publication_date
stringlengths 20
20
| doc_url
stringlengths 33
34
| id
stringlengths 33
34
| authors
stringlengths 5
20.1k
| update_date
stringlengths 20
20
| category_all
stringlengths 5
138
| abstract
stringlengths 123
3.09k
| category_primary
stringclasses 78
values | title
stringlengths 10
211
|
|---|---|---|---|---|---|---|---|---|
2023-11-02T10:14:52Z | http://arxiv.org/pdf/2311.04918v1 | http://arxiv.org/abs/2311.04918v1 | Ngoc Dang Nguyen, Wei Tan, Lan Du, Wray Buntine, Richard Beare, Changyou Chen | 2023-11-02T10:14:52Z | cs.CL, cs.LG | Named entity recognition (NER), a task that identifies and categorizes named
entities such as persons or organizations from text, is traditionally framed as
a multi-class classification problem. However, this approach often overlooks
the issues of imbalanced label distributions, particularly in low-resource
settings, which is common in certain NER contexts, like biomedical NER
(bioNER). To address these issues, we propose an innovative reformulation of
the multi-class problem as a one-vs-all (OVA) learning problem and introduce a
loss function based on the area under the receiver operating characteristic
curve (AUC). To enhance the efficiency of our OVA-based approach, we propose
two training strategies: one groups labels with similar linguistic
characteristics, and another employs meta-learning. The superiority of our
approach is confirmed by its performance, which surpasses traditional NER
learning in varying NER settings. | cs.CL | Low-Resource Named Entity Recognition: Can One-vs-All AUC Maximization Help? |
2023-11-02T09:20:38Z | http://arxiv.org/pdf/2311.01108v1 | http://arxiv.org/abs/2311.01108v1 | Song Wang, Zhen Tan, Ruocheng Guo, Jundong Li | 2023-11-02T09:20:38Z | cs.CL | Adopting a two-stage paradigm of pretraining followed by fine-tuning,
Pretrained Language Models (PLMs) have achieved substantial advancements in the
field of natural language processing. However, in real-world scenarios, data
labels are often noisy due to the complex annotation process, making it
essential to develop strategies for fine-tuning PLMs with such noisy labels. To
this end, we introduce an innovative approach for fine-tuning PLMs using noisy
labels, which incorporates the guidance of Large Language Models (LLMs) like
ChatGPT. This guidance assists in accurately distinguishing between clean and
noisy samples and provides supplementary information beyond the noisy labels,
thereby boosting the learning process during fine-tuning PLMs. Extensive
experiments on synthetic and real-world noisy datasets further demonstrate the
superior advantages of our framework over the state-of-the-art baselines. | cs.CL | Noise-Robust Fine-Tuning of Pretrained Language Models via External Guidance |
2023-11-02T08:39:45Z | http://arxiv.org/pdf/2311.04917v2 | http://arxiv.org/abs/2311.04917v2 | Jinyan Su, Claire Cardie, Preslav Nakov | 2024-04-13T13:52:01Z | cs.CL, cs.AI | In the age of large language models (LLMs) and the widespread adoption of
AI-driven content creation, the landscape of information dissemination has
witnessed a paradigm shift. With the proliferation of both human-written and
machine-generated real and fake news, robustly and effectively discerning the
veracity of news articles has become an intricate challenge. While substantial
research has been dedicated to fake news detection, this either assumes that
all news articles are human-written or abruptly assumes that all
machine-generated news are fake. Thus, a significant gap exists in
understanding the interplay between machine-(paraphrased) real news,
machine-generated fake news, human-written fake news, and human-written real
news. In this paper, we study this gap by conducting a comprehensive evaluation
of fake news detectors trained in various scenarios. Our primary objectives
revolve around the following pivotal question: How to adapt fake news detectors
to the era of LLMs? Our experiments reveal an interesting pattern that
detectors trained exclusively on human-written articles can indeed perform well
at detecting machine-generated fake news, but not vice versa. Moreover, due to
the bias of detectors against machine-generated texts \cite{su2023fake}, they
should be trained on datasets with a lower machine-generated news ratio than
the test set. Building on our findings, we provide a practical strategy for the
development of robust fake news detectors. | cs.CL | Adapting Fake News Detection to the Era of Large Language Models |
2023-11-02T08:37:30Z | http://arxiv.org/pdf/2311.01070v3 | http://arxiv.org/abs/2311.01070v3 | Thomas Palmeira Ferraz, Marcely Zanon Boito, Caroline Brun, Vassilina Nikoulina | 2024-03-12T14:50:30Z | cs.CL, cs.SD, eess.AS | Whisper is a multitask and multilingual speech model covering 99 languages.
It yields commendable automatic speech recognition (ASR) results in a subset of
its covered languages, but the model still underperforms on a non-negligible
number of under-represented languages, a problem exacerbated in smaller model
versions. In this work, we propose DistilWhisper, an approach able to bridge
the performance gap in ASR for these languages while retaining the advantages
of multitask and multilingual capabilities. Our approach involves two key
strategies: lightweight modular ASR fine-tuning of whisper-small using
language-specific experts, and knowledge distillation from whisper-large-v2.
This dual approach allows us to effectively boost ASR performance while keeping
the robustness inherited from the multitask and multilingual pre-training.
Results demonstrate that our approach is more effective than standard
fine-tuning or LoRA adapters, boosting performance in the targeted languages
for both in- and out-of-domain test sets, while introducing only a negligible
parameter overhead at inference. | cs.CL | Multilingual DistilWhisper: Efficient Distillation of Multi-task Speech Models via Language-Specific Experts |
2023-11-02T07:50:43Z | http://arxiv.org/pdf/2311.01049v1 | http://arxiv.org/abs/2311.01049v1 | Thanh Nguyen Ngoc, Quang Nhat Tran, Arthur Tang, Bao Nguyen, Thuy Nguyen, Thanh Pham | 2023-11-02T07:50:43Z | cs.CL, cs.AI | Data is a cornerstone for fine-tuning large language models, yet acquiring
suitable data remains challenging. Challenges encompassed data scarcity,
linguistic diversity, and domain-specific content. This paper presents lessons
learned while crawling and refining data tailored for fine-tuning Vietnamese
language models. Crafting such a dataset, while accounting for linguistic
intricacies and striking a balance between inclusivity and accuracy, demands
meticulous planning. Our paper presents a multidimensional strategy including
leveraging existing datasets in the English language and developing customized
data-crawling scripts with the assistance of generative AI tools. A fine-tuned
LLM model for the Vietnamese language, which was produced using resultant
datasets, demonstrated good performance while generating Vietnamese news
articles from prompts. The study offers practical solutions and guidance for
future fine-tuning models in languages like Vietnamese. | cs.CL | Multi-dimensional data refining strategy for effective fine-tuning LLMs |
2023-11-02T07:20:49Z | http://arxiv.org/pdf/2311.01041v3 | http://arxiv.org/abs/2311.01041v3 | Lang Cao | 2024-05-29T09:19:35Z | cs.CL, cs.AI | Large language models (LLMs) have demonstrated impressive language
understanding and generation capabilities, enabling them to answer a wide range
of questions across various domains. However, these models are not flawless and
often produce responses that contain errors or misinformation. These
inaccuracies, commonly referred to as hallucinations, render LLMs unreliable
and even unusable in many scenarios. In this paper, our focus is on mitigating
the issue of hallucination in LLMs, particularly in the context of
question-answering. Instead of attempting to answer all questions, we explore a
refusal mechanism that instructs LLMs to refuse to answer challenging questions
in order to avoid errors. We then propose a simple yet effective solution
called Learn to Refuse (L2R), which incorporates the refusal mechanism to
enable LLMs to recognize and refuse to answer questions that they find
difficult to address. To achieve this, we utilize a structured knowledge base
to represent all the LLM's understanding of the world, enabling it to provide
traceable gold knowledge. This knowledge base is separate from the LLM and
initially empty. It can be filled with validated knowledge and progressively
expanded. When an LLM encounters questions outside its domain, the system
recognizes its knowledge scope and determines whether it can answer the
question independently. Additionally, we introduce a method for automatically
and efficiently expanding the knowledge base of LLMs. Through qualitative and
quantitative analysis, we demonstrate that our approach enhances the
controllability and reliability of LLMs. | cs.CL | Learn to Refuse: Making Large Language Models More Controllable and Reliable through Knowledge Scope Limitation and Refusal Mechanism |
2023-11-02T07:03:25Z | http://arxiv.org/pdf/2311.01036v1 | http://arxiv.org/abs/2311.01036v1 | JB. Kim, Hazel Kim, Joonghyuk Hahn, Yo-Sub Han | 2023-11-02T07:03:25Z | cs.CL, cs.AI, I.2.7; I.2.3; F.4.1 | Solving math word problems depends on how to articulate the problems, the
lens through which models view human linguistic expressions. Real-world
settings count on such a method even more due to the diverse practices of the
same mathematical operations. Earlier works constrain available thinking
processes by limited prediction strategies without considering their
significance in acquiring mathematical knowledge. We introduce Attention-based
THought Expansion Network Architecture (ATHENA) to tackle the challenges of
real-world practices by mimicking human thought expansion mechanisms in the
form of neural network propagation. A thought expansion recurrently generates
the candidates carrying the thoughts of possible math expressions driven from
the previous step and yields reasonable thoughts by selecting the valid
pathways to the goal. Our experiments show that ATHENA achieves a new
state-of-the-art stage toward the ideal model that is compelling in variant
questions even when the informativeness in training examples is restricted. | cs.CL | ATHENA: Mathematical Reasoning with Thought Expansion |
2023-11-02T06:43:50Z | http://arxiv.org/pdf/2311.01030v1 | http://arxiv.org/abs/2311.01030v1 | Hao Niu, Yun Xiong, Xiaosu Wang, Philip S. Yu | 2023-11-02T06:43:50Z | cs.CL, cs.AI | Aspect-based sentiment classification (ASC) aims to judge the sentiment
polarity conveyed by the given aspect term in a sentence. The sentiment
polarity is not only determined by the local context but also related to the
words far away from the given aspect term. Most recent efforts related to the
attention-based models can not sufficiently distinguish which words they should
pay more attention to in some cases. Meanwhile, graph-based models are coming
into ASC to encode syntactic dependency tree information. But these models do
not fully leverage syntactic dependency trees as they neglect to incorporate
dependency relation tag information into representation learning effectively.
In this paper, we address these problems by effectively modeling the local and
global features. Firstly, we design a local encoder containing: a Gaussian mask
layer and a covariance self-attention layer. The Gaussian mask layer tends to
adjust the receptive field around aspect terms adaptively to deemphasize the
effects of unrelated words and pay more attention to local information. The
covariance self-attention layer can distinguish the attention weights of
different words more obviously. Furthermore, we propose a dual-level graph
attention network as a global encoder by fully employing dependency tag
information to capture long-distance information effectively. Our model
achieves state-of-the-art performance on both SemEval 2014 and Twitter
datasets. | cs.CL | Joint Learning of Local and Global Features for Aspect-based Sentiment Classification |
2023-11-02T06:14:41Z | http://arxiv.org/pdf/2311.01012v3 | http://arxiv.org/abs/2311.01012v3 | Haryo Akbarianto Wibowo, Erland Hilman Fuadi, Made Nindyatama Nityasya, Radityo Eko Prasojo, Alham Fikri Aji | 2024-04-21T22:11:58Z | cs.CL, 68T50 | We present COPAL-ID, a novel, public Indonesian language common sense
reasoning dataset. Unlike the previous Indonesian COPA dataset (XCOPA-ID),
COPAL-ID incorporates Indonesian local and cultural nuances, and therefore,
provides a more natural portrayal of day-to-day causal reasoning within the
Indonesian cultural sphere. Professionally written by natives from scratch,
COPAL-ID is more fluent and free from awkward phrases, unlike the translated
XCOPA-ID. In addition, we present COPAL-ID in both standard Indonesian and in
Jakartan Indonesian-a dialect commonly used in daily conversation. COPAL-ID
poses a greater challenge for existing open-sourced and closed state-of-the-art
multilingual language models, yet is trivially easy for humans. Our findings
suggest that general multilingual models struggle to perform well, achieving
66.91% accuracy on COPAL-ID. South-East Asian-specific models achieve slightly
better performance of 73.88% accuracy. Yet, this number still falls short of
near-perfect human performance. This shows that these language models are still
way behind in comprehending the local nuances of Indonesian. | cs.CL | COPAL-ID: Indonesian Language Reasoning with Local Culture and Nuances |
2023-11-02T05:27:48Z | http://arxiv.org/pdf/2311.00998v1 | http://arxiv.org/abs/2311.00998v1 | Lucky Susanto, Ryandito Diandaru, Adila Krisnadhi, Ayu Purwarianti, Derry Wijaya | 2023-11-02T05:27:48Z | cs.CL, cs.AI | Neural machine translation (NMT) for low-resource local languages in
Indonesia faces significant challenges, including the need for a representative
benchmark and limited data availability. This work addresses these challenges
by comprehensively analyzing training NMT systems for four low-resource local
languages in Indonesia: Javanese, Sundanese, Minangkabau, and Balinese. Our
study encompasses various training approaches, paradigms, data sizes, and a
preliminary study into using large language models for synthetic low-resource
languages parallel data generation. We reveal specific trends and insights into
practical strategies for low-resource language translation. Our research
demonstrates that despite limited computational resources and textual data,
several of our NMT systems achieve competitive performances, rivaling the
translation quality of zero-shot gpt-3.5-turbo. These findings significantly
advance NMT for low-resource languages, offering valuable guidance for
researchers in similar contexts. | cs.CL | Replicable Benchmarking of Neural Machine Translation (NMT) on Low-Resource Local Languages in Indonesia |
2023-11-02T04:01:04Z | http://arxiv.org/pdf/2311.04916v2 | http://arxiv.org/abs/2311.04916v2 | Azmine Toushik Wasi | 2023-12-10T15:46:54Z | cs.CL, cs.AI, cs.IR, cs.LG, cs.SI | Islamophobic language is a prevalent challenge on online social interaction
platforms. Identifying and eliminating such hatred is a crucial step towards a
future of harmony and peace. This study presents a novel paradigm for
identifying and explaining hate speech towards Islam using graph neural
networks. Utilizing the intrinsic ability of graph neural networks to find,
extract, and use relationships across disparate data points, our model
consistently achieves outstanding performance while offering explanations for
the underlying correlations and causation. | cs.CL | Explainable Identification of Hate Speech towards Islam using Graph Neural Networks |
2023-11-02T03:32:30Z | http://arxiv.org/pdf/2311.00967v2 | http://arxiv.org/abs/2311.00967v2 | Keisuke Shirai, Cristian C. Beltran-Hernandez, Masashi Hamaya, Atsushi Hashimoto, Shohei Tanaka, Kento Kawaharazuka, Kazutoshi Tanaka, Yoshitaka Ushiku, Shinsuke Mori | 2024-02-20T03:13:30Z | cs.RO, cs.AI, cs.CL | Large language models (LLMs) are accelerating the development of
language-guided robot planners. Meanwhile, symbolic planners offer the
advantage of interpretability. This paper proposes a new task that bridges
these two trends, namely, multimodal planning problem specification. The aim is
to generate a problem description (PD), a machine-readable file used by the
planners to find a plan. By generating PDs from language instruction and scene
observation, we can drive symbolic planners in a language-guided framework. We
propose a Vision-Language Interpreter (ViLaIn), a new framework that generates
PDs using state-of-the-art LLM and vision-language models. ViLaIn can refine
generated PDs via error message feedback from the symbolic planner. Our aim is
to answer the question: How accurately can ViLaIn and the symbolic planner
generate valid robot plans? To evaluate ViLaIn, we introduce a novel dataset
called the problem description generation (ProDG) dataset. The framework is
evaluated with four new evaluation metrics. Experimental results show that
ViLaIn can generate syntactically correct problems with more than 99\% accuracy
and valid plans with more than 58\% accuracy. Our code and dataset are
available at https://github.com/omron-sinicx/ViLaIn. | cs.RO | Vision-Language Interpreter for Robot Task Planning |
2023-11-02T03:04:50Z | http://arxiv.org/pdf/2311.10749v1 | http://arxiv.org/abs/2311.10749v1 | Ashlee Kupor, Candice Morgan, Dorottya Demszky | 2023-11-02T03:04:50Z | cs.CY, cs.AI, cs.CL, cs.LG | Providing consistent, individualized feedback to teachers on their
instruction can improve student learning outcomes. Such feedback can especially
benefit novice instructors who teach on online platforms and have limited
access to instructional training. To build scalable measures of instruction, we
fine-tune RoBERTa and GPT models to identify five instructional talk moves
inspired by accountable talk theory: adding on, connecting, eliciting, probing
and revoicing students' ideas. We fine-tune these models on a newly annotated
dataset of 2500 instructor utterances derived from transcripts of small group
instruction in an online computer science course, Code in Place. Although we
find that GPT-3 consistently outperforms RoBERTa in terms of precision, its
recall varies significantly. We correlate the instructors' use of each talk
move with indicators of student engagement and satisfaction, including
students' section attendance, section ratings, and assignment completion rates.
We find that using talk moves generally correlates positively with student
outcomes, and connecting student ideas has the largest positive impact. These
results corroborate previous research on the effectiveness of accountable talk
moves and provide exciting avenues for using these models to provide
instructors with useful, scalable feedback. | cs.CY | Measuring Five Accountable Talk Moves to Improve Instruction at Scale |
2023-11-02T03:01:53Z | http://arxiv.org/pdf/2311.00958v1 | http://arxiv.org/abs/2311.00958v1 | Muhammad Dehan Al Kautsar, Rahmah Khoirussyifa' Nurdini, Samuel Cahyawijaya, Genta Indra Winata, Ayu Purwarianti | 2023-11-02T03:01:53Z | cs.CL, cs.AI | Task-oriented dialogue (ToD) systems have been mostly created for
high-resource languages, such as English and Chinese. However, there is a need
to develop ToD systems for other regional or local languages to broaden their
ability to comprehend the dialogue contexts in various languages. This paper
introduces IndoToD, an end-to-end multi domain ToD benchmark in Indonesian. We
extend two English ToD datasets to Indonesian, comprising four different
domains by delexicalization to efficiently reduce the size of annotations. To
ensure a high-quality data collection, we hire native speakers to manually
translate the dialogues. Along with the original English datasets, these new
Indonesian datasets serve as an effective benchmark for evaluating Indonesian
and English ToD systems as well as exploring the potential benefits of
cross-lingual and bilingual transfer learning approaches. | cs.CL | IndoToD: A Multi-Domain Indonesian Benchmark For End-to-End Task-Oriented Dialogue Systems |
2023-11-02T02:42:41Z | http://arxiv.org/pdf/2311.00953v1 | http://arxiv.org/abs/2311.00953v1 | Wanyu Du, Yangfeng Ji | 2023-11-02T02:42:41Z | cs.CL | The development of trustworthy conversational information-seeking systems
relies on dialogue models that can generate faithful and accurate responses
based on relevant knowledge texts. However, two main challenges hinder this
task. Firstly, language models may generate hallucinations due to data biases
present in their pretraining corpus. Secondly, knowledge texts often contain
redundant and irrelevant information that distracts the model's attention from
the relevant text span. Previous works use additional data annotations on the
knowledge texts to learn a knowledge identification module in order to bypass
irrelevant information, but collecting such high-quality span annotations can
be costly. In this work, we leverage reinforcement learning algorithms to
overcome the above challenges by introducing a novel reward function. Our
reward function combines an accuracy metric and a faithfulness metric to
provide a balanced quality judgment of generated responses, which can be used
as a cost-effective approximation to a human preference reward model when only
a few preference annotations are available. Empirical experiments on two
conversational information-seeking datasets demonstrate that our method can
compete with other strong supervised learning baselines. | cs.CL | Blending Reward Functions via Few Expert Demonstrations for Faithful and Accurate Knowledge-Grounded Dialogue Generation |
2023-11-02T02:22:21Z | http://arxiv.org/pdf/2311.00945v1 | http://arxiv.org/abs/2311.00945v1 | Yuan Gao, Nobuyuki Morioka, Yu Zhang, Nanxin Chen | 2023-11-02T02:22:21Z | cs.SD, cs.CL, cs.LG, eess.AS | We propose Easy End-to-End Diffusion-based Text to Speech, a simple and
efficient end-to-end text-to-speech model based on diffusion. E3 TTS directly
takes plain text as input and generates an audio waveform through an iterative
refinement process. Unlike many prior work, E3 TTS does not rely on any
intermediate representations like spectrogram features or alignment
information. Instead, E3 TTS models the temporal structure of the waveform
through the diffusion process. Without relying on additional conditioning
information, E3 TTS could support flexible latent structure within the given
audio. This enables E3 TTS to be easily adapted for zero-shot tasks such as
editing without any additional training. Experiments show that E3 TTS can
generate high-fidelity audio, approaching the performance of a state-of-the-art
neural TTS system. Audio samples are available at https://e3tts.github.io. | cs.SD | E3 TTS: Easy End-to-End Diffusion-based Text to Speech |
2023-11-02T02:21:39Z | http://arxiv.org/pdf/2311.04915v2 | http://arxiv.org/abs/2311.04915v2 | Yoon Kyung Lee, Inju Lee, Minjung Shin, Seoyeon Bae, Sowon Hahn | 2023-12-14T01:30:01Z | cs.CL, cs.AI, cs.HC | We present a novel method, the Chain of Empathy (CoE) prompting, that
utilizes insights from psychotherapy to induce Large Language Models (LLMs) to
reason about human emotional states. This method is inspired by various
psychotherapy approaches including Cognitive Behavioral Therapy (CBT),
Dialectical Behavior Therapy (DBT), Person Centered Therapy (PCT), and Reality
Therapy (RT), each leading to different patterns of interpreting clients'
mental states. LLMs without reasoning generated predominantly exploratory
responses. However, when LLMs used CoE reasoning, we found a more comprehensive
range of empathetic responses aligned with the different reasoning patterns of
each psychotherapy model. The CBT based CoE resulted in the most balanced
generation of empathetic responses. The findings underscore the importance of
understanding the emotional context and how it affects human and AI
communication. Our research contributes to understanding how psychotherapeutic
models can be incorporated into LLMs, facilitating the development of
context-specific, safer, and empathetic AI. | cs.CL | Chain of Empathy: Enhancing Empathetic Response of Large Language Models Based on Psychotherapy Models |
2023-11-02T01:17:29Z | http://arxiv.org/pdf/2311.00915v1 | http://arxiv.org/abs/2311.00915v1 | Zedian Xiao, William Held, Yanchen Liu, Diyi Yang | 2023-11-02T01:17:29Z | cs.CL | Large Language Models (LLMs) are trained on corpora disproportionally
weighted in favor of Standard American English. As a result, speakers of other
dialects experience significantly more failures when interacting with these
technologies. In practice, these speakers often accommodate their speech to be
better understood. Our work shares the belief that language technologies should
be designed to accommodate the diversity in English dialects and not the other
way around. However, prior works on dialect struggle with generalizing to
evolving and emerging dialects in a scalable manner. To fill this gap, our
method, HyperLoRA, leverages expert linguistic knowledge to enable
resource-efficient adaptation via hypernetworks. By disentangling
dialect-specific and cross-dialectal information, HyperLoRA improves
generalization to unseen dialects in a task-agnostic fashion. Not only is
HyperLoRA more scalable in the number of parameters, but it also achieves the
best or most competitive performance across 5 dialects in a zero-shot setting.
In this way, our approach facilitates access to language technology for
billions of English dialect speakers who are traditionally underrepresented. | cs.CL | Task-Agnostic Low-Rank Adapters for Unseen English Dialects |
2023-11-02T01:00:46Z | http://arxiv.org/pdf/2311.00913v1 | http://arxiv.org/abs/2311.00913v1 | Megh Thakkar, Tolga Bolukbasi, Sriram Ganapathy, Shikhar Vashishth, Sarath Chandar, Partha Talukdar | 2023-11-02T01:00:46Z | cs.CL | Language Models (LMs) pre-trained with self-supervision on large text corpora
have become the default starting point for developing models for various NLP
tasks. Once the pre-training corpus has been assembled, all data samples in the
corpus are treated with equal importance during LM pre-training. However, due
to varying levels of relevance and quality of data, equal importance to all the
data samples may not be the optimal choice. While data reweighting has been
explored in the context of task-specific supervised learning and LM
fine-tuning, model-driven reweighting for pre-training data has not been
explored. We fill this important gap and propose PRESENCE, a method for jointly
reweighting samples by leveraging self-influence (SI) scores as an indicator of
sample importance and pre-training. PRESENCE promotes novelty and stability for
model pre-training. Through extensive analysis spanning multiple model sizes,
datasets, and tasks, we present PRESENCE as an important first step in the
research direction of sample reweighting for pre-training language models. | cs.CL | Self-Influence Guided Data Reweighting for Language Model Pre-training |
2023-11-02T00:19:02Z | http://arxiv.org/pdf/2311.00906v1 | http://arxiv.org/abs/2311.00906v1 | Haocheng Luo, Wei Tan, Ngoc Dang Nguyen, Lan Du | 2023-11-02T00:19:02Z | cs.CL, cs.LG | Active learning, a widely adopted technique for enhancing machine learning
models in text and image classification tasks with limited annotation
resources, has received relatively little attention in the domain of Named
Entity Recognition (NER). The challenge of data imbalance in NER has hindered
the effectiveness of active learning, as sequence labellers lack sufficient
learning signals. To address these challenges, this paper presents a novel
reweighting-based active learning strategy that assigns dynamic smoothed
weights to individual tokens. This adaptable strategy is compatible with
various token-level acquisition functions and contributes to the development of
robust active learners. Experimental results on multiple corpora demonstrate
the substantial performance improvement achieved by incorporating our
re-weighting strategy into existing acquisition functions, validating its
practical efficacy. | cs.CL | Re-weighting Tokens: A Simple and Effective Active Learning Strategy for Named Entity Recognition |
2023-11-01T23:33:25Z | http://arxiv.org/pdf/2311.00897v1 | http://arxiv.org/abs/2311.00897v1 | Ernie Chang, Sidd Srinivasan, Mahi Luthra, Pin-Jie Lin, Varun Nagaraja, Forrest Iandola, Zechun Liu, Zhaoheng Ni, Changsheng Zhao, Yangyang Shi, Vikas Chandra | 2023-11-01T23:33:25Z | cs.SD, cs.CL, eess.AS | Text-to-audio generation (TTA) produces audio from a text description,
learning from pairs of audio samples and hand-annotated text. However,
commercializing audio generation is challenging as user-input prompts are often
under-specified when compared to text descriptions used to train TTA models. In
this work, we treat TTA models as a ``blackbox'' and address the user prompt
challenge with two key insights: (1) User prompts are generally
under-specified, leading to a large alignment gap between user prompts and
training prompts. (2) There is a distribution of audio descriptions for which
TTA models are better at generating higher quality audio, which we refer to as
``audionese''. To this end, we rewrite prompts with instruction-tuned models
and propose utilizing text-audio alignment as feedback signals via margin
ranking learning for audio improvements. On both objective and subjective human
evaluations, we observed marked improvements in both text-audio alignment and
music audio quality. | cs.SD | On The Open Prompt Challenge In Conditional Audio Generation |
2023-11-01T23:31:51Z | http://arxiv.org/pdf/2311.00895v1 | http://arxiv.org/abs/2311.00895v1 | Ernie Chang, Pin-Jie Lin, Yang Li, Sidd Srinivasan, Gael Le Lan, David Kant, Yangyang Shi, Forrest Iandola, Vikas Chandra | 2023-11-01T23:31:51Z | cs.SD, cs.CL, eess.AS | Distributional shift is a central challenge in the deployment of machine
learning models as they can be ill-equipped for real-world data. This is
particularly evident in text-to-audio generation where the encoded
representations are easily undermined by unseen prompts, which leads to the
degradation of generated audio -- the limited set of the text-audio pairs
remains inadequate for conditional audio generation in the wild as user prompts
are under-specified. In particular, we observe a consistent audio quality
degradation in generated audio samples with user prompts, as opposed to
training set prompts. To this end, we present a retrieval-based in-context
prompt editing framework that leverages the training captions as demonstrative
exemplars to revisit the user prompts. We show that the framework enhanced the
audio quality across the set of collected user prompts, which were edited with
reference to the training captions as exemplars. | cs.SD | In-Context Prompt Editing For Conditional Audio Generation |
2023-11-01T21:41:08Z | http://arxiv.org/pdf/2311.00871v1 | http://arxiv.org/abs/2311.00871v1 | Steve Yadlowsky, Lyric Doshi, Nilesh Tripuraneni | 2023-11-01T21:41:08Z | cs.LG, cs.CL, stat.ML | Transformer models, notably large language models (LLMs), have the remarkable
ability to perform in-context learning (ICL) -- to perform new tasks when
prompted with unseen input-output examples without any explicit model training.
In this work, we study how effectively transformers can bridge between their
pretraining data mixture, comprised of multiple distinct task families, to
identify and learn new tasks in-context which are both inside and outside the
pretraining distribution. Building on previous work, we investigate this
question in a controlled setting, where we study transformer models trained on
sequences of $(x, f(x))$ pairs rather than natural language. Our empirical
results show transformers demonstrate near-optimal unsupervised model selection
capabilities, in their ability to first in-context identify different task
families and in-context learn within them when the task families are
well-represented in their pretraining data. However when presented with tasks
or functions which are out-of-domain of their pretraining data, we demonstrate
various failure modes of transformers and degradation of their generalization
for even simple extrapolation tasks. Together our results highlight that the
impressive ICL abilities of high-capacity sequence models may be more closely
tied to the coverage of their pretraining data mixtures than inductive biases
that create fundamental generalization capabilities. | cs.LG | Pretraining Data Mixtures Enable Narrow Model Selection Capabilities in Transformer Models |
2023-11-01T21:36:39Z | http://arxiv.org/pdf/2311.00867v1 | http://arxiv.org/abs/2311.00867v1 | Amrit Romana, Kazuhito Koishida, Emily Mower Provost | 2023-11-01T21:36:39Z | eess.AS, cs.CL | Speech disfluencies, such as filled pauses or repetitions, are disruptions in
the typical flow of speech. Stuttering is a speech disorder characterized by a
high rate of disfluencies, but all individuals speak with some disfluencies and
the rates of disfluencies may by increased by factors such as cognitive load.
Clinically, automatic disfluency detection may help in treatment planning for
individuals who stutter. Outside of the clinic, automatic disfluency detection
may serve as a pre-processing step to improve natural language understanding in
downstream applications. With this wide range of applications in mind, we
investigate language, acoustic, and multimodal methods for frame-level
automatic disfluency detection and categorization. Each of these methods relies
on audio as an input. First, we evaluate several automatic speech recognition
(ASR) systems in terms of their ability to transcribe disfluencies, measured
using disfluency error rates. We then use these ASR transcripts as input to a
language-based disfluency detection model. We find that disfluency detection
performance is largely limited by the quality of transcripts and alignments. We
find that an acoustic-based approach that does not require transcription as an
intermediate step outperforms the ASR language approach. Finally, we present
multimodal architectures which we find improve disfluency detection performance
over the unimodal approaches. Ultimately, this work introduces novel approaches
for automatic frame-level disfluency and categorization. In the long term, this
will help researchers incorporate automatic disfluency detection into a range
of applications. | eess.AS | Automatic Disfluency Detection from Untranscribed Speech |
2023-11-01T21:32:51Z | http://arxiv.org/pdf/2311.00863v1 | http://arxiv.org/abs/2311.00863v1 | Lucia Quirke, Lovis Heindrich, Wes Gurnee, Neel Nanda | 2023-11-01T21:32:51Z | cs.LG, cs.AI, cs.CL | Prior work has shown the existence of contextual neurons in language models,
including a neuron that activates on German text. We show that this neuron
exists within a broader contextual n-gram circuit: we find late layer neurons
which recognize and continue n-grams common in German text, but which only
activate if the German neuron is active. We investigate the formation of this
circuit throughout training and find that it is an example of what we call a
second-order circuit. In particular, both the constituent n-gram circuits and
the German detection circuit which culminates in the German neuron form with
independent functions early in training - the German detection circuit
partially through modeling German unigram statistics, and the n-grams by
boosting appropriate completions. Only after both circuits have already formed
do they fit together into a second-order circuit. Contrary to the hypotheses
presented in prior work, we find that the contextual n-gram circuit forms
gradually rather than in a sudden phase transition. We further present a range
of anomalous observations such as a simultaneous phase transition in many tasks
coinciding with the learning rate warm-up, and evidence that many context
neurons form simultaneously early in training but are later unlearned. | cs.LG | Training Dynamics of Contextual N-Grams in Language Models |
2023-11-01T20:39:12Z | http://arxiv.org/pdf/2311.00835v1 | http://arxiv.org/abs/2311.00835v1 | Yanlin Feng, Adithya Pratapa, David R Mortensen | 2023-11-01T20:39:12Z | cs.CL | Ultra-fine entity typing plays a crucial role in information extraction by
predicting fine-grained semantic types for entity mentions in text. However,
this task poses significant challenges due to the massive number of entity
types in the output space. The current state-of-the-art approaches, based on
standard multi-label classifiers or cross-encoder models, suffer from poor
generalization performance or inefficient inference. In this paper, we present
CASENT, a seq2seq model designed for ultra-fine entity typing that predicts
ultra-fine types with calibrated confidence scores. Our model takes an entity
mention as input and employs constrained beam search to generate multiple types
autoregressively. The raw sequence probabilities associated with the predicted
types are then transformed into confidence scores using a novel calibration
method. We conduct extensive experiments on the UFET dataset which contains
over 10k types. Our method outperforms the previous state-of-the-art in terms
of F1 score and calibration error, while achieving an inference speedup of over
50 times. Additionally, we demonstrate the generalization capabilities of our
model by evaluating it in zero-shot and few-shot settings on five specialized
domain entity typing datasets that are unseen during training. Remarkably, our
model outperforms large language models with 10 times more parameters in the
zero-shot setting, and when fine-tuned on 50 examples, it significantly
outperforms ChatGPT on all datasets. Our code, models and demo are available at
https://github.com/yanlinf/CASENT. | cs.CL | Calibrated Seq2seq Models for Efficient and Generalizable Ultra-fine Entity Typing |
2023-11-01T19:21:55Z | http://arxiv.org/pdf/2311.00790v2 | http://arxiv.org/abs/2311.00790v2 | Joanne Boisson, Luis Espinosa-Anke, Jose Camacho-Collados | 2023-11-15T19:11:40Z | cs.CL | Metaphor identification aims at understanding whether a given expression is
used figuratively in context. However, in this paper we show how existing
metaphor identification datasets can be gamed by fully ignoring the potential
metaphorical expression or the context in which it occurs. We test this
hypothesis in a variety of datasets and settings, and show that metaphor
identification systems based on language models without complete information
can be competitive with those using the full context. This is due to the
construction procedures to build such datasets, which introduce unwanted biases
for positive and negative classes. Finally, we test the same hypothesis on
datasets that are carefully sampled from natural corpora and where this bias is
not present, making these datasets more challenging and reliable. | cs.CL | Construction Artifacts in Metaphor Identification Datasets |
2023-11-01T18:41:50Z | http://arxiv.org/pdf/2311.04913v2 | http://arxiv.org/abs/2311.04913v2 | Suhaima Jamal, Hayden Wimmer | 2023-11-12T16:32:16Z | cs.CL, cs.AI | Phishing and spam detection is long standing challenge that has been the
subject of much academic research. Large Language Models (LLM) have vast
potential to transform society and provide new and innovative approaches to
solve well-established challenges. Phishing and spam have caused financial
hardships and lost time and resources to email users all over the world and
frequently serve as an entry point for ransomware threat actors. While
detection approaches exist, especially heuristic-based approaches, LLMs offer
the potential to venture into a new unexplored area for understanding and
solving this challenge. LLMs have rapidly altered the landscape from business,
consumers, and throughout academia and demonstrate transformational potential
for the potential of society. Based on this, applying these new and innovative
approaches to email detection is a rational next step in academic research. In
this work, we present IPSDM, our model based on fine-tuning the BERT family of
models to specifically detect phishing and spam email. We demonstrate our
fine-tuned version, IPSDM, is able to better classify emails in both unbalanced
and balanced datasets. This work serves as an important first step towards
employing LLMs to improve the security of our information systems. | cs.CL | An Improved Transformer-based Model for Detecting Phishing, Spam, and Ham: A Large Language Model Approach |
2023-11-01T18:31:02Z | http://arxiv.org/pdf/2311.04911v1 | http://arxiv.org/abs/2311.04911v1 | Samyar Janatian, Hannes Westermann, Jinzhe Tan, Jaromir Savelka, Karim Benyekhlef | 2023-11-01T18:31:02Z | cs.CL, cs.AI, cs.HC | Encoding legislative text in a formal representation is an important
prerequisite to different tasks in the field of AI & Law. For example,
rule-based expert systems focused on legislation can support laypeople in
understanding how legislation applies to them and provide them with helpful
context and information. However, the process of analyzing legislation and
other sources to encode it in the desired formal representation can be
time-consuming and represents a bottleneck in the development of such systems.
Here, we investigate to what degree large language models (LLMs), such as
GPT-4, are able to automatically extract structured representations from
legislation. We use LLMs to create pathways from legislation, according to the
JusticeBot methodology for legal decision support systems, evaluate the
pathways and compare them to manually created pathways. The results are
promising, with 60% of generated pathways being rated as equivalent or better
than manually created ones in a blind comparison. The approach suggests a
promising path to leverage the capabilities of LLMs to ease the costly
development of systems based on symbolic approaches that are transparent and
explainable. | cs.CL | From Text to Structure: Using Large Language Models to Support the Development of Legal Expert Systems |
2023-11-01T18:23:12Z | http://arxiv.org/pdf/2311.00768v2 | http://arxiv.org/abs/2311.00768v2 | Yurong Hu, Manuel Burger, Gunnar Rätsch, Rita Kuznetsova | 2024-02-06T16:33:48Z | cs.LG, cs.CL | In research areas with scarce data, representation learning plays a
significant role. This work aims to enhance representation learning for
clinical time series by deriving universal embeddings for clinical features,
such as heart rate and blood pressure. We use self-supervised training
paradigms for language models to learn high-quality clinical feature
embeddings, achieving a finer granularity than existing time-step and
patient-level representation learning. We visualize the learnt embeddings via
unsupervised dimension reduction techniques and observe a high degree of
consistency with prior clinical knowledge. We also evaluate the model
performance on the MIMIC-III benchmark and demonstrate the effectiveness of
using clinical feature embeddings. We publish our code online for replication. | cs.LG | Language Model Training Paradigms for Clinical Feature Embeddings |
2023-11-01T18:08:44Z | http://arxiv.org/pdf/2311.00762v1 | http://arxiv.org/abs/2311.00762v1 | Carol Neidle | 2023-11-01T18:08:44Z | cs.CV, cs.CL | There have been recent advances in computer-based recognition of isolated,
citation-form signs from video. There are many challenges for such a task, not
least the naturally occurring inter- and intra- signer synchronic variation in
sign production, including sociolinguistic variation in the realization of
certain signs. However, there are several significant factors that make
recognition of signs from continuous signing an even more difficult problem.
This article presents an overview of such challenges, based in part on findings
from a large corpus of linguistically annotated video data for American Sign
Language (ASL). Some linguistic regularities in the structure of signs that can
boost handshape and sign recognition are also discussed. | cs.CV | Challenges for Linguistically-Driven Computer-Based Sign Recognition from Continuous Signing for American Sign Language |
2023-11-01T17:55:09Z | http://arxiv.org/pdf/2311.00697v1 | http://arxiv.org/abs/2311.00697v1 | Juan Zuluaga-Gomez, Zhaocheng Huang, Xing Niu, Rohit Paturi, Sundararajan Srinivasan, Prashant Mathur, Brian Thompson, Marcello Federico | 2023-11-01T17:55:09Z | cs.CL, eess.AS | Conventional speech-to-text translation (ST) systems are trained on
single-speaker utterances, and they may not generalize to real-life scenarios
where the audio contains conversations by multiple speakers. In this paper, we
tackle single-channel multi-speaker conversational ST with an end-to-end and
multi-task training model, named Speaker-Turn Aware Conversational Speech
Translation, that combines automatic speech recognition, speech translation and
speaker turn detection using special tokens in a serialized labeling format. We
run experiments on the Fisher-CALLHOME corpus, which we adapted by merging the
two single-speaker channels into one multi-speaker channel, thus representing
the more realistic and challenging scenario with multi-speaker turns and
cross-talk. Experimental results across single- and multi-speaker conditions
and against conventional ST systems, show that our model outperforms the
reference systems on the multi-speaker condition, while attaining comparable
performance on the single-speaker condition. We release scripts for data
processing and model training. | cs.CL | End-to-End Single-Channel Speaker-Turn Aware Conversational Speech Translation |
2023-11-01T17:52:15Z | http://arxiv.org/pdf/2311.00694v2 | http://arxiv.org/abs/2311.00694v2 | Zhan Ling, Yunhao Fang, Xuanlin Li, Tongzhou Mu, Mingu Lee, Reza Pourreza, Roland Memisevic, Hao Su | 2023-12-05T20:44:45Z | cs.AI, cs.CL | Large Language Models (LLMs) have achieved tremendous progress, yet they
still often struggle with challenging reasoning problems. Current approaches
address this challenge by sampling or searching detailed and low-level
reasoning chains. However, these methods are still limited in their exploration
capabilities, making it challenging for correct solutions to stand out in the
huge solution space. In this work, we unleash LLMs' creative potential for
exploring multiple diverse problem solving strategies by framing an LLM as a
hierarchical policy via in-context learning. This policy comprises of a
visionary leader that proposes multiple diverse high-level problem-solving
tactics as hints, accompanied by a follower that executes detailed
problem-solving processes following each of the high-level instruction. The
follower uses each of the leader's directives as a guide and samples multiple
reasoning chains to tackle the problem, generating a solution group for each
leader proposal. Additionally, we propose an effective and efficient
tournament-based approach to select among these explored solution groups to
reach the final answer. Our approach produces meaningful and inspiring hints,
enhances problem-solving strategy exploration, and improves the final answer
accuracy on challenging problems in the MATH dataset. Code will be released at
https://github.com/lz1oceani/LLM-As-Hierarchical-Policy. | cs.AI | Unleashing the Creative Mind: Language Model As Hierarchical Policy For Improved Exploration on Challenging Problem Solving |
2023-11-01T17:44:50Z | http://arxiv.org/pdf/2311.00687v2 | http://arxiv.org/abs/2311.00687v2 | Ryan Liu, Howard Yen, Raja Marjieh, Thomas L. Griffiths, Ranjay Krishna | 2023-11-03T13:17:55Z | cs.AI, cs.CL, cs.HC, cs.LG | How do we communicate with others to achieve our goals? We use our prior
experience or advice from others, or construct a candidate utterance by
predicting how it will be received. However, our experiences are limited and
biased, and reasoning about potential outcomes can be difficult and cognitively
challenging. In this paper, we explore how we can leverage Large Language Model
(LLM) simulations to help us communicate better. We propose the
Explore-Generate-Simulate (EGS) framework, which takes as input any scenario
where an individual is communicating to an audience with a goal they want to
achieve. EGS (1) explores the solution space by producing a diverse set of
advice relevant to the scenario, (2) generates communication candidates
conditioned on subsets of the advice, and (3) simulates the reactions from
various audiences to determine both the best candidate and advice to use. We
evaluate the framework on eight scenarios spanning the ten fundamental
processes of interpersonal communication. For each scenario, we collect a
dataset of human evaluations across candidates and baselines, and showcase that
our framework's chosen candidate is preferred over popular generation
mechanisms including Chain-of-Thought. We also find that audience simulations
achieve reasonably high agreement with human raters across 5 of the 8
scenarios. Finally, we demonstrate the generality of our framework by applying
it to real-world scenarios described by users on web forums. Through
evaluations and demonstrations, we show that EGS enhances the effectiveness and
outcomes of goal-oriented communication across a variety of situations, thus
opening up new possibilities for the application of large language models in
revolutionizing communication and decision-making processes. | cs.AI | Improving Interpersonal Communication by Simulating Audiences with Language Models |
2023-11-01T17:44:35Z | http://arxiv.org/pdf/2311.00686v1 | http://arxiv.org/abs/2311.00686v1 | Neema Kotonya, Saran Krishnasamy, Joel Tetreault, Alejandro Jaimes | 2023-11-01T17:44:35Z | cs.CL | This paper describes and analyzes our participation in the 2023 Eval4NLP
shared task, which focuses on assessing the effectiveness of prompt-based
techniques to empower Large Language Models to handle the task of quality
estimation, particularly in the context of evaluating machine translations and
summaries. We conducted systematic experiments with various prompting
techniques, including standard prompting, prompts informed by annotator
instructions, and innovative chain-of-thought prompting. In addition, we
integrated these approaches with zero-shot and one-shot learning methods to
maximize the efficacy of our evaluation procedures. Our work reveals that
combining these approaches using a "small", open source model (orca_mini_v3_7B)
yields competitive results. | cs.CL | Little Giants: Exploring the Potential of Small LLMs as Evaluation Metrics in Summarization in the Eval4NLP 2023 Shared Task |
2023-11-01T17:43:35Z | http://arxiv.org/pdf/2311.00684v2 | http://arxiv.org/abs/2311.00684v2 | Ta-Chung Chi, Ting-Han Fan, Alexander I. Rudnicky | 2023-11-15T15:55:02Z | cs.CL, cs.LG | An ideal length-extrapolatable Transformer language model can handle
sequences longer than the training length without any fine-tuning. Such
long-context utilization capability relies heavily on a flexible positional
embedding design. Upon investigating the flexibility of existing large
pre-trained Transformer language models, we find that the T5 family deserves a
closer look, as its positional embeddings capture rich and flexible attention
patterns. However, T5 suffers from the dispersed attention issue: the longer
the input sequence, the flatter the attention distribution. To alleviate the
issue, we propose two attention alignment strategies via temperature scaling.
Our findings show improvement on the long-context utilization capability of T5
on language modeling, retrieval, multi-document question answering, and code
completion tasks without any fine-tuning. This suggests that a flexible
positional embedding design and attention alignment can go a long way toward
Transformer length extrapolation. | cs.CL | Attention Alignment and Flexible Positional Embeddings Improve Transformer Length Extrapolation |
2023-11-01T17:42:45Z | http://arxiv.org/pdf/2311.00681v1 | http://arxiv.org/abs/2311.00681v1 | Xue-Yong Fu, Md Tahmid Rahman Laskar, Cheng Chen, Shashi Bhushan TN | 2023-11-01T17:42:45Z | cs.CL | In recent years, Large Language Models (LLMs) have gained immense attention
due to their notable emergent capabilities, surpassing those seen in earlier
language models. A particularly intriguing application of LLMs is their role as
evaluators for texts produced by various generative models.
In this study, we delve into the potential of LLMs as reliable assessors of
factual consistency in summaries generated by text-generation models.
Initially, we introduce an innovative approach for factuality assessment using
LLMs. This entails employing a singular LLM for the entirety of the
question-answering-based factuality scoring process. Following this, we examine
the efficacy of various LLMs in direct factuality scoring, benchmarking them
against traditional measures and human annotations.
Contrary to initial expectations, our results indicate a lack of significant
correlations between factuality metrics and human evaluations, specifically for
GPT-4 and PaLM-2. Notable correlations were only observed with GPT-3.5 across
two factuality subcategories. These consistent findings across various factual
error categories suggest a fundamental limitation in the current LLMs'
capability to accurately gauge factuality.
This version presents the information more concisely while maintaining the
main points and findings of the original text. | cs.CL | Are Large Language Models Reliable Judges? A Study on the Factuality Evaluation Capabilities of LLMs |
2023-11-01T17:21:09Z | http://arxiv.org/pdf/2311.00671v1 | http://arxiv.org/abs/2311.00671v1 | Zhiwei Liu, Tianlin Zhang, Kailai Yang, Paul Thompson, Zeping Yu, Sophia Ananiadou | 2023-11-01T17:21:09Z | cs.CL | With the advent of social media, an increasing number of netizens are sharing
and reading posts and news online. However, the huge volumes of misinformation
(e.g., fake news and rumors) that flood the internet can adversely affect
people's lives, and have resulted in the emergence of rumor and fake news
detection as a hot research topic. The emotions and sentiments of netizens, as
expressed in social media posts and news, constitute important factors that can
help to distinguish fake news from genuine news and to understand the spread of
rumors. This article comprehensively reviews emotion-based methods for
misinformation detection. We begin by explaining the strong links between
emotions and misinformation. We subsequently provide a detailed analysis of a
range of misinformation detection methods that employ a variety of emotion,
sentiment and stance-based features, and describe their strengths and
weaknesses. Finally, we discuss a number of ongoing challenges in emotion-based
misinformation detection based on large language models and suggest future
research directions, including data collection (multi-platform, multilingual),
annotation, benchmark, multimodality, and interpretability. | cs.CL | Emotion Detection for Misinformation: A Review |
2023-11-01T17:02:49Z | http://arxiv.org/pdf/2311.00658v1 | http://arxiv.org/abs/2311.00658v1 | Eylon Gueta, Omer Goldman, Reut Tsarfaty | 2023-11-01T17:02:49Z | cs.CL | Pre-trained language models (PLMs) have shown remarkable successes in
acquiring a wide range of linguistic knowledge, relying solely on
self-supervised training on text streams. Nevertheless, the effectiveness of
this language-agnostic approach has been frequently questioned for its
sub-optimal performance when applied to morphologically-rich languages (MRLs).
We investigate the hypothesis that incorporating explicit morphological
knowledge in the pre-training phase can improve the performance of PLMs for
MRLs. We propose various morphologically driven tokenization methods enabling
the model to leverage morphological cues beyond raw text. We pre-train multiple
language models utilizing the different methods and evaluate them on Hebrew, a
language with complex and highly ambiguous morphology. Our experiments show
that morphologically driven tokenization demonstrates improved results compared
to a standard language-agnostic tokenization, on a benchmark of both semantic
and morphologic tasks. These findings suggest that incorporating morphological
knowledge holds the potential for further improving PLMs for morphologically
rich languages. | cs.CL | Explicit Morphological Knowledge Improves Pre-training of Language Models for Hebrew |
2023-11-01T16:28:38Z | http://arxiv.org/pdf/2311.00629v1 | http://arxiv.org/abs/2311.00629v1 | Kamila Barylska, Anna Gogolinska, Lukasz Mikulski, Anna Philippou, Marcin Piatkowski, Kyriaki Psara | 2023-11-01T16:28:38Z | cs.LO, cs.CL, 03, F.2; G.0 | Reversible computation is an emerging computing paradigm that allows any
sequence of operations to be executed in reverse order at any point during
computation. Its appeal lies in its potential for lowpower computation and its
relevance to a wide array of applications such as chemical reactions, quantum
computation, robotics, and distributed systems. Reversing Petri nets are a
recently-proposed extension of Petri nets that implements the three main forms
of reversibility, namely, backtracking, causal reversing, and
out-of-causal-order reversing. Their distinguishing feature is the use of named
tokens that can be combined together to form bonds. Named tokens along with a
history function, constitute the means of remembering past behaviour, thus,
enabling reversal. In recent work, we have proposed a structural translation
from a subclass of RPNs to the model of Coloured Petri Nets (CPNs), an
extension of traditional Petri nets where tokens carry data values. In this
paper, we extend the translation to handle RPNs with token multiplicity under
the individual-token interpretation, a model which allows multiple tokens of
the same type to exist in a system. To support the three types of
reversibility, tokens are associated with their causal history and, while
tokens of the same type are equally eligible to fire a transition when going
forward, when going backwards they are able to reverse only the transitions
they have previously fired. The new translation, in addition to lifting the
restriction on token uniqueness, presents a refined approach for transforming
RPNs to CPNs through a unifying approach that allows instantiating each of the
three types of reversibility. The paper also reports on a tool that implements
this translation, paving the way for automated translations and analysis of
reversible systems using CPN Tools. | cs.LO | Formal Translation from Reversing Petri Nets to Coloured Petri Nets |
2023-11-01T15:33:38Z | http://arxiv.org/pdf/2311.00588v1 | http://arxiv.org/abs/2311.00588v1 | Yu Yang, Xiaotong Shen | 2023-11-01T15:33:38Z | cs.CL, cs.AI | This paper presents FlowSUM, a normalizing flows-based variational
encoder-decoder framework for Transformer-based summarization. Our approach
tackles two primary challenges in variational summarization: insufficient
semantic information in latent representations and posterior collapse during
training. To address these challenges, we employ normalizing flows to enable
flexible latent posterior modeling, and we propose a controlled alternate
aggressive training (CAAT) strategy with an improved gate mechanism.
Experimental results show that FlowSUM significantly enhances the quality of
generated summaries and unleashes the potential for knowledge distillation with
minimal impact on inference time. Furthermore, we investigate the issue of
posterior collapse in normalizing flows and analyze how the summary quality is
affected by the training strategy, gate initialization, and the type and number
of normalizing flows used, offering valuable insights for future research. | cs.CL | Boosting Summarization with Normalizing Flows and Aggressive Training |
2023-11-01T15:32:50Z | http://arxiv.org/pdf/2311.00587v2 | http://arxiv.org/abs/2311.00587v2 | Xiaoqian Li, Ercong Nie, Sheng Liang | 2023-12-02T16:54:23Z | cs.CL | The promise of Large Language Models (LLMs) in Natural Language Processing
has often been overshadowed by their limited performance in low-resource
languages such as Bangla. To address this, our paper presents a pioneering
approach that utilizes cross-lingual retrieval augmented in-context learning.
By strategically sourcing semantically similar prompts from high-resource
language, we enable multilingual pretrained language models (MPLMs), especially
the generative model BLOOMZ, to successfully boost performance on Bangla tasks.
Our extensive evaluation highlights that the cross-lingual retrieval augmented
prompts bring steady improvements to MPLMs over the zero-shot performance. | cs.CL | Crosslingual Retrieval Augmented In-context Learning for Bangla |
2023-11-01T15:14:46Z | http://arxiv.org/pdf/2311.00739v1 | http://arxiv.org/abs/2311.00739v1 | Naiqing Guan, Kaiwen Chen, Nick Koudas | 2023-11-01T15:14:46Z | cs.CL, cs.DB, cs.LG, H.2.8; I.5.4 | Programmatic weak supervision methodologies facilitate the expedited labeling
of extensive datasets through the use of label functions (LFs) that encapsulate
heuristic data sources. Nonetheless, the creation of precise LFs necessitates
domain expertise and substantial endeavors. Recent advances in pre-trained
language models (PLMs) have exhibited substantial potential across diverse
tasks. However, the capacity of PLMs to autonomously formulate accurate LFs
remains an underexplored domain. In this research, we address this gap by
introducing DataSculpt, an interactive framework that harnesses PLMs for the
automated generation of LFs. Within DataSculpt, we incorporate an array of
prompting techniques, instance selection strategies, and LF filtration methods
to explore the expansive design landscape. Ultimately, we conduct a thorough
assessment of DataSculpt's performance on 12 real-world datasets, encompassing
a range of tasks. This evaluation unveils both the strengths and limitations of
contemporary PLMs in LF design. | cs.CL | Can Large Language Models Design Accurate Label Functions? |
2023-11-01T15:13:43Z | http://arxiv.org/pdf/2311.00571v1 | http://arxiv.org/abs/2311.00571v1 | Wei-Ge Chen, Irina Spiridonova, Jianwei Yang, Jianfeng Gao, Chunyuan Li | 2023-11-01T15:13:43Z | cs.CV, cs.AI, cs.CL, cs.HC, cs.MM | LLaVA-Interactive is a research prototype for multimodal human-AI
interaction. The system can have multi-turn dialogues with human users by
taking multimodal user inputs and generating multimodal responses. Importantly,
LLaVA-Interactive goes beyond language prompt, where visual prompt is enabled
to align human intents in the interaction. The development of LLaVA-Interactive
is extremely cost-efficient as the system combines three multimodal skills of
pre-built AI models without additional model training: visual chat of LLaVA,
image segmentation from SEEM, as well as image generation and editing from
GLIGEN. A diverse set of application scenarios is presented to demonstrate the
promises of LLaVA-Interactive and to inspire future research in multimodal
interactive systems. | cs.CV | LLaVA-Interactive: An All-in-One Demo for Image Chat, Segmentation, Generation and Editing |
2023-11-01T14:20:18Z | http://arxiv.org/pdf/2311.00541v5 | http://arxiv.org/abs/2311.00541v5 | Schyan Zafar, Geoff K. Nicholls | 2024-06-25T16:13:12Z | cs.CL, stat.ME | Word meanings change over time, and word senses evolve, emerge or die out in
the process. For ancient languages, where the corpora are often small and
sparse, modelling such changes accurately proves challenging, and quantifying
uncertainty in sense-change estimates consequently becomes important. GASC
(Genre-Aware Semantic Change) and DiSC (Diachronic Sense Change) are existing
generative models that have been used to analyse sense change for target words
from an ancient Greek text corpus, using unsupervised learning without the help
of any pre-training. These models represent the senses of a given target word
such as "kosmos" (meaning decoration, order or world) as distributions over
context words, and sense prevalence as a distribution over senses. The models
are fitted using Markov Chain Monte Carlo (MCMC) methods to measure temporal
changes in these representations. This paper introduces EDiSC, an Embedded DiSC
model, which combines word embeddings with DiSC to provide superior model
performance. It is shown empirically that EDiSC offers improved predictive
accuracy, ground-truth recovery and uncertainty quantification, as well as
better sampling efficiency and scalability properties with MCMC methods. The
challenges of fitting these models are also discussed. | cs.CL | An Embedded Diachronic Sense Change Model with a Case Study from Ancient Greek |
2023-11-01T13:49:31Z | http://arxiv.org/pdf/2311.00522v1 | http://arxiv.org/abs/2311.00522v1 | Jonas F. Lotz, Elizabeth Salesky, Phillip Rust, Desmond Elliott | 2023-11-01T13:49:31Z | cs.CL | Pixel-based language models process text rendered as images, which allows
them to handle any script, making them a promising approach to open vocabulary
language modelling. However, recent approaches use text renderers that produce
a large set of almost-equivalent input patches, which may prove sub-optimal for
downstream tasks, due to redundancy in the input representations. In this
paper, we investigate four approaches to rendering text in the PIXEL model
(Rust et al., 2023), and find that simple character bigram rendering brings
improved performance on sentence-level tasks without compromising performance
on token-level or multilingual tasks. This new rendering strategy also makes it
possible to train a more compact model with only 22M parameters that performs
on par with the original 86M parameter model. Our analyses show that character
bigram rendering leads to a consistently better model but with an anisotropic
patch embedding space, driven by a patch frequency bias, highlighting the
connections between image patch- and tokenization-based language models. | cs.CL | Text Rendering Strategies for Pixel Language Models |
2023-11-01T13:36:20Z | http://arxiv.org/pdf/2311.00513v1 | http://arxiv.org/abs/2311.00513v1 | Atsushi Shirafuji, Taku Matsumoto, Md Faizul Ibne Amin, Yutaka Watanobe | 2023-11-01T13:36:20Z | cs.SE, cs.CL, cs.PL | Finding and fixing errors is a time-consuming task not only for novice
programmers but also for expert programmers. Prior work has identified frequent
error patterns among various levels of programmers. However, the differences in
the tendencies between novices and experts have yet to be revealed. From the
knowledge of the frequent errors in each level of programmers, instructors will
be able to provide helpful advice for each level of learners. In this paper, we
propose a rule-based error classification tool to classify errors in code pairs
consisting of wrong and correct programs. We classify errors for 95,631 code
pairs and identify 3.47 errors on average, which are submitted by various
levels of programmers on an online judge system. The classified errors are used
to analyze the differences in frequent errors between novice and expert
programmers. The analyzed results show that, as for the same introductory
problems, errors made by novices are due to the lack of knowledge in
programming, and the mistakes are considered an essential part of the learning
process. On the other hand, errors made by experts are due to misunderstandings
caused by the carelessness of reading problems or the challenges of solving
problems differently than usual. The proposed tool can be used to create
error-labeled datasets and for further code-related educational research. | cs.SE | Rule-Based Error Classification for Analyzing Differences in Frequent Errors |
2023-11-01T13:14:23Z | http://arxiv.org/pdf/2311.00508v1 | http://arxiv.org/abs/2311.00508v1 | Yichen Huang, Timothy Baldwin | 2023-11-01T13:14:23Z | cs.CL | We investigate MT evaluation metric performance on adversarially-synthesized
texts, to shed light on metric robustness. We experiment with word- and
character-level attacks on three popular machine translation metrics:
BERTScore, BLEURT, and COMET. Our human experiments validate that automatic
metrics tend to overpenalize adversarially-degraded translations. We also
identify inconsistencies in BERTScore ratings, where it judges the original
sentence and the adversarially-degraded one as similar, while judging the
degraded translation as notably worse than the original with respect to the
reference. We identify patterns of brittleness that motivate more robust metric
development. | cs.CL | Robustness Tests for Automatic Machine Translation Metrics with Adversarial Attacks |
2023-11-01T13:08:50Z | http://arxiv.org/pdf/2311.00502v2 | http://arxiv.org/abs/2311.00502v2 | Haihao Shen, Hanwen Chang, Bo Dong, Yu Luo, Hengyu Meng | 2023-12-07T12:16:42Z | cs.LG, cs.AI, cs.CL | Large language models (LLMs) have demonstrated remarkable performance and
tremendous potential across a wide range of tasks. However, deploying these
models has been challenging due to the astronomical amount of model parameters,
which requires a demand for large memory capacity and high memory bandwidth. In
this paper, we propose an effective approach that can make the deployment of
LLMs more efficiently. We support an automatic INT4 weight-only quantization
flow and design a special LLM runtime with highly-optimized kernels to
accelerate the LLM inference on CPUs. We demonstrate the general applicability
of our approach on popular LLMs including Llama2, Llama, GPT-NeoX, and showcase
the extreme inference efficiency on CPUs. The code is publicly available at:
https://github.com/intel/intel-extension-for-transformers. | cs.LG | Efficient LLM Inference on CPUs |
2023-11-01T12:42:14Z | http://arxiv.org/pdf/2311.00488v1 | http://arxiv.org/abs/2311.00488v1 | Hugo Fry, Seamus Fallows, Ian Fan, Jamie Wright, Nandi Schoots | 2023-11-01T12:42:14Z | cs.LG, cs.CL | We investigate the optimization target of Contrast-Consistent Search (CCS),
which aims to recover the internal representations of truth of a large language
model. We present a new loss function that we call the Midpoint-Displacement
(MD) loss function. We demonstrate that for a certain hyper-parameter value
this MD loss function leads to a prober with very similar weights to CCS. We
further show that this hyper-parameter is not optimal and that with a better
hyper-parameter the MD loss function attains a higher test accuracy than CCS. | cs.LG | Comparing Optimization Targets for Contrast-Consistent Search |
2023-11-01T12:21:53Z | http://arxiv.org/pdf/2311.00475v1 | http://arxiv.org/abs/2311.00475v1 | Gilles Nawezi, Lucie Flek, Charles Welch | 2023-11-01T12:21:53Z | cs.CL | Recent language models have been improved by the addition of external memory.
Nearest neighbor language models retrieve similar contexts to assist in word
prediction. The addition of locality levels allows a model to learn how to
weight neighbors based on their relative location to the current text in source
documents, and have been shown to further improve model performance. Nearest
neighbor models have been explored for controllable generation but have not
examined the use of locality levels. We present a novel approach for this
purpose and evaluate it using automatic and human evaluation on politeness,
formality, supportiveness, and toxicity textual data. We find that our model is
successfully able to control style and provides a better fluency-style
trade-off than previous work. | cs.CL | Style Locality for Controllable Generation with kNN Language Models |
2023-11-01T11:38:19Z | http://arxiv.org/pdf/2311.00451v1 | http://arxiv.org/abs/2311.00451v1 | Yingxue Fu | 2023-11-01T11:38:19Z | cs.CL | Existing discourse formalisms use different taxonomies of discourse
relations, which require expert knowledge to understand, posing a challenge for
annotation and automatic classification. We show that discourse relations can
be effectively captured by some simple cognitively inspired dimensions proposed
by Sanders et al.(2018). Our experiments on cross-framework discourse relation
classification (PDTB & RST) demonstrate that it is possible to transfer
knowledge of discourse relations for one framework to another framework by
means of these dimensions, in spite of differences in discourse segmentation of
the two frameworks. This manifests the effectiveness of these dimensions in
characterizing discourse relations across frameworks. Ablation studies reveal
that different dimensions influence different types of discourse relations. The
patterns can be explained by the role of dimensions in characterizing and
distinguishing different relations. We also report our experimental results on
automatic prediction of these dimensions. | cs.CL | Discourse Relations Classification and Cross-Framework Discourse Relation Classification Through the Lens of Cognitive Dimensions: An Empirical Investigation |
2023-11-01T11:13:06Z | http://arxiv.org/pdf/2311.00445v2 | http://arxiv.org/abs/2311.00445v2 | Tiwalayo Eisape, MH Tessler, Ishita Dasgupta, Fei Sha, Sjoerd van Steenkiste, Tal Linzen | 2024-04-11T16:49:57Z | cs.CL, cs.AI, cs.LG | A central component of rational behavior is logical inference: the process of
determining which conclusions follow from a set of premises. Psychologists have
documented several ways in which humans' inferences deviate from the rules of
logic. Do language models, which are trained on text generated by humans,
replicate such human biases, or are they able to overcome them? Focusing on the
case of syllogisms -- inferences from two simple premises -- we show that,
within the PaLM2 family of transformer language models, larger models are more
logical than smaller ones, and also more logical than humans. At the same time,
even the largest models make systematic errors, some of which mirror human
reasoning biases: they show sensitivity to the (irrelevant) ordering of the
variables in the syllogism, and draw confident but incorrect inferences from
particular syllogisms (syllogistic fallacies). Overall, we find that language
models often mimic the human biases included in their training data, but are
able to overcome them in some cases. | cs.CL | A Systematic Comparison of Syllogistic Reasoning in Humans and Language Models |
2023-11-01T10:45:07Z | http://arxiv.org/pdf/2311.00430v1 | http://arxiv.org/abs/2311.00430v1 | Sanchit Gandhi, Patrick von Platen, Alexander M. Rush | 2023-11-01T10:45:07Z | cs.CL, cs.SD, eess.AS | As the size of pre-trained speech recognition models increases, running these
large models in low-latency or resource-constrained environments becomes
challenging. In this work, we leverage pseudo-labelling to assemble a
large-scale open-source dataset which we use to distill the Whisper model into
a smaller variant, called Distil-Whisper. Using a simple word error rate (WER)
heuristic, we select only the highest quality pseudo-labels for training. The
distilled model is 5.8 times faster with 51% fewer parameters, while performing
to within 1% WER on out-of-distribution test data in a zero-shot transfer
setting. Distil-Whisper maintains the robustness of the Whisper model to
difficult acoustic conditions, while being less prone to hallucination errors
on long-form audio. Distil-Whisper is designed to be paired with Whisper for
speculative decoding, yielding a 2 times speed-up while mathematically ensuring
the same outputs as the original model. To facilitate further research in this
domain, we make our training code, inference code and models publicly
accessible. | cs.CL | Distil-Whisper: Robust Knowledge Distillation via Large-Scale Pseudo Labelling |
2023-11-01T10:18:23Z | http://arxiv.org/pdf/2311.00416v1 | http://arxiv.org/abs/2311.00416v1 | Cong Guan, Lichao Zhang, Chunpeng Fan, Yichen Li, Feng Chen, Lihe Li, Yunjia Tian, Lei Yuan, Yang Yu | 2023-11-01T10:18:23Z | cs.LG, cs.CL | Developing intelligent agents capable of seamless coordination with humans is
a critical step towards achieving artificial general intelligence. Existing
methods for human-AI coordination typically train an agent to coordinate with a
diverse set of policies or with human models fitted from real human data.
However, the massively diverse styles of human behavior present obstacles for
AI systems with constrained capacity, while high quality human data may not be
readily available in real-world scenarios. In this study, we observe that prior
to coordination, humans engage in communication to establish conventions that
specify individual roles and actions, making their coordination proceed in an
orderly manner. Building upon this observation, we propose employing the large
language model (LLM) to develop an action plan (or equivalently, a convention)
that effectively guides both human and AI. By inputting task requirements,
human preferences, the number of agents, and other pertinent information into
the LLM, it can generate a comprehensive convention that facilitates a clear
understanding of tasks and responsibilities for all parties involved.
Furthermore, we demonstrate that decomposing the convention formulation problem
into sub-problems with multiple new sessions being sequentially employed and
human feedback, will yield a more efficient coordination convention.
Experimental evaluations conducted in the Overcooked-AI environment, utilizing
a human proxy model, highlight the superior performance of our proposed method
compared to existing learning-based approaches. When coordinating with real
humans, our method achieves better alignment with human preferences and an
average performance improvement of 15% compared to the state-of-the-art. | cs.LG | Efficient Human-AI Coordination via Preparatory Language-based Convention |
2023-11-01T10:00:15Z | http://arxiv.org/pdf/2311.00408v1 | http://arxiv.org/abs/2311.00408v1 | Yongxin Huang, Kexin Wang, Sourav Dutta, Raj Nath Patel, Goran Glavaš, Iryna Gurevych | 2023-11-01T10:00:15Z | cs.CL | Recent work has found that few-shot sentence classification based on
pre-trained Sentence Encoders (SEs) is efficient, robust, and effective. In
this work, we investigate strategies for domain-specialization in the context
of few-shot sentence classification with SEs. We first establish that
unsupervised Domain-Adaptive Pre-Training (DAPT) of a base Pre-trained Language
Model (PLM) (i.e., not an SE) substantially improves the accuracy of few-shot
sentence classification by up to 8.4 points. However, applying DAPT on SEs, on
the one hand, disrupts the effects of their (general-domain) Sentence Embedding
Pre-Training (SEPT). On the other hand, applying general-domain SEPT on top of
a domain-adapted base PLM (i.e., after DAPT) is effective but inefficient,
since the computationally expensive SEPT needs to be executed on top of a
DAPT-ed PLM of each domain. As a solution, we propose AdaSent, which decouples
SEPT from DAPT by training a SEPT adapter on the base PLM. The adapter can be
inserted into DAPT-ed PLMs from any domain. We demonstrate AdaSent's
effectiveness in extensive experiments on 17 different few-shot sentence
classification datasets. AdaSent matches or surpasses the performance of full
SEPT on DAPT-ed PLM, while substantially reducing the training costs. The code
for AdaSent is available. | cs.CL | AdaSent: Efficient Domain-Adapted Sentence Embeddings for Few-Shot Classification |
2023-11-01T09:50:55Z | http://arxiv.org/pdf/2311.00399v1 | http://arxiv.org/abs/2311.00399v1 | Qingqiu Li, Jilan Xu, Runtian Yuan, Mohan Chen, Yuejie Zhang, Rui Feng, Xiaobo Zhang, Shang Gao | 2023-11-01T09:50:55Z | cs.CV, cs.CL | Automatic generation of radiology reports holds crucial clinical value, as it
can alleviate substantial workload on radiologists and remind less experienced
ones of potential anomalies. Despite the remarkable performance of various
image captioning methods in the natural image field, generating accurate
reports for medical images still faces challenges, i.e., disparities in visual
and textual data, and lack of accurate domain knowledge. To address these
issues, we propose an enhanced knowledge injection framework, which utilizes
two branches to extract different types of knowledge. The Weighted Concept
Knowledge (WCK) branch is responsible for introducing clinical medical concepts
weighted by TF-IDF scores. The Multimodal Retrieval Knowledge (MRK) branch
extracts triplets from similar reports, emphasizing crucial clinical
information related to entity positions and existence. By integrating this
finer-grained and well-structured knowledge with the current image, we are able
to leverage the multi-source knowledge gain to ultimately facilitate more
accurate report generation. Extensive experiments have been conducted on two
public benchmarks, demonstrating that our method achieves superior performance
over other state-of-the-art methods. Ablation studies further validate the
effectiveness of two extracted knowledge sources. | cs.CV | Enhanced Knowledge Injection for Radiology Report Generation |
2023-11-01T08:38:08Z | http://arxiv.org/pdf/2311.00367v1 | http://arxiv.org/abs/2311.00367v1 | Chenxu Wang, Ping Jian, Mu Huang | 2023-11-01T08:38:08Z | cs.CL, cs.AI | Implicit Discourse Relation Recognition (IDRR), which infers discourse
relations without the help of explicit connectives, is still a crucial and
challenging task for discourse parsing. Recent works tend to exploit the
hierarchical structure information from the annotated senses, which demonstrate
enhanced discourse relation representations can be obtained by integrating
sense hierarchy. Nevertheless, the performance and robustness for IDRR are
significantly constrained by the availability of annotated data. Fortunately,
there is a wealth of unannotated utterances with explicit connectives, that can
be utilized to acquire enriched discourse relation features. In light of such
motivation, we propose a Prompt-based Logical Semantics Enhancement (PLSE)
method for IDRR. Essentially, our method seamlessly injects knowledge relevant
to discourse relation into pre-trained language models through prompt-based
connective prediction. Furthermore, considering the prompt-based connective
prediction exhibits local dependencies due to the deficiency of masked language
model (MLM) in capturing global semantics, we design a novel self-supervised
learning objective based on mutual information maximization to derive enhanced
representations of logical semantics for IDRR. Experimental results on PDTB 2.0
and CoNLL16 datasets demonstrate that our method achieves outstanding and
consistent performance against the current state-of-the-art models. | cs.CL | Prompt-based Logical Semantics Enhancement for Implicit Discourse Relation Recognition |
2023-11-01T07:42:34Z | http://arxiv.org/pdf/2311.04910v1 | http://arxiv.org/abs/2311.04910v1 | Oleksandr Palagin, Mykola Petrenko, Sergii Kryvyi, Mykola Boyko, Kyrylo Malakhov | 2023-11-01T07:42:34Z | cs.DL, cs.AI, cs.CL | The monograph discusses certain aspects of modern real-world problems facing
humanity, which are much more challenging than scientific ones. Modern science
is unable to solve them in a fundamental way. Vernadsky's noosphere thesis, in
fact, appeals to the scientific worldview that needs to be built in a way that
overcomes the interdisciplinary barriers and increases the effectiveness of
interdisciplinary interaction and modern science overall. We are talking about
the general transdisciplinary knowledge. In world practice, there is still no
systematic methodology and a specific form of generally accepted valid
scientific theory that would provide transdisciplinary knowledge. Non-linear
interdisciplinary interaction is the standard of evolution of modern science.
At the same time, a new transdisciplinary theory (domain of scientific
research) is being de facto created and the process is repeated many times:
from an individual or group of disciplines, through interdisciplinary
interaction, in a direction that brings us closer to creating a holistic
general scientific worldview. | cs.DL | Ontology-Driven Processing of Transdisciplinary Domain Knowledge |
2023-11-01T07:41:23Z | http://arxiv.org/pdf/2311.00732v1 | http://arxiv.org/abs/2311.00732v1 | Anna Glazkova | 2023-11-01T07:41:23Z | cs.CL, cs.AI, cs.LG, cs.SI, 68T50, I.2.7; I.7.m; H.3.3; J.3 | The paper describes a system developed for Task 1 at SMM4H 2023. The goal of
the task is to automatically distinguish tweets that self-report a COVID-19
diagnosis (for example, a positive test, clinical diagnosis, or
hospitalization) from those that do not. We investigate the use of different
techniques for preprocessing tweets using four transformer-based models. The
ensemble of fine-tuned language models obtained an F1-score of 84.5%, which is
4.1% higher than the average value. | cs.CL | tmn at #SMM4H 2023: Comparing Text Preprocessing Techniques for Detecting Tweets Self-reporting a COVID-19 Diagnosis |
2023-11-01T06:09:54Z | http://arxiv.org/pdf/2311.00321v2 | http://arxiv.org/abs/2311.00321v2 | Yongjin Yang, Joonkee Kim, Yujin Kim, Namgyu Ho, James Thorne, Se-young Yun | 2023-11-22T09:08:03Z | cs.CL | With the proliferation of social media, accurate detection of hate speech has
become critical to ensure safety online. To combat nuanced forms of hate
speech, it is important to identify and thoroughly explain hate speech to help
users understand its harmful effects. Recent benchmarks have attempted to
tackle this issue by training generative models on free-text annotations of
implications in hateful text. However, we find significant reasoning gaps in
the existing annotations schemes, which may hinder the supervision of detection
models. In this paper, we introduce a hate speech detection framework, HARE,
which harnesses the reasoning capabilities of large language models (LLMs) to
fill these gaps in explanations of hate speech, thus enabling effective
supervision of detection models. Experiments on SBIC and Implicit Hate
benchmarks show that our method, using model-generated data, consistently
outperforms baselines, using existing free-text human annotations. Analysis
demonstrates that our method enhances the explanation quality of trained models
and improves generalization to unseen datasets. Our code is available at
https://github.com/joonkeekim/hare-hate-speech.git. | cs.CL | HARE: Explainable Hate Speech Detection with Step-by-Step Reasoning |
2023-11-01T06:01:22Z | http://arxiv.org/pdf/2311.00317v1 | http://arxiv.org/abs/2311.00317v1 | Yiqing Xie, Atharva Naik, Daniel Fried, Carolyn Rose | 2023-11-01T06:01:22Z | cs.CL, cs.LG, cs.SE | One major challenge of translating code between programming languages is that
parallel training data is often limited. To overcome this challenge, we present
two data augmentation techniques, one that builds comparable corpora (i.e.,
code pairs with similar functionality), and another that augments existing
parallel data with multiple reference translations. Specifically, we build and
analyze multiple types of comparable corpora, including programs generated from
natural language documentation using a code generation model. Furthermore, to
reduce overfitting to a single reference translation, we automatically generate
additional translation references for available parallel data and filter the
translations by unit tests, which increases variation in target translations.
Experiments show that our data augmentation techniques significantly improve
CodeT5 for translation between Java, Python, and C++ by an average of 7.5%
Computational Accuracy (CA@1), which verifies the correctness of translations
by execution. The code is available at https://github.com/Veronicium/CMTrans. | cs.CL | Data Augmentation for Code Translation with Comparable Corpora and Multiple References |
2023-11-01T05:48:05Z | http://arxiv.org/pdf/2311.00310v1 | http://arxiv.org/abs/2311.00310v1 | Takashi Wada, Timothy Baldwin, Jey Han Lau | 2023-11-01T05:48:05Z | cs.CL, cs.AI | We propose a new unsupervised lexical simplification method that uses only
monolingual data and pre-trained language models. Given a target word and its
context, our method generates substitutes based on the target context and also
additional contexts sampled from monolingual data. We conduct experiments in
English, Portuguese, and Spanish on the TSAR-2022 shared task, and show that
our model substantially outperforms other unsupervised systems across all
languages. We also establish a new state-of-the-art by ensembling our model
with GPT-3.5. Lastly, we evaluate our model on the SWORDS lexical substitution
data set, achieving a state-of-the-art result. | cs.CL | Unsupervised Lexical Simplification with Context Augmentation |
2023-11-01T05:31:46Z | http://arxiv.org/pdf/2311.00306v1 | http://arxiv.org/abs/2311.00306v1 | Xiangjue Dong, Yibo Wang, Philip S. Yu, James Caverlee | 2023-11-01T05:31:46Z | cs.CL | Large Language Models (LLMs) can generate biased and toxic responses. Yet
most prior work on LLM gender bias evaluation requires predefined
gender-related phrases or gender stereotypes, which are challenging to be
comprehensively collected and are limited to explicit bias evaluation. In
addition, we believe that instances devoid of gender-related language or
explicit stereotypes in inputs can still induce gender bias in LLMs. Thus, in
this work, we propose a conditional text generation mechanism without the need
for predefined gender phrases and stereotypes. This approach employs three
types of inputs generated through three distinct strategies to probe LLMs,
aiming to show evidence of explicit and implicit gender biases in LLMs. We also
utilize explicit and implicit evaluation metrics to evaluate gender bias in
LLMs under different strategies. Our experiments demonstrate that an increased
model size does not consistently lead to enhanced fairness and all tested LLMs
exhibit explicit and/or implicit gender bias, even when explicit gender
stereotypes are absent in the inputs. | cs.CL | Probing Explicit and Implicit Gender Bias through LLM Conditional Text Generation |
2023-11-01T05:05:49Z | http://arxiv.org/pdf/2311.00301v1 | http://arxiv.org/abs/2311.00301v1 | Wang Weiying, Nakajima Akinori | 2023-11-01T05:05:49Z | cs.SD, cs.CL, eess.AS | One precondition of effective oral communication is that words should be
pronounced clearly, especially for non-native speakers. Word stress is the key
to clear and correct English, and misplacement of syllable stress may lead to
misunderstandings. Thus, knowing the stress level is important for English
speakers and learners. This paper presents a self-attention model to identify
the stress level for each syllable of spoken English. Various prosodic and
categorical features, including the pitch level, intensity, duration and type
of the syllable and its nuclei (the vowel of the syllable), are explored. These
features are input to the self-attention model, and syllable-level stresses are
predicted. The simplest model yields an accuracy of over 88% and 93% on
different datasets, while more advanced models provide higher accuracy. Our
study suggests that the self-attention model can be promising in stress-level
detection. These models could be applied to various scenarios, such as online
meetings and English learning. | cs.SD | Detecting Syllable-Level Pronunciation Stress with A Self-Attention Model |
2023-11-01T05:04:55Z | http://arxiv.org/pdf/2311.00300v1 | http://arxiv.org/abs/2311.00300v1 | Runze Fang, Yawen Li, Yingxia Shao, Zeli Guan, Zhe Xue | 2023-11-01T05:04:55Z | cs.CL | The entity alignment of science and technology patents aims to link the
equivalent entities in the knowledge graph of different science and technology
patent data sources. Most entity alignment methods only use graph neural
network to obtain the embedding of graph structure or use attribute text
description to obtain semantic representation, ignoring the process of
multi-information fusion in science and technology patents. In order to make
use of the graphic structure and auxiliary information such as the name,
description and attribute of the patent entity, this paper proposes an entity
alignment method based on the graph convolution network for science and
technology patent information fusion. Through the graph convolution network and
BERT model, the structure information and entity attribute information of the
science and technology patent knowledge graph are embedded and represented to
achieve multi-information fusion, thus improving the performance of entity
alignment. Experiments on three benchmark data sets show that the proposed
method Hit@K The evaluation indicators are better than the existing methods. | cs.CL | Entity Alignment Method of Science and Technology Patent based on Graph Convolution Network and Information Fusion |
2023-11-01T05:00:44Z | http://arxiv.org/pdf/2311.00296v1 | http://arxiv.org/abs/2311.00296v1 | Hongrui Gao, Yawen Li, Meiyu Liang, Zeli Guan, Zhe Xue | 2023-11-01T05:00:44Z | cs.CL | Because most of the scientific literature data is unmarked, it makes semantic
representation learning based on unsupervised graph become crucial. At the same
time, in order to enrich the features of scientific literature, a learning
method of semantic representation of scientific literature based on adaptive
features and graph neural network is proposed. By introducing the adaptive
feature method, the features of scientific literature are considered globally
and locally. The graph attention mechanism is used to sum the features of
scientific literature with citation relationship, and give each scientific
literature different feature weights, so as to better express the correlation
between the features of different scientific literature. In addition, an
unsupervised graph neural network semantic representation learning method is
proposed. By comparing the mutual information between the positive and negative
local semantic representation of scientific literature and the global graph
semantic representation in the potential space, the graph neural network can
capture the local and global information, thus improving the learning ability
of the semantic representation of scientific literature. The experimental
results show that the proposed learning method of semantic representation of
scientific literature based on adaptive feature and graph neural network is
competitive on the basis of scientific literature classification, and has
achieved good results. | cs.CL | Semantic Representation Learning of Scientific Literature based on Adaptive Feature and Graph Neural Network |
2023-11-01T04:50:38Z | http://arxiv.org/pdf/2311.00292v1 | http://arxiv.org/abs/2311.00292v1 | Xiaoyue Wang, Xin Liu, Lijie Wang, Yaoxiang Wang, Jinsong Su, Hua Wu | 2023-11-01T04:50:38Z | cs.CL | As commonly-used methods for debiasing natural language understanding (NLU)
models, dataset refinement approaches heavily rely on manual data analysis, and
thus maybe unable to cover all the potential biased features. In this paper, we
propose IBADR, an Iterative Bias-Aware Dataset Refinement framework, which
debiases NLU models without predefining biased features. We maintain an
iteratively expanded sample pool. Specifically, at each iteration, we first
train a shallow model to quantify the bias degree of samples in the pool. Then,
we pair each sample with a bias indicator representing its bias degree, and use
these extended samples to train a sample generator. In this way, this generator
can effectively learn the correspondence relationship between bias indicators
and samples. Furthermore, we employ the generator to produce pseudo samples
with fewer biased features by feeding specific bias indicators. Finally, we
incorporate the generated pseudo samples into the pool. Experimental results
and in-depth analyses on two NLU tasks show that IBADR not only significantly
outperforms existing dataset refinement approaches, achieving SOTA, but also is
compatible with model-centric methods. | cs.CL | IBADR: an Iterative Bias-Aware Dataset Refinement Framework for Debiasing NLU models |
2023-11-01T04:40:05Z | http://arxiv.org/pdf/2311.00288v1 | http://arxiv.org/abs/2311.00288v1 | Po-Nien Kung, Fan Yin, Di Wu, Kai-Wei Chang, Nanyun Peng | 2023-11-01T04:40:05Z | cs.CL, cs.AI | Instruction tuning (IT) achieves impressive zero-shot generalization results
by training large language models (LLMs) on a massive amount of diverse tasks
with instructions. However, how to select new tasks to improve the performance
and generalizability of IT models remains an open question. Training on all
existing tasks is impractical due to prohibiting computation requirements, and
randomly selecting tasks can lead to suboptimal performance. In this work, we
propose active instruction tuning based on prompt uncertainty, a novel
framework to identify informative tasks, and then actively tune the models on
the selected tasks. We represent the informativeness of new tasks with the
disagreement of the current model outputs over perturbed prompts. Our
experiments on NIV2 and Self-Instruct datasets demonstrate that our method
consistently outperforms other baseline strategies for task selection,
achieving better out-of-distribution generalization with fewer training tasks.
Additionally, we introduce a task map that categorizes and diagnoses tasks
based on prompt uncertainty and prediction probability. We discover that
training on ambiguous (prompt-uncertain) tasks improves generalization while
training on difficult (prompt-certain and low-probability) tasks offers no
benefit, underscoring the importance of task selection for instruction tuning. | cs.CL | Active Instruction Tuning: Improving Cross-Task Generalization by Training on Prompt Sensitive Tasks |
2023-11-01T04:37:28Z | http://arxiv.org/pdf/2311.00287v1 | http://arxiv.org/abs/2311.00287v1 | Ran Xu, Hejie Cui, Yue Yu, Xuan Kan, Wenqi Shi, Yuchen Zhuang, Wei Jin, Joyce Ho, Carl Yang | 2023-11-01T04:37:28Z | cs.CL, cs.AI, cs.LG, q-bio.QM | Clinical natural language processing requires methods that can address
domain-specific challenges, such as complex medical terminology and clinical
contexts. Recently, large language models (LLMs) have shown promise in this
domain. Yet, their direct deployment can lead to privacy issues and are
constrained by resources. To address this challenge, we delve into synthetic
clinical text generation using LLMs for clinical NLP tasks. We propose an
innovative, resource-efficient approach, ClinGen, which infuses knowledge into
the process. Our model involves clinical knowledge extraction and
context-informed LLM prompting. Both clinical topics and writing styles are
drawn from external domain-specific knowledge graphs and LLMs to guide data
generation. Our extensive empirical study across 7 clinical NLP tasks and 16
datasets reveals that ClinGen consistently enhances performance across various
tasks, effectively aligning the distribution of real datasets and significantly
enriching the diversity of generated training instances. We will publish our
code and all the generated data in \url{https://github.com/ritaranx/ClinGen}. | cs.CL | Knowledge-Infused Prompting: Assessing and Advancing Clinical Text Data Generation with Large Language Models |
2023-11-01T04:36:45Z | http://arxiv.org/pdf/2311.00286v3 | http://arxiv.org/abs/2311.00286v3 | Mi Zhang, Xudong Pan, Min Yang | 2023-12-10T13:58:24Z | cs.CL, cs.AI, cs.CR, cs.LG | In this paper, we present JADE, a targeted linguistic fuzzing platform which
strengthens the linguistic complexity of seed questions to simultaneously and
consistently break a wide range of widely-used LLMs categorized in three
groups: eight open-sourced Chinese, six commercial Chinese and four commercial
English LLMs. JADE generates three safety benchmarks for the three groups of
LLMs, which contain unsafe questions that are highly threatening: the questions
simultaneously trigger harmful generation of multiple LLMs, with an average
unsafe generation ratio of $70\%$ (please see the table below), while are still
natural questions, fluent and preserving the core unsafe semantics. We release
the benchmark demos generated for commercial English LLMs and open-sourced
English LLMs in the following link: https://github.com/whitzard-ai/jade-db. For
readers who are interested in evaluating on more questions generated by JADE,
please contact us.
JADE is based on Noam Chomsky's seminal theory of transformational-generative
grammar. Given a seed question with unsafe intention, JADE invokes a sequence
of generative and transformational rules to increment the complexity of the
syntactic structure of the original question, until the safety guardrail is
broken. Our key insight is: Due to the complexity of human language, most of
the current best LLMs can hardly recognize the invariant evil from the infinite
number of different syntactic structures which form an unbound example space
that can never be fully covered. Technically, the generative/transformative
rules are constructed by native speakers of the languages, and, once developed,
can be used to automatically grow and transform the parse tree of a given
question, until the guardrail is broken. For more evaluation results and demo,
please check our website: https://whitzard-ai.github.io/jade.html. | cs.CL | JADE: A Linguistics-based Safety Evaluation Platform for Large Language Models |
2023-11-01T03:49:52Z | http://arxiv.org/pdf/2311.00273v1 | http://arxiv.org/abs/2311.00273v1 | Yirong Chen, Xiaofen Xing, Jingkai Lin, Huimin Zheng, Zhenyu Wang, Qi Liu, Xiangmin Xu | 2023-11-01T03:49:52Z | cs.CL | Large language models (LLMs) have been widely applied in various fields due
to their excellent capability for memorizing knowledge and chain of thought
(CoT). When these language models are applied in the field of psychological
counseling, they often rush to provide universal advice. However, when users
seek psychological support, they need to gain empathy, trust, understanding and
comfort, rather than just reasonable advice. To this end, we constructed a
multi-turn empathetic conversation dataset of more than 2 million samples, in
which the input is the multi-turn conversation context, and the target is
empathetic responses that cover expressions such as questioning, comfort,
recognition, listening, trust, emotional support, etc. Experiments have shown
that the empathy ability of LLMs can be significantly enhanced when finetuning
by using multi-turn dialogue history and responses that are closer to the
expression of a psychological consultant. | cs.CL | SoulChat: Improving LLMs' Empathy, Listening, and Comfort Abilities through Fine-tuning with Multi-turn Empathy Conversations |
2023-11-01T03:32:46Z | http://arxiv.org/pdf/2311.00268v1 | http://arxiv.org/abs/2311.00268v1 | Luke Gessler, Nathan Schneider | 2023-11-01T03:32:46Z | cs.CL | A line of work on Transformer-based language models such as BERT has
attempted to use syntactic inductive bias to enhance the pretraining process,
on the theory that building syntactic structure into the training process
should reduce the amount of data needed for training. But such methods are
often tested for high-resource languages such as English. In this work, we
investigate whether these methods can compensate for data sparseness in
low-resource languages, hypothesizing that they ought to be more effective for
low-resource languages. We experiment with five low-resource languages: Uyghur,
Wolof, Maltese, Coptic, and Ancient Greek. We find that these syntactic
inductive bias methods produce uneven results in low-resource settings, and
provide surprisingly little benefit in most cases. | cs.CL | Syntactic Inductive Bias in Transformer Language Models: Especially Helpful for Low-Resource Languages? |
2023-11-01T03:20:16Z | http://arxiv.org/pdf/2311.00262v2 | http://arxiv.org/abs/2311.00262v2 | Yang Deng, Wenxuan Zhang, Wai Lam, See-Kiong Ng, Tat-Seng Chua | 2024-03-11T08:30:31Z | cs.CL, cs.AI | Proactive dialogues serve as a practical yet challenging dialogue problem in
the era of large language models (LLMs), where the dialogue policy planning is
the key to improving the proactivity of LLMs. Most existing studies enable the
dialogue policy planning of LLMs using various prompting schemes or iteratively
enhance this capability in handling the given case with verbal AI feedback.
However, these approaches are either bounded by the policy planning capability
of the frozen LLMs or hard to be transferred to new cases. In this work, we
introduce a new dialogue policy planning paradigm to strategize LLMs for
proactive dialogue problems with a tunable language model plug-in as a
plug-and-play dialogue policy planner, named PPDPP. Specifically, we develop a
novel training framework to facilitate supervised fine-tuning over available
human-annotated data as well as reinforcement learning from goal-oriented AI
feedback with dynamic interaction data collected by the LLM-based self-play
simulation. In this manner, the LLM-powered dialogue agent can not only be
generalized to different cases after the training, but also be applicable to
different applications by just substituting the learned plug-in. In addition,
we propose to evaluate the policy planning capability of dialogue systems under
the interactive setting. Experimental results demonstrate that PPDPP
consistently and substantially outperforms existing approaches on three
different proactive dialogue applications, including negotiation, emotional
support, and tutoring dialogues. | cs.CL | Plug-and-Play Policy Planner for Large Language Model Powered Dialogue Agents |
2023-11-01T03:15:05Z | http://arxiv.org/pdf/2311.00258v1 | http://arxiv.org/abs/2311.00258v1 | Hongyi Zheng, Abulhair Saparov | 2023-11-01T03:15:05Z | cs.CL, cs.LG | Recent advances in prompt engineering enable large language models (LLMs) to
solve multi-hop logical reasoning problems with impressive accuracy. However,
there is little existing work investigating the robustness of LLMs with
few-shot prompting techniques. Therefore, we introduce a systematic approach to
test the robustness of LLMs in multi-hop reasoning tasks via domain-agnostic
perturbations. We include perturbations at multiple levels of abstractions
(e.g. lexical perturbations such as typos, and semantic perturbations such as
the inclusion of intermediate reasoning steps in the questions) to conduct
behavioral analysis on the LLMs. Throughout our experiments, we find that
models are more sensitive to certain perturbations such as replacing words with
their synonyms. We also demonstrate that increasing the proportion of perturbed
exemplars in the prompts improves the robustness of few-shot prompting methods. | cs.CL | Noisy Exemplars Make Large Language Models More Robust: A Domain-Agnostic Behavioral Analysis |
2023-11-01T02:40:42Z | http://arxiv.org/pdf/2311.00237v2 | http://arxiv.org/abs/2311.00237v2 | Yuxiang Zhou, Jiazheng Li, Yanzheng Xiang, Hanqi Yan, Lin Gui, Yulan He | 2024-02-16T00:55:48Z | cs.CL | Understanding in-context learning (ICL) capability that enables large
language models (LLMs) to excel in proficiency through demonstration examples
is of utmost importance. This importance stems not only from the better
utilization of this capability across various tasks, but also from the
proactive identification and mitigation of potential risks, including concerns
regarding truthfulness, bias, and toxicity, that may arise alongside the
capability. In this paper, we present a thorough survey on the interpretation
and analysis of in-context learning. First, we provide a concise introduction
to the background and definition of in-context learning. Then, we give an
overview of advancements from two perspectives: 1) a theoretical perspective,
emphasizing studies on mechanistic interpretability and delving into the
mathematical foundations behind ICL; and 2) an empirical perspective,
concerning studies that empirically analyze factors associated with ICL. We
conclude by highlighting the challenges encountered and suggesting potential
avenues for future research. We believe that our work establishes the basis for
further exploration into the interpretation of in-context learning.
Additionally, we have created a repository containing the resources referenced
in our survey. | cs.CL | The Mystery of In-Context Learning: A Comprehensive Survey on Interpretation and Analysis |
2023-11-01T02:31:35Z | http://arxiv.org/pdf/2311.00233v2 | http://arxiv.org/abs/2311.00233v2 | Taehyeon Kim, Joonkee Kim, Gihun Lee, Se-Young Yun | 2024-02-17T09:00:29Z | cs.CL | While instruction-tuned language models have demonstrated impressive
zero-shot generalization, these models often struggle to generate accurate
responses when faced with instructions that fall outside their training set.
This paper presents Instructive Decoding (ID), a simple yet effective approach
that augments the efficacy of instruction-tuned models. Specifically, ID
adjusts the logits for next-token prediction in a contrastive manner, utilizing
predictions generated from a manipulated version of the original instruction,
referred to as a noisy instruction. This noisy instruction aims to elicit
responses that could diverge from the intended instruction yet remain
plausible. We conduct experiments across a spectrum of such noisy instructions,
ranging from those that insert semantic noise via random words to others like
'opposite' that elicit the deviated responses. Our approach achieves
considerable performance gains across various instruction-tuned models and
tasks without necessitating any additional parameter updates. Notably,
utilizing 'opposite' as the noisy instruction in ID, which exhibits the maximum
divergence from the original instruction, consistently produces the most
significant performance gains across multiple models and tasks. | cs.CL | Instructive Decoding: Instruction-Tuned Large Language Models are Self-Refiner from Noisy Instructions |
2023-11-01T01:57:48Z | http://arxiv.org/pdf/2311.00223v1 | http://arxiv.org/abs/2311.00223v1 | Jingjing Wang, Joshua Luo, Grace Yang, Allen Hong, Feng Luo | 2023-11-01T01:57:48Z | cs.CL, cs.MM | Large Language Models (LLMs), representing a significant achievement in
artificial intelligence (AI) research, have demonstrated their ability in a
multitude of tasks. This project aims to explore the capabilities of GPT-3.5, a
leading example of LLMs, in processing the sentiment analysis of Internet
memes. Memes, which include both verbal and visual aspects, act as a powerful
yet complex tool for expressing ideas and sentiments, demanding an
understanding of societal norms and cultural contexts. Notably, the detection
and moderation of hateful memes pose a significant challenge due to their
implicit offensive nature. This project investigates GPT's proficiency in such
subjective tasks, revealing its strengths and potential limitations. The tasks
include the classification of meme sentiment, determination of humor type, and
detection of implicit hate in memes. The performance evaluation, using datasets
from SemEval-2020 Task 8 and Facebook hateful memes, offers a comparative
understanding of GPT responses against human annotations. Despite GPT's
remarkable progress, our findings underscore the challenges faced by these
models in handling subjective tasks, which are rooted in their inherent
limitations including contextual understanding, interpretation of implicit
meanings, and data biases. This research contributes to the broader discourse
on the applicability of AI in handling complex, context-dependent tasks, and
offers valuable insights for future advancements. | cs.CL | Is GPT Powerful Enough to Analyze the Emotions of Memes? |
2023-11-01T00:38:26Z | http://arxiv.org/pdf/2311.00208v2 | http://arxiv.org/abs/2311.00208v2 | Lena Strobl, William Merrill, Gail Weiss, David Chiang, Dana Angluin | 2024-05-06T19:48:28Z | cs.LG, cs.CL, cs.FL, cs.LO | As transformers have gained prominence in natural language processing, some
researchers have investigated theoretically what problems they can and cannot
solve, by treating problems as formal languages. Exploring such questions can
help clarify the power of transformers relative to other models of computation,
their fundamental capabilities and limits, and the impact of architectural
choices. Work in this subarea has made considerable progress in recent years.
Here, we undertake a comprehensive survey of this work, documenting the diverse
assumptions that underlie different results and providing a unified framework
for harmonizing seemingly contradictory findings. | cs.LG | What Formal Languages Can Transformers Express? A Survey |
2023-11-01T00:18:00Z | http://arxiv.org/pdf/2311.00204v1 | http://arxiv.org/abs/2311.00204v1 | Zhen Guo, Yining Hua | 2023-11-01T00:18:00Z | cs.CL, cs.AI | Large language models exhibit promising general capabilities but often lack
specialized knowledge for domain-specific tasks. Developing domain experts from
a base model enables a range of applications without prohibitive training
costs. This work demonstrates a method using continuous training and
instruction fine-tuning to rapidly adapt Llama 2 base models to the Chinese
medical domain. We first conduct continuous training on 1B tokens from Chinese
medical references to teach relevant vocabulary and knowledge. The models are
then fine-tuned on 54K examples sourced from the Chinese National Medical
Licensing Examination. Experiments on Chinese medical data confirm the
effectiveness of this approach, producing a model comparable to GPT-3.5-turbo
while using way less computational resource. The resulting domain-specific
model could be useful for various Chinese medical applications. More broadly,
this provides a template for domain-specific training of large language models
in areas where pre-trained models lack the required expertise, such as law,
science, and engineering. | cs.CL | Continuous Training and Fine-tuning for Domain-Specific Language Models in Medical Question Answering |
2023-10-31T23:24:22Z | http://arxiv.org/pdf/2311.00189v1 | http://arxiv.org/abs/2311.00189v1 | Daniel Hajialigol, Hanwen Liu, Xuan Wang | 2023-10-31T23:24:22Z | cs.CL, cs.AI | Text classification aims to effectively categorize documents into pre-defined
categories. Traditional methods for text classification often rely on large
amounts of manually annotated training data, making the process time-consuming
and labor-intensive. To address this issue, recent studies have focused on
weakly-supervised and extremely weakly-supervised settings, which require
minimal or no human annotation, respectively. In previous methods of weakly
supervised text classification, pseudo-training data is generated by assigning
pseudo-labels to documents based on their alignment (e.g., keyword matching)
with specific classes. However, these methods ignore the importance of
incorporating the explanations of the generated pseudo-labels, or saliency of
individual words, as additional guidance during the text classification
training process. To address this limitation, we propose XAI-CLASS, a novel
explanation-enhanced extremely weakly-supervised text classification method
that incorporates word saliency prediction as an auxiliary task. XAI-CLASS
begins by employing a multi-round question-answering process to generate
pseudo-training data that promotes the mutual enhancement of class labels and
corresponding explanation word generation. This pseudo-training data is then
used to train a multi-task framework that simultaneously learns both text
classification and word saliency prediction. Extensive experiments on several
weakly-supervised text classification datasets show that XAI-CLASS outperforms
other weakly-supervised text classification methods significantly. Moreover,
experiments demonstrate that XAI-CLASS enhances both model performance and
explainability. | cs.CL | XAI-CLASS: Explanation-Enhanced Text Classification with Extremely Weak Supervision |
2023-10-31T23:14:54Z | http://arxiv.org/pdf/2311.01472v1 | http://arxiv.org/abs/2311.01472v1 | Jaeff Hong, Duong Dung, Danielle Hutchinson, Zubair Akhtar, Rosalie Chen, Rebecca Dawson, Aditya Joshi, Samsung Lim, C Raina MacIntyre, Deepti Gurdasani | 2023-10-31T23:14:54Z | cs.CL, cs.AI | Relation Extraction from News Articles (RENA) is a browser-based tool
designed to extract key entities and their semantic relationships in English
language news articles related to infectious diseases. Constructed using the
React framework, this system presents users with an elegant and user-friendly
interface. It enables users to input a news article and select from a choice of
two models to generate a comprehensive list of relations within the provided
text. As a result, RENA allows real-time parsing of news articles to extract
key information for epidemic surveillance, contributing to EPIWATCH, an
open-source intelligence-based epidemic warning system. | cs.CL | Relation Extraction from News Articles (RENA): A Tool for Epidemic Surveillance |
2023-10-31T22:35:58Z | http://arxiv.org/pdf/2311.00176v5 | http://arxiv.org/abs/2311.00176v5 | Mingjie Liu, Teodor-Dumitru Ene, Robert Kirby, Chris Cheng, Nathaniel Pinckney, Rongjian Liang, Jonah Alben, Himyanshu Anand, Sanmitra Banerjee, Ismet Bayraktaroglu, Bonita Bhaskaran, Bryan Catanzaro, Arjun Chaudhuri, Sharon Clay, Bill Dally, Laura Dang, Parikshit Deshpande, Siddhanth Dhodhi, Sameer Halepete, Eric Hill, Jiashang Hu, Sumit Jain, Ankit Jindal, Brucek Khailany, George Kokai, Kishor Kunal, Xiaowei Li, Charley Lind, Hao Liu, Stuart Oberman, Sujeet Omar, Ghasem Pasandi, Sreedhar Pratty, Jonathan Raiman, Ambar Sarkar, Zhengjiang Shao, Hanfei Sun, Pratik P Suthar, Varun Tej, Walker Turner, Kaizhe Xu, Haoxing Ren | 2024-04-04T20:18:57Z | cs.CL | ChipNeMo aims to explore the applications of large language models (LLMs) for
industrial chip design. Instead of directly deploying off-the-shelf commercial
or open-source LLMs, we instead adopt the following domain adaptation
techniques: domain-adaptive tokenization, domain-adaptive continued
pretraining, model alignment with domain-specific instructions, and
domain-adapted retrieval models. We evaluate these methods on three selected
LLM applications for chip design: an engineering assistant chatbot, EDA script
generation, and bug summarization and analysis. Our evaluations demonstrate
that domain-adaptive pretraining of language models, can lead to superior
performance in domain related downstream tasks compared to their base LLaMA2
counterparts, without degradations in generic capabilities. In particular, our
largest model, ChipNeMo-70B, outperforms the highly capable GPT-4 on two of our
use cases, namely engineering assistant chatbot and EDA scripts generation,
while exhibiting competitive performance on bug summarization and analysis.
These results underscore the potential of domain-specific customization for
enhancing the effectiveness of large language models in specialized
applications. | cs.CL | ChipNeMo: Domain-Adapted LLMs for Chip Design |
2023-10-31T22:22:10Z | http://arxiv.org/pdf/2311.00172v1 | http://arxiv.org/abs/2311.00172v1 | Jinhwa Kim, Ali Derakhshan, Ian G. Harris | 2023-10-31T22:22:10Z | cs.CL, cs.AI | Large Language Models' safety remains a critical concern due to their
vulnerability to adversarial attacks, which can prompt these systems to produce
harmful responses. In the heart of these systems lies a safety classifier, a
computational model trained to discern and mitigate potentially harmful,
offensive, or unethical outputs. However, contemporary safety classifiers,
despite their potential, often fail when exposed to inputs infused with
adversarial noise. In response, our study introduces the Adversarial Prompt
Shield (APS), a lightweight model that excels in detection accuracy and
demonstrates resilience against adversarial prompts. Additionally, we propose
novel strategies for autonomously generating adversarial training datasets,
named Bot Adversarial Noisy Dialogue (BAND) datasets. These datasets are
designed to fortify the safety classifier's robustness, and we investigate the
consequences of incorporating adversarial examples into the training process.
Through evaluations involving Large Language Models, we demonstrate that our
classifier has the potential to decrease the attack success rate resulting from
adversarial attacks by up to 60%. This advancement paves the way for the next
generation of more reliable and resilient conversational agents. | cs.CL | Robust Safety Classifier for Large Language Models: Adversarial Prompt Shield |
2023-10-31T21:33:46Z | http://arxiv.org/pdf/2311.00161v1 | http://arxiv.org/abs/2311.00161v1 | Jimin Mun, Emily Allaway, Akhila Yerukola, Laura Vianna, Sarah-Jane Leslie, Maarten Sap | 2023-10-31T21:33:46Z | cs.CL, cs.AI | Counterspeech, i.e., responses to counteract potential harms of hateful
speech, has become an increasingly popular solution to address online hate
speech without censorship. However, properly countering hateful language
requires countering and dispelling the underlying inaccurate stereotypes
implied by such language. In this work, we draw from psychology and philosophy
literature to craft six psychologically inspired strategies to challenge the
underlying stereotypical implications of hateful language. We first examine the
convincingness of each of these strategies through a user study, and then
compare their usages in both human- and machine-generated counterspeech
datasets. Our results show that human-written counterspeech uses countering
strategies that are more specific to the implied stereotype (e.g., counter
examples to the stereotype, external factors about the stereotype's origins),
whereas machine-generated counterspeech uses less specific strategies (e.g.,
generally denouncing the hatefulness of speech). Furthermore, machine-generated
counterspeech often employs strategies that humans deem less convincing
compared to human-produced counterspeech. Our findings point to the importance
of accounting for the underlying stereotypical implications of speech when
generating counterspeech and for better machine reasoning about
anti-stereotypical examples. | cs.CL | Beyond Denouncing Hate: Strategies for Countering Implied Biases and Stereotypes in Language |
2023-10-31T21:32:11Z | http://arxiv.org/pdf/2311.00159v1 | http://arxiv.org/abs/2311.00159v1 | Xinting Huang, Jiajing Wan, Ioannis Kritikos, Nora Hollenstein | 2023-10-31T21:32:11Z | cs.CL | Humans read texts at a varying pace, while machine learning models treat each
token in the same way in terms of a computational process. Therefore, we ask,
does it help to make models act more like humans? In this paper, we convert
this intuition into a set of novel models with fixation-guided parallel RNNs or
layers and conduct various experiments on language modeling and sentiment
analysis tasks to test their effectiveness, thus providing empirical validation
for this intuition. Our proposed models achieve good performance on the
language modeling task, considerably surpassing the baseline model. In
addition, we find that, interestingly, the fixation duration predicted by
neural networks bears some resemblance to humans' fixation. Without any
explicit guidance, the model makes similar choices to humans. We also
investigate the reasons for the differences between them, which explain why
"model fixations" are often more suitable than human fixations, when used to
guide language models. | cs.CL | Longer Fixations, More Computation: Gaze-Guided Recurrent Neural Networks |
2023-10-31T20:31:41Z | http://arxiv.org/pdf/2311.00143v1 | http://arxiv.org/abs/2311.00143v1 | Fatemeh Rajabi, Ali Mohades | 2023-10-31T20:31:41Z | cs.LG, cs.AI, cs.CL | In elections around the world, the candidates may turn their campaigns toward
negativity due to the prospect of failure and time pressure. In the digital
age, social media platforms such as Twitter are rich sources of political
discourse. Therefore, despite the large amount of data that is published on
Twitter, the automatic system for campaign negativity detection can play an
essential role in understanding the strategy of candidates and parties in their
campaigns. In this paper, we propose a hybrid model for detecting campaign
negativity consisting of a two-stage classifier that combines the strengths of
two machine learning models. Here, we have collected Persian tweets from 50
political users, including candidates and government officials. Then we
annotated 5,100 of them that were published during the year before the 2021
presidential election in Iran. In the proposed model, first, the required
datasets of two classifiers based on the cosine similarity of tweet embeddings
with axis embeddings (which are the average of embedding in positive and
negative classes of tweets) from the training set (85\%) are made, and then
these datasets are considered the training set of the two classifiers in the
hybrid model. Finally, our best model (RF-RF) was able to achieve 79\% for the
macro F1 score and 82\% for the weighted F1 score. By running the best model on
the rest of the tweets of 50 political users that were published one year
before the election and with the help of statistical models, we find that the
publication of a tweet by a candidate has nothing to do with the negativity of
that tweet, and the presence of the names of political persons and political
organizations in the tweet is directly related to its negativity. | cs.LG | Two-Stage Classifier for Campaign Negativity Detection using Axis Embeddings: A Case Study on Tweets of Political Users during 2021 Presidential Election in Iran |
2023-10-31T20:05:30Z | http://arxiv.org/pdf/2311.00128v2 | http://arxiv.org/abs/2311.00128v2 | Mattia Opper, J. Morrison, N. Siddharth | 2023-11-03T16:42:33Z | cs.CL | This work explores the degree to which grammar acquisition is driven by
language `simplicity' and the source modality (speech vs. text) of data. Using
BabyBERTa as a probe, we find that grammar acquisition is largely driven by
exposure to speech data, and in particular through exposure to two of the
BabyLM training corpora: AO-Childes and Open Subtitles. We arrive at this
finding by examining various ways of presenting input data to our model. First,
we assess the impact of various sequence-level complexity based curricula. We
then examine the impact of learning over `blocks' -- covering spans of text
that are balanced for the number of tokens in each of the source corpora
(rather than number of lines). Finally, we explore curricula that vary the
degree to which the model is exposed to different corpora. In all cases, we
find that over-exposure to AO-Childes and Open Subtitles significantly drives
performance. We verify these findings through a comparable control dataset in
which exposure to these corpora, and speech more generally, is limited by
design. Our findings indicate that it is not the proportion of tokens occupied
by high-utility data that aids acquisition, but rather the proportion of
training steps assigned to such data. We hope this encourages future research
into the use of more developmentally plausible linguistic data (which tends to
be more scarce) to augment general purpose pre-training regimes. | cs.CL | On the effect of curriculum learning with developmental data for grammar acquisition |
2023-10-31T19:45:15Z | http://arxiv.org/pdf/2311.00117v3 | http://arxiv.org/abs/2311.00117v3 | Pranav Gade, Simon Lermen, Charlie Rogers-Smith, Jeffrey Ladish | 2024-05-28T10:33:03Z | cs.CL | Llama 2-Chat is a collection of large language models that Meta developed and
released to the public. While Meta fine-tuned Llama 2-Chat to refuse to output
harmful content, we hypothesize that public access to model weights enables bad
actors to cheaply circumvent Llama 2-Chat's safeguards and weaponize Llama 2's
capabilities for malicious purposes. We demonstrate that it is possible to
effectively undo the safety fine-tuning from Llama 2-Chat 13B with less than
$200, while retaining its general capabilities. Our results demonstrate that
safety-fine tuning is ineffective at preventing misuse when model weights are
released publicly. Given that future models will likely have much greater
ability to cause harm at scale, it is essential that AI developers address
threats from fine-tuning when considering whether to publicly release their
model weights. | cs.CL | BadLlama: cheaply removing safety fine-tuning from Llama 2-Chat 13B |
2023-10-31T19:44:50Z | http://arxiv.org/pdf/2311.00116v1 | http://arxiv.org/abs/2311.00116v1 | Aarohi Srivastava, David Chiang | 2023-10-31T19:44:50Z | cs.CL | Real-world NLP applications often deal with nonstandard text (e.g.,
dialectal, informal, or misspelled text). However, language models like BERT
deteriorate in the face of dialect variation or noise. How do we push BERT's
modeling capabilities to encompass nonstandard text? Fine-tuning helps, but it
is designed for specializing a model to a task and does not seem to bring about
the deeper, more pervasive changes needed to adapt a model to nonstandard
language. In this paper, we introduce the novel idea of sandwiching BERT's
encoder stack between additional encoder layers trained to perform masked
language modeling on noisy text. We find that our approach, paired with recent
work on including character-level noise in fine-tuning data, can promote
zero-shot transfer to dialectal text, as well as reduce the distance in the
embedding space between words and their noisy counterparts. | cs.CL | BERTwich: Extending BERT's Capabilities to Model Dialectal and Noisy Text |
2023-10-31T18:19:28Z | http://arxiv.org/pdf/2311.14677v1 | http://arxiv.org/abs/2311.14677v1 | Tomo Lazovich | 2023-10-31T18:19:28Z | cs.CY, cs.CL, cs.LG | Echoing the history of search engines and social media content rankings, the
advent of large language models (LLMs) has led to a push for increased
personalization of model outputs to individual users. In the past, personalized
recommendations and ranking systems have been linked to the development of
filter bubbles (serving content that may confirm a user's existing biases) and
affective polarization (strong negative sentiment towards those with differing
views). In this work, we explore how prompting a leading large language model,
ChatGPT-3.5, with a user's political affiliation prior to asking factual
questions about public figures and organizations leads to differing results. We
observe that left-leaning users tend to receive more positive statements about
left-leaning political figures and media outlets, while right-leaning users see
more positive statements about right-leaning entities. This pattern holds
across presidential candidates, members of the U.S. Senate, and media
organizations with ratings from AllSides. When qualitatively evaluating some of
these outputs, there is evidence that particular facts are included or excluded
based on the user's political affiliation. These results illustrate that
personalizing LLMs based on user demographics carry the same risks of affective
polarization and filter bubbles that have been seen in other personalized
internet technologies. This ``failure mode" should be monitored closely as
there are more attempts to monetize and personalize these models. | cs.CY | Filter bubbles and affective polarization in user-personalized large language model outputs |
2023-10-31T18:07:07Z | http://arxiv.org/pdf/2311.00059v1 | http://arxiv.org/abs/2311.00059v1 | Peter West, Ximing Lu, Nouha Dziri, Faeze Brahman, Linjie Li, Jena D. Hwang, Liwei Jiang, Jillian Fisher, Abhilasha Ravichander, Khyathi Chandu, Benjamin Newman, Pang Wei Koh, Allyson Ettinger, Yejin Choi | 2023-10-31T18:07:07Z | cs.AI, cs.CL, cs.CV, cs.LG | The recent wave of generative AI has sparked unprecedented global attention,
with both excitement and concern over potentially superhuman levels of
artificial intelligence: models now take only seconds to produce outputs that
would challenge or exceed the capabilities even of expert humans. At the same
time, models still show basic errors in understanding that would not be
expected even in non-expert humans. This presents us with an apparent paradox:
how do we reconcile seemingly superhuman capabilities with the persistence of
errors that few humans would make? In this work, we posit that this tension
reflects a divergence in the configuration of intelligence in today's
generative models relative to intelligence in humans. Specifically, we propose
and test the Generative AI Paradox hypothesis: generative models, having been
trained directly to reproduce expert-like outputs, acquire generative
capabilities that are not contingent upon -- and can therefore exceed -- their
ability to understand those same types of outputs. This contrasts with humans,
for whom basic understanding almost always precedes the ability to generate
expert-level outputs. We test this hypothesis through controlled experiments
analyzing generation vs. understanding in generative models, across both
language and image modalities. Our results show that although models can
outperform humans in generation, they consistently fall short of human
capabilities in measures of understanding, as well as weaker correlation
between generation and understanding performance, and more brittleness to
adversarial inputs. Our findings support the hypothesis that models' generative
capability may not be contingent upon understanding capability, and call for
caution in interpreting artificial intelligence by analogy to human
intelligence. | cs.AI | The Generative AI Paradox: "What It Can Create, It May Not Understand" |
2023-10-31T18:01:11Z | http://arxiv.org/pdf/2311.00047v1 | http://arxiv.org/abs/2311.00047v1 | Yichi Zhang, Jiayi Pan, Yuchen Zhou, Rui Pan, Joyce Chai | 2023-10-31T18:01:11Z | cs.AI, cs.CL, cs.CV, cs.LG | Vision-Language Models (VLMs) are trained on vast amounts of data captured by
humans emulating our understanding of the world. However, known as visual
illusions, human's perception of reality isn't always faithful to the physical
world. This raises a key question: do VLMs have the similar kind of illusions
as humans do, or do they faithfully learn to represent reality? To investigate
this question, we build a dataset containing five types of visual illusions and
formulate four tasks to examine visual illusions in state-of-the-art VLMs. Our
findings have shown that although the overall alignment is low, larger models
are closer to human perception and more susceptible to visual illusions. Our
dataset and initial findings will promote a better understanding of visual
illusions in humans and machines and provide a stepping stone for future
computational models that can better align humans and machines in perceiving
and communicating about the shared visual world. The code and data are
available at https://github.com/vl-illusion/dataset. | cs.AI | Grounding Visual Illusions in Language: Do Vision-Language Models Perceive Illusions Like Humans? |
2023-10-31T17:59:38Z | http://arxiv.org/pdf/2310.20707v2 | http://arxiv.org/abs/2310.20707v2 | Yanai Elazar, Akshita Bhagia, Ian Magnusson, Abhilasha Ravichander, Dustin Schwenk, Alane Suhr, Pete Walsh, Dirk Groeneveld, Luca Soldaini, Sameer Singh, Hanna Hajishirzi, Noah A. Smith, Jesse Dodge | 2024-03-05T20:02:31Z | cs.CL, cs.LG | Large text corpora are the backbone of language models. However, we have a
limited understanding of the content of these corpora, including general
statistics, quality, social factors, and inclusion of evaluation data
(contamination). In this work, we propose What's In My Big Data? (WIMBD), a
platform and a set of sixteen analyses that allow us to reveal and compare the
contents of large text corpora. WIMBD builds on two basic capabilities -- count
and search -- at scale, which allows us to analyze more than 35 terabytes on a
standard compute node. We apply WIMBD to ten different corpora used to train
popular language models, including C4, The Pile, and RedPajama. Our analysis
uncovers several surprising and previously undocumented findings about these
corpora, including the high prevalence of duplicate, synthetic, and low-quality
content, personally identifiable information, toxic language, and benchmark
contamination. For instance, we find that about 50% of the documents in
RedPajama and LAION-2B-en are duplicates. In addition, several datasets used
for benchmarking models trained on such corpora are contaminated with respect
to important benchmarks, including the Winograd Schema Challenge and parts of
GLUE and SuperGLUE. We open-source WIMBD's code and artifacts to provide a
standard set of evaluations for new text-based corpora and to encourage more
analyses and transparency around them. | cs.CL | What's In My Big Data? |
2023-10-31T17:59:05Z | http://arxiv.org/pdf/2310.20703v3 | http://arxiv.org/abs/2310.20703v3 | Noam Razin, Hattie Zhou, Omid Saremi, Vimal Thilak, Arwen Bradley, Preetum Nakkiran, Joshua Susskind, Etai Littwin | 2024-03-14T08:05:18Z | cs.LG, cs.AI, cs.CL, stat.ML | Pretrained language models are commonly aligned with human preferences and
downstream tasks via reinforcement finetuning (RFT), which refers to maximizing
a (possibly learned) reward function using policy gradient algorithms. This
work identifies a fundamental optimization obstacle in RFT: we prove that the
expected gradient for an input vanishes when its reward standard deviation
under the model is small, even if the expected reward is far from optimal.
Through experiments on an RFT benchmark and controlled environments, as well as
a theoretical analysis, we then demonstrate that vanishing gradients due to
small reward standard deviation are prevalent and detrimental, leading to
extremely slow reward maximization. Lastly, we explore ways to overcome
vanishing gradients in RFT. We find the common practice of an initial
supervised finetuning (SFT) phase to be the most promising candidate, which
sheds light on its importance in an RFT pipeline. Moreover, we show that a
relatively small number of SFT optimization steps on as few as 1% of the input
samples can suffice, indicating that the initial SFT phase need not be
expensive in terms of compute and data labeling efforts. Overall, our results
emphasize that being mindful for inputs whose expected gradient vanishes, as
measured by the reward standard deviation, is crucial for successful execution
of RFT. | cs.LG | Vanishing Gradients in Reinforcement Finetuning of Language Models |
2023-10-31T17:56:51Z | http://arxiv.org/pdf/2310.20697v1 | http://arxiv.org/abs/2310.20697v1 | Victoria Lin, Louis-Philippe Morency, Eli Ben-Michael | 2023-10-31T17:56:51Z | cs.CL, stat.ME | As language technologies gain prominence in real-world settings, it is
important to understand how changes to language affect reader perceptions. This
can be formalized as the causal effect of varying a linguistic attribute (e.g.,
sentiment) on a reader's response to the text. In this paper, we introduce
Text-Transport, a method for estimation of causal effects from natural language
under any text distribution. Current approaches for valid causal effect
estimation require strong assumptions about the data, meaning the data from
which one can estimate valid causal effects often is not representative of the
actual target domain of interest. To address this issue, we leverage the notion
of distribution shift to describe an estimator that transports causal effects
between domains, bypassing the need for strong assumptions in the target
domain. We derive statistical guarantees on the uncertainty of this estimator,
and we report empirical results and analyses that support the validity of
Text-Transport across data settings. Finally, we use Text-Transport to study a
realistic setting--hate speech on social media--in which causal effects do
shift significantly between text domains, demonstrating the necessity of
transport when conducting causal inference on natural language. | cs.CL | Text-Transport: Toward Learning Causal Effects of Natural Language |
2023-10-31T17:52:22Z | http://arxiv.org/pdf/2310.20689v4 | http://arxiv.org/abs/2310.20689v4 | Shengnan An, Zexiong Ma, Zeqi Lin, Nanning Zheng, Jian-Guang Lou, Weizhu Chen | 2024-03-29T07:17:39Z | cs.CL, cs.AI | Large language models (LLMs) recently exhibited remarkable reasoning
capabilities on solving math problems. To further improve their reasoning
capabilities, this work explores whether LLMs can LEarn from MistAkes (LEMA),
akin to the human learning process. Consider a human student who failed to
solve a math problem, he will learn from what mistake he has made and how to
correct it. Mimicking this error-driven learning process, LEMA incorporates
mistake-correction data pairs during fine-tuning LLMs. Specifically, we first
collect inaccurate reasoning paths from various LLMs, and then employ GPT-4 as
a ''corrector'' to identify the mistake step, explain the reason for the
mistake, correct the mistake and generate the final answer. In addition, we
apply a correction-centric evolution strategy that effectively expands the
question set for generating correction data. Experiments across various LLMs
and reasoning tasks show that LEMA effectively improves CoT-alone fine-tuning.
Our further ablations shed light on the non-homogeneous effectiveness between
CoT data and correction data. These results suggest a significant potential for
LLMs to improve through learning from their mistakes. Our code, models and
prompts are publicly available at https://github.com/microsoft/LEMA. | cs.CL | Learning From Mistakes Makes LLM Better Reasoner |
2023-10-31T17:25:07Z | http://arxiv.org/pdf/2310.20656v1 | http://arxiv.org/abs/2310.20656v1 | Verna Dankers, Christopher G. Lucas | 2023-10-31T17:25:07Z | cs.CL | When natural language phrases are combined, their meaning is often more than
the sum of their parts. In the context of NLP tasks such as sentiment analysis,
where the meaning of a phrase is its sentiment, that still applies. Many NLP
studies on sentiment analysis, however, focus on the fact that sentiment
computations are largely compositional. We, instead, set out to obtain
non-compositionality ratings for phrases with respect to their sentiment. Our
contributions are as follows: a) a methodology for obtaining those
non-compositionality ratings, b) a resource of ratings for 259 phrases --
NonCompSST -- along with an analysis of that resource, and c) an evaluation of
computational models for sentiment analysis using this new resource. | cs.CL | Non-Compositionality in Sentiment: New Data and Analyses |
2023-10-31T17:02:33Z | http://arxiv.org/pdf/2310.20633v1 | http://arxiv.org/abs/2310.20633v1 | Sha Li, Chi Han, Pengfei Yu, Carl Edwards, Manling Li, Xingyao Wang, Yi R. Fung, Charles Yu, Joel R. Tetreault, Eduard H. Hovy, Heng Ji | 2023-10-31T17:02:33Z | cs.CL | The recent explosion of performance of large language models (LLMs) has
changed the field of Natural Language Processing (NLP) more abruptly and
seismically than any other shift in the field's 80-year history. This has
resulted in concerns that the field will become homogenized and
resource-intensive. The new status quo has put many academic researchers,
especially PhD students, at a disadvantage. This paper aims to define a new NLP
playground by proposing 20+ PhD-dissertation-worthy research directions,
covering theoretical analysis, new and challenging problems, learning
paradigms, and interdisciplinary applications. | cs.CL | Defining a New NLP Playground |
2023-10-31T16:53:10Z | http://arxiv.org/pdf/2310.20620v2 | http://arxiv.org/abs/2310.20620v2 | Evgeniia Tokarchuk, Vlad Niculae | 2024-04-02T15:56:46Z | cs.CL, cs.LG | Continuous-output neural machine translation (CoNMT) replaces the discrete
next-word prediction problem with an embedding prediction. The semantic
structure of the target embedding space (i.e., closeness of related words) is
intuitively believed to be crucial. We challenge this assumption and show that
completely random output embeddings can outperform laboriously pretrained ones,
especially on larger datasets. Further investigation shows this surprising
effect is strongest for rare words, due to the geometry of their embeddings. We
shed further light on this finding by designing a mixed strategy that combines
random and pre-trained embeddings for different tokens. | cs.CL | The Unreasonable Effectiveness of Random Target Embeddings for Continuous-Output Neural Machine Translation |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.