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
|
|---|---|---|---|---|---|---|---|---|
2022-12-20T05:58:32Z | http://arxiv.org/pdf/2212.10011v2 | http://arxiv.org/abs/2212.10011v2 | Jianfeng Chi, Wasi Uddin Ahmad, Yuan Tian, Kai-Wei Chang | 2023-05-12T07:38:29Z | cs.CL | Privacy policies provide individuals with information about their rights and
how their personal information is handled. Natural language understanding (NLU)
technologies can support individuals and practitioners to understand better
privacy practices described in lengthy and complex documents. However, existing
efforts that use NLU technologies are limited by processing the language in a
way exclusive to a single task focusing on certain privacy practices. To this
end, we introduce the Privacy Policy Language Understanding Evaluation (PLUE)
benchmark, a multi-task benchmark for evaluating the privacy policy language
understanding across various tasks. We also collect a large corpus of privacy
policies to enable privacy policy domain-specific language model pre-training.
We evaluate several generic pre-trained language models and continue
pre-training them on the collected corpus. We demonstrate that domain-specific
continual pre-training offers performance improvements across all tasks. | cs.CL | PLUE: Language Understanding Evaluation Benchmark for Privacy Policies in English |
2022-12-20T05:51:47Z | http://arxiv.org/pdf/2212.10008v2 | http://arxiv.org/abs/2212.10008v2 | Miaoran Li, Baolin Peng, Michel Galley, Jianfeng Gao, Zhu Zhang | 2023-07-30T16:04:23Z | cs.CL, cs.AI | Many efforts have been made to construct dialog systems for different types
of conversations, such as task-oriented dialog (TOD) and open-domain dialog
(ODD). To better mimic human-level conversations that usually fuse various
dialog modes, it is essential to build a system that can effectively handle
both TOD and ODD and access different knowledge sources. To address the lack of
available data for the fused task, we propose a framework for automatically
generating dialogues that combine knowledge-grounded ODDs and TODs in various
settings. Additionally, we introduce a unified model PivotBot that is capable
of appropriately adopting TOD and ODD modes and accessing different knowledge
sources in order to effectively tackle the fused task. Evaluation results
demonstrate the superior ability of the proposed model to switch seamlessly
between TOD and ODD tasks. | cs.CL | Enhancing Task Bot Engagement with Synthesized Open-Domain Dialog |
2022-12-20T05:48:09Z | http://arxiv.org/pdf/2212.10007v2 | http://arxiv.org/abs/2212.10007v2 | Yangruibo Ding, Zijian Wang, Wasi Uddin Ahmad, Murali Krishna Ramanathan, Ramesh Nallapati, Parminder Bhatia, Dan Roth, Bing Xiang | 2023-05-24T06:56:45Z | cs.CL, cs.SE | While pre-trained language models (LM) for code have achieved great success
in code completion, they generate code conditioned only on the contents within
the file, i.e., in-file context, but ignore the rich semantics in other files
within the same project, i.e., cross-file context, a critical source of
information that is especially useful in modern modular software development.
Such overlooking constrains code language models' capacity in code completion,
leading to unexpected behaviors such as generating hallucinated class member
functions or function calls with unexpected arguments. In this work, we develop
a cross-file context finder tool, CCFINDER, that effectively locates and
retrieves the most relevant cross-file context. We propose CoCoMIC, a framework
that incorporates cross-file context to learn the in-file and cross-file
context jointly on top of pretrained code LMs. CoCoMIC successfully improves
the existing code LM with a 33.94% relative increase in exact match and a
28.69% relative increase in identifier matching for code completion when the
cross-file context is provided. | cs.CL | CoCoMIC: Code Completion By Jointly Modeling In-file and Cross-file Context |
2022-12-20T05:25:12Z | http://arxiv.org/pdf/2212.10003v2 | http://arxiv.org/abs/2212.10003v2 | Najoung Kim, Phu Mon Htut, Samuel R. Bowman, Jackson Petty | 2023-08-29T19:36:32Z | cs.CL | Naturally occurring information-seeking questions often contain questionable
assumptions -- assumptions that are false or unverifiable. Questions containing
questionable assumptions are challenging because they require a distinct answer
strategy that deviates from typical answers for information-seeking questions.
For instance, the question "When did Marie Curie discover Uranium?" cannot be
answered as a typical "when" question without addressing the false assumption
"Marie Curie discovered Uranium". In this work, we propose (QA)$^2$ (Question
Answering with Questionable Assumptions), an open-domain evaluation dataset
consisting of naturally occurring search engine queries that may or may not
contain questionable assumptions. To be successful on (QA)$^2$, systems must be
able to detect questionable assumptions and also be able to produce adequate
responses for both typical information-seeking questions and ones with
questionable assumptions. Through human rater acceptability on end-to-end QA
with (QA)$^2$, we find that current models do struggle with handling
questionable assumptions, leaving substantial headroom for progress. | cs.CL | (QA)$^2$: Question Answering with Questionable Assumptions |
2022-12-20T05:25:01Z | http://arxiv.org/pdf/2212.10002v3 | http://arxiv.org/abs/2212.10002v3 | Orion Weller, Aleem Khan, Nathaniel Weir, Dawn Lawrie, Benjamin Van Durme | 2024-02-26T20:52:59Z | cs.CL, cs.IR | Recent work in open-domain question answering (ODQA) has shown that
adversarial poisoning of the search collection can cause large drops in
accuracy for production systems. However, little to no work has proposed
methods to defend against these attacks. To do so, we rely on the intuition
that redundant information often exists in large corpora. To find it, we
introduce a method that uses query augmentation to search for a diverse set of
passages that could answer the original question but are less likely to have
been poisoned. We integrate these new passages into the model through the
design of a novel confidence method, comparing the predicted answer to its
appearance in the retrieved contexts (what we call Confidence from Answer
Redundancy, i.e. CAR). Together these methods allow for a simple but effective
way to defend against poisoning attacks that provides gains of nearly 20% exact
match across varying levels of data poisoning/knowledge conflicts. | cs.CL | Defending Against Disinformation Attacks in Open-Domain Question Answering |
2022-12-20T05:20:54Z | http://arxiv.org/pdf/2212.10001v2 | http://arxiv.org/abs/2212.10001v2 | Boshi Wang, Sewon Min, Xiang Deng, Jiaming Shen, You Wu, Luke Zettlemoyer, Huan Sun | 2023-06-01T05:38:00Z | cs.CL | Chain-of-Thought (CoT) prompting can dramatically improve the multi-step
reasoning abilities of large language models (LLMs). CoT explicitly encourages
the LLM to generate intermediate rationales for solving a problem, by providing
a series of reasoning steps in the demonstrations. Despite its success, there
is still little understanding of what makes CoT prompting effective and which
aspects of the demonstrated reasoning steps contribute to its performance. In
this paper, we show that CoT reasoning is possible even with invalid
demonstrations - prompting with invalid reasoning steps can achieve over 80-90%
of the performance obtained using CoT under various metrics, while still
generating coherent lines of reasoning during inference. Further experiments
show that other aspects of the rationales, such as being relevant to the query
and correctly ordering the reasoning steps, are much more important for
effective CoT reasoning. Overall, these findings both deepen our understanding
of CoT prompting, and open up new questions regarding LLMs' capability to learn
to reason in context. | cs.CL | Towards Understanding Chain-of-Thought Prompting: An Empirical Study of What Matters |
2022-12-20T04:38:23Z | http://arxiv.org/pdf/2212.09994v1 | http://arxiv.org/abs/2212.09994v1 | Xinyu Pi, Bing Wang, Yan Gao, Jiaqi Guo, Zhoujun Li, Jian-Guang Lou | 2022-12-20T04:38:23Z | cs.CL | The robustness of Text-to-SQL parsers against adversarial perturbations plays
a crucial role in delivering highly reliable applications. Previous studies
along this line primarily focused on perturbations in the natural language
question side, neglecting the variability of tables. Motivated by this, we
propose the Adversarial Table Perturbation (ATP) as a new attacking paradigm to
measure the robustness of Text-to-SQL models. Following this proposition, we
curate ADVETA, the first robustness evaluation benchmark featuring natural and
realistic ATPs. All tested state-of-the-art models experience dramatic
performance drops on ADVETA, revealing models' vulnerability in real-world
practices. To defend against ATP, we build a systematic adversarial training
example generation framework tailored for better contextualization of tabular
data. Experiments show that our approach not only brings the best robustness
improvement against table-side perturbations but also substantially empowers
models against NL-side perturbations. We release our benchmark and code at:
https://github.com/microsoft/ContextualSP. | cs.CL | Towards Robustness of Text-to-SQL Models Against Natural and Realistic Adversarial Table Perturbation |
2022-12-20T03:54:44Z | http://arxiv.org/pdf/2212.09982v1 | http://arxiv.org/abs/2212.09982v1 | Mozhdeh Gheini, Tatiana Likhomanenko, Matthias Sperber, Hendra Setiawan | 2022-12-20T03:54:44Z | cs.CL, cs.SD, eess.AS | Self-training has been shown to be helpful in addressing data scarcity for
many domains, including vision, speech, and language. Specifically,
self-training, or pseudo-labeling, labels unsupervised data and adds that to
the training pool. In this work, we investigate and use pseudo-labeling for a
recently proposed novel setup: joint transcription and translation of speech,
which suffers from an absence of sufficient data resources. We show that under
such data-deficient circumstances, the unlabeled data can significantly vary in
domain from the supervised data, which results in pseudo-label quality
degradation. We investigate two categories of remedies that require no
additional supervision and target the domain mismatch: pseudo-label filtering
and data augmentation. We show that pseudo-label analysis and processing as
such results in additional gains on top of the vanilla pseudo-labeling setup
resulting in total improvements of up to 0.6% absolute WER and 2.2 BLEU points. | cs.CL | Joint Speech Transcription and Translation: Pseudo-Labeling with Out-of-Distribution Data |
2022-12-20T02:47:37Z | http://arxiv.org/pdf/2212.09968v2 | http://arxiv.org/abs/2212.09968v2 | Yixin Liu, Budhaditya Deb, Milagro Teruel, Aaron Halfaker, Dragomir Radev, Ahmed H. Awadallah | 2023-10-16T04:31:22Z | cs.CL | Despite the recent progress in language generation models, their outputs may
not always meet user expectations. In this work, we study whether informational
feedback in natural language can be leveraged to improve generation quality and
user preference alignment. To this end, we consider factual consistency in
summarization, the quality that the summary should only contain information
supported by the input documents, as the user-expected preference. We collect a
high-quality dataset, DeFacto, containing human demonstrations and
informational natural language feedback consisting of corrective instructions,
edited summaries, and explanations with respect to the factual consistency of
the summary. Using our dataset, we study three natural language generation
tasks: (1) editing a summary by following the human feedback, (2) generating
human feedback for editing the original summary, and (3) revising the initial
summary to correct factual errors by generating both the human feedback and
edited summary. We show that DeFacto can provide factually consistent
human-edited summaries and further insights into summarization factual
consistency thanks to its informational natural language feedback. We further
demonstrate that fine-tuned language models can leverage our dataset to improve
the summary factual consistency, while large language models lack the zero-shot
learning ability in our proposed tasks that require controllable text
generation. | cs.CL | On Improving Summarization Factual Consistency from Natural Language Feedback |
2022-12-20T02:17:30Z | http://arxiv.org/pdf/2212.09955v2 | http://arxiv.org/abs/2212.09955v2 | Liang Ma, Shuyang Cao, Robert L. Logan IV, Di Lu, Shihao Ran, Ke Zhang, Joel Tetreault, Alejandro Jaimes | 2023-06-05T01:29:40Z | cs.CL | The proliferation of automatic faithfulness metrics for summarization has
produced a need for benchmarks to evaluate them. While existing benchmarks
measure the correlation with human judgements of faithfulness on
model-generated summaries, they are insufficient for diagnosing whether metrics
are: 1) consistent, i.e., indicate lower faithfulness as errors are introduced
into a summary, 2) effective on human-written texts, and 3) sensitive to
different error types (as summaries can contain multiple errors). To address
these needs, we present a benchmark of unfaithful minimal pairs (BUMP), a
dataset of 889 human-written, minimally different summary pairs, where a single
error is introduced to a summary from the CNN/DailyMail dataset to produce an
unfaithful summary. We find BUMP complements existing benchmarks in a number of
ways: 1) the summaries in BUMP are harder to discriminate and less probable
under SOTA summarization models, 2) unlike non-pair-based datasets, BUMP can be
used to measure the consistency of metrics, and reveals that the most
discriminative metrics tend not to be the most consistent, and 3) unlike
datasets containing generated summaries with multiple errors, BUMP enables the
measurement of metrics' performance on individual error types. | cs.CL | BUMP: A Benchmark of Unfaithful Minimal Pairs for Meta-Evaluation of Faithfulness Metrics |
2022-12-20T01:53:26Z | http://arxiv.org/pdf/2212.09947v1 | http://arxiv.org/abs/2212.09947v1 | Brian D. Zimmerman, Gaurav Sahu, Olga Vechtomova | 2022-12-20T01:53:26Z | cs.CL, cs.AI, cs.LG | Recent advances in deep learning research, such as transformers, have
bolstered the ability for automated agents to generate creative texts similar
to those that a human would write. By default, transformer decoders can only
generate new text with respect to previously generated text. The output
distribution of candidate tokens at any position is conditioned on previously
selected tokens using a self-attention mechanism to emulate the property of
autoregression. This is inherently limiting for tasks such as controllable
story generation where it may be necessary to condition on future plot events
when writing a story. In this work, we propose Future Sight, a method for
finetuning a pretrained generative transformer on the task of future
conditioning. Transformer decoders are typically pretrained on the task of
completing a context, one token at a time, by means of self-attention. Future
Sight additionally enables a decoder to attend to an encoded future plot event.
This motivates the decoder to expand on the context in a way that logically
concludes with the provided future. During inference, the future plot event can
be written by a human author to steer the narrative being generated in a
certain direction. We evaluate the efficacy of our approach on a story
generation task with human evaluators. | cs.CL | Future Sight: Dynamic Story Generation with Large Pretrained Language Models |
2022-12-20T01:52:46Z | http://arxiv.org/pdf/2212.09946v1 | http://arxiv.org/abs/2212.09946v1 | Raphael Shu, Elman Mansimov, Tamer Alkhouli, Nikolaos Pappas, Salvatore Romeo, Arshit Gupta, Saab Mansour, Yi Zhang, Dan Roth | 2022-12-20T01:52:46Z | cs.CL | Functionality and dialogue experience are two important factors of
task-oriented dialogue systems. Conventional approaches with closed schema
(e.g., conversational semantic parsing) often fail as both the functionality
and dialogue experience are strongly constrained by the underlying schema. We
introduce a new paradigm for task-oriented dialogue - Dialog2API - to greatly
expand the functionality and provide seamless dialogue experience. The
conversational model interacts with the environment by generating and executing
programs triggering a set of pre-defined APIs. The model also manages the
dialogue policy and interact with the user through generating appropriate
natural language responses. By allowing generating free-form programs,
Dialog2API supports composite goals by combining different APIs, whereas
unrestricted program revision provides natural and robust dialogue experience.
To facilitate Dialog2API, the core model is provided with API documents, an
execution environment and optionally some example dialogues annotated with
programs. We propose an approach tailored for the Dialog2API, where the
dialogue states are represented by a stack of programs, with most recently
mentioned program on the top of the stack. Dialog2API can work with many
application scenarios such as software automation and customer service. In this
paper, we construct a dataset for AWS S3 APIs and present evaluation results of
in-context learning baselines. | cs.CL | Dialog2API: Task-Oriented Dialogue with API Description and Example Programs |
2022-12-20T01:23:01Z | http://arxiv.org/pdf/2212.09939v2 | http://arxiv.org/abs/2212.09939v2 | Jeffrey Zhao, Yuan Cao, Raghav Gupta, Harrison Lee, Abhinav Rastogi, Mingqiu Wang, Hagen Soltau, Izhak Shafran, Yonghui Wu | 2023-02-13T18:26:37Z | cs.CL | We propose AnyTOD, an end-to-end, zero-shot task-oriented dialog (TOD) system
capable of handling unseen tasks without task-specific training. We view TOD as
a program executed by a language model (LM), where program logic and ontology
is provided by a designer as a schema. To enable generalization to unseen
schemas and programs without prior training, AnyTOD adopts a neuro-symbolic
approach. A neural LM keeps track of events occurring during a conversation and
a symbolic program implementing the dialog policy is executed to recommend next
actions AnyTOD should take. This approach drastically reduces data annotation
and model training requirements, addressing the enduring challenge of rapidly
adapting a TOD system to unseen tasks and domains. We demonstrate
state-of-the-art results on STAR, ABCD and SGD benchmarks. We also demonstrate
strong zero-shot transfer ability in low-resource settings, such as zero-shot
on MultiWOZ. In addition, we release STARv2, an updated version of the STAR
dataset with richer annotations, for benchmarking zero-shot end-to-end TOD
models. | cs.CL | AnyTOD: A Programmable Task-Oriented Dialog System |
2022-12-20T00:33:11Z | http://arxiv.org/pdf/2212.09928v2 | http://arxiv.org/abs/2212.09928v2 | Kundan Krishna, Yao Zhao, Jie Ren, Balaji Lakshminarayanan, Jiaming Luo, Mohammad Saleh, Peter J. Liu | 2023-12-04T16:18:33Z | cs.CL, cs.LG | The evaluation of abstractive summarization models typically uses test data
that is identically distributed as training data. In real-world practice,
documents to be summarized may contain input noise caused by text extraction
artifacts or data pipeline bugs. The robustness of model performance under
distribution shift caused by such noise is relatively under-studied. We present
a large empirical study quantifying the sometimes severe loss in performance
(up to 12 ROUGE-1 points) from different types of input noise for a range of
datasets and model sizes. We then propose a light-weight method for detecting
and removing such noise in the input during model inference without requiring
any extra training, auxiliary models, or even prior knowledge of the type of
noise. Our proposed approach effectively mitigates the loss in performance,
recovering a large fraction of the performance drop, sometimes as large as 11
ROUGE-1 points. | cs.CL | Improving the Robustness of Summarization Models by Detecting and Removing Input Noise |
2022-12-19T23:45:05Z | http://arxiv.org/pdf/2212.09917v2 | http://arxiv.org/abs/2212.09917v2 | Yu Fu, Deyi Xiong, Yue Dong | 2023-12-05T01:06:17Z | cs.CL | We introduce inverse reinforcement learning (IRL) as an effective paradigm
for training abstractive summarization models, imitating human summarization
behaviors. Our IRL model estimates the reward function using a suite of
important sub-rewards for summarization and concurrently optimizes the policy
network. Experimental results across datasets in different domains
(CNN/DailyMail and WikiHow) and various model sizes (BART-base and BART-large)
demonstrate the superiority of our proposed IRL model for summarization over
MLE and RL baselines. The resulting summaries exhibit greater similarity to
human-crafted gold references, outperforming MLE and RL baselines on metrics
such as ROUGE, coverage, novelty, compression ratio, factuality, and human
evaluations. | cs.CL | Inverse Reinforcement Learning for Text Summarization |
2022-12-19T23:33:21Z | http://arxiv.org/pdf/2212.09912v2 | http://arxiv.org/abs/2212.09912v2 | Kaiser Sun, Peng Qi, Yuhao Zhang, Lan Liu, William Yang Wang, Zhiheng Huang | 2023-10-24T20:59:33Z | cs.CL | Generative models have been widely applied to solve extractive tasks, where
parts of the input is extracted to form the desired output, and achieved
significant success. For example, in extractive question answering (QA),
generative models have constantly yielded state-of-the-art results. In this
work, we identify the issue of tokenization inconsistency that is commonly
neglected in training these models. This issue damages the extractive nature of
these tasks after the input and output are tokenized inconsistently by the
tokenizer, and thus leads to performance drop as well as hallucination. We
propose a simple yet effective fix to this issue and conduct a case study on
extractive QA. We show that, with consistent tokenization, the model performs
better in both in-domain and out-of-domain datasets, with a notable average of
+1.7 F2 gain when a BART model is trained on SQuAD and evaluated on 8 QA
datasets. Further, the model converges faster, and becomes less likely to
generate out-of-context answers. With these findings, we would like to call for
more attention on how tokenization should be done when solving extractive tasks
and recommend applying consistent tokenization during training. | cs.CL | Tokenization Consistency Matters for Generative Models on Extractive NLP Tasks |
2022-12-19T22:37:46Z | http://arxiv.org/pdf/2212.09897v2 | http://arxiv.org/abs/2212.09897v2 | Jing Huang, Zhengxuan Wu, Kyle Mahowald, Christopher Potts | 2023-12-19T13:05:12Z | cs.CL | Language tasks involving character-level manipulations (e.g., spelling
corrections, arithmetic operations, word games) are challenging for models
operating on subword units. To address this, we develop a causal intervention
framework to learn robust and interpretable character representations inside
subword-based language models. Our method treats each character as a typed
variable in a causal model and learns such causal structures by adapting the
interchange intervention training method of Geiger et al. (2021). We
additionally introduce a suite of character-level tasks that systematically
vary in their dependence on meaning and sequence-level context. While
character-level models still perform best on purely form-based tasks like
string reversal, our method outperforms character-level models on more complex
tasks that blend form, meaning, and context, such as spelling correction in
context and word search games. Compared with standard subword-based models, our
approach also significantly improves robustness on unseen token sequences and
leads to human-interpretable internal representations of characters. | cs.CL | Inducing Character-level Structure in Subword-based Language Models with Type-level Interchange Intervention Training |
2022-12-19T22:36:53Z | http://arxiv.org/pdf/2212.09895v1 | http://arxiv.org/abs/2212.09895v1 | Arya D. McCarthy, Hao Zhang, Shankar Kumar, Felix Stahlberg, Axel H. Ng | 2022-12-19T22:36:53Z | cs.CL | A challenge in spoken language translation is that plenty of spoken content
is long-form, but short units are necessary for obtaining high-quality
translations. To address this mismatch, we fine-tune a general-purpose, large
language model to split long ASR transcripts into segments that can be
independently translated so as to maximize the overall translation quality. We
compare to several segmentation strategies and find that our approach improves
BLEU score on three languages by an average of 2.7 BLEU overall compared to an
automatic punctuation baseline. Further, we demonstrate the effectiveness of
two constrained decoding strategies to improve well-formedness of the model
output from above 99% to 100%. | cs.CL | Improved Long-Form Spoken Language Translation with Large Language Models |
2022-12-19T22:08:36Z | http://arxiv.org/pdf/2212.09885v2 | http://arxiv.org/abs/2212.09885v2 | Haau-Sing Li, Mohsen Mesgar, André F. T. Martins, Iryna Gurevych | 2023-05-26T16:03:08Z | cs.CL | Code generation from text requires understanding the user's intent from a
natural language description and generating an executable code snippet that
satisfies this intent. While recent pretrained language models demonstrate
remarkable performance for this task, these models fail when the given natural
language description is under-specified. In this work, we introduce a novel and
more realistic setup for this task. We hypothesize that the under-specification
of a natural language description can be resolved by asking clarification
questions. Therefore, we collect and introduce a new dataset named CodeClarQA
containing pairs of natural language descriptions and code with created
synthetic clarification questions and answers. The empirical results of our
evaluation of pretrained language model performance on code generation show
that clarifications result in more precisely generated code, as shown by the
substantial improvement of model performance in all evaluation metrics.
Alongside this, our task and dataset introduce new challenges to the community,
including when and what clarification questions should be asked. Our code and
dataset are available on GitHub. | cs.CL | Python Code Generation by Asking Clarification Questions |
2022-12-19T21:59:22Z | http://arxiv.org/pdf/2212.09879v1 | http://arxiv.org/abs/2212.09879v1 | Lenka Jungmannová, Petr Plecháč | 2022-12-19T21:59:22Z | cs.CL | In addition to being a widely recognised novelist, Milan Kundera has also
authored three pieces for theatre: The Owners of the Keys (Majitel\'e
kl\'i\v{c}\r{u}, 1961), The Blunder (Pt\'akovina, 1967), and Jacques and his
Master (Jakub a jeho p\'an, 1971). In recent years, however, the hypothesis has
been raised that Kundera is the true author of a fourth play: Juro
J\'ano\v{s}\'ik, first performed in a 1974 production under the name of Karel
Steigerwald, who was Kundera's student at the time. In this study, we make use
of supervised machine learning to settle the question of authorship attribution
in the case of Juro J\'ano\v{s}\'ik, with results strongly supporting the
hypothesis of Kundera's authorship. | cs.CL | Unsigned Play by Milan Kundera? An Authorship Attribution Study |
2022-12-19T21:50:36Z | http://arxiv.org/pdf/2212.09873v2 | http://arxiv.org/abs/2212.09873v2 | Karin de Langis, Dongyeop Kang | 2023-10-22T20:28:18Z | cs.CL | There is growing interest in incorporating eye-tracking data and other
implicit measures of human language processing into natural language processing
(NLP) pipelines. The data from human language processing contain unique insight
into human linguistic understanding that could be exploited by language models.
However, many unanswered questions remain about the nature of this data and how
it can best be utilized in downstream NLP tasks. In this paper, we present
eyeStyliency, an eye-tracking dataset for human processing of stylistic text
(e.g., politeness). We develop a variety of methods to derive style saliency
scores over text using the collected eye dataset. We further investigate how
this saliency data compares to both human annotation methods and model-based
interpretability metrics. We find that while eye-tracking data is unique, it
also intersects with both human annotations and model-based importance scores,
providing a possible bridge between human- and machine-based perspectives. We
propose utilizing this type of data to evaluate the cognitive plausibility of
models that interpret style. Our eye-tracking data and processing code are
publicly available. | cs.CL | A Comparative Study on Textual Saliency of Styles from Eye Tracking, Annotations, and Language Models |
2022-12-19T21:37:36Z | http://arxiv.org/pdf/2212.09867v1 | http://arxiv.org/abs/2212.09867v1 | Daniel N. Sosa, Malavika Suresh, Christopher Potts, Russ B. Altman | 2022-12-19T21:37:36Z | cs.CL | The COVID-19 pandemic created a deluge of questionable and contradictory
scientific claims about drug efficacy -- an "infodemic" with lasting
consequences for science and society. In this work, we argue that NLP models
can help domain experts distill and understand the literature in this complex,
high-stakes area. Our task is to automatically identify contradictory claims
about COVID-19 drug efficacy. We frame this as a natural language inference
problem and offer a new NLI dataset created by domain experts. The NLI framing
allows us to create curricula combining existing datasets and our own. The
resulting models are useful investigative tools. We provide a case study of how
these models help a domain expert summarize and assess evidence concerning
remdisivir and hydroxychloroquine. | cs.CL | Detecting Contradictory COVID-19 Drug Efficacy Claims from Biomedical Literature |
2022-12-19T21:34:26Z | http://arxiv.org/pdf/2212.09865v2 | http://arxiv.org/abs/2212.09865v2 | Xinxi Lyu, Sewon Min, Iz Beltagy, Luke Zettlemoyer, Hannaneh Hajishirzi | 2023-06-03T22:51:39Z | cs.CL, cs.AI | Although large language models can be prompted for both zero- and few-shot
learning, performance drops significantly when no demonstrations are available.
In this paper, we introduce Z-ICL, a new zero-shot method that closes the gap
by constructing pseudo-demonstrations for a given test input using a raw text
corpus. Concretely, pseudo-demonstrations are constructed by (1) finding the
nearest neighbors to the test input from the corpus and pairing them with
random task labels, and (2) applying a set of techniques to reduce the amount
of direct copying the model does from the resulting demonstrations. Evaluation
on nine classification datasets shows that Z-ICL outperforms previous zero-shot
methods by a significant margin, and is on par with in-context learning with
labeled training data in the few-shot setting. Overall, Z-ICL provides a
significantly higher estimate of the zero-shot performance levels of a model,
and supports future efforts to develop better pseudo-demonstrations that
further improve zero-shot results. | cs.CL | Z-ICL: Zero-Shot In-Context Learning with Pseudo-Demonstrations |
2022-12-19T21:34:00Z | http://arxiv.org/pdf/2212.09864v2 | http://arxiv.org/abs/2212.09864v2 | Zexue He, Graeme Blackwood, Rameswar Panda, Julian McAuley, Rogerio Feris | 2023-05-31T01:34:54Z | cs.CL, cs.AI | Pre-training models with large crawled corpora can lead to issues such as
toxicity and bias, as well as copyright and privacy concerns. A promising way
of alleviating such concerns is to conduct pre-training with synthetic tasks
and data, since no real-world information is ingested by the model. Our goal in
this paper is to understand the factors that contribute to the effectiveness of
pre-training models when using synthetic resources, particularly in the context
of neural machine translation. We propose several novel approaches to
pre-training translation models that involve different levels of lexical and
structural knowledge, including: 1) generating obfuscated data from a large
parallel corpus 2) concatenating phrase pairs extracted from a small
word-aligned corpus, and 3) generating synthetic parallel data without real
human language corpora. Our experiments on multiple language pairs reveal that
pre-training benefits can be realized even with high levels of obfuscation or
purely synthetic parallel data. We hope the findings from our comprehensive
empirical analysis will shed light on understanding what matters for NMT
pre-training, as well as pave the way for the development of more efficient and
less toxic models. | cs.CL | Synthetic Pre-Training Tasks for Neural Machine Translation |
2022-12-19T21:07:27Z | http://arxiv.org/pdf/2212.09858v1 | http://arxiv.org/abs/2212.09858v1 | Michael R. Lindstrom, Xiaofu Ding, Feng Liu, Anand Somayajula, Deanna Needell | 2022-12-19T21:07:27Z | cs.CL, cs.LG | Nonnegative matrix factorization can be used to automatically detect topics
within a corpus in an unsupervised fashion. The technique amounts to an
approximation of a nonnegative matrix as the product of two nonnegative
matrices of lower rank. In this paper, we show this factorization can be
combined with regression on a continuous response variable. In practice, the
method performs better than regression done after topics are identified and
retrains interpretability. | cs.CL | Continuous Semi-Supervised Nonnegative Matrix Factorization |
2022-12-19T20:57:45Z | http://arxiv.org/pdf/2212.09855v1 | http://arxiv.org/abs/2212.09855v1 | Xiaofei Li, Daniel Wiechmann, Yu Qiao, Elma Kerz | 2022-12-19T20:57:45Z | cs.CL | In this paper we present our contribution to the TSAR-2022 Shared Task on
Lexical Simplification of the EMNLP 2022 Workshop on Text Simplification,
Accessibility, and Readability. Our approach builds on and extends the
unsupervised lexical simplification system with pretrained encoders (LSBert)
system in the following ways: For the subtask of simplification candidate
selection, it utilizes a RoBERTa transformer language model and expands the
size of the generated candidate list. For subsequent substitution ranking, it
introduces a new feature weighting scheme and adopts a candidate filtering
method based on textual entailment to maximize semantic similarity between the
target word and its simplification. Our best-performing system improves LSBert
by 5.9% accuracy and achieves second place out of 33 ranked solutions. | cs.CL | MANTIS at TSAR-2022 Shared Task: Improved Unsupervised Lexical Simplification with Pretrained Encoders |
2022-12-19T20:46:43Z | http://arxiv.org/pdf/2212.09849v5 | http://arxiv.org/abs/2212.09849v5 | Xisen Jin, Xiang Ren, Daniel Preotiuc-Pietro, Pengxiang Cheng | 2023-10-12T19:02:38Z | cs.CL, cs.LG | Fine-tuning pre-trained language models has become the prevalent paradigm for
building downstream NLP models. Oftentimes fine-tuned models are readily
available but their training data is not, due to data privacy or intellectual
property concerns. This creates a barrier to fusing knowledge across individual
models to yield a better single model. In this paper, we study the problem of
merging individual models built on different training data sets to obtain a
single model that performs well both across all data set domains and can
generalize on out-of-domain data. We propose a dataless knowledge fusion method
that merges models in their parameter space, guided by weights that minimize
prediction differences between the merged model and the individual models. Over
a battery of evaluation settings, we show that the proposed method
significantly outperforms baselines such as Fisher-weighted averaging or model
ensembling. Further, we find that our method is a promising alternative to
multi-task learning that can preserve or sometimes improve over the individual
models without access to the training data. Finally, model merging is more
efficient than training a multi-task model, thus making it applicable to a
wider set of scenarios. | cs.CL | Dataless Knowledge Fusion by Merging Weights of Language Models |
2022-12-19T20:46:21Z | http://arxiv.org/pdf/2212.09848v1 | http://arxiv.org/abs/2212.09848v1 | Yu Qiao, Xiaofei Li, Daniel Wiechmann, Elma Kerz | 2022-12-19T20:46:21Z | cs.CL | State-of-the-art text simplification (TS) systems adopt end-to-end neural
network models to directly generate the simplified version of the input text,
and usually function as a blackbox. Moreover, TS is usually treated as an
all-purpose generic task under the assumption of homogeneity, where the same
simplification is suitable for all. In recent years, however, there has been
increasing recognition of the need to adapt the simplification techniques to
the specific needs of different target groups. In this work, we aim to advance
current research on explainable and controllable TS in two ways: First,
building on recently proposed work to increase the transparency of TS systems,
we use a large set of (psycho-)linguistic features in combination with
pre-trained language models to improve explainable complexity prediction.
Second, based on the results of this preliminary task, we extend a
state-of-the-art Seq2Seq TS model, ACCESS, to enable explicit control of ten
attributes. The results of experiments show (1) that our approach improves the
performance of state-of-the-art models for predicting explainable complexity
and (2) that explicitly conditioning the Seq2Seq model on ten attributes leads
to a significant improvement in performance in both within-domain and
out-of-domain settings. | cs.CL | (Psycho-)Linguistic Features Meet Transformer Models for Improved Explainable and Controllable Text Simplification |
2022-12-19T20:31:47Z | http://arxiv.org/pdf/2212.09839v1 | http://arxiv.org/abs/2212.09839v1 | Sourabh Zanwar, Daniel Wiechmann, Yu Qiao, Elma Kerz | 2022-12-19T20:31:47Z | cs.CL | In recent years, there has been a surge of interest in research on automatic
mental health detection (MHD) from social media data leveraging advances in
natural language processing and machine learning techniques. While significant
progress has been achieved in this interdisciplinary research area, the vast
majority of work has treated MHD as a binary classification task. The
multiclass classification setup is, however, essential if we are to uncover the
subtle differences among the statistical patterns of language use associated
with particular mental health conditions. Here, we report on experiments aimed
at predicting six conditions (anxiety, attention deficit hyperactivity
disorder, bipolar disorder, post-traumatic stress disorder, depression, and
psychological stress) from Reddit social media posts. We explore and compare
the performance of hybrid and ensemble models leveraging transformer-based
architectures (BERT and RoBERTa) and BiLSTM neural networks trained on
within-text distributions of a diverse set of linguistic features. This set
encompasses measures of syntactic complexity, lexical sophistication and
diversity, readability, and register-specific ngram frequencies, as well as
sentiment and emotion lexicons. In addition, we conduct feature ablation
experiments to investigate which types of features are most indicative of
particular mental health conditions. | cs.CL | Exploring Hybrid and Ensemble Models for Multiclass Prediction of Mental Health Status on Social Media |
2022-12-19T19:53:14Z | http://arxiv.org/pdf/2212.09825v2 | http://arxiv.org/abs/2212.09825v2 | Abhilasha Sancheti, Aparna Garimella, Balaji Vasan Srinivasan, Rachel Rudinger | 2023-10-24T20:43:40Z | cs.CL | Reviewing and comprehending key obligations, entitlements, and prohibitions
in legal contracts can be a tedious task due to their length and
domain-specificity. Furthermore, the key rights and duties requiring review
vary for each contracting party. In this work, we propose a new task of
party-specific extractive summarization for legal contracts to facilitate
faster reviewing and improved comprehension of rights and duties. To facilitate
this, we curate a dataset comprising of party-specific pairwise importance
comparisons annotated by legal experts, covering ~293K sentence pairs that
include obligations, entitlements, and prohibitions extracted from lease
agreements. Using this dataset, we train a pairwise importance ranker and
propose a pipeline-based extractive summarization system that generates a
party-specific contract summary. We establish the need for incorporating
domain-specific notion of importance during summarization by comparing our
system against various baselines using both automatic and human evaluation
methods | cs.CL | What to Read in a Contract? Party-Specific Summarization of Legal Obligations, Entitlements, and Prohibitions |
2022-12-19T19:29:40Z | http://arxiv.org/pdf/2212.09811v3 | http://arxiv.org/abs/2212.09811v3 | Yeskendir Koishekenov, Alexandre Berard, Vassilina Nikoulina | 2023-07-07T09:53:20Z | cs.CL, cs.AI, cs.LG | The recently released NLLB-200 is a set of multilingual Neural Machine
Translation models that cover 202 languages. The largest model is based on a
Mixture of Experts architecture and achieves SoTA results across many language
pairs. It contains 54.5B parameters and requires at least four 32GB GPUs just
for inference. In this work, we propose a pruning method that enables the
removal of up to 80% of experts without further finetuning and with a
negligible loss in translation quality, which makes it feasible to run the
model on a single 32GB GPU. Further analysis suggests that our pruning metrics
can identify language-specific experts. | cs.CL | Memory-efficient NLLB-200: Language-specific Expert Pruning of a Massively Multilingual Machine Translation Model |
2022-12-19T19:16:29Z | http://arxiv.org/pdf/2212.09803v3 | http://arxiv.org/abs/2212.09803v3 | Mengzhou Xia, Mikel Artetxe, Chunting Zhou, Xi Victoria Lin, Ramakanth Pasunuru, Danqi Chen, Luke Zettlemoyer, Ves Stoyanov | 2023-05-30T03:33:27Z | cs.CL, cs.AI, cs.LG | Scaling up language models has led to unprecedented performance gains, but
little is understood about how the training dynamics change as models get
larger. How do language models of different sizes learn during pre-training?
Why do larger language models demonstrate more desirable behaviors? In this
paper, we analyze the intermediate training checkpoints of differently sized
OPT models (Zhang et al.,2022)--from 125M to 175B parameters--on next-token
prediction, sequence-level generation, and downstream tasks. We find that 1) at
a given perplexity and independent of model sizes, a similar subset of training
tokens see the most significant reduction in loss, with the rest stagnating or
showing double-descent behavior; 2) early in training, all models learn to
reduce the perplexity of grammatical sequences that contain hallucinations,
with small models halting at this suboptimal distribution and larger ones
eventually learning to assign these sequences lower probabilities; 3)
perplexity is a strong predictor of in-context learning performance on 74
multiple-choice tasks from BIG-Bench, and this holds independent of the model
size. Together, these results show that perplexity is more predictive of model
behaviors than model size or training computation. | cs.CL | Training Trajectories of Language Models Across Scales |
2022-12-19T19:11:27Z | http://arxiv.org/pdf/2212.09750v1 | http://arxiv.org/abs/2212.09750v1 | Jiaao Chen, Mohan Dodda, Diyi Yang | 2022-12-19T19:11:27Z | cs.CL, cs.AI | Abstractive dialogue summarization has received increasing attention
recently. Despite the fact that most of the current dialogue summarization
systems are trained to maximize the likelihood of human-written summaries and
have achieved significant results, there is still a huge gap in generating
high-quality summaries as determined by humans, such as coherence and
faithfulness, partly due to the misalignment in maximizing a single
human-written summary. To this end, we propose to incorporate different levels
of human feedback into the training process. This will enable us to guide the
models to capture the behaviors humans care about for summaries. Specifically,
we ask humans to highlight the salient information to be included in summaries
to provide the local feedback , and to make overall comparisons among summaries
in terms of coherence, accuracy, coverage, concise and overall quality, as the
global feedback. We then combine both local and global feedback to fine-tune
the dialog summarization policy with Reinforcement Learning. Experiments
conducted on multiple datasets demonstrate the effectiveness and generalization
of our methods over the state-of-the-art supervised baselines, especially in
terms of human judgments. | cs.CL | Human-in-the-loop Abstractive Dialogue Summarization |
2022-12-19T18:59:56Z | http://arxiv.org/pdf/2212.09747v2 | http://arxiv.org/abs/2212.09747v2 | Shuheng Liu, Alan Ritter | 2023-07-12T02:41:46Z | cs.CL | The CoNLL-2003 English named entity recognition (NER) dataset has been widely
used to train and evaluate NER models for almost 20 years. However, it is
unclear how well models that are trained on this 20-year-old data and developed
over a period of decades using the same test set will perform when applied on
modern data. In this paper, we evaluate the generalization of over 20 different
models trained on CoNLL-2003, and show that NER models have very different
generalization. Surprisingly, we find no evidence of performance degradation in
pre-trained Transformers, such as RoBERTa and T5, even when fine-tuned using
decades-old data. We investigate why some models generalize well to new data
while others do not, and attempt to disentangle the effects of temporal drift
and overfitting due to test reuse. Our analysis suggests that most
deterioration is due to temporal mismatch between the pre-training corpora and
the downstream test sets. We found that four factors are important for good
generalization: model architecture, number of parameters, time period of the
pre-training corpus, in addition to the amount of fine-tuning data. We suggest
current evaluation methods have, in some sense, underestimated progress on NER
over the past 20 years, as NER models have not only improved on the original
CoNLL-2003 test set, but improved even more on modern data. Our datasets can be
found at https://github.com/ShuhengL/acl2023_conllpp. | cs.CL | Do CoNLL-2003 Named Entity Taggers Still Work Well in 2023? |
2022-12-19T18:59:45Z | http://arxiv.org/pdf/2212.09746v5 | http://arxiv.org/abs/2212.09746v5 | Mina Lee, Megha Srivastava, Amelia Hardy, John Thickstun, Esin Durmus, Ashwin Paranjape, Ines Gerard-Ursin, Xiang Lisa Li, Faisal Ladhak, Frieda Rong, Rose E. Wang, Minae Kwon, Joon Sung Park, Hancheng Cao, Tony Lee, Rishi Bommasani, Michael Bernstein, Percy Liang | 2024-01-05T22:09:26Z | cs.CL | Many real-world applications of language models (LMs), such as writing
assistance and code autocomplete, involve human-LM interaction. However, most
benchmarks are non-interactive in that a model produces output without human
involvement. To evaluate human-LM interaction, we develop a new framework,
Human-AI Language-based Interaction Evaluation (HALIE), that defines the
components of interactive systems and dimensions to consider when designing
evaluation metrics. Compared to standard, non-interactive evaluation, HALIE
captures (i) the interactive process, not only the final output; (ii) the
first-person subjective experience, not just a third-party assessment; and
(iii) notions of preference beyond quality (e.g., enjoyment and ownership). We
then design five tasks to cover different forms of interaction: social
dialogue, question answering, crossword puzzles, summarization, and metaphor
generation. With four state-of-the-art LMs (three variants of OpenAI's GPT-3
and AI21 Labs' Jurassic-1), we find that better non-interactive performance
does not always translate to better human-LM interaction. In particular, we
highlight three cases where the results from non-interactive and interactive
metrics diverge and underscore the importance of human-LM interaction for LM
evaluation. | cs.CL | Evaluating Human-Language Model Interaction |
2022-12-19T18:59:34Z | http://arxiv.org/pdf/2212.09744v3 | http://arxiv.org/abs/2212.09744v3 | Sanket Vaibhav Mehta, Jai Gupta, Yi Tay, Mostafa Dehghani, Vinh Q. Tran, Jinfeng Rao, Marc Najork, Emma Strubell, Donald Metzler | 2023-12-08T05:20:31Z | cs.CL, cs.AI, cs.IR, cs.LG | Differentiable Search Indices (DSIs) encode a corpus of documents in model
parameters and use the same model to answer user queries directly. Despite the
strong performance of DSI models, deploying them in situations where the corpus
changes over time is computationally expensive because reindexing the corpus
requires re-training the model. In this work, we introduce DSI++, a continual
learning challenge for DSI to incrementally index new documents while being
able to answer queries related to both previously and newly indexed documents.
Across different model scales and document identifier representations, we show
that continual indexing of new documents leads to considerable forgetting of
previously indexed documents. We also hypothesize and verify that the model
experiences forgetting events during training, leading to unstable learning. To
mitigate these issues, we investigate two approaches. The first focuses on
modifying the training dynamics. Flatter minima implicitly alleviate
forgetting, so we optimize for flatter loss basins and show that the model
stably memorizes more documents ($+12\%$). Next, we introduce a generative
memory to sample pseudo-queries for documents and supplement them during
continual indexing to prevent forgetting for the retrieval task. Extensive
experiments on novel continual indexing benchmarks based on Natural Questions
(NQ) and MS MARCO demonstrate that our proposed solution mitigates forgetting
significantly. Concretely, it improves the average Hits@10 by $+21.1\%$ over
competitive baselines for NQ and requires $6$ times fewer model updates
compared to re-training the DSI model for incrementally indexing five corpora
in a sequence. | cs.CL | DSI++: Updating Transformer Memory with New Documents |
2022-12-19T18:57:05Z | http://arxiv.org/pdf/2212.09741v3 | http://arxiv.org/abs/2212.09741v3 | Hongjin Su, Weijia Shi, Jungo Kasai, Yizhong Wang, Yushi Hu, Mari Ostendorf, Wen-tau Yih, Noah A. Smith, Luke Zettlemoyer, Tao Yu | 2023-05-30T15:22:50Z | cs.CL | We introduce INSTRUCTOR, a new method for computing text embeddings given
task instructions: every text input is embedded together with instructions
explaining the use case (e.g., task and domain descriptions). Unlike encoders
from prior work that are more specialized, INSTRUCTOR is a single embedder that
can generate text embeddings tailored to different downstream tasks and
domains, without any further training. We first annotate instructions for 330
diverse tasks and train INSTRUCTOR on this multitask mixture with a contrastive
loss. We evaluate INSTRUCTOR on 70 embedding evaluation tasks (66 of which are
unseen during training), ranging from classification and information retrieval
to semantic textual similarity and text generation evaluation. INSTRUCTOR,
while having an order of magnitude fewer parameters than the previous best
model, achieves state-of-the-art performance, with an average improvement of
3.4% compared to the previous best results on the 70 diverse datasets. Our
analysis suggests that INSTRUCTOR is robust to changes in instructions, and
that instruction finetuning mitigates the challenge of training a single model
on diverse datasets. Our model, code, and data are available at
https://instructor-embedding.github.io. | cs.CL | One Embedder, Any Task: Instruction-Finetuned Text Embeddings |
2022-12-19T18:56:52Z | http://arxiv.org/pdf/2212.09739v4 | http://arxiv.org/abs/2212.09739v4 | Mounica Maddela, Yao Dou, David Heineman, Wei Xu | 2023-07-07T20:41:23Z | cs.CL | Training learnable metrics using modern language models has recently emerged
as a promising method for the automatic evaluation of machine translation.
However, existing human evaluation datasets for text simplification have
limited annotations that are based on unitary or outdated models, making them
unsuitable for this approach. To address these issues, we introduce the
SimpEval corpus that contains: SimpEval_past, comprising 12K human ratings on
2.4K simplifications of 24 past systems, and SimpEval_2022, a challenging
simplification benchmark consisting of over 1K human ratings of 360
simplifications including GPT-3.5 generated text. Training on SimpEval, we
present LENS, a Learnable Evaluation Metric for Text Simplification. Extensive
empirical results show that LENS correlates much better with human judgment
than existing metrics, paving the way for future progress in the evaluation of
text simplification. We also introduce Rank and Rate, a human evaluation
framework that rates simplifications from several models in a list-wise manner
using an interactive interface, which ensures both consistency and accuracy in
the evaluation process and is used to create the SimpEval datasets. | cs.CL | LENS: A Learnable Evaluation Metric for Text Simplification |
2022-12-19T18:55:21Z | http://arxiv.org/pdf/2212.09736v2 | http://arxiv.org/abs/2212.09736v2 | Yu Gu, Xiang Deng, Yu Su | 2023-05-03T04:32:35Z | cs.CL, cs.AI, I.2.7 | A key missing capacity of current language models (LMs) is grounding to
real-world environments. Most existing work for grounded language understanding
uses LMs to directly generate plans that can be executed in the environment to
achieve the desired effects. It thereby casts the burden of ensuring
grammaticality, faithfulness, and controllability all on the LMs. We propose
Pangu, a generic framework for grounded language understanding that capitalizes
on the discriminative ability of LMs instead of their generative ability. Pangu
consists of a symbolic agent and a neural LM working in a concerted fashion:
The agent explores the environment to incrementally construct valid plans, and
the LM evaluates the plausibility of the candidate plans to guide the search
process. A case study on the challenging problem of knowledge base question
answering (KBQA), which features a massive environment, demonstrates the
remarkable effectiveness and flexibility of Pangu: A BERT-base LM is sufficient
for setting a new record on standard KBQA datasets, and larger LMs further
bring substantial gains. Pangu also enables, for the first time, effective
few-shot in-context learning for KBQA with large LMs such as Codex. | cs.CL | Don't Generate, Discriminate: A Proposal for Grounding Language Models to Real-World Environments |
2022-12-19T18:53:04Z | http://arxiv.org/pdf/2212.09730v2 | http://arxiv.org/abs/2212.09730v2 | Gallil Maimon, Yossi Adi | 2023-10-18T19:23:27Z | cs.SD, cs.CL, cs.LG, eess.AS | We introduce DISSC, a novel, lightweight method that converts the rhythm,
pitch contour and timbre of a recording to a target speaker in a textless
manner. Unlike DISSC, most voice conversion (VC) methods focus primarily on
timbre, and ignore people's unique speaking style (prosody). The proposed
approach uses a pretrained, self-supervised model for encoding speech to
discrete units, which makes it simple, effective, and fast to train. All
conversion modules are only trained on reconstruction like tasks, thus suitable
for any-to-many VC with no paired data. We introduce a suite of quantitative
and qualitative evaluation metrics for this setup, and empirically demonstrate
that DISSC significantly outperforms the evaluated baselines. Code and samples
are available at https://pages.cs.huji.ac.il/adiyoss-lab/dissc/. | cs.SD | Speaking Style Conversion in the Waveform Domain Using Discrete Self-Supervised Units |
2022-12-19T18:51:06Z | http://arxiv.org/pdf/2212.09726v2 | http://arxiv.org/abs/2212.09726v2 | Asish Ghoshal, Arash Einolghozati, Ankit Arun, Haoran Li, Lili Yu, Vera Gor, Yashar Mehdad, Scott Wen-tau Yih, Asli Celikyilmaz | 2024-01-18T19:27:04Z | cs.CL, cs.AI, cs.LG | Lack of factual correctness is an issue that still plagues state-of-the-art
summarization systems despite their impressive progress on generating seemingly
fluent summaries. In this paper, we show that factual inconsistency can be
caused by irrelevant parts of the input text, which act as confounders. To that
end, we leverage information-theoretic measures of causal effects to quantify
the amount of confounding and precisely quantify how they affect the
summarization performance. Based on insights derived from our theoretical
results, we design a simple multi-task model to control such confounding by
leveraging human-annotated relevant sentences when available. Crucially, we
give a principled characterization of data distributions where such confounding
can be large thereby necessitating the use of human annotated relevant
sentences to generate factual summaries. Our approach improves faithfulness
scores by 20\% over strong baselines on AnswerSumm
\citep{fabbri2021answersumm}, a conversation summarization dataset where lack
of faithfulness is a significant issue due to the subjective nature of the
task. Our best method achieves the highest faithfulness score while also
achieving state-of-the-art results on standard metrics like ROUGE and METEOR.
We corroborate these improvements through human evaluation. | cs.CL | Improving Faithfulness of Abstractive Summarization by Controlling Confounding Effect of Irrelevant Sentences |
2022-12-19T18:50:54Z | http://arxiv.org/pdf/2212.09724v3 | http://arxiv.org/abs/2212.09724v3 | Vardaan Pahuja, Boshi Wang, Hugo Latapie, Jayanth Srinivasa, Yu Su | 2023-10-22T17:57:24Z | cs.CL, cs.AI | Knowledge graph (KG) link prediction aims to infer new facts based on
existing facts in the KG. Recent studies have shown that using the graph
neighborhood of a node via graph neural networks (GNNs) provides more useful
information compared to just using the query information. Conventional GNNs for
KG link prediction follow the standard message-passing paradigm on the entire
KG, which leads to superfluous computation, over-smoothing of node
representations, and also limits their expressive power. On a large scale, it
becomes computationally expensive to aggregate useful information from the
entire KG for inference. To address the limitations of existing KG link
prediction frameworks, we propose a novel retrieve-and-read framework, which
first retrieves a relevant subgraph context for the query and then jointly
reasons over the context and the query with a high-capacity reader. As part of
our exemplar instantiation for the new framework, we propose a novel
Transformer-based GNN as the reader, which incorporates graph-based attention
structure and cross-attention between query and context for deep fusion. This
simple yet effective design enables the model to focus on salient context
information relevant to the query. Empirical results on two standard KG link
prediction datasets demonstrate the competitive performance of the proposed
method. Furthermore, our analysis yields valuable insights for designing
improved retrievers within the framework. | cs.CL | A Retrieve-and-Read Framework for Knowledge Graph Link Prediction |
2022-12-19T18:49:50Z | http://arxiv.org/pdf/2212.09723v1 | http://arxiv.org/abs/2212.09723v1 | Shashank Sonkar, Zichao Wang, Richard G. Baraniuk | 2022-12-19T18:49:50Z | cs.CL | This paper investigates the problem of Named Entity Recognition (NER) for
extreme low-resource languages with only a few hundred tagged data samples. NER
is a fundamental task in Natural Language Processing (NLP). A critical driver
accelerating NER systems' progress is the existence of large-scale language
corpora that enable NER systems to achieve outstanding performance in languages
such as English and French with abundant training data. However, NER for
low-resource languages remains relatively unexplored. In this paper, we
introduce Mask Augmented Named Entity Recognition (MANER), a new methodology
that leverages the distributional hypothesis of pre-trained masked language
models (MLMs) for NER. The <mask> token in pre-trained MLMs encodes valuable
semantic contextual information. MANER re-purposes the <mask> token for NER
prediction. Specifically, we prepend the <mask> token to every word in a
sentence for which we would like to predict the named entity tag. During
training, we jointly fine-tune the MLM and a new NER prediction head attached
to each <mask> token. We demonstrate that MANER is well-suited for NER in
low-resource languages; our experiments show that for 100 languages with as few
as 100 training examples, it improves on state-of-the-art methods by up to 48%
and by 12% on average on F1 score. We also perform detailed analyses and
ablation studies to understand the scenarios that are best-suited to MANER. | cs.CL | MANER: Mask Augmented Named Entity Recognition for Extreme Low-Resource Languages |
2022-12-19T18:49:09Z | http://arxiv.org/pdf/2212.09721v2 | http://arxiv.org/abs/2212.09721v2 | Aaron Chan, Zhiyuan Zeng, Wyatt Lake, Brihi Joshi, Hanjie Chen, Xiang Ren | 2023-05-22T00:19:21Z | cs.CL, cs.AI, cs.LG | Language models (LMs) have yielded impressive results on many language
reasoning tasks, but their unexpected errors raise doubts about their reasoning
abilities. In light of this, there is growing interest in finetuning/prompting
LMs with both task instances and their associated free-text rationales (FTRs),
which explain the correct reasoning process for predicting the correct task
output (i.e., how to be "right for the right reasons"). However, existing
finetuning methods fail to improve LM performance, while prompting needs
prohibitively large (i.e., >50B) LMs to work well. We propose KNIFE, which
shows that reasoning knowledge can be effectively distilled from FTRs into a
small (i.e., <1B) LM and improve the LM's performance. First, KNIFE finetunes a
teacher LM (given task input and FTR) to predict the task output, transferring
reasoning knowledge from the FTRs to the teacher's hidden states. Second, KNIFE
finetunes a student LM (given task input only) such that its hidden states are
aligned with the teacher's. Thus, the student is endowed with reasoning
knowledge but can be used for inference without direct FTR input. On two
question-answering datasets, KNIFE outperforms various finetuning and prompting
baselines in fully-supervised and low-resource settings. Also, we observe that
FTR quality is crucial to KNIFE's performance. | cs.CL | KNIFE: Distilling Reasoning Knowledge From Free-Text Rationales |
2022-12-19T18:39:43Z | http://arxiv.org/pdf/2212.09710v2 | http://arxiv.org/abs/2212.09710v2 | Alane Suhr, Yoav Artzi | 2023-12-05T21:22:11Z | cs.CL, cs.AI, cs.LG | We propose and deploy an approach to continually train an
instruction-following agent from feedback provided by users during
collaborative interactions. During interaction, human users instruct an agent
using natural language, and provide realtime binary feedback as they observe
the agent following their instructions. We design a contextual bandit learning
approach, converting user feedback to immediate reward. We evaluate through
thousands of human-agent interactions, demonstrating 15.4% absolute improvement
in instruction execution accuracy over time. We also show our approach is
robust to several design variations, and that the feedback signal is roughly
equivalent to the learning signal of supervised demonstration data. | cs.CL | Continual Learning for Instruction Following from Realtime Feedback |
2022-12-19T18:30:36Z | http://arxiv.org/pdf/2212.09702v3 | http://arxiv.org/abs/2212.09702v3 | William Gantt, Reno Kriz, Yunmo Chen, Siddharth Vashishtha, Aaron Steven White | 2023-10-20T21:26:29Z | cs.CL, cs.AI, cs.LG | As information extraction (IE) systems have grown more adept at processing
whole documents, the classic task of template filling has seen renewed interest
as benchmark for document-level IE. In this position paper, we call into
question the suitability of template filling for this purpose. We argue that
the task demands definitive answers to thorny questions of event individuation
-- the problem of distinguishing distinct events -- about which even human
experts disagree. Through an annotation study and error analysis, we show that
this raises concerns about the usefulness of template filling metrics, the
quality of datasets for the task, and the ability of models to learn it.
Finally, we consider possible solutions. | cs.CL | On Event Individuation for Document-Level Information Extraction |
2022-12-19T18:30:26Z | http://arxiv.org/pdf/2212.09701v1 | http://arxiv.org/abs/2212.09701v1 | Mina Samizadeh | 2022-12-19T18:30:26Z | cs.CL, cs.LG | In the past few decades, there has been an explosion in the amount of
available data produced from various sources with different topics. The
availability of this enormous data necessitates us to adopt effective
computational tools to explore the data. This leads to an intense growing
interest in the research community to develop computational methods focused on
processing this text data. A line of study focused on condensing the text so
that we are able to get a higher level of understanding in a shorter time. The
two important tasks to do this are keyword extraction and text summarization.
In keyword extraction, we are interested in finding the key important words
from a text. This makes us familiar with the general topic of a text. In text
summarization, we are interested in producing a short-length text which
includes important information about the document. The TextRank algorithm, an
unsupervised learning method that is an extension of the PageRank (algorithm
which is the base algorithm of Google search engine for searching pages and
ranking them) has shown its efficacy in large-scale text mining, especially for
text summarization and keyword extraction. this algorithm can automatically
extract the important parts of a text (keywords or sentences) and declare them
as the result. However, this algorithm neglects the semantic similarity between
the different parts. In this work, we improved the results of the TextRank
algorithm by incorporating the semantic similarity between parts of the text.
Aside from keyword extraction and text summarization, we develop a topic
clustering algorithm based on our framework which can be used individually or
as a part of generating the summary to overcome coverage problems. | cs.CL | Graph-based Semantical Extractive Text Analysis |
2022-12-19T18:30:09Z | http://arxiv.org/pdf/2212.09700v2 | http://arxiv.org/abs/2212.09700v2 | John Licato, Logan Fields, Zaid Marji | 2023-02-03T16:41:06Z | cs.CL | Open-textured terms in written rules are typically settled through
interpretive argumentation. Ongoing work has attempted to catalogue the schemes
used in such interpretive argumentation. But how can the use of these schemes
affect the way in which people actually use and reason over the proper
interpretations of open-textured terms? Using the interpretive
argument-eliciting game Aporia as our framework, we carried out an empirical
study to answer this question. Differing from previous work, we did not allow
participants to argue for interpretations arbitrarily, but to only use
arguments that fit with a given set of interpretive argument templates.
Finally, we analyze the results captured by this new dataset, specifically
focusing on practical implications for the development of
interpretation-capable artificial reasoners. | cs.CL | Resolving Open-textured Rules with Templated Interpretive Arguments |
2022-12-19T18:29:31Z | http://arxiv.org/pdf/2212.09699v3 | http://arxiv.org/abs/2212.09699v3 | Ioannis Tsiamas, José A. R. Fonollosa, Marta R. Costa-jussà | 2023-11-01T14:18:40Z | cs.CL, cs.SD, eess.AS | End-to-end Speech Translation is hindered by a lack of available data
resources. While most of them are based on documents, a sentence-level version
is available, which is however single and static, potentially impeding the
usefulness of the data. We propose a new data augmentation strategy,
SegAugment, to address this issue by generating multiple alternative
sentence-level versions of a dataset. Our method utilizes an Audio Segmentation
system, which re-segments the speech of each document with different length
constraints, after which we obtain the target text via alignment methods.
Experiments demonstrate consistent gains across eight language pairs in MuST-C,
with an average increase of 2.5 BLEU points, and up to 5 BLEU for low-resource
scenarios in mTEDx. Furthermore, when combined with a strong system, SegAugment
establishes new state-of-the-art results in MuST-C. Finally, we show that the
proposed method can also successfully augment sentence-level datasets, and that
it enables Speech Translation models to close the gap between the manual and
automatic segmentation at inference time. | cs.CL | SegAugment: Maximizing the Utility of Speech Translation Data with Segmentation-based Augmentations |
2022-12-19T18:21:00Z | http://arxiv.org/pdf/2212.09689v1 | http://arxiv.org/abs/2212.09689v1 | Or Honovich, Thomas Scialom, Omer Levy, Timo Schick | 2022-12-19T18:21:00Z | cs.CL, cs.AI, cs.LG | Instruction tuning enables pretrained language models to perform new tasks
from inference-time natural language descriptions. These approaches rely on
vast amounts of human supervision in the form of crowdsourced datasets or user
interactions. In this work, we introduce Unnatural Instructions: a large
dataset of creative and diverse instructions, collected with virtually no human
labor. We collect 64,000 examples by prompting a language model with three seed
examples of instructions and eliciting a fourth. This set is then expanded by
prompting the model to rephrase each instruction, creating a total of
approximately 240,000 examples of instructions, inputs, and outputs.
Experiments show that despite containing a fair amount of noise, training on
Unnatural Instructions rivals the effectiveness of training on open-source
manually-curated datasets, surpassing the performance of models such as T0++
and Tk-Instruct across various benchmarks. These results demonstrate the
potential of model-generated data as a cost-effective alternative to
crowdsourcing for dataset expansion and diversification. | cs.CL | Unnatural Instructions: Tuning Language Models with (Almost) No Human Labor |
2022-12-19T18:14:36Z | http://arxiv.org/pdf/2212.09686v2 | http://arxiv.org/abs/2212.09686v2 | Clara Meister, Wojciech Stokowiec, Tiago Pimentel, Lei Yu, Laura Rimell, Adhiguna Kuncoro | 2023-06-23T05:59:16Z | cs.CL | After just a few hundred training updates, a standard probabilistic model for
language generation has likely not yet learnt many semantic or syntactic rules
of natural language, making it difficult to estimate the probability
distribution over next tokens. Yet around this point, these models have
identified a simple, loss-minimising behaviour: to output the unigram
distribution of the target training corpus. The use of such a heuristic raises
the question: Can we initialise our models with this behaviour and save
precious compute resources and model capacity? Here we show that we can
effectively endow standard neural language generation models with a separate
module that reflects unigram frequency statistics as prior knowledge, simply by
initialising the bias term in a model's final linear layer with the log-unigram
distribution. We use neural machine translation as a test bed for this simple
technique and observe that it: (i) improves learning efficiency; (ii) achieves
better overall performance; and perhaps most importantly (iii) appears to
disentangle strong frequency effects by encouraging the model to specialise in
non-frequency-related aspects of language. | cs.CL | A Natural Bias for Language Generation Models |
2022-12-19T18:11:10Z | http://arxiv.org/pdf/2212.09683v4 | http://arxiv.org/abs/2212.09683v4 | Ethan Mendes, Yang Chen, Wei Xu, Alan Ritter | 2023-07-03T07:04:45Z | cs.CL | We present a human-in-the-loop evaluation framework for fact-checking novel
misinformation claims and identifying social media messages that support them.
Our approach extracts check-worthy claims, which are aggregated and ranked for
review. Stance classifiers are then used to identify tweets supporting novel
misinformation claims, which are further reviewed to determine whether they
violate relevant policies. To demonstrate the feasibility of our approach, we
develop a baseline system based on modern NLP methods for human-in-the-loop
fact-checking in the domain of COVID-19 treatments. We make our data and
detailed annotation guidelines available to support the evaluation of
human-in-the-loop systems that identify novel misinformation directly from raw
user-generated content. | cs.CL | Human-in-the-loop Evaluation for Early Misinformation Detection: A Case Study of COVID-19 Treatments |
2022-12-19T18:10:23Z | http://arxiv.org/pdf/2212.09682v1 | http://arxiv.org/abs/2212.09682v1 | Matan Eyal, Hila Noga, Roee Aharoni, Idan Szpektor, Reut Tsarfaty | 2022-12-19T18:10:23Z | cs.CL | Recent work attributes progress in NLP to large language models (LMs) with
increased model size and large quantities of pretraining data. Despite this,
current state-of-the-art LMs for Hebrew are both under-parameterized and
under-trained compared to LMs in other languages. Additionally, previous work
on pretrained Hebrew LMs focused on encoder-only models. While the encoder-only
architecture is beneficial for classification tasks, it does not cater well for
sub-word prediction tasks, such as Named Entity Recognition, when considering
the morphologically rich nature of Hebrew. In this paper we argue that
sequence-to-sequence generative architectures are more suitable for LLMs in the
case of morphologically rich languages (MRLs) such as Hebrew. We demonstrate
that by casting tasks in the Hebrew NLP pipeline as text-to-text tasks, we can
leverage powerful multilingual, pretrained sequence-to-sequence models as mT5,
eliminating the need for a specialized, morpheme-based, separately fine-tuned
decoder. Using this approach, our experiments show substantial improvements
over previously published results on existing Hebrew NLP benchmarks. These
results suggest that multilingual sequence-to-sequence models present a
promising building block for NLP for MRLs. | cs.CL | Multilingual Sequence-to-Sequence Models for Hebrew NLP |
2022-12-19T18:01:06Z | http://arxiv.org/pdf/2212.09676v2 | http://arxiv.org/abs/2212.09676v2 | Li Lucy, Jesse Dodge, David Bamman, Katherine A. Keith | 2023-05-23T02:24:54Z | cs.CL, cs.DL | Scholarly text is often laden with jargon, or specialized language that can
facilitate efficient in-group communication within fields but hinder
understanding for out-groups. In this work, we develop and validate an
interpretable approach for measuring scholarly jargon from text. Expanding the
scope of prior work which focuses on word types, we use word sense induction to
also identify words that are widespread but overloaded with different meanings
across fields. We then estimate the prevalence of these discipline-specific
words and senses across hundreds of subfields, and show that word senses
provide a complementary, yet unique view of jargon alongside word types. We
demonstrate the utility of our metrics for science of science and computational
sociolinguistics by highlighting two key social implications. First, though
most fields reduce their use of jargon when writing for general-purpose venues,
and some fields (e.g., biological sciences) do so less than others. Second, the
direction of correlation between jargon and citation rates varies among fields,
but jargon is nearly always negatively correlated with interdisciplinary
impact. Broadly, our findings suggest that though multidisciplinary venues
intend to cater to more general audiences, some fields' writing norms may act
as barriers rather than bridges, and thus impede the dispersion of scholarly
ideas. | cs.CL | Words as Gatekeepers: Measuring Discipline-specific Terms and Meanings in Scholarly Publications |
2022-12-19T18:00:09Z | http://arxiv.org/pdf/2212.09674v2 | http://arxiv.org/abs/2212.09674v2 | Chester Palen-Michel, Constantine Lignos | 2023-10-26T19:50:29Z | cs.CL | This preprint describes work in progress on LR-Sum, a new
permissively-licensed dataset created with the goal of enabling further
research in automatic summarization for less-resourced languages. LR-Sum
contains human-written summaries for 40 languages, many of which are
less-resourced. We describe our process for extracting and filtering the
dataset from the Multilingual Open Text corpus (Palen-Michel et al., 2022). The
source data is public domain newswire collected from from Voice of America
websites, and LR-Sum is released under a Creative Commons license (CC BY 4.0),
making it one of the most openly-licensed multilingual summarization datasets.
We describe how we plan to use the data for modeling experiments and discuss
limitations of the dataset. | cs.CL | LR-Sum: Summarization for Less-Resourced Languages |
2022-12-19T17:59:18Z | http://arxiv.org/pdf/2212.09670v1 | http://arxiv.org/abs/2212.09670v1 | Kangchen Zhu, Zhiliang Tian, Ruifeng Luo, Xiaoguang Mao | 2022-12-19T17:59:18Z | cs.CL | Text style transfer aims to alter the style of a sentence while preserving
its content. Due to the lack of parallel corpora, most recent work focuses on
unsupervised methods and often uses cycle construction to train models. Since
cycle construction helps to improve the style transfer ability of the model by
rebuilding transferred sentences back to original-style sentences, it brings
about a content loss in unsupervised text style transfer tasks. In this paper,
we propose a novel disentanglement-based style transfer model StyleFlow to
enhance content preservation. Instead of the typical encoder-decoder scheme,
StyleFlow can not only conduct the forward process to obtain the output, but
also infer to the input through the output. We design an attention-aware
coupling layers to disentangle the content representations and the style
representations of a sentence. Besides, we propose a data augmentation method
based on Normalizing Flow to improve the robustness of the model. Experiment
results demonstrate that our model preserves content effectively and achieves
the state-of-the-art performance on the most metrics. | cs.CL | StyleFlow: Disentangle Latent Representations via Normalizing Flow for Unsupervised Text Style Transfer |
2022-12-19T17:51:47Z | http://arxiv.org/pdf/2212.09667v2 | http://arxiv.org/abs/2212.09667v2 | Alex Mei, Sharon Levy, William Yang Wang | 2023-05-19T05:19:18Z | cs.CL, cs.AI, cs.LG | Users' physical safety is an increasing concern as the market for intelligent
systems continues to grow, where unconstrained systems may recommend users
dangerous actions that can lead to serious injury. Covertly unsafe text is an
area of particular interest, as such text may arise from everyday scenarios and
are challenging to detect as harmful. We propose FARM, a novel framework
leveraging external knowledge for trustworthy rationale generation in the
context of safety. In particular, FARM foveates on missing knowledge to qualify
the information required to reason in specific scenarios and retrieves this
information with attribution to trustworthy sources. This knowledge is used to
both classify the safety of the original text and generate human-interpretable
rationales, shedding light on the risk of systems to specific user groups and
helping both stakeholders manage the risks of their systems and policymakers to
provide concrete safeguards for consumer safety. Our experiments show that FARM
obtains state-of-the-art results on the SafeText dataset, showing absolute
improvement in safety classification accuracy by 5.9%. | cs.CL | Foveate, Attribute, and Rationalize: Towards Physically Safe and Trustworthy AI |
2022-12-19T17:50:05Z | http://arxiv.org/pdf/2212.09666v1 | http://arxiv.org/abs/2212.09666v1 | Zi Gong, Yinpeng Guo, Pingyi Zhou, Cuiyun Gao, Yasheng Wang, Zenglin Xu | 2022-12-19T17:50:05Z | cs.CL | Code completion is a valuable topic in both academia and industry. Recently,
large-scale mono-programming-lingual (MonoPL) pre-training models have been
proposed to boost the performance of code completion. However, the code
completion on low-resource programming languages (PL) is difficult for the
data-driven paradigm, while there are plenty of developers using low-resource
PLs. On the other hand, there are few studies exploring the effects of
multi-programming-lingual (MultiPL) pre-training for the code completion,
especially the impact on low-resource programming languages. To this end, we
propose the MultiCoder to enhance the low-resource code completion via MultiPL
pre-training and MultiPL Mixture-of-Experts (MoE) layers. We further propose a
novel PL-level MoE routing strategy (PL-MoE) for improving the code completion
on all PLs. Experimental results on CodeXGLUE and MultiCC demonstrate that 1)
the proposed MultiCoder significantly outperforms the MonoPL baselines on
low-resource programming languages, and 2) the PL-MoE module further boosts the
performance on six programming languages. In addition, we analyze the effects
of the proposed method in details and explore the effectiveness of our method
in a variety of scenarios. | cs.CL | MultiCoder: Multi-Programming-Lingual Pre-Training for Low-Resource Code Completion |
2022-12-19T17:45:07Z | http://arxiv.org/pdf/2212.09663v3 | http://arxiv.org/abs/2212.09663v3 | Momose Oyama, Sho Yokoi, Hidetoshi Shimodaira | 2023-11-02T16:01:32Z | cs.CL | Distributed representations of words encode lexical semantic information, but
what type of information is encoded and how? Focusing on the skip-gram with
negative-sampling method, we found that the squared norm of static word
embedding encodes the information gain conveyed by the word; the information
gain is defined by the Kullback-Leibler divergence of the co-occurrence
distribution of the word to the unigram distribution. Our findings are
explained by the theoretical framework of the exponential family of probability
distributions and confirmed through precise experiments that remove spurious
correlations arising from word frequency. This theory also extends to
contextualized word embeddings in language models or any neural networks with
the softmax output layer. We also demonstrate that both the KL divergence and
the squared norm of embedding provide a useful metric of the informativeness of
a word in tasks such as keyword extraction, proper-noun discrimination, and
hypernym discrimination. | cs.CL | Norm of Word Embedding Encodes Information Gain |
2022-12-19T17:44:54Z | http://arxiv.org/pdf/2212.09662v2 | http://arxiv.org/abs/2212.09662v2 | Fangyu Liu, Francesco Piccinno, Syrine Krichene, Chenxi Pang, Kenton Lee, Mandar Joshi, Yasemin Altun, Nigel Collier, Julian Martin Eisenschlos | 2023-05-23T18:21:27Z | cs.CL, cs.AI, cs.CV | Visual language data such as plots, charts, and infographics are ubiquitous
in the human world. However, state-of-the-art vision-language models do not
perform well on these data. We propose MatCha (Math reasoning and Chart
derendering pretraining) to enhance visual language models' capabilities in
jointly modeling charts/plots and language data. Specifically, we propose
several pretraining tasks that cover plot deconstruction and numerical
reasoning which are the key capabilities in visual language modeling.
We perform the MatCha pretraining starting from Pix2Struct, a recently
proposed image-to-text visual language model. On standard benchmarks such as
PlotQA and ChartQA, the MatCha model outperforms state-of-the-art methods by as
much as nearly 20%. We also examine how well MatCha pretraining transfers to
domains such as screenshots, textbook diagrams, and document figures and
observe overall improvement, verifying the usefulness of MatCha pretraining on
broader visual language tasks. | cs.CL | MatCha: Enhancing Visual Language Pretraining with Math Reasoning and Chart Derendering |
2022-12-19T17:42:07Z | http://arxiv.org/pdf/2212.09660v2 | http://arxiv.org/abs/2212.09660v2 | Genta Indra Winata, Alham Fikri Aji, Zheng-Xin Yong, Thamar Solorio | 2023-05-24T19:07:20Z | cs.CL | Code-Switching, a common phenomenon in written text and conversation, has
been studied over decades by the natural language processing (NLP) research
community. Initially, code-switching is intensively explored by leveraging
linguistic theories and, currently, more machine-learning oriented approaches
to develop models. We introduce a comprehensive systematic survey on
code-switching research in natural language processing to understand the
progress of the past decades and conceptualize the challenges and tasks on the
code-switching topic. Finally, we summarize the trends and findings and
conclude with a discussion for future direction and open questions for further
investigation. | cs.CL | The Decades Progress on Code-Switching Research in NLP: A Systematic Survey on Trends and Challenges |
2022-12-19T17:39:07Z | http://arxiv.org/pdf/2212.09656v1 | http://arxiv.org/abs/2212.09656v1 | Jayr Pereira, Robson Fidalgo, Roberto Lotufo, Rodrigo Nogueira | 2022-12-19T17:39:07Z | cs.CL, cs.IR | This paper proposes a question-answering system that can answer questions
whose supporting evidence is spread over multiple (potentially long) documents.
The system, called Visconde, uses a three-step pipeline to perform the task:
decompose, retrieve, and aggregate. The first step decomposes the question into
simpler questions using a few-shot large language model (LLM). Then, a
state-of-the-art search engine is used to retrieve candidate passages from a
large collection for each decomposed question. In the final step, we use the
LLM in a few-shot setting to aggregate the contents of the passages into the
final answer. The system is evaluated on three datasets: IIRC, Qasper, and
StrategyQA. Results suggest that current retrievers are the main bottleneck and
that readers are already performing at the human level as long as relevant
passages are provided. The system is also shown to be more effective when the
model is induced to give explanations before answering a question. Code is
available at \url{https://github.com/neuralmind-ai/visconde}. | cs.CL | Visconde: Multi-document QA with GPT-3 and Neural Reranking |
2022-12-19T17:29:37Z | http://arxiv.org/pdf/2212.09651v4 | http://arxiv.org/abs/2212.09651v4 | Ercong Nie, Sheng Liang, Helmut Schmid, Hinrich Schütze | 2023-07-10T22:27:15Z | cs.CL | Multilingual Pretrained Language Models (MPLMs) have shown their strong
multilinguality in recent empirical cross-lingual transfer studies. In this
paper, we propose the Prompts Augmented by Retrieval Crosslingually (PARC)
pipeline to improve the zero-shot performance on low-resource languages (LRLs)
by augmenting the context with semantically similar sentences retrieved from a
high-resource language (HRL) as prompts. PARC improves the zero-shot
performance on three downstream tasks (binary sentiment classification, topic
categorization and natural language inference) with multilingual parallel test
sets across 10 LRLs covering 6 language families in both unlabeled settings
(+5.1%) and labeled settings (+16.3%). PARC-labeled also outperforms the
finetuning baseline by 3.7%. We find a significant positive correlation between
cross-lingual transfer performance on one side, and the similarity between the
high- and low-resource languages as well as the amount of low-resource
pretraining data on the other side. A robustness analysis suggests that PARC
has the potential to achieve even stronger performance with more powerful
MPLMs. | cs.CL | Cross-Lingual Retrieval Augmented Prompt for Low-Resource Languages |
2022-12-19T17:28:22Z | http://arxiv.org/pdf/2212.09648v4 | http://arxiv.org/abs/2212.09648v4 | Samuel Cahyawijaya, Holy Lovenia, Alham Fikri Aji, Genta Indra Winata, Bryan Wilie, Rahmad Mahendra, Christian Wibisono, Ade Romadhony, Karissa Vincentio, Fajri Koto, Jennifer Santoso, David Moeljadi, Cahya Wirawan, Frederikus Hudi, Ivan Halim Parmonangan, Ika Alfina, Muhammad Satrio Wicaksono, Ilham Firdausi Putra, Samsul Rahmadani, Yulianti Oenang, Ali Akbar Septiandri, James Jaya, Kaustubh D. Dhole, Arie Ardiyanti Suryani, Rifki Afina Putri, Dan Su, Keith Stevens, Made Nindyatama Nityasya, Muhammad Farid Adilazuarda, Ryan Ignatius, Ryandito Diandaru, Tiezheng Yu, Vito Ghifari, Wenliang Dai, Yan Xu, Dyah Damapuspita, Cuk Tho, Ichwanul Muslim Karo Karo, Tirana Noor Fatyanosa, Ziwei Ji, Pascale Fung, Graham Neubig, Timothy Baldwin, Sebastian Ruder, Herry Sujaini, Sakriani Sakti, Ayu Purwarianti | 2023-07-21T14:44:45Z | cs.CL, cs.AI | We present NusaCrowd, a collaborative initiative to collect and unify
existing resources for Indonesian languages, including opening access to
previously non-public resources. Through this initiative, we have brought
together 137 datasets and 118 standardized data loaders. The quality of the
datasets has been assessed manually and automatically, and their value is
demonstrated through multiple experiments. NusaCrowd's data collection enables
the creation of the first zero-shot benchmarks for natural language
understanding and generation in Indonesian and the local languages of
Indonesia. Furthermore, NusaCrowd brings the creation of the first multilingual
automatic speech recognition benchmark in Indonesian and the local languages of
Indonesia. Our work strives to advance natural language processing (NLP)
research for languages that are under-represented despite being widely spoken. | cs.CL | NusaCrowd: Open Source Initiative for Indonesian NLP Resources |
2022-12-19T17:06:58Z | http://arxiv.org/pdf/2212.09631v2 | http://arxiv.org/abs/2212.09631v2 | Nuno M. Guerreiro, Pierre Colombo, Pablo Piantanida, André F. T. Martins | 2023-05-19T16:36:12Z | cs.CL, cs.LG | Neural machine translation (NMT) has become the de-facto standard in
real-world machine translation applications. However, NMT models can
unpredictably produce severely pathological translations, known as
hallucinations, that seriously undermine user trust. It becomes thus crucial to
implement effective preventive strategies to guarantee their proper
functioning. In this paper, we address the problem of hallucination detection
in NMT by following a simple intuition: as hallucinations are detached from the
source content, they exhibit encoder-decoder attention patterns that are
statistically different from those of good quality translations. We frame this
problem with an optimal transport formulation and propose a fully unsupervised,
plug-in detector that can be used with any attention-based NMT model.
Experimental results show that our detector not only outperforms all previous
model-based detectors, but is also competitive with detectors that employ large
models trained on millions of samples. | cs.CL | Optimal Transport for Unsupervised Hallucination Detection in Neural Machine Translation |
2022-12-19T17:00:54Z | http://arxiv.org/pdf/2212.09621v1 | http://arxiv.org/abs/2212.09621v1 | Haoli Bai, Zhiguang Liu, Xiaojun Meng, Wentao Li, Shuang Liu, Nian Xie, Rongfu Zheng, Liangwei Wang, Lu Hou, Jiansheng Wei, Xin Jiang, Qun Liu | 2022-12-19T17:00:54Z | cs.CL, cs.CV | Unsupervised pre-training on millions of digital-born or scanned documents
has shown promising advances in visual document understanding~(VDU). While
various vision-language pre-training objectives are studied in existing
solutions, the document textline, as an intrinsic granularity in VDU, has
seldom been explored so far. A document textline usually contains words that
are spatially and semantically correlated, which can be easily obtained from
OCR engines. In this paper, we propose Wukong-Reader, trained with new
pre-training objectives to leverage the structural knowledge nested in document
textlines. We introduce textline-region contrastive learning to achieve
fine-grained alignment between the visual regions and texts of document
textlines. Furthermore, masked region modeling and textline-grid matching are
also designed to enhance the visual and layout representations of textlines.
Experiments show that our Wukong-Reader has superior performance on various VDU
tasks such as information extraction. The fine-grained alignment over textlines
also empowers Wukong-Reader with promising localization ability. | cs.CL | Wukong-Reader: Multi-modal Pre-training for Fine-grained Visual Document Understanding |
2022-12-19T16:50:41Z | http://arxiv.org/pdf/2212.09611v2 | http://arxiv.org/abs/2212.09611v2 | Yaru Hao, Zewen Chi, Li Dong, Furu Wei | 2023-12-29T10:15:15Z | cs.CL, cs.CV | Well-designed prompts can guide text-to-image models to generate amazing
images. However, the performant prompts are often model-specific and misaligned
with user input. Instead of laborious human engineering, we propose prompt
adaptation, a general framework that automatically adapts original user input
to model-preferred prompts. Specifically, we first perform supervised
fine-tuning with a pretrained language model on a small collection of manually
engineered prompts. Then we use reinforcement learning to explore better
prompts. We define a reward function that encourages the policy to generate
more aesthetically pleasing images while preserving the original user
intentions. Experimental results on Stable Diffusion show that our method
outperforms manual prompt engineering in terms of both automatic metrics and
human preference ratings. Moreover, reinforcement learning further boosts
performance, especially on out-of-domain prompts. The pretrained checkpoints
are available at https://aka.ms/promptist. The demo can be found at
https://aka.ms/promptist-demo. | cs.CL | Optimizing Prompts for Text-to-Image Generation |
2022-12-19T16:41:19Z | http://arxiv.org/pdf/2212.09603v2 | http://arxiv.org/abs/2212.09603v2 | Qintong Li, Zhiyong Wu, Lingpeng Kong, Wei Bi | 2023-07-11T05:17:19Z | cs.CL | Explaining the black-box predictions of NLP models naturally and accurately
is an important open problem in natural language generation. These free-text
explanations are expected to contain sufficient and carefully-selected evidence
to form supportive arguments for predictions. Due to the superior generative
capacity of large pretrained language models, recent work built on prompt
engineering enables explanation generation without specific training. However,
explanation generated through single-pass prompting often lacks sufficiency and
conciseness. To address this problem, we develop an information bottleneck
method EIB to produce refined explanations that are sufficient and concise. Our
approach regenerates the free-text explanation by polishing the single-pass
output from the pretrained language model but retaining the information that
supports the contents being explained. Experiments on two out-of-domain tasks
verify the effectiveness of EIB through automatic evaluation and
thoroughly-conducted human evaluation. | cs.CL | Explanation Regeneration via Information Bottleneck |
2022-12-19T16:32:42Z | http://arxiv.org/pdf/2212.09597v8 | http://arxiv.org/abs/2212.09597v8 | Shuofei Qiao, Yixin Ou, Ningyu Zhang, Xiang Chen, Yunzhi Yao, Shumin Deng, Chuanqi Tan, Fei Huang, Huajun Chen | 2023-09-18T10:47:13Z | cs.CL, cs.AI, cs.CV, cs.IR, cs.LG | Reasoning, as an essential ability for complex problem-solving, can provide
back-end support for various real-world applications, such as medical
diagnosis, negotiation, etc. This paper provides a comprehensive survey of
cutting-edge research on reasoning with language model prompting. We introduce
research works with comparisons and summaries and provide systematic resources
to help beginners. We also discuss the potential reasons for emerging such
reasoning abilities and highlight future research directions. Resources are
available at https://github.com/zjunlp/Prompt4ReasoningPapers (updated
periodically). | cs.CL | Reasoning with Language Model Prompting: A Survey |
2022-12-19T16:29:26Z | http://arxiv.org/pdf/2212.09593v3 | http://arxiv.org/abs/2212.09593v3 | Mathieu Ravaut, Shafiq Joty, Nancy Chen | 2023-05-26T05:26:23Z | cs.CL | With the rise of task-specific pre-training objectives, abstractive
summarization models like PEGASUS offer appealing zero-shot performance on
downstream summarization tasks. However, the performance of such unsupervised
models still lags significantly behind their supervised counterparts. Similarly
to the supervised setup, we notice a very high variance in quality among
summary candidates from these models while only one candidate is kept as the
summary output. In this paper, we propose to re-rank summary candidates in an
unsupervised manner, aiming to close the performance gap between unsupervised
and supervised models. Our approach improves the unsupervised PEGASUS by up to
7.27% and ChatGPT by up to 6.86% relative mean ROUGE across four widely-adopted
summarization benchmarks ; and achieves relative gains of 7.51% (up to 23.73%
from XSum to WikiHow) averaged over 30 zero-shot transfer setups (finetuning on
a dataset, evaluating on another). | cs.CL | Unsupervised Summarization Re-ranking |
2022-12-19T16:21:05Z | http://arxiv.org/pdf/2212.09588v2 | http://arxiv.org/abs/2212.09588v2 | Mingzhu Cai, Siqi Bao, Xin Tian, Huang He, Fan Wang, Hua Wu | 2023-05-26T11:02:13Z | cs.CL | In this paper, we propose an unsupervised query enhanced approach for
knowledge-intensive conversations, namely QKConv. There are three modules in
QKConv: a query generator, an off-the-shelf knowledge selector, and a response
generator. QKConv is optimized through joint training, which produces the
response by exploring multiple candidate queries and leveraging corresponding
selected knowledge. The joint training solely relies on the dialogue context
and target response, getting exempt from extra query annotations or knowledge
provenances. To evaluate the effectiveness of the proposed QKConv, we conduct
experiments on three representative knowledge-intensive conversation datasets:
conversational question-answering, task-oriented dialogue, and
knowledge-grounded conversation. Experimental results reveal that QKConv
performs better than all unsupervised methods across three datasets and
achieves competitive performance compared to supervised methods. | cs.CL | Query Enhanced Knowledge-Intensive Conversation via Unsupervised Joint Modeling |
2022-12-19T16:13:52Z | http://arxiv.org/pdf/2212.09580v1 | http://arxiv.org/abs/2212.09580v1 | Tomáš Musil, David Mareček | 2022-12-19T16:13:52Z | cs.CL | Independent Component Analysis (ICA) is an algorithm originally developed for
finding separate sources in a mixed signal, such as a recording of multiple
people in the same room speaking at the same time. It has also been used to
find linguistic features in distributional representations. In this paper, we
used ICA to analyze words embeddings. We have found that ICA can be used to
find semantic features of the words and these features can easily be combined
to search for words that satisfy the combination. We show that only some of the
independent components represent such features, but those that do are stable
with regard to random initialization of the algorithm. | cs.CL | Independent Components of Word Embeddings Represent Semantic Features |
2022-12-19T16:10:56Z | http://arxiv.org/pdf/2212.09577v3 | http://arxiv.org/abs/2212.09577v3 | Martin Funkquist, Ilia Kuznetsov, Yufang Hou, Iryna Gurevych | 2023-11-03T19:55:56Z | cs.CL | Science progresses by building upon the prior body of knowledge documented in
scientific publications. The acceleration of research makes it hard to stay
up-to-date with the recent developments and to summarize the ever-growing body
of prior work. To address this, the task of citation text generation aims to
produce accurate textual summaries given a set of papers-to-cite and the citing
paper context. Due to otherwise rare explicit anchoring of cited documents in
the citing paper, citation text generation provides an excellent opportunity to
study how humans aggregate and synthesize textual knowledge from sources. Yet,
existing studies are based upon widely diverging task definitions, which makes
it hard to study this task systematically. To address this challenge, we
propose CiteBench: a benchmark for citation text generation that unifies
multiple diverse datasets and enables standardized evaluation of citation text
generation models across task designs and domains. Using the new benchmark, we
investigate the performance of multiple strong baselines, test their
transferability between the datasets, and deliver new insights into the task
definition and evaluation to guide future research in citation text generation.
We make the code for CiteBench publicly available at
https://github.com/UKPLab/citebench. | cs.CL | CiteBench: A benchmark for Scientific Citation Text Generation |
2022-12-19T16:06:45Z | http://arxiv.org/pdf/2212.09573v1 | http://arxiv.org/abs/2212.09573v1 | Vinayshekhar Bannihatti Kumar, Rashmi Gangadharaiah, Dan Roth | 2022-12-19T16:06:45Z | cs.CL | Regulations introduced by General Data Protection Regulation (GDPR) in the EU
or California Consumer Privacy Act (CCPA) in the US have included provisions on
the \textit{right to be forgotten} that mandates industry applications to
remove data related to an individual from their systems. In several real world
industry applications that use Machine Learning to build models on user data,
such mandates require significant effort both in terms of data cleansing as
well as model retraining while ensuring the models do not deteriorate in
prediction quality due to removal of data. As a result, continuous removal of
data and model retraining steps do not scale if these applications receive such
requests at a very high frequency. Recently, a few researchers proposed the
idea of \textit{Machine Unlearning} to tackle this challenge. Despite the
significant importance of this task, the area of Machine Unlearning is
under-explored in Natural Language Processing (NLP) tasks. In this paper, we
explore the Unlearning framework on various GLUE tasks \cite{Wang:18}, such as,
QQP, SST and MNLI. We propose computationally efficient approaches (SISA-FC and
SISA-A) to perform \textit{guaranteed} Unlearning that provides significant
reduction in terms of both memory (90-95\%), time (100x) and space consumption
(99\%) in comparison to the baselines while keeping model performance constant. | cs.CL | Privacy Adhering Machine Un-learning in NLP |
2022-12-19T15:53:12Z | http://arxiv.org/pdf/2212.09563v2 | http://arxiv.org/abs/2212.09563v2 | M. Yin, B. Wang, Y. Dong, C. Ling | 2024-03-17T19:06:38Z | cs.CL | Most previous unsupervised domain adaptation (UDA) methods for question
answering(QA) require access to source domain data while fine-tuning the model
for the target domain. Source domain data may, however, contain sensitive
information and may be restricted. In this study, we investigate a more
challenging setting, source-free UDA, in which we have only the pretrained
source model and target domain data, without access to source domain data. We
propose a novel self-training approach to QA models that integrates a unique
mask module for domain adaptation. The mask is auto-adjusted to extract key
domain knowledge while trained on the source domain. To maintain previously
learned domain knowledge, certain mask weights are frozen during adaptation,
while other weights are adjusted to mitigate domain shifts with pseudo-labeled
samples generated in the target domain. %As part of the self-training process,
we generate pseudo-labeled samples in the target domain based on models trained
in the source domain. Our empirical results on four benchmark datasets suggest
that our approach significantly enhances the performance of pretrained QA
models on the target domain, and even outperforms models that have access to
the source data during adaptation. | cs.CL | Source-Free Domain Adaptation for Question Answering with Masked Self-training |
2022-12-19T15:51:52Z | http://arxiv.org/pdf/2212.09561v5 | http://arxiv.org/abs/2212.09561v5 | Yixuan Weng, Minjun Zhu, Fei Xia, Bin Li, Shizhu He, Shengping Liu, Bin Sun, Kang Liu, Jun Zhao | 2023-10-19T12:24:42Z | cs.AI, cs.CL | Recently, with the chain of thought (CoT) prompting, large language models
(LLMs), e.g., GPT-3, have shown strong reasoning ability in several natural
language processing tasks such as arithmetic, commonsense, and logical
reasoning. However, LLMs with CoT require multi-step prompting and multi-token
prediction, which is highly sensitive to individual mistakes and vulnerable to
error accumulation. The above issues make the LLMs need the ability to verify
the answers. In fact, after inferring conclusions in some thinking decision
tasks, people often check them by re-verifying steps to avoid some mistakes. In
this paper, we propose and prove that LLMs also have similar self-verification
abilities. We take the conclusion obtained by CoT as one of the conditions for
solving the original problem. By performing a backward verification of the
answers that LLM deduced for itself, we can obtain interpretable answer
validation scores to select the candidate answer with the highest score.
Experimental results demonstrate that the proposed method can improve the
reasoning performance on various arithmetic, commonsense, and logical reasoning
datasets. Our code is publicly available at:
https://github.com/WENGSYX/Self-Verification. | cs.AI | Large Language Models are Better Reasoners with Self-Verification |
2022-12-19T15:45:36Z | http://arxiv.org/pdf/2212.09553v2 | http://arxiv.org/abs/2212.09553v2 | Yong Cheng, Yu Zhang, Melvin Johnson, Wolfgang Macherey, Ankur Bapna | 2023-06-27T01:18:45Z | cs.CL, cs.SD, eess.AS | We present Mu$^{2}$SLAM, a multilingual sequence-to-sequence model
pre-trained jointly on unlabeled speech, unlabeled text and supervised data
spanning Automatic Speech Recognition (ASR), Automatic Speech Translation (AST)
and Machine Translation (MT), in over 100 languages. By leveraging a quantized
representation of speech as a target, Mu$^{2}$SLAM trains the speech-text
models with a sequence-to-sequence masked denoising objective similar to T5 on
the decoder and a masked language modeling (MLM) objective on the encoder, for
both unlabeled speech and text, while utilizing the supervised tasks to improve
cross-lingual and cross-modal representation alignment within the model. On
CoVoST AST, Mu$^{2}$SLAM establishes a new state-of-the-art for models trained
on public datasets, improving on xx-en translation over the previous best by
1.9 BLEU points and on en-xx translation by 1.1 BLEU points. On Voxpopuli ASR,
our model matches the performance of an mSLAM model fine-tuned with an RNN-T
decoder, despite using a relatively weaker sequence-to-sequence architecture.
On text understanding tasks, our model improves by more than 6\% over mSLAM on
XNLI, getting closer to the performance of mT5 models of comparable capacity on
XNLI and TydiQA, paving the way towards a single model for all speech and text
understanding tasks. | cs.CL | Mu$^{2}$SLAM: Multitask, Multilingual Speech and Language Models |
2022-12-19T15:24:45Z | http://arxiv.org/pdf/2212.09535v3 | http://arxiv.org/abs/2212.09535v3 | Zheng-Xin Yong, Hailey Schoelkopf, Niklas Muennighoff, Alham Fikri Aji, David Ifeoluwa Adelani, Khalid Almubarak, M Saiful Bari, Lintang Sutawika, Jungo Kasai, Ahmed Baruwa, Genta Indra Winata, Stella Biderman, Edward Raff, Dragomir Radev, Vassilina Nikoulina | 2023-05-27T05:48:38Z | cs.CL, cs.AI, cs.LG | The BLOOM model is a large publicly available multilingual language model,
but its pretraining was limited to 46 languages. To extend the benefits of
BLOOM to other languages without incurring prohibitively large costs, it is
desirable to adapt BLOOM to new languages not seen during pretraining. In this
work, we apply existing language adaptation strategies to BLOOM and benchmark
its zero-shot prompting performance on eight new languages in a
resource-constrained setting. We find language adaptation to be effective at
improving zero-shot performance in new languages. Surprisingly, we find that
adapter-based finetuning is more effective than continued pretraining for large
models. In addition, we discover that prompting performance is not
significantly affected by language specifics, such as the writing system. It is
primarily determined by the size of the language adaptation data. We also add
new languages to BLOOMZ, which is a multitask finetuned version of BLOOM
capable of following task instructions zero-shot. We find including a new
language in the multitask fine-tuning mixture to be the most effective method
to teach BLOOMZ a new language. We conclude that with sufficient training data
language adaptation can generalize well to diverse languages. Our code is
available at https://github.com/bigscience-workshop/multilingual-modeling. | cs.CL | BLOOM+1: Adding Language Support to BLOOM for Zero-Shot Prompting |
2022-12-19T14:48:08Z | http://arxiv.org/pdf/2212.09512v3 | http://arxiv.org/abs/2212.09512v3 | Zhangyue Yin, Yuxin Wang, Xiannian Hu, Yiguang Wu, Hang Yan, Xinyu Zhang, Zhao Cao, Xuanjing Huang, Xipeng Qiu | 2023-12-13T16:27:16Z | cs.CL | Multi-Hop Question Answering (MHQA) is a significant area in question
answering, requiring multiple reasoning components, including document
retrieval, supporting sentence prediction, and answer span extraction. In this
work, we analyze the primary factors limiting the performance of multi-hop
reasoning and introduce label smoothing into the MHQA task. This is aimed at
enhancing the generalization capabilities of MHQA systems and mitigating
overfitting of answer spans and reasoning paths in training set. We propose a
novel label smoothing technique, F1 Smoothing, which incorporates uncertainty
into the learning process and is specifically tailored for Machine Reading
Comprehension (MRC) tasks. Inspired by the principles of curriculum learning,
we introduce the Linear Decay Label Smoothing Algorithm (LDLA), which
progressively reduces uncertainty throughout the training process. Experiment
on the HotpotQA dataset demonstrates the effectiveness of our methods in
enhancing performance and generalizability in multi-hop reasoning, achieving
new state-of-the-art results on the leaderboard. | cs.CL | Rethinking Label Smoothing on Multi-hop Question Answering |
2022-12-19T13:58:48Z | http://arxiv.org/pdf/2212.09465v1 | http://arxiv.org/abs/2212.09465v1 | Sourabh Zanwar, Daniel Wiechmann, Yu Qiao, Elma Kerz | 2022-12-19T13:58:48Z | cs.CL | In recent years, there has been increased interest in building predictive
models that harness natural language processing and machine learning techniques
to detect emotions from various text sources, including social media posts,
micro-blogs or news articles. Yet, deployment of such models in real-world
sentiment and emotion applications faces challenges, in particular poor
out-of-domain generalizability. This is likely due to domain-specific
differences (e.g., topics, communicative goals, and annotation schemes) that
make transfer between different models of emotion recognition difficult. In
this work we propose approaches for text-based emotion detection that leverage
transformer models (BERT and RoBERTa) in combination with Bidirectional Long
Short-Term Memory (BiLSTM) networks trained on a comprehensive set of
psycholinguistic features. First, we evaluate the performance of our models
within-domain on two benchmark datasets: GoEmotion and ISEAR. Second, we
conduct transfer learning experiments on six datasets from the Unified Emotion
Dataset to evaluate their out-of-domain robustness. We find that the proposed
hybrid models improve the ability to generalize to out-of-distribution data
compared to a standard transformer-based approach. Moreover, we observe that
these models perform competitively on in-domain data. | cs.CL | Improving the Generalizability of Text-Based Emotion Detection by Leveraging Transformers with Psycholinguistic Features |
2022-12-19T13:57:06Z | http://arxiv.org/pdf/2212.09462v2 | http://arxiv.org/abs/2212.09462v2 | Justin Lovelace, Varsha Kishore, Chao Wan, Eliot Shekhtman, Kilian Q. Weinberger | 2023-11-07T15:35:45Z | cs.CL, cs.LG | Diffusion models have achieved great success in modeling continuous data
modalities such as images, audio, and video, but have seen limited use in
discrete domains such as language. Recent attempts to adapt diffusion to
language have presented diffusion as an alternative to existing pretrained
language models. We view diffusion and existing language models as
complementary. We demonstrate that encoder-decoder language models can be
utilized to efficiently learn high-quality language autoencoders. We then
demonstrate that continuous diffusion models can be learned in the latent space
of the language autoencoder, enabling us to sample continuous latent
representations that can be decoded into natural language with the pretrained
decoder. We validate the effectiveness of our approach for unconditional,
class-conditional, and sequence-to-sequence language generation. We demonstrate
across multiple diverse data sets that our latent language diffusion models are
significantly more effective than previous diffusion language models. | cs.CL | Latent Diffusion for Language Generation |
2022-12-19T12:57:11Z | http://arxiv.org/pdf/2212.09422v1 | http://arxiv.org/abs/2212.09422v1 | Anton Thielmann, Christoph Weisser, Benjamin Säfken | 2022-12-19T12:57:11Z | cs.CL | Few-shot methods for accurate modeling under sparse label-settings have
improved significantly. However, the applications of few-shot modeling in
natural language processing remain solely in the field of document
classification. With recent performance improvements, supervised few-shot
methods, combined with a simple topic extraction method pose a significant
challenge to unsupervised topic modeling methods. Our research shows that
supervised few-shot learning, combined with a simple topic extraction method,
can outperform unsupervised topic modeling techniques in terms of generating
coherent topics, even when only a few labeled documents per class are used. | cs.CL | Human in the loop: How to effectively create coherent topics by manually labeling only a few documents per class |
2022-12-19T12:55:32Z | http://arxiv.org/pdf/2212.09420v2 | http://arxiv.org/abs/2212.09420v2 | Daoguang Zan, Bei Chen, Fengji Zhang, Dianjie Lu, Bingchao Wu, Bei Guan, Yongji Wang, Jian-Guang Lou | 2023-05-08T10:25:41Z | cs.SE, cs.AI, cs.CL, cs.PL | The task of generating code from a natural language description, or NL2Code,
is considered a pressing and significant challenge in code intelligence. Thanks
to the rapid development of pre-training techniques, surging large language
models are being proposed for code, sparking the advances in NL2Code. To
facilitate further research and applications in this field, in this paper, we
present a comprehensive survey of 27 existing large language models for
NL2Code, and also review benchmarks and metrics. We provide an intuitive
comparison of all existing models on the HumanEval benchmark. Through in-depth
observation and analysis, we provide some insights and conclude that the key
factors contributing to the success of large language models for NL2Code are
"Large Size, Premium Data, Expert Tuning". In addition, we discuss challenges
and opportunities regarding the gap between models and humans. We also create a
website https://nl2code.github.io to track the latest progress through
crowd-sourcing. To the best of our knowledge, this is the first survey of large
language models for NL2Code, and we believe it will contribute to the ongoing
development of the field. | cs.SE | Large Language Models Meet NL2Code: A Survey |
2022-12-19T12:44:25Z | http://arxiv.org/pdf/2212.09412v3 | http://arxiv.org/abs/2212.09412v3 | Zhujin Gao, Junliang Guo, Xu Tan, Yongxin Zhu, Fang Zhang, Jiang Bian, Linli Xu | 2024-04-22T09:50:44Z | cs.CL, cs.AI, cs.LG | Diffusion models have achieved state-of-the-art synthesis quality on both
visual and audio tasks, and recent works further adapt them to textual data by
diffusing on the embedding space. In this paper, we conduct systematic studies
of the optimization challenges encountered with both the embedding space and
the denoising model, which have not been carefully explored. Firstly, the data
distribution is learnable for embeddings, which may lead to the collapse of the
embedding space and unstable training. To alleviate this problem, we propose a
new objective called the anchor loss which is more efficient than previous
methods. Secondly, we find the noise levels of conventional schedules are
insufficient for training a desirable denoising model while introducing varying
degrees of degeneration in consequence. To address this challenge, we propose a
novel framework called noise rescaling. Based on the above analysis, we propose
Difformer, an embedding diffusion model based on Transformer. Experiments on
varieties of seminal text generation tasks show the effectiveness of the
proposed methods and the superiority of Difformer over previous
state-of-the-art embedding diffusion baselines. | cs.CL | Empowering Diffusion Models on the Embedding Space for Text Generation |
2022-12-19T12:40:18Z | http://arxiv.org/pdf/2212.09409v2 | http://arxiv.org/abs/2212.09409v2 | Dustin Wright, Isabelle Augenstein | 2023-05-23T14:44:24Z | cs.CL | Selecting an effective training signal for tasks in natural language
processing is difficult: expert annotations are expensive, and crowd-sourced
annotations may not be reliable. At the same time, recent work in NLP has
demonstrated that learning from a distribution over labels acquired from crowd
annotations can be effective. However, there are many ways to acquire such a
distribution, and the performance allotted by any one method can fluctuate
based on the task and the amount of available crowd annotations, making it
difficult to know a priori which distribution is best. This paper
systematically analyzes this in the out-of-domain setting, adding to the NLP
literature which has focused on in-domain evaluation, and proposes new methods
for acquiring soft-labels from crowd-annotations by aggregating the
distributions produced by existing methods. In particular, we propose to
aggregate multiple-views of crowd annotations via temperature scaling and
finding their Jensen-Shannon centroid. We demonstrate that these aggregation
methods lead to the most consistent performance across four NLP tasks on
out-of-domain test sets, mitigating fluctuations in performance from the
individual distributions. Additionally, aggregation results in the most
consistently well-calibrated uncertainty estimation. We argue that aggregating
different views of crowd-annotations is an effective and minimal intervention
to acquire soft-labels which induce robust classifiers despite the
inconsistency of the individual soft-labeling methods. | cs.CL | Multi-View Knowledge Distillation from Crowd Annotations for Out-of-Domain Generalization |
2022-12-19T12:40:18Z | http://arxiv.org/pdf/2212.09410v1 | http://arxiv.org/abs/2212.09410v1 | Zhiying Jiang, Matthew Y. R. Yang, Mikhail Tsirlin, Raphael Tang, Jimmy Lin | 2022-12-19T12:40:18Z | cs.CL | Deep neural networks (DNNs) are often used for text classification tasks as
they usually achieve high levels of accuracy. However, DNNs can be
computationally intensive with billions of parameters and large amounts of
labeled data, which can make them expensive to use, to optimize and to transfer
to out-of-distribution (OOD) cases in practice. In this paper, we propose a
non-parametric alternative to DNNs that's easy, light-weight and universal in
text classification: a combination of a simple compressor like gzip with a
$k$-nearest-neighbor classifier. Without any training, pre-training or
fine-tuning, our method achieves results that are competitive with
non-pretrained deep learning methods on six in-distributed datasets. It even
outperforms BERT on all five OOD datasets, including four low-resource
languages. Our method also performs particularly well in few-shot settings
where labeled data are too scarce for DNNs to achieve a satisfying accuracy. | cs.CL | Less is More: Parameter-Free Text Classification with Gzip |
2022-12-19T12:24:32Z | http://arxiv.org/pdf/2212.09400v3 | http://arxiv.org/abs/2212.09400v3 | Peng Gao, Feng Gao, Jian-Cheng Ni, Yu Wang, Fei Wang | 2024-02-23T01:45:32Z | cs.CL | Drug-drug interaction prediction is a crucial issue in molecular biology.
Traditional methods of observing drug-drug interactions through medical
experiments require significant resources and labor. This paper presents a
medical knowledge graph question answering model, dubbed MedKGQA, that predicts
drug-drug interaction by employing machine reading comprehension from
closed-domain literature and constructing a knowledge graph of drug-protein
triplets from open-domain documents. The model vectorizes the drug-protein
target attributes in the graph using entity embeddings and establishes directed
connections between drug and protein entities based on the metabolic
interaction pathways of protein targets in the human body. This aligns multiple
external knowledge and applies it to learn the graph neural network. Without
bells and whistles, the proposed model achieved a 4.5% improvement in terms of
drug-drug interaction prediction accuracy compared to previous state-of-the-art
models on the Qangaroo MedHop dataset. Experimental results demonstrate the
efficiency and effectiveness of the model and verify the feasibility of
integrating external knowledge in machine reading comprehension tasks. | cs.CL | Medical Knowledge Graph QA for Drug-Drug Interaction Prediction based on Multi-hop Machine Reading Comprehension |
2022-12-19T11:53:59Z | http://arxiv.org/pdf/2212.09387v2 | http://arxiv.org/abs/2212.09387v2 | Xuancheng Huang, Zijun Liu, Peng Li, Tao Li, Maosong Sun, Yang Liu | 2023-05-28T14:12:48Z | cs.CL, cs.AI | Recently, multi-aspect controllable text generation that controls the
generated text in multiple aspects (e.g., sentiment, topic, and keywords) has
attracted increasing attention. Although methods based on parameter efficient
tuning like prefix-tuning could achieve multi-aspect controlling in a
plug-and-play way, the mutual interference of multiple prefixes leads to
significant degeneration of constraints and limits their extensibility to
training-time unseen aspect combinations. In this work, we provide a
theoretical lower bound for the interference and empirically found that the
interference grows with the number of layers where prefixes are inserted. Based
on these analyses, we propose using trainable gates to normalize the
intervention of prefixes to restrain the growing interference. As a result,
controlling training-time unseen combinations of aspects can be realized by
simply concatenating corresponding plugins such that new constraints can be
extended at a lower cost. In addition, we propose a unified way to process both
categorical and free-form constraints. Experiments on text generation and
machine translation demonstrate the superiority of our approach over baselines
on constraint accuracy, text quality, and extensibility. | cs.CL | An Extensible Plug-and-Play Method for Multi-Aspect Controllable Text Generation |
2022-12-19T11:26:23Z | http://arxiv.org/pdf/2212.09376v1 | http://arxiv.org/abs/2212.09376v1 | Alessandro Temperoni, Maria Biryukov, Martin Theobald | 2022-12-19T11:26:23Z | cs.CL, cs.IR, cs.LG | Relation extraction (RE) is a sub-discipline of information extraction (IE)
which focuses on the prediction of a relational predicate from a
natural-language input unit (such as a sentence, a clause, or even a short
paragraph consisting of multiple sentences and/or clauses). Together with
named-entity recognition (NER) and disambiguation (NED), RE forms the basis for
many advanced IE tasks such as knowledge-base (KB) population and verification.
In this work, we explore how recent approaches for open information extraction
(OpenIE) may help to improve the task of RE by encoding structured information
about the sentences' principal units, such as subjects, objects, verbal
phrases, and adverbials, into various forms of vectorized (and hence
unstructured) representations of the sentences. Our main conjecture is that the
decomposition of long and possibly convoluted sentences into multiple smaller
clauses via OpenIE even helps to fine-tune context-sensitive language models
such as BERT (and its plethora of variants) for RE. Our experiments over two
annotated corpora, KnowledgeNet and FewRel, demonstrate the improved accuracy
of our enriched models compared to existing RE approaches. Our best results
reach 92% and 71% of F1 score for KnowledgeNet and FewRel, respectively,
proving the effectiveness of our approach on competitive benchmarks. | cs.CL | Enriching Relation Extraction with OpenIE |
2022-12-19T10:49:35Z | http://arxiv.org/pdf/2212.09359v3 | http://arxiv.org/abs/2212.09359v3 | Siqi Ouyang, Rong Ye, Lei Li | 2023-07-07T04:56:14Z | cs.CL, cs.SD, eess.AS | End-to-end Speech Translation (E2E ST) aims to directly translate source
speech into target text. Existing ST methods perform poorly when only extremely
small speech-text data are available for training. We observe that an ST
model's performance closely correlates with its embedding similarity between
speech and source transcript. In this paper, we propose Word-Aligned
COntrastive learning (WACO), a simple and effective method for extremely
low-resource speech-to-text translation. Our key idea is bridging word-level
representations for both speech and text modalities via contrastive learning.
We evaluate WACO and other methods on the MuST-C dataset, a widely used ST
benchmark, and on a low-resource direction Maltese-English from IWSLT 2023. Our
experiments demonstrate that WACO outperforms the best baseline by 9+ BLEU
points with only 1-hour parallel ST data. Code is available at
https://github.com/owaski/WACO. | cs.CL | WACO: Word-Aligned Contrastive Learning for Speech Translation |
2022-12-19T10:38:30Z | http://arxiv.org/pdf/2212.09353v1 | http://arxiv.org/abs/2212.09353v1 | Xiao Zhang, Heyan Huang, Zewen Chi, Xian-Ling Mao | 2022-12-19T10:38:30Z | cs.CL | Open-retrieval conversational machine reading comprehension (OCMRC) simulates
real-life conversational interaction scenes. Machines are required to make a
decision of "Yes/No/Inquire" or generate a follow-up question when the decision
is "Inquire" based on retrieved rule texts, user scenario, user question, and
dialogue history. Recent studies explored the methods to reduce the information
gap between decision-making and question generation and thus improve the
performance of generation. However, the information gap still exists because
these pipeline structures are still limited in decision-making, span
extraction, and question rephrasing three stages. Decision-making and
generation are reasoning separately, and the entailment reasoning utilized in
decision-making is hard to share through all stages. To tackle the above
problem, we proposed a novel one-stage end-to-end framework, called Entailment
Fused-T5 (EFT), to bridge the information gap between decision-making and
generation in a global understanding manner. The extensive experimental results
demonstrate that our proposed framework achieves new state-of-the-art
performance on the OR-ShARC benchmark. | cs.CL | Bridging The Gap: Entailment Fused-T5 for Open-retrieval Conversational Machine Reading Comprehension |
2022-12-19T09:03:32Z | http://arxiv.org/pdf/2212.09306v1 | http://arxiv.org/abs/2212.09306v1 | Ting Wai Terence Au, Ingemar J. Cox, Vasileios Lampos | 2022-12-19T09:03:32Z | cs.CL | Identifying named entities such as a person, location or organization, in
documents can highlight key information to readers. Training Named Entity
Recognition (NER) models requires an annotated data set, which can be a
time-consuming labour-intensive task. Nevertheless, there are publicly
available NER data sets for general English. Recently there has been interest
in developing NER for legal text. However, prior work and experimental results
reported here indicate that there is a significant degradation in performance
when NER methods trained on a general English data set are applied to legal
text. We describe a publicly available legal NER data set, called E-NER, based
on legal company filings available from the US Securities and Exchange
Commission's EDGAR data set. Training a number of different NER algorithms on
the general English CoNLL-2003 corpus but testing on our test collection
confirmed significant degradations in accuracy, as measured by the F1-score, of
between 29.4\% and 60.4\%, compared to training and testing on the E-NER
collection. | cs.CL | E-NER -- An Annotated Named Entity Recognition Corpus of Legal Text |
2022-12-19T09:02:16Z | http://arxiv.org/pdf/2212.09305v2 | http://arxiv.org/abs/2212.09305v2 | Wenda Xu, Xian Qian, Mingxuan Wang, Lei Li, William Yang Wang | 2023-07-07T17:49:31Z | cs.CL | Is it possible to train a general metric for evaluating text generation
quality without human annotated ratings? Existing learned metrics either
perform unsatisfactorily across text generation tasks or require human ratings
for training on specific tasks. In this paper, we propose SESCORE2, a
self-supervised approach for training a model-based metric for text generation
evaluation. The key concept is to synthesize realistic model mistakes by
perturbing sentences retrieved from a corpus. The primary advantage of the
SESCORE2 is its ease of extension to many other languages while providing
reliable severity estimation. We evaluate SESCORE2 and previous methods on four
text generation tasks across three languages. SESCORE2 outperforms unsupervised
metric PRISM on four text generation evaluation benchmarks, with a Kendall
improvement of 0.078. Surprisingly, SESCORE2 even outperforms the supervised
BLEURT and COMET on multiple text generation tasks. The code and data are
available at https://github.com/xu1998hz/SEScore2. | cs.CL | SESCORE2: Learning Text Generation Evaluation via Synthesizing Realistic Mistakes |
2022-12-19T08:15:16Z | http://arxiv.org/pdf/2212.09292v1 | http://arxiv.org/abs/2212.09292v1 | Teo Susnjak | 2022-12-19T08:15:16Z | cs.AI, cs.CL | This study evaluated the ability of ChatGPT, a recently developed artificial
intelligence (AI) agent, to perform high-level cognitive tasks and produce text
that is indistinguishable from human-generated text. This capacity raises
concerns about the potential use of ChatGPT as a tool for academic misconduct
in online exams. The study found that ChatGPT is capable of exhibiting critical
thinking skills and generating highly realistic text with minimal input, making
it a potential threat to the integrity of online exams, particularly in
tertiary education settings where such exams are becoming more prevalent.
Returning to invigilated and oral exams could form part of the solution, while
using advanced proctoring techniques and AI-text output detectors may be
effective in addressing this issue, they are not likely to be foolproof
solutions. Further research is needed to fully understand the implications of
large language models like ChatGPT and to devise strategies for combating the
risk of cheating using these tools. It is crucial for educators and
institutions to be aware of the possibility of ChatGPT being used for cheating
and to investigate measures to address it in order to maintain the fairness and
validity of online exams for all students. | cs.AI | ChatGPT: The End of Online Exam Integrity? |
2022-12-19T07:41:39Z | http://arxiv.org/pdf/2212.09284v1 | http://arxiv.org/abs/2212.09284v1 | Shelly Jain, Priyanshi Pal, Anil Vuppala, Prasanta Ghosh, Chiranjeevi Yarra | 2022-12-19T07:41:39Z | cs.CL | Speech systems are sensitive to accent variations. This is especially
challenging in the Indian context, with an abundance of languages but a dearth
of linguistic studies characterising pronunciation variations. The growing
number of L2 English speakers in India reinforces the need to study accents and
L1-L2 interactions. We investigate the accents of Indian English (IE) speakers
and report in detail our observations, both specific and common to all regions.
In particular, we observe the phonemic variations and phonotactics occurring in
the speakers' native languages and apply this to their English pronunciations.
We demonstrate the influence of 18 Indian languages on IE by comparing the
native language pronunciations with IE pronunciations obtained jointly from
existing literature studies and phonetically annotated speech of 80 speakers.
Consequently, we are able to validate the intuitions of Indian language
influences on IE pronunciations by justifying pronunciation rules from the
perspective of Indian language phonology. We obtain a comprehensive description
in terms of universal and region-specific characteristics of IE, which
facilitates accent conversion and adaptation of existing ASR and TTS systems to
different Indian accents. | cs.CL | An Investigation of Indian Native Language Phonemic Influences on L2 English Pronunciations |
2022-12-19T07:40:02Z | http://arxiv.org/pdf/2212.09282v2 | http://arxiv.org/abs/2212.09282v2 | Soumya Sanyal, Yichong Xu, Shuohang Wang, Ziyi Yang, Reid Pryzant, Wenhao Yu, Chenguang Zhu, Xiang Ren | 2023-06-05T00:56:11Z | cs.CL, cs.AI, cs.LG | Logical reasoning of text is an important ability that requires understanding
the information present in the text, their interconnections, and then reasoning
through them to infer new conclusions. Prior works on improving the logical
reasoning ability of language models require complex processing of training
data (e.g., aligning symbolic knowledge to text), yielding task-specific data
augmentation solutions that restrict the learning of general logical reasoning
skills. In this work, we propose APOLLO, an adaptively pretrained language
model that has improved logical reasoning abilities. We select a subset of
Wikipedia, based on a set of logical inference keywords, for continued
pretraining of a language model. We use two self-supervised loss functions: a
modified masked language modeling loss where only specific parts-of-speech
words, that would likely require more reasoning than basic language
understanding, are masked, and a sentence-level classification loss that
teaches the model to distinguish between entailment and contradiction types of
sentences. The proposed training paradigm is both simple and independent of
task formats. We demonstrate the effectiveness of APOLLO by comparing it with
prior baselines on two logical reasoning datasets. APOLLO performs comparably
on ReClor and outperforms baselines on LogiQA. The code base has been made
publicly available. | cs.CL | APOLLO: A Simple Approach for Adaptive Pretraining of Language Models for Logical Reasoning |
2022-12-19T07:14:32Z | http://arxiv.org/pdf/2212.09278v1 | http://arxiv.org/abs/2212.09278v1 | Yingwen Fu, Wenjie Ou, Zhou Yu, Yue Lin | 2022-12-19T07:14:32Z | cs.CL, cs.AI | Conversational text-to-SQL is designed to translate multi-turn natural
language questions into their corresponding SQL queries. Most state-of-the-art
conversational text- to-SQL methods are incompatible with generative
pre-trained language models (PLMs), such as T5. In this paper, we present a
two-stage unified MultI-task Generation frAmework (MIGA) that leverages PLMs'
ability to tackle conversational text-to-SQL. In the pre-training stage, MIGA
first decomposes the main task into several related sub-tasks and then unifies
them into the same sequence-to-sequence (Seq2Seq) paradigm with task-specific
natural language prompts to boost the main task from multi-task training. Later
in the fine-tuning stage, we propose four SQL perturbations to alleviate the
error propagation problem. MIGA tends to achieve state-of-the-art performance
on two benchmarks (SparC and CoSQL). We also provide extensive analyses and
discussions to shed light on some new perspectives for conversational
text-to-SQL. | cs.CL | MIGA: A Unified Multi-task Generation Framework for Conversational Text-to-SQL |
2022-12-19T06:55:42Z | http://arxiv.org/pdf/2212.09272v1 | http://arxiv.org/abs/2212.09272v1 | Chengwen Wang, Qingxiu Dong, Xiaochen Wang, Haitao Wang, Zhifang Sui | 2022-12-19T06:55:42Z | cs.CL | Datasets serve as crucial training resources and model performance trackers.
However, existing datasets have exposed a plethora of problems, inducing biased
models and unreliable evaluation results. In this paper, we propose a
model-agnostic dataset evaluation framework for automatic dataset quality
evaluation. We seek the statistical properties of the datasets and address
three fundamental dimensions: reliability, difficulty, and validity, following
a classical testing theory. Taking the Named Entity Recognition (NER) datasets
as a case study, we introduce $9$ statistical metrics for a statistical dataset
evaluation framework. Experimental results and human evaluation validate that
our evaluation framework effectively assesses various aspects of the dataset
quality. Furthermore, we study how the dataset scores on our statistical
metrics affect the model performance, and appeal for dataset quality evaluation
or targeted dataset improvement before training or testing models. | cs.CL | Statistical Dataset Evaluation: Reliability, Difficulty, and Validity |
2022-12-19T06:04:54Z | http://arxiv.org/pdf/2212.09257v2 | http://arxiv.org/abs/2212.09257v2 | Bairu Hou, Joe O'Connor, Jacob Andreas, Shiyu Chang, Yang Zhang | 2023-07-03T02:28:27Z | cs.CL | We describe PromptBoosting, a query-efficient procedure for building a text
classifier from a neural language model (LM) without access to the LM's
parameters, gradients, or hidden representations. This form of "black-box"
classifier training has become increasingly important as the cost of training
and inference in large-scale LMs grows. But existing black-box LM classifier
learning approaches are themselves computationally inefficient, typically
specializing LMs to the target task by searching in a large space of (discrete
or continuous) prompts using zeroth-order optimization methods. Instead of
directly optimizing in prompt space, PromptBoosting obtains a small pool of
prompts via a gradient-free approach and then constructs a large pool of weak
learners by pairing these prompts with different elements of the LM's output
distribution. These weak learners are then ensembled using the AdaBoost
algorithm. The entire learning process requires only a small number of forward
passes and no backward pass. Experiments show that PromptBoosting achieves
state-of-the-art performance in multiple black-box few-shot classification
tasks, and matches or outperforms full fine-tuning in both few-shot and
standard learning paradigms, while training 10x faster than existing black-box
methods. | cs.CL | PromptBoosting: Black-Box Text Classification with Ten Forward Passes |
2022-12-19T06:03:04Z | http://arxiv.org/pdf/2212.09255v1 | http://arxiv.org/abs/2212.09255v1 | Lester James Miranda, Ákos Kádár, Adriane Boyd, Sofie Van Landeghem, Anders Søgaard, Matthew Honnibal | 2022-12-19T06:03:04Z | cs.CL, I.2.7 | The distributed representation of symbols is one of the key technologies in
machine learning systems today, playing a pivotal role in modern natural
language processing. Traditional word embeddings associate a separate vector
with each word. While this approach is simple and leads to good performance, it
requires a lot of memory for representing a large vocabulary. To reduce the
memory footprint, the default embedding layer in spaCy is a hash embeddings
layer. It is a stochastic approximation of traditional embeddings that provides
unique vectors for a large number of words without explicitly storing a
separate vector for each of them. To be able to compute meaningful
representations for both known and unknown words, hash embeddings represent
each word as a summary of the normalized word form, subword information and
word shape. Together, these features produce a multi-embedding of a word. In
this technical report we lay out a bit of history and introduce the embedding
methods in spaCy in detail. Second, we critically evaluate the hash embedding
architecture with multi-embeddings on Named Entity Recognition datasets from a
variety of domains and languages. The experiments validate most key design
choices behind spaCy's embedders, but we also uncover a few surprising results. | cs.CL | Multi hash embeddings in spaCy |
Subsets and Splits