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
|
|---|---|---|---|---|---|---|---|---|
2022-12-23T16:29:34Z | http://arxiv.org/pdf/2212.12454v1 | http://arxiv.org/abs/2212.12454v1 | Yuting Guo, Swati Rajwal, Sahithi Lakamana, Chia-Chun Chiang, Paul C. Menell, Adnan H. Shahid, Yi-Chieh Chen, Nikita Chhabra, Wan-Ju Chao, Chieh-Ju Chao, Todd J. Schwedt, Imon Banerjee, Abeed Sarker | 2022-12-23T16:29:34Z | cs.CL | Migraine is a high-prevalence and disabling neurological disorder. However,
information migraine management in real-world settings could be limited to
traditional health information sources. In this paper, we (i) verify that there
is substantial migraine-related chatter available on social media (Twitter and
Reddit), self-reported by migraine sufferers; (ii) develop a
platform-independent text classification system for automatically detecting
self-reported migraine-related posts, and (iii) conduct analyses of the
self-reported posts to assess the utility of social media for studying this
problem. We manually annotated 5750 Twitter posts and 302 Reddit posts. Our
system achieved an F1 score of 0.90 on Twitter and 0.93 on Reddit. Analysis of
information posted by our 'migraine cohort' revealed the presence of a plethora
of relevant information about migraine therapies and patient sentiments
associated with them. Our study forms the foundation for conducting an in-depth
analysis of migraine-related information using social media data. | cs.CL | Generalizable Natural Language Processing Framework for Migraine Reporting from Social Media |
2022-12-23T14:36:43Z | http://arxiv.org/pdf/2212.12363v1 | http://arxiv.org/abs/2212.12363v1 | Zhitong Yang, Xing Ma, Anqi Liu, Zheyu Zhang | 2022-12-23T14:36:43Z | cs.CL | Task-oriented dialog(TOD) aims to assist users in achieving specific goals
through multi-turn conversation. Recently, good results have been obtained
based on large pre-trained models. However, the labeled-data scarcity hinders
the efficient development of TOD systems at scale. In this work, we constructed
a weakly supervised dataset based on a teacher/student paradigm that leverages
a large collection of unlabelled dialogues. Furthermore, we built a modular
dialogue system and integrated coarse-to-fine grained classification for user
intent detection. Experiments show that our method can reach the dialog goal
with a higher success rate and generate more coherent responses. | cs.CL | Discovering Customer-Service Dialog System with Semi-Supervised Learning and Coarse-to-Fine Intent Detection |
2022-12-23T10:20:58Z | http://arxiv.org/pdf/2212.12238v1 | http://arxiv.org/abs/2212.12238v1 | Oren Sultan, Rayen Dhahri, Yauheni Mardan, Tobias Eder, Georg Groh | 2022-12-23T10:20:58Z | cs.CL, cs.AI | Key Point Analysis(KPA) is a relatively new task in NLP that combines
summarization and classification by extracting argumentative key points (KPs)
for a topic from a collection of texts and categorizing their closeness to the
different arguments. In our work, we focus on the legal domain and develop
methods that identify and extract KPs from premises derived from texts of
judgments. The first method is an adaptation to an existing state-of-the-art
method, and the two others are new methods that we developed from scratch. We
present our methods and examples of their outputs, as well a comparison between
them. The full evaluation of our results is done in the matching task -- match
between the generated KPs to arguments (premises). | cs.CL | From Judgement's Premises Towards Key Points |
2022-12-23T08:59:30Z | http://arxiv.org/pdf/2212.12213v1 | http://arxiv.org/abs/2212.12213v1 | Ishita Goyal, Priyank Bhandia, Sanjana Dulam | 2022-12-23T08:59:30Z | cs.CL, cs.AI | Sarcasm is a form of irony that involves saying or writing something that is
opposite or opposite to what one really means, often in a humorous or mocking
way. It is often used to mock or mock someone or something, or to be humorous
or amusing. Sarcasm is usually conveyed through tone of voice, facial
expressions, or other forms of nonverbal communication, but it can also be
indicated by the use of certain words or phrases that are typically associated
with irony or humor. Sarcasm detection is difficult because it relies on
context and non-verbal cues. It can also be culturally specific, subjective and
ambiguous. In this work, we fine-tune the RoBERTa based sarcasm detection model
presented in Abaskohi et al. [2022] to get to within 0.02 F1 of the
state-of-the-art (Hercog et al. [2022]) on the iSarcasm dataset (Oprea and
Magdy [2019]). This performance is achieved by augmenting iSarcasm with a
pruned version of the Self Annotated Reddit Corpus (SARC) (Khodak et al.
[2017]). Our pruned version is 100 times smaller than the subset of SARC used
to train the state-of-the-art model. | cs.CL | Finetuning for Sarcasm Detection with a Pruned Dataset |
2022-12-23T08:23:32Z | http://arxiv.org/pdf/2212.12192v3 | http://arxiv.org/abs/2212.12192v3 | Minh-Tien Nguyen, Nhung Bui, Manh Tran-Tien, Linh Le, Huy-The Vu | 2024-03-15T16:03:46Z | cs.CL | Patent classification is the task that assigns each input patent into several
codes (classes). Due to its high demand, several datasets and methods have been
introduced. However, the lack of both systematic performance comparison of
baselines and access to some datasets creates a gap for the task. To fill the
gap, we introduce two new datasets in English and Japanese collected by using
CPC codes. The English dataset includes 45,131 patent documents with 425 labels
and the Japanese dataset contains 54,657 documents with 523 labels. To
facilitate the next studies, we compare the performance of strong multi-label
text classification methods on the two datasets. Experimental results show that
AttentionXML is consistently better than other strong baselines. The ablation
study is also conducted in two aspects: the contribution of different parts
(title, abstract, description, and claims) of a patent and the behavior of
baselines in terms of performance with different training data segmentation. We
release the two new datasets with the code of the baselines. | cs.CL | CinPatent: Datasets for Patent Classification |
2022-12-23T04:12:52Z | http://arxiv.org/pdf/2212.12137v1 | http://arxiv.org/abs/2212.12137v1 | William Brannon, Yogesh Virkar, Brian Thompson | 2022-12-23T04:12:52Z | cs.CL | We investigate how humans perform the task of dubbing video content from one
language into another, leveraging a novel corpus of 319.57 hours of video from
54 professionally produced titles. This is the first such large-scale study we
are aware of. The results challenge a number of assumptions commonly made in
both qualitative literature on human dubbing and machine-learning literature on
automatic dubbing, arguing for the importance of vocal naturalness and
translation quality over commonly emphasized isometric (character length) and
lip-sync constraints, and for a more qualified view of the importance of
isochronic (timing) constraints. We also find substantial influence of the
source-side audio on human dubs through channels other than the words of the
translation, pointing to the need for research on ways to preserve speech
characteristics, as well as semantic transfer such as emphasis/emotion, in
automatic dubbing systems. | cs.CL | Dubbing in Practice: A Large Scale Study of Human Localization With Insights for Automatic Dubbing |
2022-12-23T03:57:54Z | http://arxiv.org/pdf/2212.12131v1 | http://arxiv.org/abs/2212.12131v1 | Byung-Doh Oh, William Schuler | 2022-12-23T03:57:54Z | cs.CL | This work presents a detailed linguistic analysis into why larger
Transformer-based pre-trained language models with more parameters and lower
perplexity nonetheless yield surprisal estimates that are less predictive of
human reading times. First, regression analyses show a strictly monotonic,
positive log-linear relationship between perplexity and fit to reading times
for the more recently released five GPT-Neo variants and eight OPT variants on
two separate datasets, replicating earlier results limited to just GPT-2 (Oh et
al., 2022). Subsequently, analysis of residual errors reveals a systematic
deviation of the larger variants, such as underpredicting reading times of
named entities and making compensatory overpredictions for reading times of
function words such as modals and conjunctions. These results suggest that the
propensity of larger Transformer-based models to 'memorize' sequences during
training makes their surprisal estimates diverge from humanlike expectations,
which warrants caution in using pre-trained language models to study human
language processing. | cs.CL | Why Does Surprisal From Larger Transformer-Based Language Models Provide a Poorer Fit to Human Reading Times? |
2022-12-22T22:27:26Z | http://arxiv.org/pdf/2212.12061v3 | http://arxiv.org/abs/2212.12061v3 | Alina Petukhova, Nuno Fachada | 2023-04-23T14:49:44Z | cs.CL, cs.AI, cs.LG | This article presents a dataset of 10,917 news articles with hierarchical
news categories collected between 1 January 2019 and 31 December 2019. We
manually labeled the articles based on a hierarchical taxonomy with 17
first-level and 109 second-level categories. This dataset can be used to train
machine learning models for automatically classifying news articles by topic.
This dataset can be helpful for researchers working on news structuring,
classification, and predicting future events based on released news. | cs.CL | MN-DS: A Multilabeled News Dataset for News Articles Hierarchical Classification |
2022-12-22T21:48:29Z | http://arxiv.org/pdf/2212.12048v1 | http://arxiv.org/abs/2212.12048v1 | Pooja Chitkara, Morgane Riviere, Jade Copet, Frank Zhang, Yatharth Saraf | 2022-12-22T21:48:29Z | cs.CL, cs.SD, eess.AS | Speech to text models tend to be trained and evaluated against a single
target accent. This is especially true for English for which native speakers
from the United States became the main benchmark. In this work, we are going to
show how two simple methods: pre-trained embeddings and auxiliary
classification losses can improve the performance of ASR systems. We are
looking for upgrades as universal as possible and therefore we will explore
their impact on several models architectures and several languages. | cs.CL | Pushing the performances of ASR models on English and Spanish accents |
2022-12-22T21:27:12Z | http://arxiv.org/pdf/2212.12043v2 | http://arxiv.org/abs/2212.12043v2 | Yutaro Yamada, Yingtian Tang, Yoyo Zhang, Ilker Yildirim | 2024-04-13T17:02:25Z | cs.CV, cs.CL, cs.LG | Large-scale vision-language models such as CLIP have shown impressive
performance on zero-shot image classification and image-to-text retrieval.
However, such performance does not realize in tasks that require a
finer-grained correspondence between vision and language, such as Visual
Question Answering (VQA). As a potential cause of the difficulty of applying
these models to VQA and similar tasks, we report an interesting phenomenon of
vision-language models, which we call the Concept Association Bias (CAB). We
find that models with CAB tend to treat input as a bag of concepts and attempt
to fill in the other missing concept crossmodally, leading to an unexpected
zero-shot prediction. We demonstrate CAB by showing that CLIP's zero-shot
classification performance greatly suffers when there is a strong concept
association between an object (e.g. eggplant) and an attribute (e.g. color
purple). We also show that the strength of CAB predicts the performance on VQA.
We observe that CAB is prevalent in vision-language models trained with
contrastive losses, even when autoregressive losses are jointly employed.
However, a model that solely relies on autoregressive loss seems to exhibit
minimal or no signs of CAB. | cs.CV | When are Lemons Purple? The Concept Association Bias of Vision-Language Models |
2022-12-22T19:56:09Z | http://arxiv.org/pdf/2212.12017v3 | http://arxiv.org/abs/2212.12017v3 | Srinivasan Iyer, Xi Victoria Lin, Ramakanth Pasunuru, Todor Mihaylov, Daniel Simig, Ping Yu, Kurt Shuster, Tianlu Wang, Qing Liu, Punit Singh Koura, Xian Li, Brian O'Horo, Gabriel Pereyra, Jeff Wang, Christopher Dewan, Asli Celikyilmaz, Luke Zettlemoyer, Ves Stoyanov | 2023-01-30T05:25:59Z | cs.CL | Recent work has shown that fine-tuning large pre-trained language models on a
collection of tasks described via instructions, a.k.a. instruction-tuning,
improves their zero and few-shot generalization to unseen tasks. However, there
is a limited understanding of the performance trade-offs of different decisions
made during the instruction-tuning process. These decisions include the scale
and diversity of the instruction-tuning benchmark, different task sampling
strategies, fine-tuning with and without demonstrations, training using
specialized datasets for reasoning and dialogue, and finally, the fine-tuning
objectives themselves. In this paper, we characterize the effect of
instruction-tuning decisions on downstream task performance when scaling both
model and benchmark sizes. To this end, we create OPT-IML Bench: a large
benchmark for Instruction Meta-Learning (IML) of 2000 NLP tasks consolidated
into task categories from 8 existing benchmarks, and prepare an evaluation
framework to measure three types of model generalizations: to tasks from fully
held-out categories, to held-out tasks from seen categories, and to held-out
instances from seen tasks. Through the lens of this framework, we first present
insights about instruction-tuning decisions as applied to OPT-30B and further
exploit these insights to train OPT-IML 30B and 175B, which are
instruction-tuned versions of OPT. OPT-IML demonstrates all three
generalization abilities at both scales on four different evaluation benchmarks
with diverse tasks and input formats -- PromptSource, FLAN,
Super-NaturalInstructions, and UnifiedSKG. Not only does it significantly
outperform OPT on all benchmarks but is also highly competitive with existing
models fine-tuned on each specific benchmark. We release OPT-IML at both
scales, together with the OPT-IML Bench evaluation framework. | cs.CL | OPT-IML: Scaling Language Model Instruction Meta Learning through the Lens of Generalization |
2022-12-22T18:51:02Z | http://arxiv.org/pdf/2212.12442v1 | http://arxiv.org/abs/2212.12442v1 | Ehsan Variani, Ke Wu, David Rybach, Cyril Allauzen, Michael Riley | 2022-12-22T18:51:02Z | cs.CL, cs.LG | Existing training criteria in automatic speech recognition(ASR) permit the
model to freely explore more than one time alignments between the feature and
label sequences. In this paper, we use entropy to measure a model's
uncertainty, i.e. how it chooses to distribute the probability mass over the
set of allowed alignments. Furthermore, we evaluate the effect of entropy
regularization in encouraging the model to distribute the probability mass only
on a smaller subset of allowed alignments. Experiments show that entropy
regularization enables a much simpler decoding method without sacrificing word
error rate, and provides better time alignment quality. | cs.CL | Alignment Entropy Regularization |
2022-12-22T18:16:58Z | http://arxiv.org/pdf/2212.11937v1 | http://arxiv.org/abs/2212.11937v1 | Christopher J. MacLellan, Peter Matsakis, Pat Langley | 2022-12-22T18:16:58Z | cs.CL, cs.AI, cs.LG | This paper presents a novel approach to the acquisition of language models
from corpora. The framework builds on Cobweb, an early system for constructing
taxonomic hierarchies of probabilistic concepts that used a tabular,
attribute-value encoding of training cases and concepts, making it unsuitable
for sequential input like language. In response, we explore three new
extensions to Cobweb -- the Word, Leaf, and Path variants. These systems encode
each training case as an anchor word and surrounding context words, and they
store probabilistic descriptions of concepts as distributions over anchor and
context information. As in the original Cobweb, a performance element sorts a
new instance downward through the hierarchy and uses the final node to predict
missing features. Learning is interleaved with performance, updating concept
probabilities and hierarchy structure as classification occurs. Thus, the new
approaches process training cases in an incremental, online manner that it very
different from most methods for statistical language learning. We examine how
well the three variants place synonyms together and keep homonyms apart, their
ability to recall synonyms as a function of training set size, and their
training efficiency. Finally, we discuss related work on incremental learning
and directions for further research. | cs.CL | Efficient Induction of Language Models Via Probabilistic Concept Formation |
2022-12-22T16:42:21Z | http://arxiv.org/pdf/2212.11856v1 | http://arxiv.org/abs/2212.11856v1 | Víctor Suárez-Paniagua, Steven Derby, Tri Kurniawan Wijaya | 2022-12-22T16:42:21Z | cs.CL, cs.AI | Early detection of relevant locations in a piece of news is especially
important in extreme events such as environmental disasters, war conflicts,
disease outbreaks, or political turmoils. Additionally, this detection also
helps recommender systems to promote relevant news based on user locations.
Note that, when the relevant locations are not mentioned explicitly in the
text, state-of-the-art methods typically fail to recognize them because these
methods rely on syntactic recognition. In contrast, by incorporating a
knowledge base and connecting entities with their locations, our system
successfully infers the relevant locations even when they are not mentioned
explicitly in the text. To evaluate the effectiveness of our approach, and due
to the lack of datasets in this area, we also contribute to the research
community with a gold-standard multilingual news-location dataset, NewsLOC. It
contains the annotation of the relevant locations (and their WikiData IDs) of
600+ Wikinews articles in five different languages: English, French, German,
Italian, and Spanish. Through experimental evaluations, we show that our
proposed system outperforms the baselines and the fine-tuned version of the
model using semi-supervised data that increases the classification rate. The
source code and the NewsLOC dataset are publicly available for being used by
the research community at https://github.com/vsuarezpaniagua/NewsLocation. | cs.CL | Multilingual News Location Detection using an Entity-Based Siamese Network with Semi-Supervised Contrastive Learning and Knowledge Base |
2022-12-22T15:22:36Z | http://arxiv.org/pdf/2212.11791v1 | http://arxiv.org/abs/2212.11791v1 | Vahid Partovi Nia, Eyyüb Sari, Vanessa Courville, Masoud Asgharian | 2022-12-22T15:22:36Z | cs.LG, cs.CL, I.2.6 | Recurrent neural networks (RNN) are the backbone of many text and speech
applications. These architectures are typically made up of several
computationally complex components such as; non-linear activation functions,
normalization, bi-directional dependence and attention. In order to maintain
good accuracy, these components are frequently run using full-precision
floating-point computation, making them slow, inefficient and difficult to
deploy on edge devices. In addition, the complex nature of these operations
makes them challenging to quantize using standard quantization methods without
a significant performance drop. We present a quantization-aware training method
for obtaining a highly accurate integer-only recurrent neural network (iRNN).
Our approach supports layer normalization, attention, and an adaptive piecewise
linear (PWL) approximation of activation functions, to serve a wide range of
state-of-the-art RNNs. The proposed method enables RNN-based language models to
run on edge devices with $2\times$ improvement in runtime, and $4\times$
reduction in model size while maintaining similar accuracy as its
full-precision counterpart. | cs.LG | Training Integer-Only Deep Recurrent Neural Networks |
2022-12-22T14:27:33Z | http://arxiv.org/pdf/2301.06859v1 | http://arxiv.org/abs/2301.06859v1 | Thilo Hagendorff, Sarah Fabi | 2022-12-22T14:27:33Z | cs.HC, cs.AI, cs.CL, cs.LG | The field of artificial intelligence (AI) alignment aims to investigate
whether AI technologies align with human interests and values and function in a
safe and ethical manner. AI alignment is particularly relevant for large
language models (LLMs), which have the potential to exhibit unintended behavior
due to their ability to learn and adapt in ways that are difficult to predict.
In this paper, we discuss methodological challenges for the alignment problem
specifically in the context of LLMs trained to summarize texts. In particular,
we focus on methods for collecting reliable human feedback on summaries to
train a reward model which in turn improves the summarization model. We
conclude by suggesting specific improvements in the experimental design of
alignment studies for LLMs' summarization capabilities. | cs.HC | Methodological reflections for AI alignment research using human feedback |
2022-12-22T13:17:11Z | http://arxiv.org/pdf/2212.11685v2 | http://arxiv.org/abs/2212.11685v2 | Zhenghao Lin, Yeyun Gong, Yelong Shen, Tong Wu, Zhihao Fan, Chen Lin, Nan Duan, Weizhu Chen | 2023-02-17T17:14:52Z | cs.CL, cs.LG | In this paper, we introduce a novel dIffusion language modEl pre-training
framework for text generation, which we call GENIE. GENIE is a large-scale
pretrained diffusion language model that consists of an encoder and a
diffusion-based decoder, which can generate text by gradually transforming a
random noise sequence into a coherent text sequence. To pre-train GENIE on a
large-scale language corpus, we design a new continuous paragraph denoise
objective, which encourages the diffusion-decoder to reconstruct a clean text
paragraph from a corrupted version, while preserving the semantic and syntactic
coherence. We evaluate GENIE on four downstream text generation benchmarks,
namely XSum, CNN/DailyMail, Gigaword, and CommonGen. Our experimental results
show that GENIE achieves comparable performance with the state-of-the-art
autoregressive models on these benchmarks, and generates more diverse text
samples. The code and models of GENIE are available at
https://github.com/microsoft/ProphetNet/tree/master/GENIE. | cs.CL | Text Generation with Diffusion Language Models: A Pre-training Approach with Continuous Paragraph Denoise |
2022-12-22T07:06:35Z | http://arxiv.org/pdf/2212.11985v1 | http://arxiv.org/abs/2212.11985v1 | Noga Mudrik, Adam S. Charles | 2022-12-22T07:06:35Z | cs.CL | While recent advancements in artificial intelligence (AI) language models
demonstrate cutting-edge performance when working with English texts,
equivalent models do not exist in other languages or do not reach the same
performance level. This undesired effect of AI advancements increases the gap
between access to new technology from different populations across the world.
This unsought bias mainly discriminates against individuals whose English
skills are less developed, e.g., non-English speakers children. Following
significant advancements in AI research in recent years, OpenAI has recently
presented DALL-E: a powerful tool for creating images based on English text
prompts. While DALL-E is a promising tool for many applications, its decreased
performance when given input in a different language, limits its audience and
deepens the gap between populations. An additional limitation of the current
DALL-E model is that it only allows for the creation of a few images in
response to a given input prompt, rather than a series of consecutive coherent
frames that tell a story or describe a process that changes over time. Here, we
present an easy-to-use automatic DALL-E storytelling framework that leverages
the existing DALL-E model to enable fast and coherent visualizations of
non-English songs and stories, pushing the limit of the one-step-at-a-time
option DALL-E currently offers. We show that our framework is able to
effectively visualize stories from non-English texts and portray the changes in
the plot over time. It is also able to create a narrative and maintain
interpretable changes in the description across frames. Additionally, our
framework offers users the ability to specify constraints on the story
elements, such as a specific location or context, and to maintain a consistent
style throughout the visualization. | cs.CL | Multi-Lingual DALL-E Storytime |
2022-12-22T02:19:25Z | http://arxiv.org/pdf/2212.11456v1 | http://arxiv.org/abs/2212.11456v1 | Dan DeGenaro, Jugal Kalita | 2022-12-22T02:19:25Z | cs.CL | Large language models having hundreds of millions, and even billions, of
parameters have performed extremely well on a variety of natural language
processing (NLP) tasks. Their widespread use and adoption, however, is hindered
by the lack of availability and portability of sufficiently large computational
resources. This paper proposes a knowledge distillation (KD) technique building
on the work of LightMBERT, a student model of multilingual BERT (mBERT). By
repeatedly distilling mBERT through increasingly compressed toplayer distilled
teacher assistant networks, CAMeMBERT aims to improve upon the time and space
complexities of mBERT while keeping loss of accuracy beneath an acceptable
threshold. At present, CAMeMBERT has an average accuracy of around 60.1%, which
is subject to change after future improvements to the hyperparameters used in
fine-tuning. | cs.CL | CAMeMBERT: Cascading Assistant-Mediated Multilingual BERT |
2022-12-22T02:13:30Z | http://arxiv.org/pdf/2212.11455v1 | http://arxiv.org/abs/2212.11455v1 | Xuewen Yao, Miriam Mikhelson, Megan Micheletti, Eunsol Choi, S Craig Watkins, Edison Thomaz, Kaya De Barbaro | 2022-12-22T02:13:30Z | cs.CL | Digital platforms, including online forums and helplines, have emerged as
avenues of support for caregivers suffering from postpartum mental health
distress. Understanding support seekers' experiences as shared on these
platforms could provide crucial insight into caregivers' needs during this
vulnerable time. In the current work, we provide a descriptive analysis of the
concerns, psychological states, and motivations shared by healthy and
distressed postpartum support seekers on two digital platforms, a one-on-one
digital helpline and a publicly available online forum. Using a combination of
human annotations, dictionary models and unsupervised techniques, we find stark
differences between the experiences of distressed and healthy mothers.
Distressed mothers described interpersonal problems and a lack of support, with
8.60% - 14.56% reporting severe symptoms including suicidal ideation. In
contrast, the majority of healthy mothers described childcare issues, such as
questions about breastfeeding or sleeping, and reported no severe mental health
concerns. Across the two digital platforms, we found that distressed mothers
shared similar content. However, the patterns of speech and affect shared by
distressed mothers differed between the helpline vs. the online forum,
suggesting the design of these platforms may shape meaningful measures of their
support-seeking experiences. Our results provide new insight into the
experiences of caregivers suffering from postpartum mental health distress. We
conclude by discussing methodological considerations for understanding content
shared by support seekers and design considerations for the next generation of
support tools for postpartum parents. | cs.CL | Understanding Postpartum Parents' Experiences via Two Digital Platforms |
2022-12-21T21:46:01Z | http://arxiv.org/pdf/2212.11382v1 | http://arxiv.org/abs/2212.11382v1 | Lukas Christ, Shahin Amiriparian, Manuel Milling, Ilhan Aslan, Björn W. Schuller | 2022-12-21T21:46:01Z | cs.CL | Telling stories is an integral part of human communication which can evoke
emotions and influence the affective states of the audience. Automatically
modelling emotional trajectories in stories has thus attracted considerable
scholarly interest. However, as most existing works have been limited to
unsupervised dictionary-based approaches, there is no labelled benchmark for
this task. We address this gap by introducing continuous valence and arousal
annotations for an existing dataset of children's stories annotated with
discrete emotion categories. We collect additional annotations for this data
and map the originally categorical labels to the valence and arousal space.
Leveraging recent advances in Natural Language Processing, we propose a set of
novel Transformer-based methods for predicting valence and arousal signals over
the course of written stories. We explore several strategies for fine-tuning a
pretrained ELECTRA model and study the benefits of considering a sentence's
context when inferring its emotionality. Moreover, we experiment with
additional LSTM and Transformer layers. The best configuration achieves a
Concordance Correlation Coefficient (CCC) of .7338 for valence and .6302 for
arousal on the test set, demonstrating the suitability of our proposed
approach. Our code and additional annotations are made available at
https://github.com/lc0197/emotion_modelling_stories. | cs.CL | Automatic Emotion Modelling in Written Stories |
2022-12-21T20:43:46Z | http://arxiv.org/pdf/2212.11353v1 | http://arxiv.org/abs/2212.11353v1 | Chris Lengerich, Gabriel Synnaeve, Amy Zhang, Hugh Leather, Kurt Shuster, François Charton, Charysse Redwood | 2022-12-21T20:43:46Z | cs.CL, cs.LG | Traditional approaches to RL have focused on learning decision policies
directly from episodic decisions, while slowly and implicitly learning the
semantics of compositional representations needed for generalization. While
some approaches have been adopted to refine representations via auxiliary
self-supervised losses while simultaneously learning decision policies,
learning compositional representations from hand-designed and
context-independent self-supervised losses (multi-view) still adapts relatively
slowly to the real world, which contains many non-IID subspaces requiring rapid
distribution shift in both time and spatial attention patterns at varying
levels of abstraction. In contrast, supervised language model cascades have
shown the flexibility to adapt to many diverse manifolds, and hints of
self-learning needed for autonomous task transfer. However, to date, transfer
methods for language models like few-shot learning and fine-tuning still
require human supervision and transfer learning using self-learning methods has
been underexplored. We propose a self-supervised loss policy called contrastive
distillation which manifests latent variables with high mutual information with
both source and target tasks from weights to tokens. We show how this
outperforms common methods of transfer learning and suggests a useful design
axis of trading off compute for generalizability for online transfer.
Contrastive distillation is improved through sampling from memory and suggests
a simple algorithm for more efficiently sampling negative examples for
contrastive losses than random sampling. | cs.CL | Contrastive Distillation Is a Sample-Efficient Self-Supervised Loss Policy for Transfer Learning |
2022-12-21T19:11:19Z | http://arxiv.org/pdf/2212.11311v1 | http://arxiv.org/abs/2212.11311v1 | Xiang Deng, Vasilisa Bashlovkina, Feng Han, Simon Baumgartner, Michael Bendersky | 2022-12-21T19:11:19Z | cs.CL, cs.AI, cs.LG, cs.SI | Market sentiment analysis on social media content requires knowledge of both
financial markets and social media jargon, which makes it a challenging task
for human raters. The resulting lack of high-quality labeled data stands in the
way of conventional supervised learning methods. Instead, we approach this
problem using semi-supervised learning with a large language model (LLM). Our
pipeline generates weak financial sentiment labels for Reddit posts with an LLM
and then uses that data to train a small model that can be served in
production. We find that prompting the LLM to produce Chain-of-Thought
summaries and forcing it through several reasoning paths helps generate more
stable and accurate labels, while using a regression loss further improves
distillation quality. With only a handful of prompts, the final model performs
on par with existing supervised models. Though production applications of our
model are limited by ethical considerations, the model's competitive
performance points to the great potential of using LLMs for tasks that
otherwise require skill-intensive annotation. | cs.CL | What do LLMs Know about Financial Markets? A Case Study on Reddit Market Sentiment Analysis |
2022-12-21T18:58:41Z | http://arxiv.org/pdf/2212.11270v1 | http://arxiv.org/abs/2212.11270v1 | Xueyan Zou, Zi-Yi Dou, Jianwei Yang, Zhe Gan, Linjie Li, Chunyuan Li, Xiyang Dai, Harkirat Behl, Jianfeng Wang, Lu Yuan, Nanyun Peng, Lijuan Wang, Yong Jae Lee, Jianfeng Gao | 2022-12-21T18:58:41Z | cs.CV, cs.CL | We present X-Decoder, a generalized decoding model that can predict
pixel-level segmentation and language tokens seamlessly. X-Decodert takes as
input two types of queries: (i) generic non-semantic queries and (ii) semantic
queries induced from text inputs, to decode different pixel-level and
token-level outputs in the same semantic space. With such a novel design,
X-Decoder is the first work that provides a unified way to support all types of
image segmentation and a variety of vision-language (VL) tasks. Further, our
design enables seamless interactions across tasks at different granularities
and brings mutual benefits by learning a common and rich pixel-level
visual-semantic understanding space, without any pseudo-labeling. After
pretraining on a mixed set of a limited amount of segmentation data and
millions of image-text pairs, X-Decoder exhibits strong transferability to a
wide range of downstream tasks in both zero-shot and finetuning settings.
Notably, it achieves (1) state-of-the-art results on open-vocabulary
segmentation and referring segmentation on eight datasets; (2) better or
competitive finetuned performance to other generalist and specialist models on
segmentation and VL tasks; and (3) flexibility for efficient finetuning and
novel task composition (e.g., referring captioning and image editing). Code,
demo, video, and visualization are available at https://x-decoder-vl.github.io. | cs.CV | Generalized Decoding for Pixel, Image, and Language |
2022-12-21T18:54:19Z | http://arxiv.org/pdf/2212.11261v2 | http://arxiv.org/abs/2212.11261v2 | Robert Wolfe, Yiwei Yang, Bill Howe, Aylin Caliskan | 2023-05-15T23:49:27Z | cs.CY, cs.AI, cs.CL, cs.CV, cs.LG | Nine language-vision AI models trained on web scrapes with the Contrastive
Language-Image Pretraining (CLIP) objective are evaluated for evidence of a
bias studied by psychologists: the sexual objectification of girls and women,
which occurs when a person's human characteristics, such as emotions, are
disregarded and the person is treated as a body. We replicate three experiments
in psychology quantifying sexual objectification and show that the phenomena
persist in AI. A first experiment uses standardized images of women from the
Sexual OBjectification and EMotion Database, and finds that human
characteristics are disassociated from images of objectified women: the model's
recognition of emotional state is mediated by whether the subject is fully or
partially clothed. Embedding association tests (EATs) return significant effect
sizes for both anger (d >0.80) and sadness (d >0.50), associating images of
fully clothed subjects with emotions. GRAD-CAM saliency maps highlight that
CLIP gets distracted from emotional expressions in objectified images. A second
experiment measures the effect in a representative application: an automatic
image captioner (Antarctic Captions) includes words denoting emotion less than
50% as often for images of partially clothed women than for images of fully
clothed women. A third experiment finds that images of female professionals
(scientists, doctors, executives) are likely to be associated with sexual
descriptions relative to images of male professionals. A fourth experiment
shows that a prompt of "a [age] year old girl" generates sexualized images (as
determined by an NSFW classifier) up to 73% of the time for VQGAN-CLIP and
Stable Diffusion; the corresponding rate for boys never surpasses 9%. The
evidence indicates that language-vision AI models trained on web scrapes learn
biases of sexual objectification, which propagate to downstream applications. | cs.CY | Contrastive Language-Vision AI Models Pretrained on Web-Scraped Multimodal Data Exhibit Sexual Objectification Bias |
2022-12-21T17:59:11Z | http://arxiv.org/pdf/2212.11234v1 | http://arxiv.org/abs/2212.11234v1 | Mikolaj Figurski | 2022-12-21T17:59:11Z | cs.CL, cs.LG, I.2.7 | We consider the problem of embedding character-entity relationships from the
reduced semantic space of narratives, proposing and evaluating the assumption
that these relationships hold under a reflection operation. We analyze this
assumption and compare the approach to a baseline state-of-the-art model with a
unique evaluation that simulates efficacy on a downstream clustering task with
human-created labels. Although our model creates clusters that achieve
Silhouette scores of -.084, outperforming the baseline -.227, our analysis
reveals that the models approach the task much differently and perform well on
very different examples. We conclude that our assumption might be useful for
specific types of data and should be evaluated on a wider range of tasks. | cs.CL | Improving Narrative Relationship Embeddings by Training with Additional Inverse-Relationship Constraints |
2022-12-21T17:58:01Z | http://arxiv.org/pdf/2212.11281v1 | http://arxiv.org/abs/2212.11281v1 | Buck Shlegeris, Fabien Roger, Lawrence Chan, Euan McLean | 2022-12-21T17:58:01Z | cs.CL, cs.AI, cs.LG | Current language models are considered to have sub-human capabilities at
natural language tasks like question-answering or writing code. However,
language models are not trained to perform well at these tasks, they are
trained to accurately predict the next token given previous tokes in tokenized
text. It is not clear whether language models are better or worse than humans
at next token prediction. To try to answer this question, we performed two
distinct experiments to directly compare humans and language models on this
front: one measuring top-1 accuracy and the other measuring perplexity. In both
experiments, we find humans to be consistently \emph{worse} than even
relatively small language models like GPT3-Ada at next-token prediction. | cs.CL | Language models are better than humans at next-token prediction |
2022-12-21T16:56:07Z | http://arxiv.org/pdf/2212.11185v1 | http://arxiv.org/abs/2212.11185v1 | Byung-Doh Oh, William Schuler | 2022-12-21T16:56:07Z | cs.CL | Transformer-based large language models are trained to make predictions about
the next word by aggregating representations of previous tokens through their
self-attention mechanism. In the field of cognitive modeling, such attention
patterns have recently been interpreted as embodying the process of cue-based
retrieval, in which attention over multiple targets is taken to generate
interference and latency during retrieval. Under this framework, this work
first defines an entropy-based predictor that quantifies the diffuseness of
self-attention, as well as distance-based predictors that capture the
incremental change in attention patterns across timesteps. Moreover, following
recent studies that question the informativeness of attention weights, we also
experiment with alternative methods for incorporating vector norms into
attention weights. Regression experiments using predictors calculated from the
GPT-2 language model show that these predictors deliver a substantially better
fit to held-out self-paced reading and eye-tracking data over a rigorous
baseline including GPT-2 surprisal. Additionally, the distance-based predictors
generally demonstrated higher predictive power, with effect sizes of up to 6.59
ms per standard deviation on self-paced reading times (compared to 2.82 ms for
surprisal) and 1.05 ms per standard deviation on eye-gaze durations (compared
to 3.81 ms for surprisal). | cs.CL | Entropy- and Distance-Based Predictors From GPT-2 Attention Patterns Predict Reading Times Over and Above GPT-2 Surprisal |
2022-12-21T16:52:10Z | http://arxiv.org/pdf/2212.11182v1 | http://arxiv.org/abs/2212.11182v1 | Tomasz Stanisz, Stanislaw Drozdz, Jaroslaw Kwapien | 2022-12-21T16:52:10Z | cs.CL, physics.soc-ph | The celebrated proverb that "speech is silver, silence is golden" has a long
multinational history and multiple specific meanings. In written texts
punctuation can in fact be considered one of its manifestations. Indeed, the
virtue of effectively speaking and writing involves - often decisively - the
capacity to apply the properly placed breaks. In the present study, based on a
large corpus of world-famous and representative literary texts in seven major
Western languages, it is shown that the distribution of intervals between
consecutive punctuation marks in almost all texts can universally be
characterised by only two parameters of the discrete Weibull distribution which
can be given an intuitive interpretation in terms of the so-called hazard
function. The values of these two parameters tend to be language-specific,
however, and even appear to navigate translations. The properties of the
computed hazard functions indicate that among the studied languages, English
turns out to be the least constrained by the necessity to place a consecutive
punctuation mark to partition a sequence of words. This may suggest that when
compared to other studied languages, English is more flexible, in the sense of
allowing longer uninterrupted sequences of words. Spanish reveals similar
tendency to only a bit lesser extent. | cs.CL | Universal versus system-specific features of punctuation usage patterns in~major Western~languages |
2022-12-21T16:41:11Z | http://arxiv.org/pdf/2212.13917v1 | http://arxiv.org/abs/2212.13917v1 | George Boateng | 2022-12-21T16:41:11Z | cs.HC, cs.CL, cs.LG, cs.SD, eess.SP | Couples generally manage chronic diseases together and the management takes
an emotional toll on both patients and their romantic partners. Consequently,
recognizing the emotions of each partner in daily life could provide an insight
into their emotional well-being in chronic disease management. The emotions of
partners are currently inferred in the lab and daily life using self-reports
which are not practical for continuous emotion assessment or observer reports
which are manual, time-intensive, and costly. Currently, there exists no
comprehensive overview of works on emotion recognition among couples.
Furthermore, approaches for emotion recognition among couples have (1) focused
on English-speaking couples in the U.S., (2) used data collected from the lab,
and (3) performed recognition using observer ratings rather than partner's
self-reported / subjective emotions. In this body of work contained in this
thesis (8 papers - 5 published and 3 currently under review in various
journals), we fill the current literature gap on couples' emotion recognition,
develop emotion recognition systems using 161 hours of data from a total of
1,051 individuals, and make contributions towards taking couples' emotion
recognition from the lab which is the status quo, to daily life. This thesis
contributes toward building automated emotion recognition systems that would
eventually enable partners to monitor their emotions in daily life and enable
the delivery of interventions to improve their emotional well-being. | cs.HC | Multimodal Emotion Recognition among Couples from Lab Settings to Daily Life using Smartwatches |
2022-12-21T16:00:44Z | http://arxiv.org/pdf/2212.12407v1 | http://arxiv.org/abs/2212.12407v1 | Ying Xie, Dongping Song | 2022-12-21T16:00:44Z | cs.CL, cs.LG, ACM-class: J.m | Obtaining labelled data in a particular context could be expensive and time
consuming. Although different algorithms, including unsupervised learning,
semi-supervised learning, self-learning have been adopted, the performance of
text classification varies with context. Given the lack of labelled dataset, we
proposed a novel and simple unsupervised text classification model to classify
cargo content in international shipping industry using the Standard
International Trade Classification (SITC) codes. Our method stems from
representing words using pretrained Glove Word Embeddings and finding the most
likely label using Cosine Similarity. To compare unsupervised text
classification model with supervised classification, we also applied several
Transformer models to classify cargo content. Due to lack of training data, the
SITC numerical codes and the corresponding textual descriptions were used as
training data. A small number of manually labelled cargo content data was used
to evaluate the classification performances of the unsupervised classification
and the Transformer based supervised classification. The comparison reveals
that unsupervised classification significantly outperforms Transformer based
supervised classification even after increasing the size of the training
dataset by 30%. Lacking training data is a key bottleneck that prohibits deep
learning models (such as Transformers) from successful practical applications.
Unsupervised classification can provide an alternative efficient and effective
method to classify text when there is scarce training data. | cs.CL | Text classification in shipping industry using unsupervised models and Transformer based supervised models |
2022-12-21T12:36:17Z | http://arxiv.org/pdf/2212.10983v2 | http://arxiv.org/abs/2212.10983v2 | Alba Curry, Amanda Cercas Curry | 2023-07-06T10:52:49Z | cs.CL, cs.HC | Emotions are an integral part of human cognition and they guide not only our
understanding of the world but also our actions within it. As such, whether we
soothe or flame an emotion is not inconsequential. Recent work in
conversational AI has focused on responding empathetically to users, validating
and soothing their emotions without a real basis. This AI-aided emotional
regulation can have negative consequences for users and society, tending
towards a one-noted happiness defined as only the absence of "negative"
emotions. We argue that we must carefully consider whether and how to respond
to users' emotions. | cs.CL | Computer says "No": The Case Against Empathetic Conversational AI |
2022-12-21T11:38:51Z | http://arxiv.org/pdf/2212.10947v3 | http://arxiv.org/abs/2212.10947v3 | Nir Ratner, Yoav Levine, Yonatan Belinkov, Ori Ram, Inbal Magar, Omri Abend, Ehud Karpas, Amnon Shashua, Kevin Leyton-Brown, Yoav Shoham | 2023-08-01T16:48:47Z | cs.CL | When applied to processing long text, Large Language Models (LLMs) are
limited by their context window. Existing efforts to address this limitation
involve training specialized architectures, and cannot be easily applied to
off-the-shelf LLMs. We present Parallel Context Windows (PCW), a method that
alleviates the context window restriction for any off-the-shelf LLM without
further training. The key to the approach is to carve a long context into
chunks (``windows''), restrict the attention mechanism to apply only within
each window, and re-use the positional embeddings across the windows. Our main
results test the PCW approach on in-context learning with models that range in
size between 750 million and 178 billion parameters, and show substantial
improvements for tasks with diverse input and output spaces. We show additional
benefits in other settings where long context windows may be beneficial:
multi-hop questions and retrieval-augmented question answering with multiple
retrieved documents. Our results highlight Parallel Context Windows as a
promising method for applying off-the-shelf LLMs in a range of settings that
require long text sequences. We make our code publicly available at
https://github.com/ai21labs/parallel-context-windows. | cs.CL | Parallel Context Windows for Large Language Models |
2022-12-21T11:25:41Z | http://arxiv.org/pdf/2212.10938v1 | http://arxiv.org/abs/2212.10938v1 | Minbeom Kim, Hwanhee Lee, Kang Min Yoo, Joonsuk Park, Hwaran Lee, Kyomin Jung | 2022-12-21T11:25:41Z | cs.CL | Steering language generation towards objectives or away from undesired
content has been a long-standing goal in utilizing language models (LM). Recent
work has demonstrated reinforcement learning and weighted decoding as effective
approaches to achieve a higher level of language control and quality with pros
and cons. In this work, we propose a novel critic decoding method for
controlled language generation (CriticControl) that combines the strengths of
reinforcement learning and weighted decoding. Specifically, we adopt the
actor-critic framework to train an LM-steering critic from non-differentiable
reward models. And similar to weighted decoding, our method freezes the
language model and manipulates the output token distribution using called
critic, improving training efficiency and stability. Evaluation of our method
on three controlled generation tasks, namely topic control, sentiment control,
and detoxification, shows that our approach generates more coherent and
well-controlled texts than previous methods. In addition, CriticControl
demonstrates superior generalization ability in zero-shot settings. Human
evaluation studies also corroborate our findings. | cs.CL | Critic-Guided Decoding for Controlled Text Generation |
2022-12-21T11:23:31Z | http://arxiv.org/pdf/2212.10935v2 | http://arxiv.org/abs/2212.10935v2 | Zihan Wang, Naoki Yoshinaga | 2024-05-08T07:25:48Z | cs.CL | Esports, a sports competition on video games, has become one of the most
important sporting events. Although esports play logs have been accumulated,
only a small portion of them accompany text commentaries for the audience to
retrieve and understand the plays. In this study, we therefore introduce the
task of generating game commentaries from esports' data records. We first build
large-scale esports data-to-text datasets that pair structured data and
commentaries from a popular esports game, League of Legends. We then evaluate
Transformer-based models to generate game commentaries from structured data
records, while examining the impact of the pre-trained language models.
Evaluation results on our dataset revealed the challenges of this novel task.
We will release our dataset to boost potential research in the data-to-text
generation community. | cs.CL | Commentary Generation from Data Records of Multiplayer Strategy Esports Game |
2022-12-21T11:22:38Z | http://arxiv.org/pdf/2212.10933v2 | http://arxiv.org/abs/2212.10933v2 | Mohammad Javad Hosseini, Filip Radlinski, Silvia Pareti, Annie Louis | 2023-05-26T20:17:15Z | cs.CL | Recent advances in language modeling have enabled new conversational systems.
In particular, it is often desirable for people to make choices among specified
options when using such systems. We address this problem of reference
resolution, when people use natural expressions to choose between the entities.
For example, given the choice `Should we make a Simnel cake or a Pandan cake?'
a natural response from a dialog participant may be indirect: `let's make the
green one'. Such natural expressions have been little studied for reference
resolution. We argue that robustly understanding such language has large
potential for improving naturalness in dialog, recommendation, and search
systems. We create AltEntities (Alternative Entities), a new public dataset of
42K entity pairs and expressions (referring to one entity in the pair), and
develop models for the disambiguation problem. Consisting of indirect referring
expressions across three domains, our corpus enables for the first time the
study of how language models can be adapted to this task. We find they achieve
82%-87% accuracy in realistic settings, which while reasonable also invites
further advances. | cs.CL | Resolving Indirect Referring Expressions for Entity Selection |
2022-12-21T11:18:09Z | http://arxiv.org/pdf/2212.10929v1 | http://arxiv.org/abs/2212.10929v1 | M Saiful Bari, Aston Zhang, Shuai Zheng, Xingjian Shi, Yi Zhu, Shafiq Joty, Mu Li | 2022-12-21T11:18:09Z | cs.CL, cs.AI, cs.LG | Pre-trained large language models can efficiently interpolate human-written
prompts in a natural way. Multitask prompted learning can help generalization
through a diverse set of tasks at once, thus enhancing the potential for more
effective downstream fine-tuning. To perform efficient multitask-inference in
the same batch, parameter-efficient fine-tuning methods such as prompt tuning
have been proposed. However, the existing prompt tuning methods may lack
generalization. We propose SPT, a semi-parametric prompt tuning method for
multitask prompted learning. The novel component of SPT is a memory bank from
where memory prompts are retrieved based on discrete prompts. Extensive
experiments, such as (i) fine-tuning a full language model with SPT on 31
different tasks from 8 different domains and evaluating zero-shot
generalization on 9 heldout datasets under 5 NLP task categories and (ii)
pretraining SPT on the GLUE datasets and evaluating fine-tuning on the
SuperGLUE datasets, demonstrate effectiveness of SPT. | cs.CL | SPT: Semi-Parametric Prompt Tuning for Multitask Prompted Learning |
2022-12-21T11:12:14Z | http://arxiv.org/pdf/2212.10923v3 | http://arxiv.org/abs/2212.10923v3 | Zonglin Yang, Li Dong, Xinya Du, Hao Cheng, Erik Cambria, Xiaodong Liu, Jianfeng Gao, Furu Wei | 2024-02-05T11:19:24Z | cs.CL, cs.AI | Inductive reasoning is a core component of human intelligence. In the past
research of inductive reasoning within computer science, formal language is
used as representations of knowledge (facts and rules, more specifically).
However, formal language can cause systematic problems for inductive reasoning
such as disability of handling raw input such as natural language,
sensitiveness to mislabeled data, and incapacity to handle ambiguous input. To
this end, we propose a new paradigm (task) for inductive reasoning, which is to
induce natural language rules from natural language facts, and create a dataset
termed DEER containing 1.2k rule-fact pairs for the task, where rules and facts
are written in natural language. New automatic metrics are also proposed and
analysed for the evaluation of this task. With DEER, we investigate a modern
approach for inductive reasoning where we use natural language as
representation for knowledge instead of formal language and use pretrained
language models as ''reasoners''. Moreover, we provide the first and
comprehensive analysis of how well pretrained language models can induce
natural language rules from natural language facts. We also propose a new
framework drawing insights from philosophy literature for this task, which we
show in the experiment section that surpasses baselines in both automatic and
human evaluations. We discuss about our future perspectives for inductive
reasoning in Section 7. Dataset and code are available at
https://github.com/ZonglinY/Inductive_Reasoning. | cs.CL | Language Models as Inductive Reasoners |
2022-12-21T10:20:54Z | http://arxiv.org/pdf/2212.10901v3 | http://arxiv.org/abs/2212.10901v3 | Zihao He, Weituo Hao, Wei-Tsung Lu, Changyou Chen, Kristina Lerman, Xuchen Song | 2023-10-22T02:12:16Z | cs.SD, cs.CL, cs.IR, cs.MM, eess.AS | Music captioning has gained significant attention in the wake of the rising
prominence of streaming media platforms. Traditional approaches often
prioritize either the audio or lyrics aspect of the music, inadvertently
ignoring the intricate interplay between the two. However, a comprehensive
understanding of music necessitates the integration of both these elements. In
this study, we delve into this overlooked realm by introducing a method to
systematically learn multimodal alignment between audio and lyrics through
contrastive learning. This not only recognizes and emphasizes the synergy
between audio and lyrics but also paves the way for models to achieve deeper
cross-modal coherence, thereby producing high-quality captions. We provide both
theoretical and empirical results demonstrating the advantage of the proposed
method, which achieves new state-of-the-art on two music captioning datasets. | cs.SD | ALCAP: Alignment-Augmented Music Captioner |
2022-12-21T10:15:19Z | http://arxiv.org/pdf/2212.10898v2 | http://arxiv.org/abs/2212.10898v2 | Khai Loong Aw, Mariya Toneva | 2023-03-01T03:20:08Z | cs.CL, q-bio.NC | Building systems that achieve a deeper understanding of language is one of
the central goals of natural language processing (NLP). Towards this goal,
recent works have begun to train language models on narrative datasets which
require extracting the most critical information by integrating across long
contexts. However, it is still an open question whether these models are
learning a deeper understanding of the text, or if the models are simply
learning a heuristic to complete the task. This work investigates this further
by turning to the one language processing system that truly understands complex
language: the human brain. We show that training language models for deeper
narrative understanding results in richer representations that have improved
alignment to human brain activity. We further find that the improvements in
brain alignment are larger for character names than for other discourse
features, which indicates that these models are learning important narrative
elements. Taken together, these results suggest that this type of training can
indeed lead to deeper language understanding. These findings have consequences
both for cognitive neuroscience by revealing some of the significant factors
behind brain-NLP alignment, and for NLP by highlighting that understanding of
long-range context can be improved beyond language modeling. | cs.CL | Training language models to summarize narratives improves brain alignment |
2022-12-21T09:58:14Z | http://arxiv.org/pdf/2212.10888v2 | http://arxiv.org/abs/2212.10888v2 | Chengtai Cao, Fan Zhou, Yurou Dai, Jianping Wang, Kunpeng Zhang | 2024-06-04T02:11:25Z | cs.LG, cs.CL, cs.CV | Data augmentation (DA) is indispensable in modern machine learning and deep
neural networks. The basic idea of DA is to construct new training data to
improve the model's generalization by adding slightly disturbed versions of
existing data or synthesizing new data. This survey comprehensively reviews a
crucial subset of DA techniques, namely Mix-based Data Augmentation (MixDA),
which generates novel samples by combining multiple examples. In contrast to
traditional DA approaches that operate on single samples or entire datasets,
MixDA stands out due to its effectiveness, simplicity, flexibility,
computational efficiency, theoretical foundation, and broad applicability. We
begin by introducing a novel taxonomy that categorizes MixDA into Mixup-based,
Cutmix-based, and mixture approaches based on a hierarchical perspective of the
data mixing operation. Subsequently, we provide an in-depth review of various
MixDA techniques, focusing on their underlying motivations. Owing to its
versatility, MixDA has penetrated a wide range of applications, which we also
thoroughly investigate in this survey. Moreover, we delve into the underlying
mechanisms of MixDA's effectiveness by examining its impact on model
generalization and calibration while providing insights into the model's
behavior by analyzing the inherent properties of MixDA. Finally, we
recapitulate the critical findings and fundamental challenges of current MixDA
studies while outlining the potential directions for future works. Different
from previous related surveys that focus on DA approaches in specific domains
(e.g., CV and NLP) or only review a limited subset of MixDA studies, we are the
first to provide a systematical survey of MixDA, covering its taxonomy,
methodology, application, and explainability. Furthermore, we provide promising
directions for researchers interested in this exciting area. | cs.LG | A Survey of Mix-based Data Augmentation: Taxonomy, Methods, Applications, and Explainability |
2022-12-21T09:44:08Z | http://arxiv.org/pdf/2212.10879v1 | http://arxiv.org/abs/2212.10879v1 | Ningyu Xu, Tao Gui, Ruotian Ma, Qi Zhang, Jingting Ye, Menghan Zhang, Xuanjing Huang | 2022-12-21T09:44:08Z | cs.CL | Multilingual BERT (mBERT) has demonstrated considerable cross-lingual
syntactic ability, whereby it enables effective zero-shot cross-lingual
transfer of syntactic knowledge. The transfer is more successful between some
languages, but it is not well understood what leads to this variation and
whether it fairly reflects difference between languages. In this work, we
investigate the distributions of grammatical relations induced from mBERT in
the context of 24 typologically different languages. We demonstrate that the
distance between the distributions of different languages is highly consistent
with the syntactic difference in terms of linguistic formalisms. Such
difference learnt via self-supervision plays a crucial role in the zero-shot
transfer performance and can be predicted by variation in morphosyntactic
properties between languages. These results suggest that mBERT properly encodes
languages in a way consistent with linguistic diversity and provide insights
into the mechanism of cross-lingual transfer. | cs.CL | Cross-Linguistic Syntactic Difference in Multilingual BERT: How Good is It and How Does It Affect Transfer? |
2022-12-21T09:37:05Z | http://arxiv.org/pdf/2212.10873v3 | http://arxiv.org/abs/2212.10873v3 | Hyunsoo Cho, Hyuhng Joon Kim, Junyeob Kim, Sang-Woo Lee, Sang-goo Lee, Kang Min Yoo, Taeuk Kim | 2023-06-14T03:23:44Z | cs.CL, cs.LG | Through in-context learning (ICL), large-scale language models are effective
few-shot learners without additional model fine-tuning. However, the ICL
performance does not scale well with the number of available training samples
as it is limited by the inherent input length constraint of the underlying
language model. Meanwhile, many studies have revealed that language models are
also powerful feature extractors, allowing them to be utilized in a black-box
manner and enabling the linear probing paradigm, where lightweight
discriminators are trained on top of the pre-extracted input representations.
This paper proposes prompt-augmented linear probing (PALP), a hybrid of linear
probing and ICL, which leverages the best of both worlds. PALP inherits the
scalability of linear probing and the capability of enforcing language models
to derive more meaningful representations via tailoring input into a more
conceivable form. Throughout in-depth investigations on various datasets, we
verified that PALP significantly enhances the input representations closing the
gap between ICL in the data-hungry scenario and fine-tuning in the
data-abundant scenario with little training overhead, potentially making PALP a
strong alternative in a black-box scenario. | cs.CL | Prompt-Augmented Linear Probing: Scaling beyond the Limit of Few-shot In-Context Learners |
2022-12-21T08:34:28Z | http://arxiv.org/pdf/2212.10843v1 | http://arxiv.org/abs/2212.10843v1 | Dongmin Hyun, Xiting Wang, Chanyoung Park, Xing Xie, Hwanjo Yu | 2022-12-21T08:34:28Z | cs.CL, cs.AI | Sentence summarization shortens given texts while maintaining core contents
of the texts. Unsupervised approaches have been studied to summarize texts
without human-written summaries. However, recent unsupervised models are
extractive, which remove words from texts and thus they are less flexible than
abstractive summarization. In this work, we devise an abstractive model based
on reinforcement learning without ground-truth summaries. We formulate the
unsupervised summarization based on the Markov decision process with rewards
representing the summary quality. To further enhance the summary quality, we
develop a multi-summary learning mechanism that generates multiple summaries
with varying lengths for a given text, while making the summaries mutually
enhance each other. Experimental results show that the proposed model
substantially outperforms both abstractive and extractive models, yet
frequently generating new words not contained in input texts. | cs.CL | Generating Multiple-Length Summaries via Reinforcement Learning for Unsupervised Sentence Summarization |
2022-12-21T07:53:26Z | http://arxiv.org/pdf/2212.13104v1 | http://arxiv.org/abs/2212.13104v1 | Marco Antonio Stranisci, Giuseppe Spillo, Cataldo Musto, Viviana Patti, Rossana Damiano | 2022-12-21T07:53:26Z | cs.CL | Digital media have enabled the access to unprecedented literary knowledge.
Authors, readers, and scholars are now able to discover and share an increasing
amount of information about books and their authors. Notwithstanding, digital
archives are still unbalanced: writers from non-Western countries are less
represented, and such a condition leads to the perpetration of old forms of
discrimination. In this paper, we present the Under-Represented Writers
Knowledge Graph (URW-KG), a resource designed to explore and possibly amend
this lack of representation by gathering and mapping information about works
and authors from Wikidata and three other sources: Open Library, Goodreads, and
Google Books. The experiments based on KG embeddings showed that the integrated
information encoded in the graph allows scholars and users to be more easily
exposed to non-Western literary works and authors with respect to Wikidata
alone. This opens to the development of fairer and effective tools for author
discovery and exploration. | cs.CL | The URW-KG: a Resource for Tackling the Underrepresentation of non-Western Writers |
2022-12-21T07:35:33Z | http://arxiv.org/pdf/2212.10826v1 | http://arxiv.org/abs/2212.10826v1 | Ayman Mansour, Wafaa F. Mukhtar | 2022-12-21T07:35:33Z | cs.CL | Designing a natural voice interface rely mostly on Speech recognition for
interaction between human and their modern digital life equipment. In addition,
speech recognition narrows the gap between monolingual individuals to better
exchange communication. However, the field lacks wide support for several
universal languages and their dialects, while most of the daily conversations
are carried out using them. This paper comes to inspect the viability of
designing an Automatic Speech Recognition model for the Sudanese dialect, which
is one of the Arabic Language dialects, and its complexity is a product of
historical and social conditions unique to its speakers. This condition is
reflected in both the form and content of the dialect, so this paper gives an
overview of the Sudanese dialect and the tasks of collecting represented
resources and pre-processing performed to construct a modest dataset to
overcome the lack of annotated data. Also proposed end- to-end speech
recognition model, the design of the model was formed using Convolution Neural
Networks. The Sudanese dialect dataset would be a stepping stone to enable
future Natural Language Processing research targeting the dialect. The designed
model provided some insights into the current recognition task and reached an
average Label Error Rate of 73.67%. | cs.CL | End-to-End Automatic Speech Recognition model for the Sudanese Dialect |
2022-12-21T07:30:22Z | http://arxiv.org/pdf/2212.10823v3 | http://arxiv.org/abs/2212.10823v3 | Wenxuan Zhou, Sheng Zhang, Tristan Naumann, Muhao Chen, Hoifung Poon | 2023-05-31T06:04:33Z | cs.CL, cs.AI, cs.LG | Relation extraction (RE), which has relied on structurally annotated corpora
for model training, has been particularly challenging in low-resource scenarios
and domains. Recent literature has tackled low-resource RE by self-supervised
learning, where the solution involves pretraining the entity pair embedding by
RE-based objective and finetuning on labeled data by classification-based
objective. However, a critical challenge to this approach is the gap in
objectives, which prevents the RE model from fully utilizing the knowledge in
pretrained representations. In this paper, we aim at bridging the gap and
propose to pretrain and finetune the RE model using consistent objectives of
contrastive learning. Since in this kind of representation learning paradigm,
one relation may easily form multiple clusters in the representation space, we
further propose a multi-center contrastive loss that allows one relation to
form multiple clusters to better align with pretraining. Experiments on two
document-level RE datasets, BioRED and Re-DocRED, demonstrate the effectiveness
of our method. Particularly, when using 1% end-task training data, our method
outperforms PLM-based RE classifier by 10.5% and 6.1% on the two datasets,
respectively. | cs.CL | Continual Contrastive Finetuning Improves Low-Resource Relation Extraction |
2022-12-21T07:17:32Z | http://arxiv.org/pdf/2212.10819v1 | http://arxiv.org/abs/2212.10819v1 | Wen Xiao, Lesly Miculicich, Yang Liu, Pengcheng He, Giuseppe Carenini | 2022-12-21T07:17:32Z | cs.CL | Content-Controllable Summarization generates summaries focused on the given
controlling signals. Due to the lack of large-scale training corpora for the
task, we propose a plug-and-play module RelAttn to adapt any general
summarizers to the content-controllable summarization task. RelAttn first
identifies the relevant content in the source documents, and then makes the
model attend to the right context by directly steering the attention weight. We
further apply an unsupervised online adaptive parameter searching algorithm to
determine the degree of control in the zero-shot setting, while such parameters
are learned in the few-shot setting. By applying the module to three backbone
summarization models, experiments show that our method effectively improves all
the summarizers, and outperforms the prefix-based method and a widely used
plug-and-play model in both zero- and few-shot settings. Tellingly, more
benefit is observed in the scenarios when more control is needed. | cs.CL | Attend to the Right Context: A Plug-and-Play Module for Content-Controllable Summarization |
2022-12-21T07:15:59Z | http://arxiv.org/pdf/2212.10818v2 | http://arxiv.org/abs/2212.10818v2 | Yui Sudo, Muhammad Shakeel, Brian Yan, Jiatong Shi, Shinji Watanabe | 2023-05-29T23:16:56Z | cs.SD, cs.CL, eess.AS | The network architecture of end-to-end (E2E) automatic speech recognition
(ASR) can be classified into several models, including connectionist temporal
classification (CTC), recurrent neural network transducer (RNN-T), attention
mechanism, and non-autoregressive mask-predict models. Since each of these
network architectures has pros and cons, a typical use case is to switch these
separate models depending on the application requirement, resulting in the
increased overhead of maintaining all models. Several methods for integrating
two of these complementary models to mitigate the overhead issue have been
proposed; however, if we integrate more models, we will further benefit from
these complementary models and realize broader applications with a single
system. This paper proposes four-decoder joint modeling (4D) of CTC, attention,
RNN-T, and mask-predict, which has the following three advantages: 1) The four
decoders are jointly trained so that they can be easily switched depending on
the application scenarios. 2) Joint training may bring model regularization and
improve the model robustness thanks to their complementary properties. 3) Novel
one-pass joint decoding methods using CTC, attention, and RNN-T further
improves the performance. The experimental results showed that the proposed
model consistently reduced the WER. | cs.SD | 4D ASR: Joint modeling of CTC, Attention, Transducer, and Mask-Predict decoders |
2022-12-21T07:06:55Z | http://arxiv.org/pdf/2212.10815v1 | http://arxiv.org/abs/2212.10815v1 | Dheeraj Mekala, Jason Wolfe, Subhro Roy | 2022-12-21T07:06:55Z | cs.CL | We explore the use of large language models (LLMs) for zero-shot semantic
parsing. Semantic parsing involves mapping natural language utterances to
task-specific meaning representations. Language models are generally trained on
the publicly available text and code and cannot be expected to directly
generalize to domain-specific parsing tasks in a zero-shot setting. In this
work, we propose ZEROTOP, a zero-shot task-oriented parsing method that
decomposes a semantic parsing problem into a set of abstractive and extractive
question-answering (QA) problems, enabling us to leverage the ability of LLMs
to zero-shot answer reading comprehension questions. For each utterance, we
prompt the LLM with questions corresponding to its top-level intent and a set
of slots and use the LLM generations to construct the target meaning
representation. We observe that current LLMs fail to detect unanswerable
questions; and as a result, cannot handle questions corresponding to missing
slots. To address this problem, we fine-tune a language model on public QA
datasets using synthetic negative samples. Experimental results show that our
QA-based decomposition paired with the fine-tuned LLM can correctly parse ~16%
of utterances in the MTOP dataset without requiring any annotated data. | cs.CL | ZEROTOP: Zero-Shot Task-Oriented Semantic Parsing using Large Language Models |
2022-12-21T06:22:03Z | http://arxiv.org/pdf/2212.10792v1 | http://arxiv.org/abs/2212.10792v1 | Najoung Kim, Jatin Khilnani, Alex Warstadt, Abed Qaddoumi | 2022-12-21T06:22:03Z | cs.CL | We propose reconstruction probing, a new analysis method for contextualized
representations based on reconstruction probabilities in masked language models
(MLMs). This method relies on comparing the reconstruction probabilities of
tokens in a given sequence when conditioned on the representation of a single
token that has been fully contextualized and when conditioned on only the
decontextualized lexical prior of the model. This comparison can be understood
as quantifying the contribution of contextualization towards reconstruction --
the difference in the reconstruction probabilities can only be attributed to
the representational change of the single token induced by contextualization.
We apply this analysis to three MLMs and find that contextualization boosts
reconstructability of tokens that are close to the token being reconstructed in
terms of linear and syntactic distance. Furthermore, we extend our analysis to
finer-grained decomposition of contextualized representations, and we find that
these boosts are largely attributable to static and positional embeddings at
the input layer. | cs.CL | Reconstruction Probing |
2022-12-21T06:20:28Z | http://arxiv.org/pdf/2212.10791v1 | http://arxiv.org/abs/2212.10791v1 | Annie Louis, Joshua Maynez | 2022-12-21T06:20:28Z | cs.CL, cs.AI | A typical product or place often has hundreds of reviews, and summarization
of these texts is an important and challenging problem. Recent progress on
abstractive summarization in domains such as news has been driven by supervised
systems trained on hundreds of thousands of news articles paired with
human-written summaries. However for opinion texts, such large scale datasets
are rarely available. Unsupervised methods, self-training, and few-shot
learning approaches bridge that gap. In this work, we present a novel
self-training approach, OpineSum, for abstractive opinion summarization. The
summaries in this approach are built using a novel application of textual
entailment and capture the consensus of opinions across the various reviews for
an item. This method can be used to obtain silver-standard summaries on a large
scale and train both unsupervised and few-shot abstractive summarization
systems. OpineSum achieves state-of-the-art performance in both settings. | cs.CL | OpineSum: Entailment-based self-training for abstractive opinion summarization |
2022-12-21T06:18:31Z | http://arxiv.org/pdf/2212.10789v3 | http://arxiv.org/abs/2212.10789v3 | Shengchao Liu, Weili Nie, Chengpeng Wang, Jiarui Lu, Zhuoran Qiao, Ling Liu, Jian Tang, Chaowei Xiao, Anima Anandkumar | 2024-01-29T22:09:49Z | cs.LG, cs.CL, q-bio.QM, stat.ML | There is increasing adoption of artificial intelligence in drug discovery.
However, existing studies use machine learning to mainly utilize the chemical
structures of molecules but ignore the vast textual knowledge available in
chemistry. Incorporating textual knowledge enables us to realize new drug
design objectives, adapt to text-based instructions and predict complex
biological activities. Here we present a multi-modal molecule structure-text
model, MoleculeSTM, by jointly learning molecules' chemical structures and
textual descriptions via a contrastive learning strategy. To train MoleculeSTM,
we construct a large multi-modal dataset, namely, PubChemSTM, with over 280,000
chemical structure-text pairs. To demonstrate the effectiveness and utility of
MoleculeSTM, we design two challenging zero-shot tasks based on text
instructions, including structure-text retrieval and molecule editing.
MoleculeSTM has two main properties: open vocabulary and compositionality via
natural language. In experiments, MoleculeSTM obtains the state-of-the-art
generalization ability to novel biochemical concepts across various benchmarks. | cs.LG | Multi-modal Molecule Structure-text Model for Text-based Retrieval and Editing |
2022-12-21T06:00:22Z | http://arxiv.org/pdf/2212.10786v2 | http://arxiv.org/abs/2212.10786v2 | Keming Lu, I-Hung Hsu, Wenxuan Zhou, Mingyu Derek Ma, Muhao Chen | 2023-06-05T00:43:13Z | cs.CL, cs.IR, cs.LG | Relation Extraction (RE) has been extended to cross-document scenarios
because many relations are not simply described in a single document. This
inevitably brings the challenge of efficient open-space evidence retrieval to
support the inference of cross-document relations, along with the challenge of
multi-hop reasoning on top of entities and evidence scattered in an open set of
documents. To combat these challenges, we propose MR.COD (Multi-hop evidence
retrieval for Cross-document relation extraction), which is a multi-hop
evidence retrieval method based on evidence path mining and ranking. We explore
multiple variants of retrievers to show evidence retrieval is essential in
cross-document RE. We also propose a contextual dense retriever for this
setting. Experiments on CodRED show that evidence retrieval with MR.COD
effectively acquires crossdocument evidence and boosts end-to-end RE
performance in both closed and open settings. | cs.CL | Multi-hop Evidence Retrieval for Cross-document Relation Extraction |
2022-12-21T05:59:25Z | http://arxiv.org/pdf/2212.11275v1 | http://arxiv.org/abs/2212.11275v1 | Neeraj Kumar, Ankur Narang, Brejesh Lall | 2022-12-21T05:59:25Z | cs.CL, cs.AI, cs.LG | Large pre-trained models, such as Bert, GPT, and Wav2Vec, have demonstrated
great potential for learning representations that are transferable to a wide
variety of downstream tasks . It is difficult to obtain a large quantity of
supervised data due to the limited availability of resources and time. In light
of this, a significant amount of research has been conducted in the area of
adopting large pre-trained datasets for diverse downstream tasks via fine
tuning, linear probing, or prompt tuning in low resource settings.
Normalization techniques are essential for accelerating training and improving
the generalization of deep neural networks and have been successfully used in a
wide variety of applications. A lot of normalization techniques have been
proposed but the success of normalization in low resource downstream NLP and
speech tasks is limited. One of the reasons is the inability to capture
expressiveness by rescaling parameters of normalization. We propose
KullbackLeibler(KL) Regularized normalization (KL-Norm) which make the
normalized data well behaved and helps in better generalization as it reduces
over-fitting, generalises well on out of domain distributions and removes
irrelevant biases and features with negligible increase in model parameters and
memory overheads. Detailed experimental evaluation on multiple low resource NLP
and speech tasks, demonstrates the superior performance of KL-Norm as compared
to other popular normalization and regularization techniques. | cs.CL | KL Regularized Normalization Framework for Low Resource Tasks |
2022-12-21T05:54:14Z | http://arxiv.org/pdf/2212.10785v2 | http://arxiv.org/abs/2212.10785v2 | Ife Adebara, AbdelRahim Elmadany, Muhammad Abdul-Mageed, Alcides Alcoba Inciarte | 2023-05-26T20:20:04Z | cs.CL, cs.AI | Multilingual pretrained language models (mPLMs) acquire valuable,
generalizable linguistic information during pretraining and have advanced the
state of the art on task-specific finetuning. To date, only ~31 out of ~2,000
African languages are covered in existing language models. We ameliorate this
limitation by developing SERENGETI, a massively multilingual language model
that covers 517 African languages and language varieties. We evaluate our novel
models on eight natural language understanding tasks across 20 datasets,
comparing to 4 mPLMs that cover 4-23 African languages. SERENGETI outperforms
other models on 11 datasets across the eights tasks, achieving 82.27 average
F_1. We also perform analyses of errors from our models, which allows us to
investigate the influence of language genealogy and linguistic similarity when
the models are applied under zero-shot settings. We will publicly release our
models for
research.\footnote{\href{https://github.com/UBC-NLP/serengeti}{https://github.com/UBC-NLP/serengeti}} | cs.CL | SERENGETI: Massively Multilingual Language Models for Africa |
2022-12-21T05:49:08Z | http://arxiv.org/pdf/2212.10784v3 | http://arxiv.org/abs/2212.10784v3 | Jiashu Xu, Mingyu Derek Ma, Muhao Chen | 2023-10-19T05:46:10Z | cs.CL, cs.AI, q-bio.QM | Two key obstacles in biomedical relation extraction (RE) are the scarcity of
annotations and the prevalence of instances without explicitly pre-defined
labels due to low annotation coverage. Existing approaches, which treat
biomedical RE as a multi-class classification task, often result in poor
generalization in low-resource settings and do not have the ability to make
selective prediction on unknown cases but give a guess from seen relations,
hindering the applicability of those approaches. We present NBR, which converts
biomedical RE as natural language inference formulation through indirect
supervision. By converting relations to natural language hypotheses, NBR is
capable of exploiting semantic cues to alleviate annotation scarcity. By
incorporating a ranking-based loss that implicitly calibrates abstinent
instances, NBR learns a clearer decision boundary and is instructed to abstain
on uncertain instances. Extensive experiments on three widely-used biomedical
RE benchmarks, namely ChemProt, DDI and GAD, verify the effectiveness of NBR in
both full-set and low-resource regimes. Our analysis demonstrates that indirect
supervision benefits biomedical RE even when a domain gap exists, and combining
NLI knowledge with biomedical knowledge leads to the best performance gains. | cs.CL | Can NLI Provide Proper Indirect Supervision for Low-resource Biomedical Relation Extraction? |
2022-12-21T05:17:06Z | http://arxiv.org/pdf/2212.10773v3 | http://arxiv.org/abs/2212.10773v3 | Zhiyang Xu, Ying Shen, Lifu Huang | 2023-06-10T18:33:21Z | cs.CL | Instruction tuning, a new learning paradigm that fine-tunes pre-trained
language models on tasks specified through instructions, has shown promising
zero-shot performance on various natural language processing tasks. However, it
has yet to be explored for vision and multimodal tasks. In this work, we
introduce MUL-TIINSTRUCT, the first multimodal instruction tuning benchmark
dataset that consists of 62 diverse multimodal tasks in a unified seq-to-seq
format covering 10 broad categories. The tasks are derived from 21 existing
open-source datasets and each task is equipped with 5 expert-written
instructions. We take OFA as the base pre-trained model for multimodal
instruction tuning, and to further improve its zero-shot performance, we
explore multiple transfer learning strategies to leverage the large-scale
NATURAL INSTRUCTIONS dataset. Experimental results demonstrate strong zero-shot
performance on various unseen multimodal tasks and the benefit of transfer
learning from a text-only instruction dataset. We also design a new evaluation
metric - Sensitivity, to evaluate how sensitive the model is to the variety of
instructions. Our results indicate that fine-tuning the model on a diverse set
of tasks and instructions leads to a reduced sensitivity to variations in
instructions for each task. | cs.CL | MultiInstruct: Improving Multi-Modal Zero-Shot Learning via Instruction Tuning |
2022-12-21T05:02:49Z | http://arxiv.org/pdf/2212.10770v1 | http://arxiv.org/abs/2212.10770v1 | Luke Vilnis, Zach Fisher, Bhargav Kanagal, Patrick Murray, Sumit Sanghai | 2022-12-21T05:02:49Z | cs.CL | Large language models have ushered in a golden age of semantic parsing. The
seq2seq paradigm allows for open-schema and abstractive attribute and relation
extraction given only small amounts of finetuning data. Language model
pretraining has simultaneously enabled great strides in natural language
inference, reasoning about entailment and implication in free text. These
advances motivate us to construct ImPaKT, a dataset for open-schema information
extraction, consisting of around 2500 text snippets from the C4 corpus, in the
shopping domain (product buying guides), professionally annotated with
extracted attributes, types, attribute summaries (attribute schema discovery
from idiosyncratic text), many-to-one relations between compound and atomic
attributes, and implication relations. We release this data in hope that it
will be useful in fine tuning semantic parsers for information extraction and
knowledge base construction across a variety of domains. We evaluate the power
of this approach by fine-tuning the open source UL2 language model on a subset
of the dataset, extracting a set of implication relations from a corpus of
product buying guides, and conducting human evaluations of the resulting
predictions. | cs.CL | ImPaKT: A Dataset for Open-Schema Knowledge Base Construction |
2022-12-21T05:02:08Z | http://arxiv.org/pdf/2212.10769v1 | http://arxiv.org/abs/2212.10769v1 | Najoung Kim, Tal Linzen, Paul Smolensky | 2022-12-21T05:02:08Z | cs.CL | Human linguistic capacity is often characterized by compositionality and the
generalization it enables -- human learners can produce and comprehend novel
complex expressions by composing known parts. Several benchmarks exploit
distributional control across training and test to gauge compositional
generalization, where certain lexical items only occur in limited contexts
during training. While recent work using these benchmarks suggests that
pretrained models achieve impressive generalization performance, we argue that
exposure to pretraining data may break the aforementioned distributional
control. Using the COGS benchmark of Kim and Linzen (2020), we test two
modified evaluation setups that control for this issue: (1) substituting
context-controlled lexical items with novel character sequences, and (2)
substituting them with special tokens represented by novel embeddings. We find
that both of these setups lead to lower generalization performance in T5
(Raffel et al., 2020), suggesting that previously reported results have been
overestimated due to uncontrolled lexical exposure during pretraining. The
performance degradation is more extreme with novel embeddings, and the
degradation increases with the amount of pretraining data, highlighting an
interesting case of inverse scaling. | cs.CL | Uncontrolled Lexical Exposure Leads to Overestimation of Compositional Generalization in Pretrained Models |
2022-12-21T05:01:01Z | http://arxiv.org/pdf/2212.10767v3 | http://arxiv.org/abs/2212.10767v3 | Kazuma Hashimoto, Iftekhar Naim, Karthik Raman | 2024-01-31T04:10:30Z | cs.CL | Sequence labeling is a core task in text understanding for IE/IR systems.
Text generation models have increasingly become the go-to solution for such
tasks (e.g., entity extraction and dialog slot filling). While most research
has focused on the labeling accuracy, a key aspect -- of vital practical
importance -- has slipped through the cracks: understanding model confidence.
More specifically, we lack a principled understanding of how to reliably gauge
the confidence of a model in its predictions for each labeled span. This paper
aims to provide some empirical insights on estimating model confidence for
generative sequence labeling. Most notably, we find that simply using the
decoder's output probabilities \textbf{is not} the best in realizing
well-calibrated confidence estimates. As verified over six public datasets of
different tasks, we show that our proposed approach -- which leverages
statistics from top-$k$ predictions by a beam search -- significantly reduces
calibration errors of the predictions of a generative sequence labeling model. | cs.CL | How Does Beam Search improve Span-Level Confidence Estimation in Generative Sequence Labeling? |
2022-12-21T04:57:59Z | http://arxiv.org/pdf/2212.11808v1 | http://arxiv.org/abs/2212.11808v1 | Jesus Guerrero, Gongbo Liang, Izzat Alsmadi | 2022-12-21T04:57:59Z | cs.CL, cs.LG | Many natural language related applications involve text generation, created
by humans or machines. While in many of those applications machines support
humans, yet in few others, (e.g. adversarial machine learning, social bots and
trolls) machines try to impersonate humans. In this scope, we proposed and
evaluated several mutation-based text generation approaches. Unlike
machine-based generated text, mutation-based generated text needs human text
samples as inputs. We showed examples of mutation operators but this work can
be extended in many aspects such as proposing new text-based mutation operators
based on the nature of the application. | cs.CL | A Mutation-based Text Generation for Adversarial Machine Learning Applications |
2022-12-21T04:49:55Z | http://arxiv.org/pdf/2212.10764v3 | http://arxiv.org/abs/2212.10764v3 | Ruicheng Xian, Honglei Zhuang, Zhen Qin, Hamed Zamani, Jing Lu, Ji Ma, Kai Hui, Han Zhao, Xuanhui Wang, Michael Bendersky | 2023-10-31T16:30:32Z | cs.IR, cs.AI, cs.CL, cs.LG | Domain adaptation aims to transfer the knowledge learned on (data-rich)
source domains to (low-resource) target domains, and a popular method is
invariant representation learning, which matches and aligns the data
distributions on the feature space. Although this method is studied extensively
and applied on classification and regression problems, its adoption on ranking
problems is sporadic, and the few existing implementations lack theoretical
justifications. This paper revisits invariant representation learning for
ranking. Upon reviewing prior work, we found that they implement what we call
item-level alignment, which aligns the distributions of the items being ranked
from all lists in aggregate but ignores their list structure. However, the list
structure should be leveraged, because it is intrinsic to ranking problems
where the data and the metrics are defined and computed on lists, not the items
by themselves. To close this discrepancy, we propose list-level alignment --
learning domain-invariant representations at the higher level of lists. The
benefits are twofold: it leads to the first domain adaptation generalization
bound for ranking, in turn providing theoretical support for the proposed
method, and it achieves better empirical transfer performance for unsupervised
domain adaptation on ranking tasks, including passage reranking. | cs.IR | Learning List-Level Domain-Invariant Representations for Ranking |
2022-12-21T04:43:19Z | http://arxiv.org/pdf/2301.05149v2 | http://arxiv.org/abs/2301.05149v2 | Lingjun Zhao, Khanh Nguyen, Hal Daumé III | 2023-05-28T14:34:54Z | cs.CL, cs.AI, cs.HC, cs.LG, cs.RO | Recent work studies the cognitive capabilities of language models through
psychological tests designed for humans. While these studies are helpful for
understanding the general capabilities of these models, there is no guarantee
that a model possessing sufficient capabilities to pass those tests would
actually use those capabilities in performing real-life tasks. In this work, we
formulate task-oriented cognitive capabilities, which are human-like cognitive
capabilities that language models leverage to perform tasks. These capabilities
are (i) the ability to quickly generate good candidate utterances (the search
capability) (ii) the ability to predict how a listener interprets those
utterances and choose the most appropriate one (the pragmatic capability). We
design an evaluation scheme for comparing these capabilities of a language
model with those of a human. Applying this scheme to examine various models in
a navigation instruction generation problem, we find that their pragmatic
capability is severely lacking. This insight leads us to augment them with
better models of the listener and obtain a significant boost of 11% in success
rate in guiding real humans. Our work advocates for having a principled
procedure for aligning language models with humans that involves (i)
formulating task-oriented capabilities, (ii) devising a method to quantify
their deficiency, and (iii) iteratively improving them. | cs.CL | Define, Evaluate, and Improve Task-Oriented Cognitive Capabilities for Instruction Generation Models |
2022-12-21T04:35:43Z | http://arxiv.org/pdf/2212.10758v2 | http://arxiv.org/abs/2212.10758v2 | AbdelRahim Elmadany, El Moatez Billah Nagoudi, Muhammad Abdul-Mageed | 2023-05-29T18:27:37Z | cs.CL, cs.AI | Due to their crucial role in all NLP, several benchmarks have been proposed
to evaluate pretrained language models. In spite of these efforts, no public
benchmark of diverse nature currently exists for evaluation of Arabic. This
makes it challenging to measure progress for both Arabic and multilingual
language models. This challenge is compounded by the fact that any benchmark
targeting Arabic needs to take into account the fact that Arabic is not a
single language but rather a collection of languages and varieties. In this
work, we introduce ORCA, a publicly available benchmark for Arabic language
understanding evaluation. ORCA is carefully constructed to cover diverse Arabic
varieties and a wide range of challenging Arabic understanding tasks exploiting
60 different datasets across seven NLU task clusters. To measure current
progress in Arabic NLU, we use ORCA to offer a comprehensive comparison between
18 multilingual and Arabic language models. We also provide a public
leaderboard with a unified single-number evaluation metric (ORCA score) to
facilitate future research. | cs.CL | ORCA: A Challenging Benchmark for Arabic Language Understanding |
2022-12-21T04:21:46Z | http://arxiv.org/pdf/2212.10755v2 | http://arxiv.org/abs/2212.10755v2 | El Moatez Billah Nagoudi, Muhammad Abdul-Mageed, AbdelRahim Elmadany, Alcides Alcoba Inciarte, Md Tawkat Islam Khondaker | 2023-10-24T21:03:30Z | cs.CL | Scholarship on generative pretraining (GPT) remains acutely Anglocentric,
leaving serious gaps in our understanding of the whole class of autoregressive
models. For example, we have little knowledge about the potential of these
models and their societal impacts in diverse linguistic and cultural settings.
We alleviate this issue for Arabic, a wide collection of languages and
dialectal varieties with more than 400 million population, by introducing
JASMINE. JASMINE is a suite of powerful Arabic autoregressive Transformer
language models ranging in size between 300 million-6.7 billion parameters
pretrained on a large and diverse dataset (~ 235 GB of text). We also carefully
design and release a comprehensive benchmark for both automated and human
evaluation of Arabic autoregressive models, with coverage of potential social
biases, harms, and toxicity. Using our novel benchmark, we evaluate JASMINE
extensively showing powerful performance intrinsically as well as in few-shot
learning on a wide range of NLP tasks. We aim to responsibly release our models
and evaluation benchmark with interested researchers, along with code for
experimenting with them. | cs.CL | JASMINE: Arabic GPT Models for Few-Shot Learning |
2022-12-21T04:21:35Z | http://arxiv.org/pdf/2212.10754v3 | http://arxiv.org/abs/2212.10754v3 | Yijiang River Dong, Lara J. Martin, Chris Callison-Burch | 2023-06-08T11:58:21Z | cs.CL | Story generation and understanding -- as with all NLG/NLU tasks -- has seen a
surge in neurosymbolic work. Researchers have recognized that, while large
language models (LLMs) have tremendous utility, they can be augmented with
symbolic means to be even better and to make up for any flaws that the neural
networks might have. However, symbolic methods are extremely costly in terms of
the amount of time and expertise needed to create them. In this work, we
capitalize on state-of-the-art Code-LLMs, such as Codex, to bootstrap the use
of symbolic methods for tracking the state of stories and aiding in story
understanding. We show that our CoRRPUS system and abstracted prompting
procedures can beat current state-of-the-art structured LLM techniques on
pre-existing story understanding tasks (bAbI Task 2 and Re^3) with minimal hand
engineering. We hope that this work can help highlight the importance of
symbolic representations and specialized prompting for LLMs as these models
require some guidance for performing reasoning tasks properly. | cs.CL | CoRRPUS: Code-based Structured Prompting for Neurosymbolic Story Understanding |
2022-12-21T04:03:33Z | http://arxiv.org/pdf/2212.10750v2 | http://arxiv.org/abs/2212.10750v2 | Sihao Chen, Senaka Buthpitiya, Alex Fabrikant, Dan Roth, Tal Schuster | 2023-05-24T23:19:14Z | cs.CL | The widely studied task of Natural Language Inference (NLI) requires a system
to recognize whether one piece of text is textually entailed by another, i.e.
whether the entirety of its meaning can be inferred from the other. In current
NLI datasets and models, textual entailment relations are typically defined on
the sentence- or paragraph-level. However, even a simple sentence often
contains multiple propositions, i.e. distinct units of meaning conveyed by the
sentence. As these propositions can carry different truth values in the context
of a given premise, we argue for the need to recognize the textual entailment
relation of each proposition in a sentence individually.
We propose PropSegmEnt, a corpus of over 45K propositions annotated by expert
human raters. Our dataset structure resembles the tasks of (1) segmenting
sentences within a document to the set of propositions, and (2) classifying the
entailment relation of each proposition with respect to a different yet
topically-aligned document, i.e. documents describing the same event or entity.
We establish strong baselines for the segmentation and entailment tasks.
Through case studies on summary hallucination detection and document-level NLI,
we demonstrate that our conceptual framework is potentially useful for
understanding and explaining the compositionality of NLI labels. | cs.CL | PropSegmEnt: A Large-Scale Corpus for Proposition-Level Segmentation and Entailment Recognition |
2022-12-21T03:22:24Z | http://arxiv.org/pdf/2212.10740v1 | http://arxiv.org/abs/2212.10740v1 | Hongxiao Li, Wanling Gao, Lei Wang, Jianfeng Zhan | 2022-12-21T03:22:24Z | cs.PL, cs.CC, cs.CL | Previous computation models either have equivalent abilities in representing
all computations but fail to provide primitive operators for programming
complex algorithms or lack generalized expression ability to represent
newly-added computations. This article presents a unified computation model
with generalized expression ability and a concise set of primitive operators
for programming high-level algorithms. We propose a unified data abstraction --
Tensor of List, and offer a unified computation model based on Tensor of List,
which we call the ToL model (in short, ToL). ToL introduces five atomic
computations that can represent any elementary computation by finite
composition, ensured with strict formal proof. Based on ToL, we design a
pure-functional language -- ToLang. ToLang provides a concise set of primitive
operators that can be used to program complex big data and AI algorithms. Our
evaluations show ToL has generalized expression ability and a built-in
performance indicator, born with a strictly defined computation metric --
elementary operation count (EOPs), consistent with FLOPs within a small error
range. | cs.PL | ToL: A Tensor of List-Based Unified Computation Model |
2022-12-21T02:47:52Z | http://arxiv.org/pdf/2212.10728v1 | http://arxiv.org/abs/2212.10728v1 | Soyeon Caren Han, Siqu Long, Henry Weld, Josiah Poon | 2022-12-21T02:47:52Z | cs.CL, cs.AI | When a human communicates with a machine using natural language on the web
and online, how can it understand the human's intention and semantic context of
their talk? This is an important AI task as it enables the machine to construct
a sensible answer or perform a useful action for the human. Meaning is
represented at the sentence level, identification of which is known as intent
detection, and at the word level, a labelling task called slot filling. This
dual-level joint task requires innovative thinking about natural language and
deep learning network design, and as a result, many approaches and models have
been proposed and applied.
This tutorial will discuss how the joint task is set up and introduce Spoken
Language Understanding/Natural Language Understanding (SLU/NLU) with Deep
Learning techniques. We will cover the datasets, experiments and metrics used
in the field. We will describe how the machine uses the latest NLP and Deep
Learning techniques to address the joint task, including recurrent and
attention-based Transformer networks and pre-trained models (e.g. BERT). We
will then look in detail at a network that allows the two levels of the task,
intent classification and slot filling, to interact to boost performance
explicitly. We will do a code demonstration of a Python notebook for this model
and attendees will have an opportunity to watch coding demo tasks on this joint
NLU to further their understanding. | cs.CL | Spoken Language Understanding for Conversational AI: Recent Advances and Future Direction |
2022-12-21T02:41:40Z | http://arxiv.org/pdf/2212.10726v2 | http://arxiv.org/abs/2212.10726v2 | John Wieting, Jonathan H. Clark, William W. Cohen, Graham Neubig, Taylor Berg-Kirkpatrick | 2023-06-04T22:42:10Z | cs.CL, cs.LG | Contrastive learning has been successfully used for retrieval of semantically
aligned sentences, but it often requires large batch sizes or careful
engineering to work well. In this paper, we instead propose a generative model
for learning multilingual text embeddings which can be used to retrieve or
score sentence pairs. Our model operates on parallel data in $N$ languages and,
through an approximation we introduce, efficiently encourages source separation
in this multilingual setting, separating semantic information that is shared
between translations from stylistic or language-specific variation. We show
careful large-scale comparisons between contrastive and generation-based
approaches for learning multilingual text embeddings, a comparison that has not
been done to the best of our knowledge despite the popularity of these
approaches. We evaluate this method on a suite of tasks including semantic
similarity, bitext mining, and cross-lingual question retrieval -- the last of
which we introduce in this paper. Overall, our Variational Multilingual
Source-Separation Transformer (VMSST) model outperforms both a strong
contrastive and generative baseline on these tasks. | cs.CL | Beyond Contrastive Learning: A Variational Generative Model for Multilingual Retrieval |
2022-12-21T02:28:07Z | http://arxiv.org/pdf/2212.10722v2 | http://arxiv.org/abs/2212.10722v2 | Faisal Ladhak, Esin Durmus, Tatsunori Hashimoto | 2023-07-11T17:06:19Z | cs.CL | Recent work has identified noisy and misannotated data as a core cause of
hallucinations and unfaithful outputs in Natural Language Generation (NLG)
tasks. Consequently, identifying and removing these examples is a key open
challenge in creating reliable NLG systems. In this work, we introduce a
framework to identify and remove low-quality training instances that lead to
undesirable outputs, such as faithfulness errors in text summarization. We show
that existing approaches for error tracing, such as gradient-based influence
measures, do not perform reliably for detecting faithfulness errors in NLG
datasets. We overcome the drawbacks of existing error tracing methods through a
new, contrast-based estimate that compares undesired generations to
human-corrected outputs. Our proposed method can achieve a mean average
precision of 0.93 at detecting known data errors across synthetic tasks with
known ground truth, substantially outperforming existing approaches. Using this
approach and re-training models on cleaned data leads to a 70% reduction in
entity hallucinations on the NYT dataset and a 55% reduction in semantic errors
on the E2E dataset. | cs.CL | Contrastive Error Attribution for Finetuned Language Models |
2022-12-21T02:21:37Z | http://arxiv.org/pdf/2212.10720v2 | http://arxiv.org/abs/2212.10720v2 | Hao Sun, Zhexin Zhang, Fei Mi, Yasheng Wang, Wei Liu, Jianwei Cui, Bin Wang, Qun Liu, Minlie Huang | 2023-05-26T14:27:14Z | cs.CL | Morality in dialogue systems has raised great attention in research recently.
A moral dialogue system aligned with users' values could enhance conversation
engagement and user connections. In this paper, we propose a framework,
MoralDial to train and evaluate moral dialogue systems. In our framework, we
first explore the communication mechanisms of morality and resolve expressed
morality into three parts, which indicate the roadmap for building a moral
dialogue system. Based on that, we design a simple yet effective method:
constructing moral discussions between simulated specific users and the
dialogue system. The constructed discussions consist of expressing, explaining,
revising, and inferring moral views in dialogue exchanges, which makes
conversational models learn morality well in a natural manner. Furthermore, we
propose a novel evaluation method under the framework. We evaluate the multiple
aspects of morality by judging the relation between dialogue responses and
human values in discussions, where the multifaceted nature of morality is
particularly considered. Automatic and manual experiments demonstrate that our
framework is promising to train and evaluate moral dialogue systems. | cs.CL | MoralDial: A Framework to Train and Evaluate Moral Dialogue Systems via Moral Discussions |
2022-12-21T01:26:07Z | http://arxiv.org/pdf/2212.10714v1 | http://arxiv.org/abs/2212.10714v1 | Masaki Asada | 2022-12-21T01:26:07Z | cs.CL | The development of deep neural networks has improved representation learning
in various domains, including textual, graph structural, and relational triple
representations. This development opened the door to new relation extraction
beyond the traditional text-oriented relation extraction. However, research on
the effectiveness of considering multiple heterogeneous domain information
simultaneously is still under exploration, and if a model can take an advantage
of integrating heterogeneous information, it is expected to exhibit a
significant contribution to many problems in the world. This thesis works on
Drug-Drug Interactions (DDIs) from the literature as a case study and realizes
relation extraction utilizing heterogeneous domain information. First, a deep
neural relation extraction model is prepared and its attention mechanism is
analyzed. Next, a method to combine the drug molecular structure information
and drug description information to the input sentence information is proposed,
and the effectiveness of utilizing drug molecular structures and drug
descriptions for the relation extraction task is shown. Then, in order to
further exploit the heterogeneous information, drug-related items, such as
protein entries, medical terms and pathways are collected from multiple
existing databases and a new data set in the form of a knowledge graph (KG) is
constructed. A link prediction task on the constructed data set is conducted to
obtain embedding representations of drugs that contain the heterogeneous domain
information. Finally, a method that integrates the input sentence information
and the heterogeneous KG information is proposed. The proposed model is trained
and evaluated on a widely used data set, and as a result, it is shown that
utilizing heterogeneous domain information significantly improves the
performance of relation extraction from the literature. | cs.CL | Integrating Heterogeneous Domain Information into Relation Extraction: A Case Study on Drug-Drug Interaction Extraction |
2022-12-21T00:57:24Z | http://arxiv.org/pdf/2212.10708v2 | http://arxiv.org/abs/2212.10708v2 | Bosung Kim, Hayate Iso, Nikita Bhutani, Estevam Hruschka, Ndapa Nakashole, Tom Mitchell | 2023-09-20T05:12:27Z | cs.CL | The task of triplet extraction aims to extract pairs of entities and their
corresponding relations from unstructured text. Most existing methods train an
extraction model on training data involving specific target relations, and are
incapable of extracting new relations that were not observed at training time.
Generalizing the model to unseen relations typically requires fine-tuning on
synthetic training data which is often noisy and unreliable. We show that by
reducing triplet extraction to a template infilling task over a pre-trained
language model (LM), we can equip the extraction model with zero-shot learning
capabilities and eliminate the need for additional training data. We propose a
novel framework, ZETT (ZEro-shot Triplet extraction by Template infilling),
that aligns the task objective to the pre-training objective of generative
transformers to generalize to unseen relations. Experiments on FewRel and
Wiki-ZSL datasets demonstrate that ZETT shows consistent and stable
performance, outperforming previous state-of-the-art methods, even when using
automatically generated templates. https://github.com/megagonlabs/zett/ | cs.CL | Zero-shot Triplet Extraction by Template Infilling |
2022-12-21T00:56:50Z | http://arxiv.org/pdf/2212.10707v1 | http://arxiv.org/abs/2212.10707v1 | Vinícius Camargo da Silva, João Paulo Papa, Kelton Augusto Pontara da Costa | 2022-12-21T00:56:50Z | cs.CL, cs.LG | Automatic Text Summarization (ATS) is becoming relevant with the growth of
textual data; however, with the popularization of public large-scale datasets,
some recent machine learning approaches have focused on dense models and
architectures that, despite producing notable results, usually turn out in
models difficult to interpret. Given the challenge behind interpretable
learning-based text summarization and the importance it may have for evolving
the current state of the ATS field, this work studies the application of two
modern Generalized Additive Models with interactions, namely Explainable
Boosting Machine and GAMI-Net, to the extractive summarization problem based on
linguistic features and binary classification. | cs.CL | Extractive Text Summarization Using Generalized Additive Models with Interactions for Sentence Selection |
2022-12-21T00:00:01Z | http://arxiv.org/pdf/2212.10696v2 | http://arxiv.org/abs/2212.10696v2 | Akshay Chaturvedi, Swarnadeep Bhar, Soumadeep Saha, Utpal Garain, Nicholas Asher | 2023-11-30T10:42:26Z | cs.CL, cs.LG | Transformer-based language models have been shown to be highly effective for
several NLP tasks. In this paper, we consider three transformer models, BERT,
RoBERTa, and XLNet, in both small and large versions, and investigate how
faithful their representations are with respect to the semantic content of
texts. We formalize a notion of semantic faithfulness, in which the semantic
content of a text should causally figure in a model's inferences in question
answering. We then test this notion by observing a model's behavior on
answering questions about a story after performing two novel semantic
interventions: deletion intervention and negation intervention. While
transformer models achieve high performance on standard question answering
tasks, we show that they fail to be semantically faithful once we perform these
interventions for a significant number of cases (~50% for deletion
intervention, and ~20% drop in accuracy for negation intervention). We then
propose an intervention-based training regime that can mitigate the undesirable
effects for deletion intervention by a significant margin (from ~ 50% to ~6%).
We analyze the inner-workings of the models to better understand the
effectiveness of intervention-based training for deletion intervention. But we
show that this training does not attenuate other aspects of semantic
unfaithfulness such as the models' inability to deal with negation intervention
or to capture the predicate-argument structure of texts. We also test
InstructGPT, via prompting, for its ability to handle the two interventions and
to capture predicate-argument structure. While InstructGPT models do achieve
very high performance on predicate-argument structure task, they fail to
respond adequately to our deletion and negation interventions. | cs.CL | Analyzing Semantic Faithfulness of Language Models via Input Intervention on Question Answering |
2022-12-20T23:49:37Z | http://arxiv.org/pdf/2212.10692v1 | http://arxiv.org/abs/2212.10692v1 | Dong Li, Yelong Shen, Ruoming Jin, Yi Mao, Kuan Wang, Weizhu Chen | 2022-12-20T23:49:37Z | cs.SE, cs.AI, cs.CL | Pre-trained language models have achieved promising success in code retrieval
tasks, where a natural language documentation query is given to find the most
relevant existing code snippet. However, existing models focus only on
optimizing the documentation code pairs by embedding them into latent space,
without the association of external knowledge. In this paper, we propose a
generation-augmented query expansion framework. Inspired by the human retrieval
process - sketching an answer before searching, in this work, we utilize the
powerful code generation model to benefit the code retrieval task.
Specifically, we demonstrate that rather than merely retrieving the target code
snippet according to the documentation query, it would be helpful to augment
the documentation query with its generation counterpart - generated code
snippets from the code generation model. To the best of our knowledge, this is
the first attempt that leverages the code generation model to enhance the code
retrieval task. We achieve new state-of-the-art results on the CodeSearchNet
benchmark and surpass the baselines significantly. | cs.SE | Generation-Augmented Query Expansion For Code Retrieval |
2022-12-20T23:30:47Z | http://arxiv.org/pdf/2212.10690v1 | http://arxiv.org/abs/2212.10690v1 | Daniel Lukas Rothenpieler, Shahin Amiriparian | 2022-12-20T23:30:47Z | cs.CV, cs.CL, cs.LG, I.2.10 | Automatic video captioning aims for a holistic visual scene understanding. It
requires a mechanism for capturing temporal context in video frames and the
ability to comprehend the actions and associations of objects in a given
timeframe. Such a system should additionally learn to abstract video sequences
into sensible representations as well as to generate natural written language.
While the majority of captioning models focus solely on the visual inputs,
little attention has been paid to the audiovisual modality. To tackle this
issue, we propose a novel two-fold approach. First, we implement a
reward-guided KL Divergence to train a video captioning model which is
resilient towards token permutations. Second, we utilise a Bi-Modal
Hierarchical Reinforcement Learning (BMHRL) Transformer architecture to capture
long-term temporal dependencies of the input data as a foundation for our
hierarchical captioning module. Using our BMHRL, we show the suitability of the
HRL agent in the generation of content-complete and grammatically sound
sentences by achieving $4.91$, $2.23$, and $10.80$ in BLEU3, BLEU4, and METEOR
scores, respectively on the ActivityNet Captions dataset. Finally, we make our
BMHRL framework and trained models publicly available for users and developers
at https://github.com/d-rothen/bmhrl. | cs.CV | METEOR Guided Divergence for Video Captioning |
2022-12-20T22:41:24Z | http://arxiv.org/pdf/2212.10678v1 | http://arxiv.org/abs/2212.10678v1 | Justus Mattern, Zhijing Jin, Mrinmaya Sachan, Rada Mihalcea, Bernhard Schölkopf | 2022-12-20T22:41:24Z | cs.CL, cs.LG | Generated texts from large pretrained language models have been shown to
exhibit a variety of harmful, human-like biases about various demographics.
These findings prompted large efforts aiming to understand and measure such
effects, with the goal of providing benchmarks that can guide the development
of techniques mitigating these stereotypical associations. However, as recent
research has pointed out, the current benchmarks lack a robust experimental
setup, consequently hindering the inference of meaningful conclusions from
their evaluation metrics. In this paper, we extend these arguments and
demonstrate that existing techniques and benchmarks aiming to measure
stereotypes tend to be inaccurate and consist of a high degree of experimental
noise that severely limits the knowledge we can gain from benchmarking language
models based on them. Accordingly, we propose a new framework for robustly
measuring and quantifying biases exhibited by generative language models.
Finally, we use this framework to investigate GPT-3's occupational gender bias
and propose prompting techniques for mitigating these biases without the need
for fine-tuning. | cs.CL | Understanding Stereotypes in Language Models: Towards Robust Measurement and Zero-Shot Debiasing |
2022-12-20T22:11:35Z | http://arxiv.org/pdf/2212.10670v1 | http://arxiv.org/abs/2212.10670v1 | Yukun Huang, Yanda Chen, Zhou Yu, Kathleen McKeown | 2022-12-20T22:11:35Z | cs.CL, cs.LG | Given the success with in-context learning of large pre-trained language
models, we introduce in-context learning distillation to transfer in-context
few-shot learning ability from large models to smaller models. We propose to
combine in-context learning objectives with language modeling objectives to
distill both the ability to read in-context examples and task knowledge to the
smaller models. We perform in-context learning distillation under two different
few-shot learning paradigms: Meta In-context Tuning (Meta-ICT) and Multitask
In-context Tuning (Multitask-ICT). Multitask-ICT performs better on multitask
few-shot learning but also requires more computation than Meta-ICT. Our method
shows consistent improvements for both Meta-ICT and Multitask-ICT on two
benchmarks: LAMA and CrossFit. Our extensive experiments and analysis reveal
that in-context learning objectives and language modeling objectives are
complementary under the Multitask-ICT paradigm. In-context learning objectives
achieve the best performance when combined with language modeling objectives. | cs.CL | In-context Learning Distillation: Transferring Few-shot Learning Ability of Pre-trained Language Models |
2022-12-20T20:56:52Z | http://arxiv.org/pdf/2212.10650v1 | http://arxiv.org/abs/2212.10650v1 | Ali Edalati, Marzieh Tahaei, Ivan Kobyzev, Vahid Partovi Nia, James J. Clark, Mehdi Rezagholizadeh | 2022-12-20T20:56:52Z | cs.CL | Fine-tuning a Pre-trained Language Model (PLM) on a specific downstream task
has been a well-known paradigm in Natural Language Processing. However, with
the ever-growing size of PLMs, training the entire model on several downstream
tasks becomes very expensive and resource-hungry. Recently, different Parameter
Efficient Tuning (PET) techniques are proposed to improve the efficiency of
fine-tuning PLMs. One popular category of PET methods is the low-rank
adaptation methods which insert learnable truncated SVD modules into the
original model either sequentially or in parallel. However, low-rank
decomposition suffers from limited representation power. In this work, we
address this problem using the Kronecker product instead of the low-rank
representation. We introduce KronA, a Kronecker product-based adapter module
for efficient fine-tuning of Transformer-based PLMs. We apply the proposed
methods for fine-tuning T5 on the GLUE benchmark to show that incorporating the
Kronecker-based modules can outperform state-of-the-art PET methods. | cs.CL | KronA: Parameter Efficient Tuning with Kronecker Adapter |
2022-12-20T19:52:41Z | http://arxiv.org/pdf/2212.10622v2 | http://arxiv.org/abs/2212.10622v2 | Roee Aharoni, Shashi Narayan, Joshua Maynez, Jonathan Herzig, Elizabeth Clark, Mirella Lapata | 2024-01-05T12:13:55Z | cs.CL | Abstractive summarization has enjoyed renewed interest in recent years,
thanks to pre-trained language models and the availability of large-scale
datasets. Despite promising results, current models still suffer from
generating factually inconsistent summaries, reducing their utility for
real-world application. Several recent efforts attempt to address this by
devising models that automatically detect factual inconsistencies in machine
generated summaries. However, they focus exclusively on English, a language
with abundant resources. In this work, we leverage factual consistency
evaluation models to improve multilingual summarization. We explore two
intuitive approaches to mitigate hallucinations based on the signal provided by
a multilingual NLI model, namely data filtering and controlled generation.
Experimental results in the 45 languages from the XLSum dataset show gains over
strong baselines in both automatic and human evaluation. | cs.CL | mFACE: Multilingual Summarization with Factual Consistency Evaluation |
2022-12-20T19:48:10Z | http://arxiv.org/pdf/2212.10618v2 | http://arxiv.org/abs/2212.10618v2 | Nathaniel Weir, Ryan Thomas, Randolph D'Amore, Kellie Hill, Benjamin Van Durme, Harsh Jhamtani | 2023-05-13T19:20:17Z | cs.CL | We introduce a language generation task grounded in a popular video game
environment. KNUDGE (KNowledge Constrained User-NPC Dialogue GEneration)
requires models to produce trees of dialogue between video game characters that
accurately reflect quest and entity specifications stated in natural language.
KNUDGE is constructed from side quest dialogues drawn directly from game data
of Obsidian Entertainment's The Outer Worlds, leading to real-world
complexities in generation: (1) dialogues are branching trees as opposed to
linear chains of utterances; (2) utterances must remain faithful to the game
lore -- character personas, backstories, and entity relationships; and (3) a
dialogue must accurately reveal new quest details to the human player. We
report results for a set of neural generation models using supervised and
in-context learning techniques; we find competent performance but room for
future work addressing the challenges of creating realistic, game-quality
dialogues. | cs.CL | Ontologically Faithful Generation of Non-Player Character Dialogues |
2022-12-20T19:37:20Z | http://arxiv.org/pdf/2212.11680v2 | http://arxiv.org/abs/2212.11680v2 | Josip Jukić, Jan Šnajder | 2023-10-23T14:05:49Z | cs.LG, cs.CL | Developed to alleviate prohibitive labeling costs, active learning (AL)
methods aim to reduce label complexity in supervised learning. While recent
work has demonstrated the benefit of using AL in combination with large
pre-trained language models (PLMs), it has often overlooked the practical
challenges that hinder the effectiveness of AL. We address these challenges by
leveraging representation smoothness analysis to ensure AL is feasible, that
is, both effective and practicable. Firstly, we propose an early stopping
technique that does not require a validation set -- often unavailable in
realistic AL conditions -- and observe significant improvements over random
sampling across multiple datasets and AL methods. Further, we find that task
adaptation improves AL, whereas standard short fine-tuning in AL does not
provide improvements over random sampling. Our work demonstrates the usefulness
of representation smoothness analysis for AL and introduces an AL stopping
criterion that reduces label complexity. | cs.LG | Smooth Sailing: Improving Active Learning for Pre-trained Language Models with Representation Smoothness Analysis |
2022-12-20T18:59:50Z | http://arxiv.org/pdf/2212.10564v3 | http://arxiv.org/abs/2212.10564v3 | Boyi Li, Rodolfo Corona, Karttikeya Mangalam, Catherine Chen, Daniel Flaherty, Serge Belongie, Kilian Q. Weinberger, Jitendra Malik, Trevor Darrell, Dan Klein | 2024-04-12T14:53:30Z | cs.CL, cs.AI, cs.LG | Are multimodal inputs necessary for grammar induction? Recent work has shown
that multimodal training inputs can improve grammar induction. However, these
improvements are based on comparisons to weak text-only baselines that were
trained on relatively little textual data. To determine whether multimodal
inputs are needed in regimes with large amounts of textual training data, we
design a stronger text-only baseline, which we refer to as LC-PCFG. LC-PCFG is
a C-PFCG that incorporates em-beddings from text-only large language models
(LLMs). We use a fixed grammar family to directly compare LC-PCFG to various
multi-modal grammar induction methods. We compare performance on four benchmark
datasets. LC-PCFG provides an up to 17% relative improvement in Corpus-F1
compared to state-of-the-art multimodal grammar induction methods. LC-PCFG is
also more computationally efficient, providing an up to 85% reduction in
parameter count and 8.8x reduction in training time compared to multimodal
approaches. These results suggest that multimodal inputs may not be necessary
for grammar induction, and emphasize the importance of strong vision-free
baselines for evaluating the benefit of multimodal approaches. | cs.CL | Re-evaluating the Need for Multimodal Signals in Unsupervised Grammar Induction |
2022-12-20T18:59:42Z | http://arxiv.org/pdf/2212.10563v2 | http://arxiv.org/abs/2212.10563v2 | Hadas Orgad, Yonatan Belinkov | 2023-06-12T03:55:12Z | cs.CL, 68T50, I.2.7 | Models trained on real-world data tend to imitate and amplify social biases.
Common methods to mitigate biases require prior information on the types of
biases that should be mitigated (e.g., gender or racial bias) and the social
groups associated with each data sample. In this work, we introduce BLIND, a
method for bias removal with no prior knowledge of the demographics in the
dataset. While training a model on a downstream task, BLIND detects biased
samples using an auxiliary model that predicts the main model's success, and
down-weights those samples during the training process. Experiments with racial
and gender biases in sentiment classification and occupation classification
tasks demonstrate that BLIND mitigates social biases without relying on a
costly demographic annotation process. Our method is competitive with other
methods that require demographic information and sometimes even surpasses them. | cs.CL | BLIND: Bias Removal With No Demographics |
2022-12-20T18:59:23Z | http://arxiv.org/pdf/2212.10561v3 | http://arxiv.org/abs/2212.10561v3 | Eric Zelikman, Qian Huang, Gabriel Poesia, Noah D. Goodman, Nick Haber | 2023-05-28T21:12:31Z | cs.CL, cs.AI, cs.LG | Despite recent success in large language model (LLM) reasoning, LLMs struggle
with hierarchical multi-step reasoning tasks like generating complex programs.
For these tasks, humans often start with a high-level algorithmic design and
implement each part gradually. We introduce Parsel, a framework enabling
automatic implementation and validation of complex algorithms with code LLMs.
With Parsel, we automatically decompose algorithmic tasks into hierarchical
natural language function descriptions and then search over combinations of
possible function implementations using tests. We show that Parsel can be used
across domains requiring hierarchical reasoning, including program synthesis
and robotic planning. We find that, using Parsel, LLMs solve more
competition-level problems in the APPS dataset, resulting in pass rates over
75\% higher than prior results from directly sampling AlphaCode and Codex,
while often using a smaller sample budget. Moreover, with automatically
generated tests, we find that Parsel can improve the state-of-the-art pass@1
performance on HumanEval from 67\% to 85\%. We also find that LLM-generated
robotic plans using Parsel are more than twice as likely to be considered
accurate than directly generated plans. Lastly, we explore how Parsel addresses
LLM limitations and discuss how Parsel may be useful for human programmers. We
release our code at https://github.com/ezelikman/parsel | cs.CL | Parsel: Algorithmic Reasoning with Language Models by Composing Decompositions |
2022-12-20T18:59:23Z | http://arxiv.org/pdf/2212.10562v2 | http://arxiv.org/abs/2212.10562v2 | Rosanne Liu, Dan Garrette, Chitwan Saharia, William Chan, Adam Roberts, Sharan Narang, Irina Blok, RJ Mical, Mohammad Norouzi, Noah Constant | 2023-05-03T16:36:38Z | cs.CL, cs.CV | Current image generation models struggle to reliably produce well-formed
visual text. In this paper, we investigate a key contributing factor: popular
text-to-image models lack character-level input features, making it much harder
to predict a word's visual makeup as a series of glyphs. To quantify this
effect, we conduct a series of experiments comparing character-aware vs.
character-blind text encoders. In the text-only domain, we find that
character-aware models provide large gains on a novel spelling task
(WikiSpell). Applying our learnings to the visual domain, we train a suite of
image generation models, and show that character-aware variants outperform
their character-blind counterparts across a range of novel text rendering tasks
(our DrawText benchmark). Our models set a much higher state-of-the-art on
visual spelling, with 30+ point accuracy gains over competitors on rare words,
despite training on far fewer examples. | cs.CL | Character-Aware Models Improve Visual Text Rendering |
2022-12-20T18:59:19Z | http://arxiv.org/pdf/2212.10560v2 | http://arxiv.org/abs/2212.10560v2 | Yizhong Wang, Yeganeh Kordi, Swaroop Mishra, Alisa Liu, Noah A. Smith, Daniel Khashabi, Hannaneh Hajishirzi | 2023-05-25T23:50:07Z | cs.CL, cs.AI | Large "instruction-tuned" language models (i.e., finetuned to respond to
instructions) have demonstrated a remarkable ability to generalize zero-shot to
new tasks. Nevertheless, they depend heavily on human-written instruction data
that is often limited in quantity, diversity, and creativity, therefore
hindering the generality of the tuned model. We introduce Self-Instruct, a
framework for improving the instruction-following capabilities of pretrained
language models by bootstrapping off their own generations. Our pipeline
generates instructions, input, and output samples from a language model, then
filters invalid or similar ones before using them to finetune the original
model. Applying our method to the vanilla GPT3, we demonstrate a 33% absolute
improvement over the original model on Super-NaturalInstructions, on par with
the performance of InstructGPT-001, which was trained with private user data
and human annotations. For further evaluation, we curate a set of
expert-written instructions for novel tasks, and show through human evaluation
that tuning GPT3 with Self-Instruct outperforms using existing public
instruction datasets by a large margin, leaving only a 5% absolute gap behind
InstructGPT-001. Self-Instruct provides an almost annotation-free method for
aligning pre-trained language models with instructions, and we release our
large synthetic dataset to facilitate future studies on instruction tuning. Our
code and data are available at https://github.com/yizhongw/self-instruct. | cs.CL | Self-Instruct: Aligning Language Models with Self-Generated Instructions |
2022-12-20T18:58:48Z | http://arxiv.org/pdf/2212.10559v3 | http://arxiv.org/abs/2212.10559v3 | Damai Dai, Yutao Sun, Li Dong, Yaru Hao, Shuming Ma, Zhifang Sui, Furu Wei | 2023-05-15T11:45:12Z | cs.CL | Large pretrained language models have shown surprising in-context learning
(ICL) ability. With a few demonstration input-label pairs, they can predict the
label for an unseen input without parameter updates. Despite the great success
in performance, its working mechanism still remains an open question. In this
paper, we explain language models as meta-optimizers and understand in-context
learning as implicit finetuning. Theoretically, we figure out that Transformer
attention has a dual form of gradient descent. On top of it, we understand ICL
as follows: GPT first produces meta-gradients according to the demonstration
examples, and then these meta-gradients are applied to the original GPT to
build an ICL model. We comprehensively compare the behaviors of in-context
learning and explicit finetuning on real tasks to provide empirical evidence
that supports our understanding. Experimental results show that in-context
learning behaves similarly to explicit finetuning from multiple perspectives.
Inspired by the dual form between Transformer attention and gradient descent,
we design a momentum-based attention by analogy with gradient descent with
momentum. The improved performance over vanilla attention further supports our
understanding from another perspective, and more importantly, shows the
potential to utilize our understanding for future model design. The code is
available at \url{https://aka.ms/icl}. | cs.CL | Why Can GPT Learn In-Context? Language Models Implicitly Perform Gradient Descent as Meta-Optimizers |
2022-12-20T18:58:33Z | http://arxiv.org/pdf/2212.10558v2 | http://arxiv.org/abs/2212.10558v2 | Tianqing Fang, Wenxuan Zhou, Fangyu Liu, Hongming Zhang, Yangqiu Song, Muhao Chen | 2024-01-31T13:14:02Z | cs.CL, cs.AI | Data Augmentation (DA) is frequently used to provide additional training data
without extra human annotation automatically. However, data augmentation may
introduce noisy data that impairs training. To guarantee the quality of
augmented data, existing methods either assume no noise exists in the augmented
data and adopt consistency training or use simple heuristics such as training
loss and diversity constraints to filter out "noisy" data. However, those
filtered examples may still contain useful information, and dropping them
completely causes a loss of supervision signals. In this paper, based on the
assumption that the original dataset is cleaner than the augmented data, we
propose an on-the-fly denoising technique for data augmentation that learns
from soft augmented labels provided by an organic teacher model trained on the
cleaner original data. To further prevent overfitting on noisy labels, a simple
self-regularization module is applied to force the model prediction to be
consistent across two distinct dropouts. Our method can be applied to general
augmentation techniques and consistently improve the performance on both text
classification and question-answering tasks. | cs.CL | On-the-fly Denoising for Data Augmentation in Natural Language Understanding |
2022-12-20T18:57:18Z | http://arxiv.org/pdf/2212.10557v2 | http://arxiv.org/abs/2212.10557v2 | Prakhar Gupta, Yang Liu, Di Jin, Behnam Hedayatnia, Spandana Gella, Sijia Liu, Patrick Lange, Julia Hirschberg, Dilek Hakkani-Tur | 2023-05-21T19:38:01Z | cs.CL | Dialogue models are able to generate coherent and fluent responses, but they
can still be challenging to control and may produce non-engaging, unsafe
results. This unpredictability diminishes user trust and can hinder the use of
the models in the real world. To address this, we introduce DialGuide, a novel
framework for controlling dialogue model behavior using natural language rules,
or guidelines. These guidelines provide information about the context they are
applicable to and what should be included in the response, allowing the models
to generate responses that are more closely aligned with the developer's
expectations and intent. We evaluate DialGuide on three tasks in open-domain
dialogue response generation: guideline selection, response generation, and
response entailment verification. Our dataset contains 10,737 positive and
15,467 negative dialogue context-response-guideline triplets across two domains
- chit-chat and safety. We provide baseline models for the tasks and benchmark
their performance. We also demonstrate that DialGuide is effective in the
dialogue safety domain, producing safe and engaging responses that follow
developer guidelines. | cs.CL | DialGuide: Aligning Dialogue Model Behavior with Developer Guidelines |
2022-12-20T18:56:57Z | http://arxiv.org/pdf/2212.10555v1 | http://arxiv.org/abs/2212.10555v1 | Dongfu Jiang, Bill Yuchen Lin, Xiang Ren | 2022-12-20T18:56:57Z | cs.CL, cs.AI, cs.LG | Pre-trained language models have been successful in natural language
generation (NLG) tasks. While various decoding methods have been employed, they
often produce suboptimal results. We first present an empirical analysis of
three NLG tasks: summarization, machine translation, and constrained text
generation. We found that selecting the best output from the results of
multiple decoding methods can significantly improve performance. To further
improve reranking for NLG tasks, we proposed a novel method,
\textsc{PairReranker}, which uses a single encoder and a pairwise loss function
to jointly encode a source input and a pair of candidates and compare them.
Experiments on three NLG tasks demonstrated the effectiveness and flexibility
of \textsc{PairReranker}, showing strong results, compared with previous
baselines. In addition, our \textsc{PairReranker} can generalize to
significantly improve GPT-3 (text-davinci-003) results (e.g., 24.55\% on
CommonGen and 11.35\% on WMT18 zh-en), even though our rerankers are not
trained with any GPT-3 candidates. | cs.CL | PairReranker: Pairwise Reranking for Natural Language Generation |
2022-12-20T18:56:20Z | http://arxiv.org/pdf/2212.10554v1 | http://arxiv.org/abs/2212.10554v1 | Yutao Sun, Li Dong, Barun Patra, Shuming Ma, Shaohan Huang, Alon Benhaim, Vishrav Chaudhary, Xia Song, Furu Wei | 2022-12-20T18:56:20Z | cs.CL | Position modeling plays a critical role in Transformers. In this paper, we
focus on length extrapolation, i.e., training on short texts while evaluating
longer sequences. We define attention resolution as an indicator of
extrapolation. Then we propose two designs to improve the above metric of
Transformers. Specifically, we introduce a relative position embedding to
explicitly maximize attention resolution. Moreover, we use blockwise causal
attention during inference for better resolution. We evaluate different
Transformer variants with language modeling. Experimental results show that our
model achieves strong performance in both interpolation and extrapolation
settings. The code will be available at https://aka.ms/LeX-Transformer. | cs.CL | A Length-Extrapolatable Transformer |
2022-12-20T18:54:08Z | http://arxiv.org/pdf/2212.10551v3 | http://arxiv.org/abs/2212.10551v3 | Fei Yuan, Yinquan Lu, WenHao Zhu, Lingpeng Kong, Lei Li, Yu Qiao, Jingjing Xu | 2023-07-19T05:52:32Z | cs.CL, cs.AI | Multilingual neural machine translation (MNMT) aims to build a unified model
for many language directions. Existing monolithic models for MNMT encounter two
challenges: parameter interference among languages and inefficient inference
for large models. In this paper, we revisit the classic multi-way structures
and develop a detachable model by assigning each language (or group of
languages) to an individual branch that supports plug-and-play training and
inference. To address the needs of learning representations for all languages
in a unified space, we propose a novel efficient training recipe, upon which we
build an effective detachable model, Lego-MT. For a fair comparison, we collect
data from OPUS and build a translation benchmark covering 433 languages and
1.3B parallel data. Experiments show that Lego-MT with 1.2B parameters brings
an average gain of 3.2 spBLEU. It even outperforms M2M-100 with 12B parameters.
The proposed training recipe brings a 28.2$\times$ speedup over the
conventional multi-way training method.\footnote{
\url{https://github.com/CONE-MT/Lego-MT}.} | cs.CL | Lego-MT: Learning Detachable Models for Massively Multilingual Machine Translation |
2022-12-20T18:53:14Z | http://arxiv.org/pdf/2212.10549v2 | http://arxiv.org/abs/2212.10549v2 | Rohan Pandey, Rulin Shao, Paul Pu Liang, Ruslan Salakhutdinov, Louis-Philippe Morency | 2023-07-04T13:18:29Z | cs.CL, cs.CV, cs.LG | Despite recent progress towards scaling up multimodal vision-language models,
these models are still known to struggle on compositional generalization
benchmarks such as Winoground. We find that a critical component lacking from
current vision-language models is relation-level alignment: the ability to
match directional semantic relations in text (e.g., "mug in grass") with
spatial relationships in the image (e.g., the position of the mug relative to
the grass). To tackle this problem, we show that relation alignment can be
enforced by encouraging the directed language attention from 'mug' to 'grass'
(capturing the semantic relation 'in') to match the directed visual attention
from the mug to the grass. Tokens and their corresponding objects are softly
identified using the cross-modal attention. We prove that this notion of soft
relation alignment is equivalent to enforcing congruence between vision and
language attention matrices under a 'change of basis' provided by the
cross-modal attention matrix. Intuitively, our approach projects visual
attention into the language attention space to calculate its divergence from
the actual language attention, and vice versa. We apply our Cross-modal
Attention Congruence Regularization (CACR) loss to UNITER and improve on the
state-of-the-art approach to Winoground. | cs.CL | Cross-modal Attention Congruence Regularization for Vision-Language Relation Alignment |
2022-12-20T18:51:48Z | http://arxiv.org/pdf/2212.10548v2 | http://arxiv.org/abs/2212.10548v2 | Iker García-Ferrero, Rodrigo Agerri, German Rigau | 2023-10-24T10:49:58Z | cs.CL | In the absence of readily available labeled data for a given sequence
labeling task and language, annotation projection has been proposed as one of
the possible strategies to automatically generate annotated data. Annotation
projection has often been formulated as the task of transporting, on parallel
corpora, the labels pertaining to a given span in the source language into its
corresponding span in the target language. In this paper we present
T-Projection, a novel approach for annotation projection that leverages large
pretrained text-to-text language models and state-of-the-art machine
translation technology. T-Projection decomposes the label projection task into
two subtasks: (i) A candidate generation step, in which a set of projection
candidates using a multilingual T5 model is generated and, (ii) a candidate
selection step, in which the generated candidates are ranked based on
translation probabilities. We conducted experiments on intrinsic and extrinsic
tasks in 5 Indo-European and 8 low-resource African languages. We demostrate
that T-projection outperforms previous annotation projection methods by a wide
margin. We believe that T-Projection can help to automatically alleviate the
lack of high-quality training data for sequence labeling tasks. Code and data
are publicly available. | cs.CL | T-Projection: High Quality Annotation Projection for Sequence Labeling Tasks |
2022-12-20T18:51:23Z | http://arxiv.org/pdf/2212.10547v2 | http://arxiv.org/abs/2212.10547v2 | Shubhashis Roy Dipta, Mehdi Rezaee, Francis Ferraro | 2023-05-30T17:47:55Z | cs.CL | Prior work has shown that coupling sequential latent variable models with
semantic ontological knowledge can improve the representational capabilities of
event modeling approaches. In this work, we present a novel, doubly
hierarchical, semi-supervised event modeling framework that provides structural
hierarchy while also accounting for ontological hierarchy. Our approach
consists of multiple layers of structured latent variables, where each
successive layer compresses and abstracts the previous layers. We guide this
compression through the injection of structured ontological knowledge that is
defined at the type level of events: importantly, our model allows for partial
injection of semantic knowledge and it does not depend on observing instances
at any particular level of the semantic ontology. Across two different datasets
and four different evaluation metrics, we demonstrate that our approach is able
to out-perform the previous state-of-the-art approaches by up to 8.5%,
demonstrating the benefits of structured and semantic hierarchical knowledge
for event modeling. | cs.CL | Semantically-informed Hierarchical Event Modeling |
2022-12-20T18:50:29Z | http://arxiv.org/pdf/2212.10545v2 | http://arxiv.org/abs/2212.10545v2 | Chenzhengyi Liu, Jie Huang, Kerui Zhu, Kevin Chen-Chuan Chang | 2023-05-16T06:32:51Z | cs.CL, cs.AI | In this paper, we propose DimonGen, which aims to generate diverse sentences
describing concept relationships in various everyday scenarios. To support
this, we first create a benchmark dataset for this task by adapting the
existing CommonGen dataset. We then propose a two-stage model called MoREE to
generate the target sentences. MoREE consists of a mixture of retrievers model
that retrieves diverse context sentences related to the given concepts, and a
mixture of generators model that generates diverse sentences based on the
retrieved contexts. We conduct experiments on the DimonGen task and show that
MoREE outperforms strong baselines in terms of both the quality and diversity
of the generated sentences. Our results demonstrate that MoREE is able to
generate diverse sentences that reflect different relationships between
concepts, leading to a comprehensive understanding of concept relationships. | cs.CL | DimonGen: Diversified Generative Commonsense Reasoning for Explaining Concept Relationships |
Subsets and Splits