMs) have gained significant popularity for their
impressive performance across diverse fields. However, LLMs are prone to
hallucinate untruthful or nonsensical outputs that fail to meet user
expectations in many real-world applications. Existing works for detecting
hallucinations in LLMs either rely on external knowledge for reference
retrieval or require sampling multiple responses from the LLM for consistency
verification, making these methods costly and inefficient. In this paper, we
propose a novel reference-free, uncertainty-based method for detecting
hallucinations in LLMs. Our approach imitates human focus in factuality
checking from three aspects: 1) focus on the most informative and important
keywords in the given text; 2) focus on the unreliable tokens in historical
context which may lead to a cascade of hallucinations; and 3) focus on the
token properties such as token type and token frequency. Experimental results
on relevant datasets demonstrate the effectiveness of our proposed method,
which achieves state-of-the-art performance across all the evaluation metrics
and eliminates the need for additional information. | cs.CL | Enhancing Uncertainty-Based Hallucination Detection with Stronger Focus |
2023-11-22T08:06:22Z | http://arxiv.org/pdf/2312.03716v1 | http://arxiv.org/abs/2312.03716v1 | Bowen Xing, Ivor W. Tsang | 2023-11-22T08:06:22Z | cs.CL, cs.AI | Recent graph-based models for multi-intent SLU have obtained promising
results through modeling the guidance from the prediction of intents to the
decoding of slot filling. However, existing methods (1) only model the
unidirectional guidance from intent to slot, while there are bidirectional
inter-correlations between intent and slot; (2) adopt homogeneous graphs to
model the interactions between the slot semantics nodes and intent label nodes,
which limit the performance. In this paper, we propose a novel model termed
Co-guiding Net, which implements a two-stage framework achieving the mutual
guidances between the two tasks. In the first stage, the initial estimated
labels of both tasks are produced, and then they are leveraged in the second
stage to model the mutual guidances. Specifically, we propose two heterogeneous
graph attention networks working on the proposed two heterogeneous semantics
label graphs, which effectively represent the relations among the semantics
nodes and label nodes. Besides, we further propose Co-guiding-SCL Net, which
exploits the single-task and dual-task semantics contrastive relations. For the
first stage, we propose single-task supervised contrastive learning, and for
the second stage, we propose co-guiding supervised contrastive learning, which
considers the two tasks' mutual guidances in the contrastive learning
procedure. Experiment results on multi-intent SLU show that our model
outperforms existing models by a large margin, obtaining a relative improvement
of 21.3% over the previous best model on MixATIS dataset in overall accuracy.
We also evaluate our model on the zero-shot cross-lingual scenario and the
results show that our model can relatively improve the state-of-the-art model
by 33.5% on average in terms of overall accuracy for the total 9 languages. | cs.CL | Co-guiding for Multi-intent Spoken Language Understanding |
2023-11-22T06:23:18Z | http://arxiv.org/pdf/2311.13184v3 | http://arxiv.org/abs/2311.13184v3 | Xingyu Wu, Yan Zhong, Jibin Wu, Bingbing Jiang, Kay Chen Tan | 2024-05-16T02:54:25Z | cs.LG, cs.CL | Algorithm selection, a critical process of automated machine learning, aims
to identify the most suitable algorithm for solving a specific problem prior to
execution. Mainstream algorithm selection techniques heavily rely on problem
features, while the role of algorithm features remains largely unexplored. Due
to the intrinsic complexity of algorithms, effective methods for universally
extracting algorithm information are lacking. This paper takes a significant
step towards bridging this gap by introducing Large Language Models (LLMs) into
algorithm selection for the first time. By comprehending the code text, LLM not
only captures the structural and semantic aspects of the algorithm, but also
demonstrates contextual awareness and library function understanding. The
high-dimensional algorithm representation extracted by LLM, after undergoing a
feature selection module, is combined with the problem representation and
passed to the similarity calculation module. The selected algorithm is
determined by the matching degree between a given problem and different
algorithms. Extensive experiments validate the performance superiority of the
proposed model and the efficacy of each key module. Furthermore, we present a
theoretical upper bound on model complexity, showcasing the influence of
algorithm representation and feature selection modules. This provides valuable
theoretical guidance for the practical implementation of our method. | cs.LG | Large Language Model-Enhanced Algorithm Selection: Towards Comprehensive Algorithm Representation |
2023-11-22T05:28:59Z | http://arxiv.org/pdf/2311.13171v1 | http://arxiv.org/abs/2311.13171v1 | Prateek Yadav, Leshem Choshen, Colin Raffel, Mohit Bansal | 2023-11-22T05:28:59Z | cs.LG, cs.AI, cs.CL | Parameter-efficient fine-tuning (PEFT) techniques make it possible to
efficiently adapt a language model to create "expert" models that specialize to
new tasks or domains. Recent techniques in model merging and compositional
generalization leverage these expert models by dynamically composing modules to
improve zero/few-shot generalization. Despite the efficiency of PEFT methods,
the size of expert models can make it onerous to retrieve expert models per
query over high-latency networks like the Internet or serve multiple experts on
a single GPU. To address these issues, we present ComPEFT, a novel method for
compressing fine-tuning residuals (task vectors) of PEFT based models. ComPEFT
employs sparsification and ternary quantization to reduce the size of the PEFT
module without performing any additional retraining while preserving or
enhancing model performance. In extensive evaluation across T5, T0, and
LLaMA-based models with 200M - 65B parameters, ComPEFT achieves compression
ratios of 8x - 50x. In particular, we show that ComPEFT improves with scale -
stronger models exhibit higher compressibility and better performance. For
example, we show that ComPEFT applied to LLaMA outperforms QLoRA by 4.16% on
MMLU with a storage size reduction of up to 26x. In addition, we show that the
compressed experts produced by ComPEFT maintain few-shot compositional
generalization capabilities, facilitate efficient communication and
computation, and exhibit enhanced performance when merged. Lastly, we provide
an analysis of different method components, compare it with other PEFT methods,
and test ComPEFT's efficacy for compressing the residual of full-finetuning.
Our code is available at https://github.com/prateeky2806/compeft. | cs.LG | ComPEFT: Compression for Communicating Parameter Efficient Updates via Sparsification and Quantization |
2023-11-22T03:37:01Z | http://arxiv.org/pdf/2311.13133v1 | http://arxiv.org/abs/2311.13133v1 | Aditi Jha, Sam Havens, Jeremy Dohmann, Alex Trott, Jacob Portes | 2023-11-22T03:37:01Z | cs.LG, cs.AI, cs.CL | Large Language Models are traditionally finetuned on large instruction
datasets. However recent studies suggest that small, high-quality datasets can
suffice for general purpose instruction following. This lack of consensus
surrounding finetuning best practices is in part due to rapidly diverging
approaches to LLM evaluation. In this study, we ask whether a small amount of
diverse finetuning samples can improve performance on both traditional
perplexity-based NLP benchmarks, and on open-ended, model-based evaluation. We
finetune open-source MPT-7B and MPT-30B models on instruction finetuning
datasets of various sizes ranging from 1k to 60k samples. We find that subsets
of 1k-6k instruction finetuning samples are sufficient to achieve good
performance on both (1) traditional NLP benchmarks and (2) model-based
evaluation. Finally, we show that mixing textbook-style and open-ended QA
finetuning datasets optimizes performance on both evaluation paradigms. | cs.LG | LIMIT: Less Is More for Instruction Tuning Across Evaluation Paradigms |
2023-11-22T03:36:18Z | http://arxiv.org/pdf/2311.16173v2 | http://arxiv.org/abs/2311.16173v2 | Changnan Xiao, Bing Liu | 2023-12-06T16:31:50Z | cs.AI, cs.CL, cs.LG | Reasoning is a fundamental capability of AI agents. Recently, large language
models (LLMs) have shown remarkable abilities to perform reasoning tasks.
However, numerous evaluations of the reasoning capabilities of LLMs have also
showed some limitations. An outstanding limitation is length generalization,
meaning that when trained on reasoning problems of smaller lengths or sizes,
the resulting models struggle with problems of larger sizes or lengths. This
potentially indicates some theoretical limitations of generalization in
learning reasoning skills. These evaluations and their observations motivated
us to perform a theoretical study of the length generalization problem. This
work focuses on reasoning tasks that can be formulated as Markov dynamic
processes (MDPs) and/or directed acyclic graphs (DAGs). It identifies and
proves conditions that decide whether the length generalization problem can be
solved or not for a reasoning task in a particular representation. Experiments
are also conducted to verify the theoretical results. | cs.AI | Conditions for Length Generalization in Learning Reasoning Skills |
2023-11-22T03:28:34Z | http://arxiv.org/pdf/2311.13126v1 | http://arxiv.org/abs/2311.13126v1 | Chengyu Wang, Junbing Yan, Wei Zhang, Jun Huang | 2023-11-22T03:28:34Z | cs.CL | This paper delves into the pressing need in Parameter-Efficient Fine-Tuning
(PEFT) for Large Language Models (LLMs). While LLMs possess remarkable
capabilities, their extensive parameter requirements and associated
computational demands hinder their practicality and scalability for real-world
applications. Our position paper highlights current states and the necessity of
further studying into the topic, and recognizes significant challenges and open
issues that must be addressed to fully harness the powerful abilities of LLMs.
These challenges encompass novel efficient PEFT architectures, PEFT for
different learning settings, PEFT combined with model compression techniques,
and the exploration of PEFT for multi-modal LLMs. By presenting this position
paper, we aim to stimulate further research and foster discussions surrounding
more efficient and accessible PEFT for LLMs. | cs.CL | Towards Better Parameter-Efficient Fine-Tuning for Large Language Models: A Position Paper |
2023-11-22T02:45:01Z | http://arxiv.org/pdf/2311.13118v1 | http://arxiv.org/abs/2311.13118v1 | Alejandro Rodriguez Perez, Pablo Rivas | 2023-11-22T02:45:01Z | cs.LG, cs.AI, cs.CL, cs.CY, cs.SI, 68T50, 62H30, 91C99, 68T068T50, 62H30, 91C99, 68T01, I.2.7; I.5.4; K.4.1; K.4.2 | This project tackles the pressing issue of human trafficking in online C2C
marketplaces through advanced Natural Language Processing (NLP) techniques. We
introduce a novel methodology for generating pseudo-labeled datasets with
minimal supervision, serving as a rich resource for training state-of-the-art
NLP models. Focusing on tasks like Human Trafficking Risk Prediction (HTRP) and
Organized Activity Detection (OAD), we employ cutting-edge Transformer models
for analysis. A key contribution is the implementation of an interpretability
framework using Integrated Gradients, providing explainable insights crucial
for law enforcement. This work not only fills a critical gap in the literature
but also offers a scalable, machine learning-driven approach to combat human
exploitation online. It serves as a foundation for future research and
practical applications, emphasizing the role of machine learning in addressing
complex social issues. | cs.LG | Combatting Human Trafficking in the Cyberspace: A Natural Language Processing-Based Methodology to Analyze the Language in Online Advertisements |
2023-11-22T02:23:32Z | http://arxiv.org/pdf/2311.13110v2 | http://arxiv.org/abs/2311.13110v2 | Yaodong Yu, Sam Buchanan, Druv Pai, Tianzhe Chu, Ziyang Wu, Shengbang Tong, Hao Bai, Yuexiang Zhai, Benjamin D. Haeffele, Yi Ma | 2023-11-24T09:18:44Z | cs.LG, cs.CL, cs.CV | In this paper, we contend that a natural objective of representation learning
is to compress and transform the distribution of the data, say sets of tokens,
towards a low-dimensional Gaussian mixture supported on incoherent subspaces.
The goodness of such a representation can be evaluated by a principled measure,
called sparse rate reduction, that simultaneously maximizes the intrinsic
information gain and extrinsic sparsity of the learned representation. From
this perspective, popular deep network architectures, including transformers,
can be viewed as realizing iterative schemes to optimize this measure.
Particularly, we derive a transformer block from alternating optimization on
parts of this objective: the multi-head self-attention operator compresses the
representation by implementing an approximate gradient descent step on the
coding rate of the features, and the subsequent multi-layer perceptron
sparsifies the features. This leads to a family of white-box transformer-like
deep network architectures, named CRATE, which are mathematically fully
interpretable. We show, by way of a novel connection between denoising and
compression, that the inverse to the aforementioned compressive encoding can be
realized by the same class of CRATE architectures. Thus, the so-derived
white-box architectures are universal to both encoders and decoders.
Experiments show that these networks, despite their simplicity, indeed learn to
compress and sparsify representations of large-scale real-world image and text
datasets, and achieve performance very close to highly engineered
transformer-based models: ViT, MAE, DINO, BERT, and GPT2. We believe the
proposed computational framework demonstrates great potential in bridging the
gap between theory and practice of deep learning, from a unified perspective of
data compression. Code is available at: https://ma-lab-berkeley.github.io/CRATE . | cs.LG | White-Box Transformers via Sparse Rate Reduction: Compression Is All There Is? |
2023-11-22T02:13:27Z | http://arxiv.org/pdf/2312.09457v1 | http://arxiv.org/abs/2312.09457v1 | Arturo N. Villanueva Jr., Steven J. Simske | 2023-11-22T02:13:27Z | cs.CL, cs.IR, K.3.2 | We propose multiple techniques for automatic document order generation for
(1) curriculum development and for (2) creation of optimal reading order for
use in learning, training, and other content-sequencing applications. Such
techniques could potentially be used to improve comprehension, identify areas
that need expounding, generate curricula, and improve search engine results. We
advance two main techniques: The first uses document similarities through
various methods. The second uses entropy against the backdrop of topics
generated through Latent Dirichlet Allocation (LDA). In addition, we try the
same methods on the summarized documents and compare them against the results
obtained using the complete documents. Our results showed that while the
document orders for our control document sets (biographies, novels, and
Wikipedia articles) could not be predicted using our methods, our test
documents (textbooks, courses, journal papers, dissertations) provided more
reliability. We also demonstrated that summarized documents were good stand-ins
for the complete documents for the purposes of ordering. | cs.CL | Functional Analytics for Document Ordering for Curriculum Development and Comprehension |
2023-11-22T02:12:36Z | http://arxiv.org/pdf/2311.13105v1 | http://arxiv.org/abs/2311.13105v1 | Pablo Loyola, Edison Marrese-Taylor, Andres Hoyos-Idobro | 2023-11-22T02:12:36Z | cs.CL | The need for grounding in language understanding is an active research topic.
Previous work has suggested that color perception and color language appear as
a suitable test bed to empirically study the problem, given its cognitive
significance and showing that there is considerable alignment between a defined
color space and the feature space defined by a language model. To further study
this issue, we collect a large scale source of colors and their descriptions,
containing almost a 1 million examples , and perform an empirical analysis to
compare two kinds of alignments: (i) inter-space, by learning a mapping between
embedding space and color space, and (ii) intra-space, by means of prompting
comparatives between color descriptions. Our results show that while color
space alignment holds for monolexemic, highly pragmatic color descriptions,
this alignment drops considerably in the presence of examples that exhibit
elements of real linguistic usage such as subjectivity and abstractedness,
suggesting that grounding may be required in such cases. | cs.CL | Perceptual Structure in the Absence of Grounding for LLMs: The Impact of Abstractedness and Subjectivity in Color Language |
2023-11-22T02:04:35Z | http://arxiv.org/pdf/2311.13102v1 | http://arxiv.org/abs/2311.13102v1 | Andres Pollano, Anupam Chaudhuri, Anj Simmons | 2023-11-22T02:04:35Z | cs.CL, cs.LG, math.AT | We attempt to detect out-of-distribution (OOD) text samples though applying
Topological Data Analysis (TDA) to attention maps in transformer-based language
models. We evaluate our proposed TDA-based approach for out-of-distribution
detection on BERT, a transformer-based language model, and compare the to a
more traditional OOD approach based on BERT CLS embeddings. We found that our
TDA approach outperforms the CLS embedding approach at distinguishing
in-distribution data (politics and entertainment news articles from HuffPost)
from far out-of-domain samples (IMDB reviews), but its effectiveness
deteriorates with near out-of-domain (CNN/Dailymail) or same-domain (business
news articles from HuffPost) datasets. | cs.CL | Detecting out-of-distribution text using topological features of transformer-based language models |
2023-11-22T01:51:50Z | http://arxiv.org/pdf/2311.13095v1 | http://arxiv.org/abs/2311.13095v1 | Ha-Thanh Nguyen, Wachara Fungwacharakorn, Ken Satoh | 2023-11-22T01:51:50Z | cs.CL, cs.AI | Language serves as a vehicle for conveying thought, enabling communication
among individuals. The ability to distinguish between diverse concepts,
identify fairness and injustice, and comprehend a range of legal notions
fundamentally relies on logical reasoning. Large Language Models (LLMs) attempt
to emulate human language understanding and generation, but their competency in
logical reasoning remains limited. This paper seeks to address the
philosophical question: How can we effectively teach logical reasoning to LLMs
while maintaining a deep understanding of the intricate relationship between
language and logic? By focusing on bolstering LLMs' capabilities in logical
reasoning, we aim to expand their applicability in law and other
logic-intensive disciplines. To this end, we propose a Reinforcement Learning
from Logical Feedback (RLLF) approach, which serves as a potential framework
for refining LLMs' reasoning capacities. Through RLLF and a revised evaluation
methodology, we explore new avenues for research in this domain and contribute
to the development of LLMs capable of handling complex legal reasoning tasks
while acknowledging the fundamental connection between language and logic. | cs.CL | Enhancing Logical Reasoning in Large Language Models to Facilitate Legal Applications |
2023-11-22T00:31:01Z | http://arxiv.org/pdf/2311.14737v1 | http://arxiv.org/abs/2311.14737v1 | Ruoqi Shen, Sébastien Bubeck, Ronen Eldan, Yin Tat Lee, Yuanzhi Li, Yi Zhang | 2023-11-22T00:31:01Z | cs.CL, cs.AI, cs.LG | Transformers, central to the successes in modern Natural Language Processing,
often falter on arithmetic tasks despite their vast capabilities --which
paradoxically include remarkable coding abilities. We observe that a crucial
challenge is their naive reliance on positional information to solve arithmetic
problems with a small number of digits, leading to poor performance on larger
numbers. Herein, we delve deeper into the role of positional encoding, and
propose several ways to fix the issue, either by modifying the positional
encoding directly, or by modifying the representation of the arithmetic task to
leverage standard positional encoding differently. We investigate the value of
these modifications for three tasks: (i) classical multiplication, (ii) length
extrapolation in addition, and (iii) addition in natural language context. For
(i) we train a small model on a small dataset (100M parameters and 300k
samples) with remarkable aptitude in (direct, no scratchpad) 15 digits
multiplication and essentially perfect up to 12 digits, while usual training in
this context would give a model failing at 4 digits multiplication. In the
experiments on addition, we use a mere 120k samples to demonstrate: for (ii)
extrapolation from 10 digits to testing on 12 digits numbers while usual
training would have no extrapolation, and for (iii) almost perfect accuracy up
to 5 digits while usual training would be correct only up to 3 digits (which is
essentially memorization with a training set of 120k samples). | cs.CL | Positional Description Matters for Transformers Arithmetic |
2023-11-21T23:50:33Z | http://arxiv.org/pdf/2311.13061v1 | http://arxiv.org/abs/2311.13061v1 | Aron Molnar, Jaap Jumelet, Mario Giulianelli, Arabella Sinclair | 2023-11-21T23:50:33Z | cs.CL | Language models are often used as the backbone of modern dialogue systems.
These models are pre-trained on large amounts of written fluent language.
Repetition is typically penalised when evaluating language model generations.
However, it is a key component of dialogue. Humans use local and partner
specific repetitions; these are preferred by human users and lead to more
successful communication in dialogue. In this study, we evaluate (a) whether
language models produce human-like levels of repetition in dialogue, and (b)
what are the processing mechanisms related to lexical re-use they use during
comprehension. We believe that such joint analysis of model production and
comprehension behaviour can inform the development of cognitively inspired
dialogue generation systems. | cs.CL | Attribution and Alignment: Effects of Local Context Repetition on Utterance Production and Comprehension in Dialogue |
2023-11-21T23:26:05Z | http://arxiv.org/pdf/2311.13053v1 | http://arxiv.org/abs/2311.13053v1 | Xiao Liu, Jianfeng Lin, Jiawei Zhang | 2023-11-21T23:26:05Z | cs.CL | The proliferation of Large Language Models like ChatGPT has significantly
advanced language understanding and generation, impacting a broad spectrum of
applications. However, these models predominantly excel in text-based tasks,
overlooking the complexity of real-world multimodal information. This study
introduces MultiAPI, a pioneering comprehensive large-scale API benchmark
dataset aimed at expanding LLMs' proficiency in multimodal contexts. Developed
collaboratively through ChatGPT, MultiAPI consists of 235 diverse API calls and
2,038 contextual prompts, offering a unique platform evaluation of
tool-augmented LLMs handling multimodal tasks. Through comprehensive
experiments, our findings reveal that while LLMs demonstrate proficiency in API
call decision-making, they face challenges in domain identification, function
selection, and argument generation. What's more, we surprisingly notice that
auxiliary context can actually impair the performance. An in-depth error
analysis paves the way for a new paradigm to address these challenges,
suggesting a potential direction for future LLM research. | cs.CL | Beyond Text: Unveiling Multimodal Proficiency of Large Language Models with MultiAPI Benchmark |
2023-11-21T22:30:37Z | http://arxiv.org/pdf/2311.13029v1 | http://arxiv.org/abs/2311.13029v1 | Lei Yu | 2023-11-21T22:30:37Z | cs.CL | The meaning of polysemous words often varies in a highly productive yet
predictable way. Generalizing the regularity between conventional senses to
derive novel word meaning is crucial for automated processing of non-literal
language uses such as figurative expressions. We introduce a novel task called
systematic word meta-sense extension (SWORME) to test and improve language
models' ability to extend word meaning to denote new semantic domains (also
called meta-senses) that bear regular semantic relations with existing senses.
We found that language models prefer incremental lexical semantic change toward
conceptually similar meta-senses such as logical metonymy, and are much worse
at predicting highly non-literal meaning extensions such as metaphors. We
propose a novel analogy-based method of word meaning extension, and show that
it effectively improves language model systematicity in making both gradual and
radical types of meta-sense extension. We further demonstrate that learning
systematic meta-sense extensions benefits language models on multiple
benchmarks of figurative language understanding. | cs.CL | Systematic word meta-sense extension |
2023-11-21T20:45:55Z | http://arxiv.org/pdf/2311.12986v2 | http://arxiv.org/abs/2311.12986v2 | Abdelfateh Bekkair, Slimane Bellaouar, Slimane Oulad-Naoui | 2023-11-24T22:24:50Z | cs.CL, cs.AI, 68T07, I.2.4 | Several natural phenomena and complex systems are often represented as
networks. Discovering their community structure is a fundamental task for
understanding these networks. Many algorithms have been proposed, but recently,
Graph Neural Networks (GNN) have emerged as a compelling approach for enhancing
this task.In this paper, we introduce a simple, efficient, and
clustering-oriented model based on unsupervised \textbf{G}raph Attention
\textbf{A}uto\textbf{E}ncoder for community detection in attributed networks
(GAECO). The proposed model adeptly learns representations from both the
network's topology and attribute information, simultaneously addressing dual
objectives: reconstruction and community discovery. It places a particular
emphasis on discovering compact communities by robustly minimizing clustering
errors. The model employs k-means as an objective function and utilizes a
multi-head Graph Attention Auto-Encoder for decoding the representations.
Experiments conducted on three datasets of attributed networks show that our
method surpasses state-of-the-art algorithms in terms of NMI and ARI.
Additionally, our approach scales effectively with the size of the network,
making it suitable for large-scale applications. The implications of our
findings extend beyond biological network interpretation and social network
analysis, where knowledge of the fundamental community structure is essential. | cs.CL | Unsupervised Graph Attention Autoencoder for Attributed Networks using K-means Loss |
2023-11-21T20:34:47Z | http://arxiv.org/pdf/2311.12983v1 | http://arxiv.org/abs/2311.12983v1 | Grégoire Mialon, Clémentine Fourrier, Craig Swift, Thomas Wolf, Yann LeCun, Thomas Scialom | 2023-11-21T20:34:47Z | cs.CL, cs.AI | We introduce GAIA, a benchmark for General AI Assistants that, if solved,
would represent a milestone in AI research. GAIA proposes real-world questions
that require a set of fundamental abilities such as reasoning, multi-modality
handling, web browsing, and generally tool-use proficiency. GAIA questions are
conceptually simple for humans yet challenging for most advanced AIs: we show
that human respondents obtain 92\% vs. 15\% for GPT-4 equipped with plugins.
This notable performance disparity contrasts with the recent trend of LLMs
outperforming humans on tasks requiring professional skills in e.g. law or
chemistry. GAIA's philosophy departs from the current trend in AI benchmarks
suggesting to target tasks that are ever more difficult for humans. We posit
that the advent of Artificial General Intelligence (AGI) hinges on a system's
capability to exhibit similar robustness as the average human does on such
questions. Using GAIA's methodology, we devise 466 questions and their answer.
We release our questions while retaining answers to 300 of them to power a
leader-board available at https://huggingface.co/gaia-benchmark. | cs.CL | GAIA: a benchmark for General AI Assistants |
2023-11-21T19:12:18Z | http://arxiv.org/pdf/2311.14736v2 | http://arxiv.org/abs/2311.14736v2 | Alexander Bukharin, Tuo Zhao | 2024-02-05T16:41:10Z | cs.CL, cs.LG | Recent works have shown that by curating high quality and diverse instruction
tuning datasets, we can significantly improve instruction-following
capabilities. However, creating such datasets is difficult and most works rely
on manual curation or proprietary language models. Automatic data curation is
difficult as it is still not clear how we can define diversity for instruction
tuning, how diversity and quality depend on one other, and how we can optimize
dataset quality and diversity. To resolve these issue, we propose a new
algorithm, Quality-Diversity Instruction Tuning (QDIT). QDIT provides a simple
method to simultaneously control dataset diversity and quality, allowing us to
conduct an in-depth study on the effect of diversity and quality on instruction
tuning performance. From this study we draw two key insights (1) there is a
natural tradeoff between data diversity and quality and (2) increasing data
diversity significantly improves the worst case instruction following
performance, therefore improving robustness. We validate the performance of
QDIT on several large scale instruction tuning datasets, where we find it can
substantially improve worst and average case performance compared to
quality-driven data selection. | cs.CL | Data Diversity Matters for Robust Instruction Tuning |
2023-11-21T18:41:26Z | http://arxiv.org/pdf/2311.14743v7 | http://arxiv.org/abs/2311.14743v7 | Will LeVine, Benjamin Pikus, Anthony Chen, Sean Hendryx | 2024-01-24T12:08:34Z | cs.CL, cs.LG | Foundation models, specifically Large Language Models (LLMs), have lately
gained wide-spread attention and adoption. Reinforcement Learning with Human
Feedback (RLHF) involves training a reward model to capture desired behaviors,
which is then used to align LLM's. These reward models are additionally used at
inference-time to estimate LLM responses' adherence to those desired behaviors.
However, there is little work measuring how robust these reward models are to
distribution shifts. In this work, we evaluate how reward model performance -
measured via accuracy and calibration (i.e. alignment between accuracy and
confidence) - is affected by distribution shift. We show novel calibration
patterns and accuracy drops due to OOD prompts and responses, and that the
reward model is more sensitive to shifts in responses than prompts.
Additionally, we adapt an OOD detection technique commonly used in
classification to the reward model setting to detect these distribution shifts
in prompts and responses. | cs.CL | A Baseline Analysis of Reward Models' Ability To Accurately Analyze Foundation Models Under Distribution Shift |
2023-11-21T17:21:15Z | http://arxiv.org/pdf/2311.12735v1 | http://arxiv.org/abs/2311.12735v1 | Aunabil Chakma, Masum Hasan | 2023-11-21T17:21:15Z | cs.CL | This paper describes the system of the LowResource Team for Task 2 of
BLP-2023, which involves conducting sentiment analysis on a dataset composed of
public posts and comments from diverse social media platforms. Our primary aim
is to utilize BanglaBert, a BERT model pre-trained on a large Bangla corpus,
using various strategies including fine-tuning, dropping random tokens, and
using several external datasets. Our final model is an ensemble of the three
best BanglaBert variations. Our system has achieved overall 3rd in the Test Set
among 30 participating teams with a score of 0.718. Additionally, we discuss
the promising systems that didn't perform well namely task-adaptive pertaining
and paraphrasing using BanglaT5. Training codes and external datasets which are
used for our system are publicly available at
https://github.com/Aunabil4602/bnlp-workshop-task2-2023 | cs.CL | LowResource at BLP-2023 Task 2: Leveraging BanglaBert for Low Resource Sentiment Analysis of Bangla Language |
2023-11-21T17:03:21Z | http://arxiv.org/pdf/2311.12727v2 | http://arxiv.org/abs/2311.12727v2 | Xiaodong Cui, Ashish Mittal, Songtao Lu, Wei Zhang, George Saon, Brian Kingsbury | 2023-11-24T03:27:31Z | cs.LG, cs.CL | Soft random sampling (SRS) is a simple yet effective approach for efficient
training of large-scale deep neural networks when dealing with massive data.
SRS selects a subset uniformly at random with replacement from the full data
set in each epoch. In this paper, we conduct a theoretical and empirical
analysis of SRS. First, we analyze its sampling dynamics including data
coverage and occupancy. Next, we investigate its convergence with non-convex
objective functions and give the convergence rate. Finally, we provide its
generalization performance. We empirically evaluate SRS for image recognition
on CIFAR10 and automatic speech recognition on Librispeech and an in-house
payload dataset to demonstrate its effectiveness. Compared to existing
coreset-based data selection methods, SRS offers a better accuracy-efficiency
trade-off. Especially on real-world industrial scale data sets, it is shown to
be a powerful training strategy with significant speedup and competitive
performance with almost no additional computing cost. | cs.LG | Soft Random Sampling: A Theoretical and Empirical Analysis |
2023-11-21T16:20:49Z | http://arxiv.org/pdf/2311.12707v1 | http://arxiv.org/abs/2311.12707v1 | Hye Sun Yun, Mehdi Arjmand, Phillip Raymond Sherlock, Michael Paasche-Orlow, James W. Griffith, Timothy Bickmore | 2023-11-21T16:20:49Z | cs.HC, cs.AI, cs.CL | Standardized, validated questionnaires are vital tools in HCI research and
healthcare, offering dependable self-report data. However, their repeated use
in longitudinal or pre-post studies can induce respondent fatigue, impacting
data quality via response biases and decreased response rates. We propose
utilizing large language models (LLMs) to generate diverse questionnaire
versions while retaining good psychometric properties. In a longitudinal study,
participants engaged with our agent system and responded daily for two weeks to
either a standardized depression questionnaire or one of two LLM-generated
questionnaire variants, alongside a validated depression questionnaire.
Psychometric testing revealed consistent covariation between the external
criterion and the focal measure administered across the three conditions,
demonstrating the reliability and validity of the LLM-generated variants.
Participants found the repeated administration of the standardized
questionnaire significantly more repetitive compared to the variants. Our
findings highlight the potential of LLM-generated variants to invigorate
questionnaires, fostering engagement and interest without compromising
validity. | cs.HC | Keeping Users Engaged During Repeated Administration of the Same Questionnaire: Using Large Language Models to Reliably Diversify Questions |
2023-11-21T16:03:51Z | http://arxiv.org/pdf/2311.12699v1 | http://arxiv.org/abs/2311.12699v1 | Mengyang Chen, Lingwei Wei, Han Cao, Wei Zhou, Songlin Hu | 2023-11-21T16:03:51Z | cs.CL, cs.AI, cs.CY | Large Language Models (LLMs) have garnered significant attention for their
powerful ability in natural language understanding and reasoning. In this
paper, we present a comprehensive empirical study to explore the performance of
LLMs on misinformation detection tasks. This study stands as the pioneering
investigation into the understanding capabilities of multiple LLMs regarding
both content and propagation across social media platforms. Our empirical
studies on five misinformation detection datasets show that LLMs with diverse
prompts achieve comparable performance in text-based misinformation detection
but exhibit notably constrained capabilities in comprehending propagation
structure compared to existing models in propagation-based misinformation
detection. Besides, we further design four instruction-tuned strategies to
enhance LLMs for both content and propagation-based misinformation detection.
These strategies boost LLMs to actively learn effective features from multiple
instances or hard instances, and eliminate irrelevant propagation structures,
thereby achieving better detection performance. Extensive experiments further
demonstrate LLMs would play a better capacity in content and propagation
structure under these proposed strategies and achieve promising detection
performance. These findings highlight the potential ability of LLMs to detect
misinformation. | cs.CL | Can Large Language Models Understand Content and Propagation for Misinformation Detection: An Empirical Study |
2023-11-21T15:51:06Z | http://arxiv.org/pdf/2311.12689v1 | http://arxiv.org/abs/2311.12689v1 | Thibaud Leteno, Antoine Gourru, Charlotte Laclau, Rémi Emonet, Christophe Gravier | 2023-11-21T15:51:06Z | cs.CL, cs.CY, cs.LG | Group fairness is a central research topic in text classification, where
reaching fair treatment between sensitive groups (e.g. women vs. men) remains
an open challenge. This paper presents a novel method for mitigating biases in
neural text classification, agnostic to the model architecture. Considering the
difficulty to distinguish fair from unfair information in a text encoder, we
take inspiration from adversarial training to induce Wasserstein independence
between representations learned to predict our target label and the ones
learned to predict some sensitive attribute. Our approach provides two
significant advantages. Firstly, it does not require annotations of sensitive
attributes in both testing and training data. This is more suitable for
real-life scenarios compared to existing methods that require annotations of
sensitive attributes at train time. Second, our approach exhibits a comparable
or better fairness-accuracy trade-off compared to existing methods. | cs.CL | Fair Text Classification with Wasserstein Independence |
2023-11-21T15:14:54Z | http://arxiv.org/pdf/2311.12664v2 | http://arxiv.org/abs/2311.12664v2 | Dominik Schlechtweg, Shafqat Mumtaz Virk, Pauline Sander, Emma Sköldberg, Lukas Theuer Linke, Tuo Zhang, Nina Tahmasebi, Jonas Kuhn, Sabine Schulte im Walde | 2024-02-05T12:50:23Z | cs.CL, cs.AI | We present the DURel tool that implements the annotation of semantic
proximity between uses of words into an online, open source interface. The tool
supports standardized human annotation as well as computational annotation,
building on recent advances with Word-in-Context models. Annotator judgments
are clustered with automatic graph clustering techniques and visualized for
analysis. This allows to measure word senses with simple and intuitive
micro-task judgments between use pairs, requiring minimal preparation efforts.
The tool offers additional functionalities to compare the agreement between
annotators to guarantee the inter-subjectivity of the obtained judgments and to
calculate summary statistics giving insights into sense frequency
distributions, semantic variation or changes of senses over time. | cs.CL | The DURel Annotation Tool: Human and Computational Measurement of Semantic Proximity, Sense Clusters and Semantic Change |
2023-11-21T14:49:00Z | http://arxiv.org/pdf/2311.12649v1 | http://arxiv.org/abs/2311.12649v1 | Lucy Horowitz, Valeria de Paiva | 2023-11-21T14:49:00Z | cs.CL | MathGloss is a project to create a knowledge graph (KG) for undergraduate
mathematics from text, automatically, using modern natural language processing
(NLP) tools and resources already available on the web. MathGloss is a linked
database of undergraduate concepts in mathematics. So far, it combines five
resources: (i) Wikidata, a collaboratively edited, multilingual knowledge graph
hosted by the Wikimedia Foundation, (ii) terms covered in mathematics courses
at the University of Chicago, (iii) the syllabus of the French undergraduate
mathematics curriculum which includes hyperlinks to the automated theorem
prover Lean 4, (iv) MuLiMa, a multilingual dictionary of mathematics curated by
mathematicians, and (v) the nLab, a wiki for category theory also curated by
mathematicians. MathGloss's goal is to bring together resources for learning
mathematics and to allow every mathematician to tailor their learning to their
own preferences. Moreover, by organizing different resources for learning
undergraduate mathematics alongside those for learning formal mathematics, we
hope to make it easier for mathematicians and formal tools (theorem provers,
computer algebra systems, etc) experts to "understand" each other and break
down some of the barriers to formal math. | cs.CL | MathGloss: Building mathematical glossaries from text |
2023-11-21T12:40:01Z | http://arxiv.org/pdf/2311.12574v1 | http://arxiv.org/abs/2311.12574v1 | Michal Spiegel, Dominik Macko | 2023-11-21T12:40:01Z | cs.CL, cs.AI | In the era of large language models generating high quality texts, it is a
necessity to develop methods for detection of machine-generated text to avoid
harmful use or simply due to annotation purposes. It is, however, also
important to properly evaluate and compare such developed methods. Recently, a
few benchmarks have been proposed for this purpose; however, integration of
newest detection methods is rather challenging, since new methods appear each
month and provide slightly different evaluation pipelines. In this paper, we
present the IMGTB framework, which simplifies the benchmarking of
machine-generated text detection methods by easy integration of custom (new)
methods and evaluation datasets. Its configurability and flexibility makes
research and development of new detection methods easier, especially their
comparison to the existing state-of-the-art detectors. The default set of
analyses, metrics and visualizations offered by the tool follows the
established practices of machine-generated text detection benchmarking found in
state-of-the-art literature. | cs.CL | IMGTB: A Framework for Machine-Generated Text Detection Benchmarking |
2023-11-21T11:33:03Z | http://arxiv.org/pdf/2311.12538v2 | http://arxiv.org/abs/2311.12538v2 | David Oniani, Yanshan Wang | 2023-11-22T08:44:34Z | cs.LG, cs.AI, cs.CL | Large Language Models (LLMs) have proven effective at In-Context Learning
(ICL), an ability that allows them to create predictors from labeled examples.
Few studies have explored the interplay between ICL and specific properties of
functions it attempts to approximate. In our study, we use a formal framework
to explore ICL and propose a new task of approximating functions with varying
number of minima. We implement a method that allows for producing functions
with given inputs as minima. We find that increasing the number of minima
degrades ICL performance. At the same time, our evaluation shows that ICL
outperforms 2-layer Neural Network (2NN) model. Furthermore, ICL learns faster
than 2NN in all settings. We validate the findings through a set of few-shot
experiments across various hyperparameter configurations. | cs.LG | In-Context Learning Functions with Varying Number of Minima |
2023-11-21T11:32:23Z | http://arxiv.org/pdf/2311.12537v1 | http://arxiv.org/abs/2311.12537v1 | Tong Zhou, Yubo Chen, Pengfei Cao, Kang Liu, Jun Zhao, Shengping Liu | 2023-11-21T11:32:23Z | cs.CL, cs.AI | Data is one of the most critical elements in building a large language model.
However, existing systems either fail to customize a corpus curation pipeline
or neglect to leverage comprehensive corpus assessment for iterative
optimization of the curation. To this end, we present a pretraining corpus
curation and assessment platform called Oasis -- a one-stop system for data
quality improvement and quantification with user-friendly interactive
interfaces. Specifically, the interactive modular rule filter module can devise
customized rules according to explicit feedback. The debiased neural filter
module builds the quality classification dataset in a negative-centric manner
to remove the undesired bias. The adaptive document deduplication module could
execute large-scale deduplication with limited memory resources. These three
parts constitute the customized data curation module. And in the holistic data
assessment module, a corpus can be assessed in local and global views, with
three evaluation means including human, GPT-4, and heuristic metrics. We
exhibit a complete process to use Oasis for the curation and assessment of
pretraining data. In addition, an 800GB bilingual corpus curated by Oasis is
publicly released. | cs.CL | Oasis: Data Curation and Assessment System for Pretraining of Large Language Models |
2023-11-21T11:26:26Z | http://arxiv.org/pdf/2311.12534v1 | http://arxiv.org/abs/2311.12534v1 | Simone Filice, Jason Ingyu Choi, Giuseppe Castellucci, Eugene Agichtein, Oleg Rokhlenko | 2023-11-21T11:26:26Z | cs.CL | Many Natural Language Generation (NLG) tasks aim to generate a single output
text given an input prompt. Other settings require the generation of multiple
texts, e.g., for Synthetic Traffic Generation (STG). This generation task is
crucial for training and evaluating QA systems as well as conversational
agents, where the goal is to generate multiple questions or utterances
resembling the linguistic variability of real users. In this paper, we show
that common NLG metrics, like BLEU, are not suitable for evaluating STG. We
propose and evaluate several metrics designed to compare the generated traffic
to the distribution of real user texts. We validate our metrics with an
automatic procedure to verify whether they capture different types of quality
issues of generated data; we also run human annotations to verify the
correlation with human judgements. Experiments on three tasks, i.e., Shopping
Utterance Generation, Product Question Generation and Query Auto Completion,
demonstrate that our metrics are effective for evaluating STG tasks, and
improve the agreement with human judgement up to 20% with respect to common NLG
metrics. We believe these findings can pave the way towards better solutions
for estimating the representativeness of synthetic text data. | cs.CL | Evaluation Metrics of Language Generation Models for Synthetic Traffic Generation Tasks |
2023-11-21T09:59:29Z | http://arxiv.org/pdf/2311.12489v1 | http://arxiv.org/abs/2311.12489v1 | Viktor Hangya, Silvia Severini, Radoslav Ralev, Alexander Fraser, Hinrich Schütze | 2023-11-21T09:59:29Z | cs.CL | Very low-resource languages, having only a few million tokens worth of data,
are not well-supported by multilingual NLP approaches due to poor quality
cross-lingual word representations. Recent work showed that good cross-lingual
performance can be achieved if a source language is related to the low-resource
target language. However, not all language pairs are related. In this paper, we
propose to build multilingual word embeddings (MWEs) via a novel language
chain-based approach, that incorporates intermediate related languages to
bridge the gap between the distant source and target. We build MWEs one
language at a time by starting from the resource rich source and sequentially
adding each language in the chain till we reach the target. We extend a
semi-joint bilingual approach to multiple languages in order to eliminate the
main weakness of previous works, i.e., independently trained monolingual
embeddings, by anchoring the target language around the multilingual space. We
evaluate our method on bilingual lexicon induction for 4 language families,
involving 4 very low-resource (<5M tokens) and 4 moderately low-resource (<50M)
target languages, showing improved performance in both categories.
Additionally, our analysis reveals the importance of good quality embeddings
for intermediate languages as well as the importance of leveraging anchor
points from all languages in the multilingual space. | cs.CL | Multilingual Word Embeddings for Low-Resource Languages using Anchors and a Chain of Related Languages |
2023-11-21T09:44:33Z | http://arxiv.org/pdf/2311.12480v1 | http://arxiv.org/abs/2311.12480v1 | David Gimeno-Gómez, Carlos-D. Martínez-Hinarejos | 2023-11-21T09:44:33Z | cs.CV, cs.CL, cs.SD, eess.AS | Different studies have shown the importance of visual cues throughout the
speech perception process. In fact, the development of audiovisual approaches
has led to advances in the field of speech technologies. However, although
noticeable results have recently been achieved, visual speech recognition
remains an open research problem. It is a task in which, by dispensing with the
auditory sense, challenges such as visual ambiguities and the complexity of
modeling silence must be faced. Nonetheless, some of these challenges can be
alleviated when the problem is approached from a speaker-dependent perspective.
Thus, this paper studies, using the Spanish LIP-RTVE database, how the
estimation of specialized end-to-end systems for a specific person could affect
the quality of speech recognition. First, different adaptation strategies based
on the fine-tuning technique were proposed. Then, a pre-trained CTC/Attention
architecture was used as a baseline throughout our experiments. Our findings
showed that a two-step fine-tuning process, where the VSR system is first
adapted to the task domain, provided significant improvements when the speaker
adaptation was addressed. Furthermore, results comparable to the current state
of the art were reached even when only a limited amount of data was available. | cs.CV | Speaker-Adapted End-to-End Visual Speech Recognition for Continuous Spanish |
2023-11-21T09:37:42Z | http://arxiv.org/pdf/2311.12475v2 | http://arxiv.org/abs/2311.12475v2 | Panyut Sriwirote, Jalinee Thapiang, Vasan Timtong, Attapol T. Rutherford | 2023-12-28T19:22:12Z | cs.CL, cs.AI | While WangchanBERTa has become the de facto standard in transformer-based
Thai language modeling, it still has shortcomings in regard to the
understanding of foreign words, most notably English words, which are often
borrowed without orthographic assimilation into Thai in many contexts. We
identify the lack of foreign vocabulary in WangchanBERTa's tokenizer as the
main source of these shortcomings. We then expand WangchanBERTa's vocabulary
via vocabulary transfer from XLM-R's pretrained tokenizer and pretrain a new
model using the expanded tokenizer, starting from WangchanBERTa's checkpoint,
on a new dataset that is larger than the one used to train WangchanBERTa. Our
results show that our new pretrained model, PhayaThaiBERT, outperforms
WangchanBERTa in many downstream tasks and datasets. | cs.CL | PhayaThaiBERT: Enhancing a Pretrained Thai Language Model with Unassimilated Loanwords |
2023-11-21T09:36:11Z | http://arxiv.org/pdf/2311.12474v1 | http://arxiv.org/abs/2311.12474v1 | Wojciech Kusa, Oscar E. Mendoza, Matthias Samwald, Petr Knoth, Allan Hanbury | 2023-11-21T09:36:11Z | cs.CL, cs.IR | Systematic literature reviews (SLRs) play an essential role in summarising,
synthesising and validating scientific evidence. In recent years, there has
been a growing interest in using machine learning techniques to automate the
identification of relevant studies for SLRs. However, the lack of standardised
evaluation datasets makes comparing the performance of such automated
literature screening systems difficult. In this paper, we analyse the citation
screening evaluation datasets, revealing that many of the available datasets
are either too small, suffer from data leakage or have limited applicability to
systems treating automated literature screening as a classification task, as
opposed to, for example, a retrieval or question-answering task. To address
these challenges, we introduce CSMeD, a meta-dataset consolidating nine
publicly released collections, providing unified access to 325 SLRs from the
fields of medicine and computer science. CSMeD serves as a comprehensive
resource for training and evaluating the performance of automated citation
screening models. Additionally, we introduce CSMeD-FT, a new dataset designed
explicitly for evaluating the full text publication screening task. To
demonstrate the utility of CSMeD, we conduct experiments and establish
baselines on new datasets. | cs.CL | CSMeD: Bridging the Dataset Gap in Automated Citation Screening for Systematic Literature Reviews |
2023-11-21T09:28:00Z | http://arxiv.org/pdf/2311.12468v1 | http://arxiv.org/abs/2311.12468v1 | David Gimeno-Gómez, Carlos-D. Martínez-Hinarejos | 2023-11-21T09:28:00Z | cs.CV, cs.CL | During a conversation, our brain is responsible for combining information
obtained from multiple senses in order to improve our ability to understand the
message we are perceiving. Different studies have shown the importance of
presenting visual information in these situations. Nevertheless, lipreading is
a complex task whose objective is to interpret speech when audio is not
available. By dispensing with a sense as crucial as hearing, it will be
necessary to be aware of the challenge that this lack presents. In this paper,
we propose an analysis of different speech visual features with the intention
of identifying which of them is the best approach to capture the nature of lip
movements for natural Spanish and, in this way, dealing with the automatic
visual speech recognition task. In order to estimate our system, we present an
audiovisual corpus compiled from a subset of the RTVE database, which has been
used in the Albayz\'in evaluations. We employ a traditional system based on
Hidden Markov Models with Gaussian Mixture Models. Results show that, although
the task is difficult, in restricted conditions we obtain recognition results
which determine that using eigenlips in combination with deep features is the
best visual approach. | cs.CV | Analysis of Visual Features for Continuous Lipreading in Spanish |
2023-11-21T09:12:21Z | http://arxiv.org/pdf/2311.12457v1 | http://arxiv.org/abs/2311.12457v1 | David Gimeno-Gómez, Carlos-D. Martínez-Hinarejos | 2023-11-21T09:12:21Z | cs.CV, cs.CL | Speech is considered as a multi-modal process where hearing and vision are
two fundamentals pillars. In fact, several studies have demonstrated that the
robustness of Automatic Speech Recognition systems can be improved when audio
and visual cues are combined to represent the nature of speech. In addition,
Visual Speech Recognition, an open research problem whose purpose is to
interpret speech by reading the lips of the speaker, has been a focus of
interest in the last decades. Nevertheless, in order to estimate these systems
in the currently Deep Learning era, large-scale databases are required. On the
other hand, while most of these databases are dedicated to English, other
languages lack sufficient resources. Thus, this paper presents a
semi-automatically annotated audiovisual database to deal with unconstrained
natural Spanish, providing 13 hours of data extracted from Spanish television.
Furthermore, baseline results for both speaker-dependent and
speaker-independent scenarios are reported using Hidden Markov Models, a
traditional paradigm that has been widely used in the field of Speech
Technologies. | cs.CV | LIP-RTVE: An Audiovisual Database for Continuous Spanish in the Wild |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.