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
|
|---|---|---|---|---|---|---|---|---|
2024-06-10T15:48:07Z | http://arxiv.org/pdf/2406.06393v2 | http://arxiv.org/abs/2406.06393v2 | Jiawen Chen, Muqing Zhou, Wenrong Wu, Jinwei Zhang, Yun Li, Didong Li | 2024-06-20T17:38:55Z | cs.CV, cs.CL, q-bio.GN, I.4.10; I.2.10 | Recent advances in multi-modal algorithms have driven and been driven by the
increasing availability of large image-text datasets, leading to significant
strides in various fields, including computational pathology. However, in most
existing medical image-text datasets, the text typically provides high-level
summaries that may not sufficiently describe sub-tile regions within a large
pathology image. For example, an image might cover an extensive tissue area
containing cancerous and healthy regions, but the accompanying text might only
specify that this image is a cancer slide, lacking the nuanced details needed
for in-depth analysis. In this study, we introduce STimage-1K4M, a novel
dataset designed to bridge this gap by providing genomic features for sub-tile
images. STimage-1K4M contains 1,149 images derived from spatial transcriptomics
data, which captures gene expression information at the level of individual
spatial spots within a pathology image. Specifically, each image in the dataset
is broken down into smaller sub-image tiles, with each tile paired with
15,000-30,000 dimensional gene expressions. With 4,293,195 pairs of sub-tile
images and gene expressions, STimage-1K4M offers unprecedented granularity,
paving the way for a wide range of advanced research in multi-modal data
analysis an innovative applications in computational pathology, and beyond. | cs.CV | STimage-1K4M: A histopathology image-gene expression dataset for spatial transcriptomics |
2024-06-10T15:46:25Z | http://arxiv.org/pdf/2406.06391v1 | http://arxiv.org/abs/2406.06391v1 | Junhao Zheng, Shengjie Qiu, Chengming Shi, Qianli Ma | 2024-06-10T15:46:25Z | cs.LG, cs.CL | As the applications of large language models (LLMs) expand across diverse
fields, the ability of these models to adapt to ongoing changes in data, tasks,
and user preferences becomes crucial. Traditional training methods, relying on
static datasets, are increasingly inadequate for coping with the dynamic nature
of real-world information. Lifelong learning, also known as continual or
incremental learning, addresses this challenge by enabling LLMs to learn
continuously and adaptively over their operational lifetime, integrating new
knowledge while retaining previously learned information and preventing
catastrophic forgetting. This survey delves into the sophisticated landscape of
lifelong learning, categorizing strategies into two primary groups: Internal
Knowledge and External Knowledge. Internal Knowledge includes continual
pretraining and continual finetuning, each enhancing the adaptability of LLMs
in various scenarios. External Knowledge encompasses retrieval-based and
tool-based lifelong learning, leveraging external data sources and
computational tools to extend the model's capabilities without modifying core
parameters. The key contributions of our survey are: (1) Introducing a novel
taxonomy categorizing the extensive literature of lifelong learning into 12
scenarios; (2) Identifying common techniques across all lifelong learning
scenarios and classifying existing literature into various technique groups
within each scenario; (3) Highlighting emerging techniques such as model
expansion and data selection, which were less explored in the pre-LLM era.
Through a detailed examination of these groups and their respective categories,
this survey aims to enhance the adaptability, reliability, and overall
performance of LLMs in real-world applications. | cs.LG | Towards Lifelong Learning of Large Language Models: A Survey |
2024-06-10T15:44:22Z | http://arxiv.org/pdf/2406.06385v2 | http://arxiv.org/abs/2406.06385v2 | Yelysei Bondarenko, Riccardo Del Chiaro, Markus Nagel | 2024-06-20T15:18:50Z | cs.LG, cs.AI, cs.CL | Large language models (LLMs) are omnipresent, however their practical
deployment is challenging due to their ever increasing computational and memory
demands. Quantization is one of the most effective ways to make them more
compute and memory efficient. Quantization-aware training (QAT) methods,
generally produce the best quantized performance, however it comes at the cost
of potentially long training time and excessive memory usage, making it
impractical when applying for LLMs. Inspired by parameter-efficient fine-tuning
(PEFT) and low-rank adaptation (LoRA) literature, we propose LR-QAT -- a
lightweight and memory-efficient QAT algorithm for LLMs. LR-QAT employs several
components to save memory without sacrificing predictive performance: (a)
low-rank auxiliary weights that are aware of the quantization grid; (b) a
downcasting operator using fixed-point or double-packed integers and (c)
checkpointing. Unlike most related work, our method (i) is inference-efficient,
leading to no additional overhead compared to traditional PTQ; (ii) can be seen
as a general extended pretraining framework, meaning that the resulting model
can still be utilized for any downstream task afterwards; (iii) can be applied
across a wide range of quantization settings, such as different choices
quantization granularity, activation quantization, and seamlessly combined with
many PTQ techniques. We apply LR-QAT to LLaMA-2/3 and Mistral model families
and validate its effectiveness on several downstream tasks. Our method
outperforms common post-training quantization (PTQ) approaches and reaches the
same model performance as full-model QAT at the fraction of its memory usage.
Specifically, we can train a 7B LLM on a single consumer grade GPU with 24GB of
memory. | cs.LG | Low-Rank Quantization-Aware Training for LLMs |
2024-06-10T15:42:03Z | http://arxiv.org/pdf/2406.06382v1 | http://arxiv.org/abs/2406.06382v1 | Yi Gu, Zhendong Wang, Yueqin Yin, Yujia Xie, Mingyuan Zhou | 2024-06-10T15:42:03Z | cs.CV, cs.CL, cs.LG | Aligning large language models with human preferences has emerged as a
critical focus in language modeling research. Yet, integrating preference
learning into Text-to-Image (T2I) generative models is still relatively
uncharted territory. The Diffusion-DPO technique made initial strides by
employing pairwise preference learning in diffusion models tailored for
specific text prompts. We introduce Diffusion-RPO, a new method designed to
align diffusion-based T2I models with human preferences more effectively. This
approach leverages both prompt-image pairs with identical prompts and those
with semantically related content across various modalities. Furthermore, we
have developed a new evaluation metric, style alignment, aimed at overcoming
the challenges of high costs, low reproducibility, and limited interpretability
prevalent in current evaluations of human preference alignment. Our findings
demonstrate that Diffusion-RPO outperforms established methods such as
Supervised Fine-Tuning and Diffusion-DPO in tuning Stable Diffusion versions
1.5 and XL-1.0, achieving superior results in both automated evaluations of
human preferences and style alignment. Our code is available at
https://github.com/yigu1008/Diffusion-RPO | cs.CV | Diffusion-RPO: Aligning Diffusion Models through Relative Preference Optimization |
2024-06-10T15:32:42Z | http://arxiv.org/pdf/2406.06371v3 | http://arxiv.org/abs/2406.06371v3 | Marcely Zanon Boito, Vivek Iyer, Nikolaos Lagos, Laurent Besacier, Ioan Calapodescu | 2024-06-27T07:56:48Z | cs.CL, cs.SD, eess.AS | We present mHuBERT-147, the first general-purpose massively multilingual
HuBERT speech representation model trained on 90K hours of clean, open-license
data. To scale up the multi-iteration HuBERT approach, we use faiss-based
clustering, achieving 5.2x faster label assignment than the original method. We
also apply a new multilingual batching up-sampling strategy, leveraging both
language and dataset diversity. After 3 training iterations, our compact 95M
parameter mHuBERT-147 outperforms larger models trained on substantially more
data. We rank second and first on the ML-SUPERB 10min and 1h leaderboards, with
SOTA scores for 3 tasks. Across ASR/LID tasks, our model consistently surpasses
XLS-R (300M params; 436K hours) and demonstrates strong competitiveness against
the much larger MMS (1B params; 491K hours). Our findings indicate that
mHuBERT-147 is a promising model for multilingual speech tasks, offering an
unprecedented balance between high performance and parameter efficiency. | cs.CL | mHuBERT-147: A Compact Multilingual HuBERT Model |
2024-06-10T15:30:13Z | http://arxiv.org/pdf/2406.06369v3 | http://arxiv.org/abs/2406.06369v3 | Rajiv Movva, Pang Wei Koh, Emma Pierson | 2024-06-24T23:23:22Z | cs.CL | To what extent do LLMs align with human perceptions of safety? We study this
question via *annotation alignment*, the extent to which LLMs and humans agree
when annotating the safety of user-chatbot conversations. We leverage the
recent DICES dataset (Aroyo et al., 2023), in which 350 conversations are each
rated for safety by 112 annotators spanning 10 race-gender groups. GPT-4
achieves a Pearson correlation of $r = 0.59$ with the average annotator rating,
higher than the median annotator's correlation with the average ($r=0.51$). We
show that larger datasets are needed to resolve whether GPT-4 exhibits
disparities in how well it correlates with demographic groups. Also, there is
substantial idiosyncratic variation in correlation *within* groups, suggesting
that race & gender do not fully capture differences in alignment. Finally, we
find that GPT-4 cannot predict when one demographic group finds a conversation
more unsafe than another. | cs.CL | Annotation alignment: Comparing LLM and human annotations of conversational safety |
2024-06-10T15:24:15Z | http://arxiv.org/pdf/2406.06366v2 | http://arxiv.org/abs/2406.06366v2 | Martin Courtois, Malte Ostendorff, Leonhard Hennig, Georg Rehm | 2024-06-19T10:42:15Z | cs.CL | Initially introduced as a machine translation model, the Transformer
architecture has now become the foundation for modern deep learning
architecture, with applications in a wide range of fields, from computer vision
to natural language processing. Nowadays, to tackle increasingly more complex
tasks, Transformer-based models are stretched to enormous sizes, requiring
increasingly larger training datasets, and unsustainable amount of compute
resources. The ubiquitous nature of the Transformer and its core component, the
attention mechanism, are thus prime targets for efficiency research. In this
work, we propose an alternative compatibility function for the self-attention
mechanism introduced by the Transformer architecture. This compatibility
function exploits an overlap in the learned representation of the traditional
scaled dot-product attention, leading to a symmetric with pairwise coefficient
dot-product attention. When applied to the pre-training of BERT-like models,
this new symmetric attention mechanism reaches a score of 79.36 on the GLUE
benchmark against 78.74 for the traditional implementation, leads to a
reduction of 6% in the number of trainable parameters, and reduces the number
of training steps required before convergence by half. | cs.CL | Symmetric Dot-Product Attention for Efficient Training of BERT Language Models |
2024-06-10T15:21:33Z | http://arxiv.org/pdf/2406.10256v1 | http://arxiv.org/abs/2406.10256v1 | Jovan Andonov, Octavian Ganea, Paulina Grnarova, Gary Bécigneul, Thomas Hofmann | 2024-06-10T15:21:33Z | cs.CL, cs.AI, cs.LG | Language Modelling has been a central part of Natural Language Processing for
a very long time and in the past few years LSTM-based language models have been
the go-to method for commercial language modeling. Recently, it has been shown
that when looking at language modelling from a matrix factorization point of
view, the final Softmax layer limits the expressiveness of the model, by
putting an upper bound on the rank of the resulting matrix. Additionally, a new
family of neural networks based called NeuralODEs, has been introduced as a
continuous alternative to Residual Networks. Moreover, it has been shown that
there is a connection between these models and Normalizing Flows. In this work
we propose a new family of language models based on NeuralODEs and the
continuous analogue of Normalizing Flows and manage to improve on some of the
baselines. | cs.CL | Explicit Word Density Estimation for Language Modelling |
2024-06-10T15:19:09Z | http://arxiv.org/pdf/2406.06357v1 | http://arxiv.org/abs/2406.06357v1 | Xingjian Zhang, Yutong Xie, Jin Huang, Jinge Ma, Zhaoying Pan, Qijia Liu, Ziyang Xiong, Tolga Ergen, Dongsub Shim, Honglak Lee, Qiaozhu Mei | 2024-06-10T15:19:09Z | cs.CL, cs.AI | Scientific innovation relies on detailed workflows, which include critical
steps such as analyzing literature, generating ideas, validating these ideas,
interpreting results, and inspiring follow-up research. However, scientific
publications that document these workflows are extensive and unstructured. This
makes it difficult for both human researchers and AI systems to effectively
navigate and explore the space of scientific innovation. To address this issue,
we introduce MASSW, a comprehensive text dataset on Multi-Aspect Summarization
of Scientific Workflows. MASSW includes more than 152,000 peer-reviewed
publications from 17 leading computer science conferences spanning the past 50
years. Using Large Language Models (LLMs), we automatically extract five core
aspects from these publications -- context, key idea, method, outcome, and
projected impact -- which correspond to five key steps in the research
workflow. These structured summaries facilitate a variety of downstream tasks
and analyses. The quality of the LLM-extracted summaries is validated by
comparing them with human annotations. We demonstrate the utility of MASSW
through multiple novel machine-learning tasks that can be benchmarked using
this new dataset, which make various types of predictions and recommendations
along the scientific workflow. MASSW holds significant potential for
researchers to create and benchmark new AI methods for optimizing scientific
workflows and fostering scientific innovation in the field. Our dataset is
openly available at \url{https://github.com/xingjian-zhang/massw}. | cs.CL | MASSW: A New Dataset and Benchmark Tasks for AI-Assisted Scientific Workflows |
2024-06-10T15:17:17Z | http://arxiv.org/pdf/2406.06355v1 | http://arxiv.org/abs/2406.06355v1 | Andreas Triantafyllopoulos, Anton Batliner, Wolfgang Mayr, Markus Fendler, Florian Pokorny, Maurice Gerczuk, Shahin Amiriparian, Thomas Berghaus, Björn Schuller | 2024-06-10T15:17:17Z | cs.CL | Chronic obstructive pulmonary disease (COPD) is a serious inflammatory lung
disease affecting millions of people around the world. Due to an obstructed
airflow from the lungs, it also becomes manifest in patients' vocal behaviour.
Of particular importance is the detection of an exacerbation episode, which
marks an acute phase and often requires hospitalisation and treatment. Previous
work has shown that it is possible to distinguish between a pre- and a
post-treatment state using automatic analysis of read speech. In this
contribution, we examine whether sustained vowels can provide a complementary
lens for telling apart these two states. Using a cohort of 50 patients, we show
that the inclusion of sustained vowels can improve performance to up to 79\%
unweighted average recall, from a 71\% baseline using read speech. We further
identify and interpret the most important acoustic features that characterise
the manifestation of COPD in sustained vowels. | cs.CL | Sustained Vowels for Pre- vs Post-Treatment COPD Classification |
2024-06-10T14:47:04Z | http://arxiv.org/pdf/2406.06331v2 | http://arxiv.org/abs/2406.06331v2 | Yunsoo Kim, Jinge Wu, Yusuf Abdulle, Honghan Wu | 2024-07-03T10:13:56Z | cs.CL, cs.AI | This paper introduces MedExQA, a novel benchmark in medical
question-answering, to evaluate large language models' (LLMs) understanding of
medical knowledge through explanations. By constructing datasets across five
distinct medical specialties that are underrepresented in current datasets and
further incorporating multiple explanations for each question-answer pair, we
address a major gap in current medical QA benchmarks which is the absence of
comprehensive assessments of LLMs' ability to generate nuanced medical
explanations. Our work highlights the importance of explainability in medical
LLMs, proposes an effective methodology for evaluating models beyond
classification accuracy, and sheds light on one specific domain, speech
language pathology, where current LLMs including GPT4 lack good understanding.
Our results show generation evaluation with multiple explanations aligns better
with human assessment, highlighting an opportunity for a more robust automated
comprehension assessment for LLMs. To diversify open-source medical LLMs
(currently mostly based on Llama2), this work also proposes a new medical
model, MedPhi-2, based on Phi-2 (2.7B). The model outperformed medical LLMs
based on Llama2-70B in generating explanations, showing its effectiveness in
the resource-constrained medical domain. We will share our benchmark datasets
and the trained model. | cs.CL | MedExQA: Medical Question Answering Benchmark with Multiple Explanations |
2024-06-10T14:46:07Z | http://arxiv.org/pdf/2406.06329v1 | http://arxiv.org/abs/2406.06329v1 | Wei Liu, Jingyong Hou, Dong Yang, Muyong Cao, Tan Lee | 2024-06-10T14:46:07Z | cs.CL, eess.AS | Covering all languages with a multilingual speech recognition model (MASR) is
very difficult. Performing language extension on top of an existing MASR is a
desirable choice. In this study, the MASR continual learning problem is
probabilistically decomposed into language identity prediction (LP) and
cross-lingual adaptation (XLA) sub-problems. Based on this, we propose an
architecture-based framework for language extension that can fundamentally
solve catastrophic forgetting, debudded as PELE. PELE is designed to be
parameter-efficient, incrementally incorporating an add-on module to adapt to a
new language. Specifically, different parameter-efficient fine-tuning (PEFT)
modules and their variants are explored as potential candidates to perform XLA.
Experiments are carried out on 5 new languages with a wide range of
low-resourced data sizes. The best-performing PEFT candidate can achieve
satisfactory performance across all languages and demonstrates superiority in
three of five languages over the continual joint learning setting. Notably,
PEFT methods focusing on weight parameters or input features are revealed to be
limited in performance, showing significantly inferior extension capabilities
compared to inserting a lightweight module in between layers such as an
Adapter. | cs.CL | A Parameter-efficient Language Extension Framework for Multilingual ASR |
2024-06-10T14:42:20Z | http://arxiv.org/pdf/2406.06326v3 | http://arxiv.org/abs/2406.06326v3 | Xiaoying Zhang, Baolin Peng, Ye Tian, Jingyan Zhou, Yipeng Zhang, Haitao Mi, Helen Meng | 2024-06-15T09:45:37Z | cs.CL | Large language models (LLMs) often struggle to provide up-to-date information
due to their one-time training and the constantly evolving nature of the world.
To keep LLMs current, existing approaches typically involve continued
pre-training on new documents. However, they frequently face difficulties in
extracting stored knowledge. Motivated by the remarkable success of the Feynman
Technique in efficient human learning, we introduce Self-Tuning, a learning
framework aimed at improving an LLM's ability to effectively acquire new
knowledge from raw documents through self-teaching. Specifically, we develop a
Self-Teaching strategy that augments the documents with a set of
knowledge-intensive tasks created in a self-supervised manner, focusing on
three crucial aspects: memorization, comprehension, and self-reflection. In
addition, we introduce three Wiki-Newpages-2023-QA datasets to facilitate an
in-depth analysis of an LLM's knowledge acquisition ability concerning
memorization, extraction, and reasoning. Extensive experimental results on
Llama2 family models reveal that Self-Tuning consistently exhibits superior
performance across all knowledge acquisition tasks and excels in preserving
previous knowledge. | cs.CL | Self-Tuning: Instructing LLMs to Effectively Acquire New Knowledge through Self-Teaching |
2024-06-10T14:36:44Z | http://arxiv.org/pdf/2406.10255v1 | http://arxiv.org/abs/2406.10255v1 | Weronika Borek-Marciniec, Pawel Zyblewski, Jakub Klikowski, Pawel Ksieniewicz | 2024-06-10T14:36:44Z | cs.CL, cs.SI | In the classification tasks, from raw data acquisition to the curation of a
dataset suitable for use in evaluating machine learning models, a series of
steps - often associated with high costs - are necessary. In the case of
Natural Language Processing, initial cleaning and conversion can be performed
automatically, but obtaining labels still requires the rationalized input of
human experts. As a result, even though many articles often state that "the
world is filled with data", data scientists suffer from its shortage. It is
crucial in the case of natural language applications, which is constantly
evolving and must adapt to new concepts or events. For example, the topic of
the COVID-19 pandemic and the vocabulary related to it would have been mostly
unrecognizable before 2019. For this reason, creating new datasets, also in
languages other than English, is still essential. This work presents a
collection of 3~187~105 posts in Polish about the pandemic and the war in
Ukraine published on popular social media platforms in 2022. The collection
includes not only preprocessed texts but also images so it can be used also for
multimodal recognition tasks. The labels define posts' topics and were created
using hashtags accompanying the posts. The work presents the process of
curating a dataset from acquisition to sample pattern recognition experiments. | cs.CL | WarCov -- Large multilabel and multimodal dataset from social platform |
2024-06-10T14:33:02Z | http://arxiv.org/pdf/2406.06316v1 | http://arxiv.org/abs/2406.06316v1 | Juan Manuel Zambrano Chaves, Eric Wang, Tao Tu, Eeshit Dhaval Vaishnav, Byron Lee, S. Sara Mahdavi, Christopher Semturs, David Fleet, Vivek Natarajan, Shekoofeh Azizi | 2024-06-10T14:33:02Z | cs.CL, cs.AI, cs.CE, cs.LG | Developing therapeutics is a lengthy and expensive process that requires the
satisfaction of many different criteria, and AI models capable of expediting
the process would be invaluable. However, the majority of current AI approaches
address only a narrowly defined set of tasks, often circumscribed within a
particular domain. To bridge this gap, we introduce Tx-LLM, a generalist large
language model (LLM) fine-tuned from PaLM-2 which encodes knowledge about
diverse therapeutic modalities. Tx-LLM is trained using a collection of 709
datasets that target 66 tasks spanning various stages of the drug discovery
pipeline. Using a single set of weights, Tx-LLM simultaneously processes a wide
variety of chemical or biological entities(small molecules, proteins, nucleic
acids, cell lines, diseases) interleaved with free-text, allowing it to predict
a broad range of associated properties, achieving competitive with
state-of-the-art (SOTA) performance on 43 out of 66 tasks and exceeding SOTA on
22. Among these, Tx-LLM is particularly powerful and exceeds best-in-class
performance on average for tasks combining molecular SMILES representations
with text such as cell line names or disease names, likely due to context
learned during pretraining. We observe evidence of positive transfer between
tasks with diverse drug types (e.g.,tasks involving small molecules and tasks
involving proteins), and we study the impact of model size, domain finetuning,
and prompting strategies on performance. We believe Tx-LLM represents an
important step towards LLMs encoding biochemical knowledge and could have a
future role as an end-to-end tool across the drug discovery development
pipeline. | cs.CL | Tx-LLM: A Large Language Model for Therapeutics |
2024-06-10T13:58:46Z | http://arxiv.org/pdf/2406.06279v1 | http://arxiv.org/abs/2406.06279v1 | Zifeng Cheng, Zhaoling Chen, Zhiwei Jiang, Yafeng Yin, Shiping Ge, Yuliang Liu, Qing Gu | 2024-06-10T13:58:46Z | cs.CL | Recent Pre-trained Language Models (PLMs) usually only provide users with the
inference APIs, namely the emerging Model-as-a-Service (MaaS) setting. To adapt
MaaS PLMs to downstream tasks without accessing their parameters and gradients,
some existing methods focus on the output-side adaptation of PLMs, viewing the
PLM as an encoder and then optimizing a task-specific decoder for decoding the
output hidden states and class scores of the PLM. Despite the effectiveness of
these methods, they only use a single prompt to query PLMs for decoding,
leading to a heavy reliance on the quality of the adopted prompt. In this
paper, we propose a simple yet effective Multi-Prompting Decoder (MPD)
framework for MaaS adaptation. The core idea is to query PLMs with multiple
different prompts for each sample, thereby obtaining multiple output hidden
states and class scores for subsequent decoding. Such multi-prompting decoding
paradigm can simultaneously mitigate reliance on the quality of a single
prompt, alleviate the issue of data scarcity under the few-shot setting, and
provide richer knowledge extracted from PLMs. Specifically, we propose two
decoding strategies: multi-prompting decoding with optimal transport for hidden
states and calibrated decoding for class scores. Extensive experiments
demonstrate that our method achieves new state-of-the-art results on multiple
natural language understanding datasets under the few-shot setting. | cs.CL | Multi-Prompting Decoder Helps Better Language Understanding |
2024-06-10T13:51:52Z | http://arxiv.org/pdf/2406.10254v1 | http://arxiv.org/abs/2406.10254v1 | Prateek Verma, Mert Pilanci | 2024-06-10T13:51:52Z | cs.CL, cs.LG, cs.SD, eess.AS | This paper introduces the idea of applying signal processing inside a Large
Language Model (LLM). With the recent explosion of generative AI, our work can
help bridge two fields together, namely the field of signal processing and
large language models. We draw parallels between classical Fourier-Transforms
and Fourier Transform-like learnable time-frequency representations for every
intermediate activation signal of an LLM. Once we decompose every activation
signal across tokens into a time-frequency representation, we learn how to
filter and reconstruct them, with all components learned from scratch, to
predict the next token given the previous context. We show that for GPT-like
architectures, our work achieves faster convergence and significantly increases
performance by adding a minuscule number of extra parameters when trained for
the same epochs. We hope this work paves the way for algorithms exploring
signal processing inside the signals found in neural architectures like LLMs
and beyond. | cs.CL | Towards Signal Processing In Large Language Models |
2024-06-10T13:44:29Z | http://arxiv.org/pdf/2406.06263v1 | http://arxiv.org/abs/2406.06263v1 | Amir Hossein Kargaran, François Yvon, Hinrich Schütze | 2024-06-10T13:44:29Z | cs.CL | We present MaskLID, a simple, yet effective, code-switching (CS) language
identification (LID) method. MaskLID does not require any training and is
designed to complement current high-performance sentence-level LIDs.
Sentence-level LIDs are classifiers trained on monolingual texts to provide
single labels, typically using a softmax layer to turn scores into
probabilities. However, in cases where a sentence is composed in both L1 and L2
languages, the LID classifier often only returns the dominant label L1. To
address this limitation, MaskLID employs a strategy to mask text features
associated with L1, allowing the LID to classify the text as L2 in the next
round. This method uses the LID itself to identify the features that require
masking and does not rely on any external resource. In this work, we explore
the use of MaskLID for two open-source LIDs (GlotLID and OpenLID), that are
both based on the FastText architecture. Code and demo are available at
https://github.com/cisnlp/MaskLID. | cs.CL | MaskLID: Code-Switching Language Identification through Iterative Masking |
2024-06-10T13:38:15Z | http://arxiv.org/pdf/2406.06257v1 | http://arxiv.org/abs/2406.06257v1 | Matthias Engelbach, Dennis Klau, Maximilien Kintz, Alexander Ulrich | 2024-06-10T13:38:15Z | cs.CL, cs.IR | Job descriptions are posted on many online channels, including company
websites, job boards or social media platforms. These descriptions are usually
published with varying text for the same job, due to the requirements of each
platform or to target different audiences. However, for the purpose of
automated recruitment and assistance of people working with these texts, it is
helpful to aggregate job postings across platforms and thus detect duplicate
descriptions that refer to the same job. In this work, we propose an approach
for detecting duplicates in job descriptions. We show that combining
overlap-based character similarity with text embedding and keyword matching
methods lead to convincing results. In particular, we show that although no
approach individually achieves satisfying performance, a combination of string
comparison, deep textual embeddings, and the use of curated weighted lookup
lists for specific skills leads to a significant boost in overall performance.
A tool based on our approach is being used in production and feedback from
real-life use confirms our evaluation. | cs.CL | Combining Embeddings and Domain Knowledge for Job Posting Duplicate Detection |
2024-06-10T13:31:18Z | http://arxiv.org/pdf/2406.06251v1 | http://arxiv.org/abs/2406.06251v1 | Chung-Ming Chien, Andros Tjandra, Apoorv Vyas, Matt Le, Bowen Shi, Wei-Ning Hsu | 2024-06-10T13:31:18Z | eess.AS, cs.CL | As the scale of generative models continues to grow, efficient reuse and
adaptation of pre-trained models have become crucial considerations. In this
work, we propose Voicebox Adapter, a novel approach that integrates
fine-grained conditions into a pre-trained Voicebox speech generation model
using a cross-attention module. To ensure a smooth integration of newly added
modules with pre-trained ones, we explore various efficient fine-tuning
approaches. Our experiment shows that the LoRA with bias-tuning configuration
yields the best performance, enhancing controllability without compromising
speech quality. Across three fine-grained conditional generation tasks, we
demonstrate the effectiveness and resource efficiency of Voicebox Adapter.
Follow-up experiments further highlight the robustness of Voicebox Adapter
across diverse data setups. | eess.AS | Learning Fine-Grained Controllability on Speech Generation via Efficient Fine-Tuning |
2024-06-10T13:13:39Z | http://arxiv.org/pdf/2406.06663v1 | http://arxiv.org/abs/2406.06663v1 | Sakshi Mahendru, Tejul Pandit | 2024-06-10T13:13:39Z | cs.CR, cs.CL, cs.LG | Phishing, whether through email, SMS, or malicious websites, poses a major
threat to organizations by using social engineering to trick users into
revealing sensitive information. It not only compromises company's data
security but also incurs significant financial losses. In this paper, we
investigate whether the remarkable performance of Large Language Models (LLMs)
can be leveraged for particular task like text classification, particularly
detecting malicious content and compare its results with state-of-the-art
Deberta V3 (DeBERTa using ELECTRA-Style Pre-Training with Gradient-Disentangled
Embedding Sharing) model. We systematically assess the potential and
limitations of both approaches using comprehensive public datasets comprising
diverse data sources such as email, HTML, URL, SMS, and synthetic data
generation. Additionally, we demonstrate how LLMs can generate convincing
phishing emails, making it harder to spot scams and evaluate the performance of
both models in this context. Our study delves further into the challenges
encountered by DeBERTa V3 during its training phases, fine-tuning methodology
and transfer learning processes. Similarly, we examine the challenges
associated with LLMs and assess their respective performance. Among our
experimental approaches, the transformer-based DeBERTa method emerged as the
most effective, achieving a test dataset (HuggingFace phishing dataset) recall
(sensitivity) of 95.17% closely followed by GPT-4 providing a recall of 91.04%.
We performed additional experiments with other datasets on the trained DeBERTa
V3 model and LLMs like GPT 4 and Gemini 1.5. Based on our findings, we provide
valuable insights into the effectiveness and robustness of these advanced
language models, offering a detailed comparative analysis that can inform
future research efforts in strengthening cybersecurity measures for detecting
and mitigating phishing threats. | cs.CR | SecureNet: A Comparative Study of DeBERTa and Large Language Models for Phishing Detection |
2024-06-10T12:58:56Z | http://arxiv.org/pdf/2406.10253v1 | http://arxiv.org/abs/2406.10253v1 | Revekka Kyriakoglou, Anna Pappa, Jilin He, Antoine Schoen, Patricia Laurens, Markarit Vartampetian, Philippe Laredo, Tita Kyriacopoulou | 2024-06-10T12:58:56Z | cs.CL, cs.IR, cs.LG | This paper presents the development of a lexicon centered on emerging
concepts, focusing on non-technological innovation. It introduces a four-step
methodology that combines human expertise, statistical analysis, and machine
learning techniques to establish a model that can be generalized across
multiple domains. This process includes the creation of a thematic corpus, the
development of a Gold Standard Lexicon, annotation and preparation of a
training corpus, and finally, the implementation of learning models to identify
new terms. The results demonstrate the robustness and relevance of our
approach, highlighting its adaptability to various contexts and its
contribution to lexical research. The developed methodology promises
applicability in conceptual fields. | cs.CL | Développement automatique de lexiques pour les concepts émergents : une exploration méthodologique |
2024-06-10T12:56:06Z | http://arxiv.org/pdf/2406.10252v2 | http://arxiv.org/abs/2406.10252v2 | Yidong Wang, Qi Guo, Wenjin Yao, Hongbo Zhang, Xin Zhang, Zhen Wu, Meishan Zhang, Xinyu Dai, Min Zhang, Qingsong Wen, Wei Ye, Shikun Zhang, Yue Zhang | 2024-06-18T02:11:31Z | cs.IR, cs.AI, cs.CL | This paper introduces AutoSurvey, a speedy and well-organized methodology for
automating the creation of comprehensive literature surveys in rapidly evolving
fields like artificial intelligence. Traditional survey paper creation faces
challenges due to the vast volume and complexity of information, prompting the
need for efficient survey methods. While large language models (LLMs) offer
promise in automating this process, challenges such as context window
limitations, parametric knowledge constraints, and the lack of evaluation
benchmarks remain. AutoSurvey addresses these challenges through a systematic
approach that involves initial retrieval and outline generation, subsection
drafting by specialized LLMs, integration and refinement, and rigorous
evaluation and iteration. Our contributions include a comprehensive solution to
the survey problem, a reliable evaluation method, and experimental validation
demonstrating AutoSurvey's effectiveness.We open our resources at
\url{https://github.com/AutoSurveys/AutoSurvey}. | cs.IR | AutoSurvey: Large Language Models Can Automatically Write Surveys |
2024-06-10T12:34:38Z | http://arxiv.org/pdf/2406.06220v1 | http://arxiv.org/abs/2406.06220v1 | Vladimir Bataev, Hainan Xu, Daniel Galvez, Vitaly Lavrukhin, Boris Ginsburg | 2024-06-10T12:34:38Z | eess.AS, cs.AI, cs.CL, cs.LG, cs.SD | This paper introduces a highly efficient greedy decoding algorithm for
Transducer inference. We propose a novel data structure using CUDA tensors to
represent partial hypotheses in a batch that supports parallelized hypothesis
manipulations. During decoding, our algorithm maximizes GPU parallelism by
adopting a nested-loop design, where the inner loop consumes all blank
predictions, while non-blank predictions are handled in the outer loop. Our
algorithm is general-purpose and can work with both conventional Transducers
and Token-and-Duration Transducers. Experiments show that the label-looping
algorithm can bring a speedup up to 2.0X compared to conventional batched
decoding algorithms when using batch size 32, and can be combined with other
compiler or GPU call-related techniques to bring more speedup. We will
open-source our implementation to benefit the research community. | eess.AS | Label-Looping: Highly Efficient Decoding for Transducers |
2024-06-10T12:06:15Z | http://arxiv.org/pdf/2406.07584v1 | http://arxiv.org/abs/2406.07584v1 | Wanaiu Huang | 2024-06-10T12:06:15Z | cs.CV, cs.AI, cs.CL | Semantic information is vital for human interaction, and decoding it from
brain activity enables non-invasive clinical augmentative and alternative
communication. While there has been significant progress in reconstructing
visual images, few studies have focused on the language aspect. To address this
gap, leveraging the powerful capabilities of the decoder-based vision-language
pretrained model CoCa, this paper proposes BrainChat, a simple yet effective
generative framework aimed at rapidly accomplishing semantic information
decoding tasks from brain activity, including fMRI question answering and fMRI
captioning. BrainChat employs the self-supervised approach of Masked Brain
Modeling to encode sparse fMRI data, obtaining a more compact embedding
representation in the latent space. Subsequently, BrainChat bridges the gap
between modalities by applying contrastive loss, resulting in aligned
representations of fMRI, image, and text embeddings. Furthermore, the fMRI
embeddings are mapped to the generative Brain Decoder via cross-attention
layers, where they guide the generation of textual content about fMRI in a
regressive manner by minimizing caption loss. Empirically, BrainChat exceeds
the performance of existing state-of-the-art methods in the fMRI captioning
task and, for the first time, implements fMRI question answering. Additionally,
BrainChat is highly flexible and can achieve high performance without image
data, making it better suited for real-world scenarios with limited data. | cs.CV | BrainChat: Decoding Semantic Information from fMRI using Vision-language Pretrained Models |
2024-06-10T11:50:29Z | http://arxiv.org/pdf/2406.06196v2 | http://arxiv.org/abs/2406.06196v2 | Andrew M. Bean, Simi Hellsten, Harry Mayne, Jabez Magomere, Ethan A. Chi, Ryan Chi, Scott A. Hale, Hannah Rose Kirk | 2024-06-11T10:19:52Z | cs.CL | In this paper, we present the LingOly benchmark, a novel benchmark for
advanced reasoning abilities in large language models. Using challenging
Linguistic Olympiad puzzles, we evaluate (i) capabilities for in-context
identification and generalisation of linguistic patterns in very low-resource
or extinct languages, and (ii) abilities to follow complex task instructions.
The LingOly benchmark covers more than 90 mostly low-resource languages,
minimising issues of data contamination, and contains 1,133 problems across 6
formats and 5 levels of human difficulty. We assess performance with both
direct accuracy and comparison to a no-context baseline to penalise
memorisation. Scores from 11 state-of-the-art LLMs demonstrate the benchmark to
be challenging, and models perform poorly on the higher difficulty problems. On
harder problems, even the top model only achieved 38.7% accuracy, 24.7%
improvement over the no-context baseline. Large closed models typically
outperform open models, and in general, the higher resource the language, the
better the scores. These results indicate, in absence of memorisation, true
multi-step out-of-domain reasoning remains a challenge for current language
models. | cs.CL | LINGOLY: A Benchmark of Olympiad-Level Linguistic Reasoning Puzzles in Low-Resource and Extinct Languages |
2024-06-10T11:19:28Z | http://arxiv.org/pdf/2406.06657v1 | http://arxiv.org/abs/2406.06657v1 | Mark A. Kramer, Allen Leavens, Alexander Scarlat | 2024-06-10T11:19:28Z | cs.CL, cs.AI, cs.IR | Policy documents, such as legislation, regulations, and executive orders, are
crucial in shaping society. However, their length and complexity make
interpretation and application challenging and time-consuming. Artificial
intelligence (AI), particularly large language models (LLMs), has the potential
to automate the process of analyzing these documents, improving accuracy and
efficiency. This study aims to evaluate the potential of AI in streamlining
policy analysis and to identify the strengths and limitations of current AI
approaches. The research focuses on question answering and tasks involving
content extraction from policy documents. A case study was conducted using
Executive Order 14110 on "Safe, Secure, and Trustworthy Development and Use of
Artificial Intelligence" as a test case. Four commercial AI systems were used
to analyze the document and answer a set of representative policy questions.
The performance of the AI systems was compared to manual analysis conducted by
human experts. The study found that two AI systems, Gemini 1.5 Pro and Claude 3
Opus, demonstrated significant potential for supporting policy analysis,
providing accurate and reliable information extraction from complex documents.
They performed comparably to human analysts but with significantly higher
efficiency. However, achieving reproducibility remains a challenge,
necessitating further research and development. | cs.CL | Harnessing AI for efficient analysis of complex policy documents: a case study of Executive Order 14110 |
2024-06-10T10:03:16Z | http://arxiv.org/pdf/2406.06144v2 | http://arxiv.org/abs/2406.06144v2 | Jiaming Ji, Kaile Wang, Tianyi Qiu, Boyuan Chen, Jiayi Zhou, Changye Li, Hantao Lou, Yaodong Yang | 2024-06-13T06:46:14Z | cs.CL, cs.AI | Large language models (LLMs) may exhibit undesirable behaviors. Recent
efforts have focused on aligning these models to prevent harmful generation.
Despite these efforts, studies have shown that even a well-conducted alignment
process can be easily circumvented, whether intentionally or accidentally. Do
alignment fine-tuning have robust effects on models, or are merely superficial?
In this work, we answer this question through both theoretical and empirical
means. Empirically, we demonstrate the elasticity of post-alignment models,
i.e., the tendency to revert to the behavior distribution formed during the
pre-training phase upon further fine-tuning. Using compression theory, we
formally derive that such fine-tuning process disproportionately undermines
alignment compared to pre-training, potentially by orders of magnitude. We
conduct experimental validations to confirm the presence of elasticity across
models of varying types and sizes. Specifically, we find that model performance
declines rapidly before reverting to the pre-training distribution, after which
the rate of decline drops significantly. We further reveal that elasticity
positively correlates with increased model size and the expansion of
pre-training data. Our discovery signifies the importance of taming the
inherent elasticity of LLMs, thereby overcoming the resistance of LLMs to
alignment finetuning. | cs.CL | Language Models Resist Alignment |
2024-06-10T09:53:54Z | http://arxiv.org/pdf/2406.06140v1 | http://arxiv.org/abs/2406.06140v1 | Zhiquan Tan, Lai Wei, Jindong Wang, Xing Xie, Weiran Huang | 2024-06-10T09:53:54Z | cs.CL, cs.LG | Large language models (LLMs) have achieved remarkable progress in linguistic
tasks, necessitating robust evaluation frameworks to understand their
capabilities and limitations. Inspired by Feynman's principle of understanding
through creation, we introduce a self-knowledge evaluation framework that is
easy to implement, evaluating models on their ability to comprehend and respond
to self-generated questions. Our findings, based on testing multiple models
across diverse tasks, reveal significant gaps in the model's self-knowledge
ability. Further analysis indicates these gaps may be due to misalignment with
human attention mechanisms. Additionally, fine-tuning on self-generated math
task may enhance the model's math performance, highlighting the potential of
the framework for efficient and insightful model evaluation and may also
contribute to the improvement of LLMs. | cs.CL | Can I understand what I create? Self-Knowledge Evaluation of Large Language Models |
2024-06-10T09:52:25Z | http://arxiv.org/pdf/2406.06139v1 | http://arxiv.org/abs/2406.06139v1 | Thanapat Trachu, Chawan Piansaddhayanon, Ekapol Chuangsuwanich | 2024-06-10T09:52:25Z | cs.SD, cs.AI, cs.CL, eess.AS | Diffusion-based speech enhancement has shown promising results, but can
suffer from a slower inference time. Initializing the diffusion process with
the enhanced audio generated by a regression-based model can be used to reduce
the computational steps required. However, these approaches often necessitate a
regression model, further increasing the system's complexity. We propose
Thunder, a unified regression-diffusion model that utilizes the Brownian bridge
process which can allow the model to act in both modes. The regression mode can
be accessed by setting the diffusion time step closed to 1. However, the
standard score-based diffusion modeling does not perform well in this setup due
to gradient instability. To mitigate this problem, we modify the diffusion
model to predict the clean speech instead of the score function, achieving
competitive performance with a more compact model size and fewer reverse steps. | cs.SD | Thunder : Unified Regression-Diffusion Speech Enhancement with a Single Reverse Step using Brownian Bridge |
2024-06-10T09:39:19Z | http://arxiv.org/pdf/2406.06131v1 | http://arxiv.org/abs/2406.06131v1 | Manuel Lardelli, Giuseppe Attanasio, Anne Lauscher | 2024-06-10T09:39:19Z | cs.CL | The translation of gender-neutral person-referring terms (e.g., the students)
is often non-trivial. Translating from English into German poses an interesting
case -- in German, person-referring nouns are usually gender-specific, and if
the gender of the referent(s) is unknown or diverse, the generic masculine (die
Studenten (m.)) is commonly used. This solution, however, reduces the
visibility of other genders, such as women and non-binary people. To counteract
gender discrimination, a societal movement towards using gender-fair language
exists (e.g., by adopting neosystems). However, gender-fair German is currently
barely supported in machine translation (MT), requiring post-editing or manual
translations. We address this research gap by studying gender-fair language in
English-to-German MT. Concretely, we enrich a community-created gender-fair
language dictionary and sample multi-sentence test instances from encyclopedic
text and parliamentary speeches. Using these novel resources, we conduct the
first benchmark study involving two commercial systems and six neural MT models
for translating words in isolation and natural contexts across two domains. Our
findings show that most systems produce mainly masculine forms and rarely
gender-neutral variants, highlighting the need for future research. We release
code and data at
https://github.com/g8a9/building-bridges-gender-fair-german-mt. | cs.CL | Building Bridges: A Dataset for Evaluating Gender-Fair Machine Translation into German |
2024-06-10T09:36:05Z | http://arxiv.org/pdf/2406.06127v1 | http://arxiv.org/abs/2406.06127v1 | Christos Vlachos, Themos Stafylakis, Ion Androutsopoulos | 2024-06-10T09:36:05Z | cs.CL, cs.AI | Creating effective and reliable task-oriented dialog systems (ToDSs) is
challenging, not only because of the complex structure of these systems, but
also due to the scarcity of training data, especially when several modules need
to be trained separately, each one with its own input/output training examples.
Data augmentation (DA), whereby synthetic training examples are added to the
training data, has been successful in other NLP systems, but has not been
explored as extensively in ToDSs. We empirically evaluate the effectiveness of
DA methods in an end-to-end ToDS setting, where a single system is trained to
handle all processing stages, from user inputs to system outputs. We experiment
with two ToDSs (UBAR, GALAXY) on two datasets (MultiWOZ, KVRET). We consider
three types of DA methods (word-level, sentence-level, dialog-level), comparing
eight DA methods that have shown promising results in ToDSs and other NLP
systems. We show that all DA methods considered are beneficial, and we
highlight the best ones, also providing advice to practitioners. We also
introduce a more challenging few-shot cross-domain ToDS setting, reaching
similar conclusions. | cs.CL | Comparing Data Augmentation Methods for End-to-End Task-Oriented Dialog Systems |
2024-06-10T09:32:37Z | http://arxiv.org/pdf/2406.06125v1 | http://arxiv.org/abs/2406.06125v1 | Shuyang Cao, Lu Wang | 2024-06-10T09:32:37Z | cs.CL | Verifiable generation requires large language models (LLMs) to cite source
documents supporting their outputs, thereby improve output transparency and
trustworthiness. Yet, previous work mainly targets the generation of
sentence-level citations, lacking specificity about which parts of a sentence
are backed by the cited sources. This work studies verifiable generation with
subsentence-level fine-grained citations for more precise location of generated
content supported by the cited sources. We first present a dataset, SCiFi,
comprising 10K Wikipedia paragraphs with subsentence-level citations. Each
paragraph is paired with a set of candidate source documents for citation and a
query that triggers the generation of the paragraph content. On SCiFi, we
evaluate the performance of state-of-the-art LLMs and strategies for processing
long documents designed for these models. Our experiment results reveals key
factors that could enhance the quality of citations, including the expansion of
the source documents' context accessible to the models and the implementation
of specialized model tuning. | cs.CL | Verifiable Generation with Subsentence-Level Fine-Grained Citations |
2024-06-10T09:29:08Z | http://arxiv.org/pdf/2406.06124v1 | http://arxiv.org/abs/2406.06124v1 | Aadharsh Aadhithya A, Sachin Kumar S, Soman K. P | 2024-06-10T09:29:08Z | cs.CL, cs.AI, I.2.7 | Large language models have limited context capacity, hindering reasoning over
long conversations. We propose the Hierarchical Aggregate Tree memory structure
to recursively aggregate relevant dialogue context through conditional tree
traversals. HAT encapsulates information from children nodes, enabling broad
coverage with depth control. We formulate finding best context as optimal tree
traversal. Experiments show HAT improves dialog coherence and summary quality
over baseline contexts, demonstrating the techniques effectiveness for multi
turn reasoning without exponential parameter growth. This memory augmentation
enables more consistent, grounded longform conversations from LLMs | cs.CL | Enhancing Long-Term Memory using Hierarchical Aggregate Tree for Retrieval Augmented Generation |
2024-06-10T08:50:59Z | http://arxiv.org/pdf/2406.06110v1 | http://arxiv.org/abs/2406.06110v1 | Chensen Huang, Guibo Zhu, Xuepeng Wang, Yifei Luo, Guojing Ge, Haoran Chen, Dong Yi, Jinqiao Wang | 2024-06-10T08:50:59Z | cs.CL, cs.AI | To extend the context length of Transformer-based large language models
(LLMs) and improve comprehension capabilities, we often face limitations due to
computational resources and bounded memory storage capacity. This work
introduces a method called Recurrent Context Compression (RCC), designed to
efficiently expand the context window length of LLMs within constrained storage
space. We also investigate the issue of poor model responses when both
instructions and context are compressed in downstream tasks, and propose an
instruction reconstruction method to mitigate this problem. We validated the
effectiveness of our approach on multiple tasks, achieving a compression rate
of up to 32x on text reconstruction tasks with a BLEU4 score close to 0.95, and
nearly 100\% accuracy on a passkey retrieval task with a sequence length of 1M.
Finally, our method demonstrated competitive performance in long-text
question-answering tasks compared to non-compressed methods, while
significantly saving storage resources in long-text inference tasks. Our code,
models, and demo are available at https://github.com/WUHU-G/RCC_Transformer | cs.CL | Recurrent Context Compression: Efficiently Expanding the Context Window of LLM |
2024-06-10T08:27:58Z | http://arxiv.org/pdf/2406.06097v1 | http://arxiv.org/abs/2406.06097v1 | Sara Papi, Marco Gaido, Matteo Negri, Luisa Bentivogli | 2024-06-10T08:27:58Z | cs.SD, cs.AI, cs.CL, eess.AS | Streaming speech-to-text translation (StreamST) is the task of automatically
translating speech while incrementally receiving an audio stream. Unlike
simultaneous ST (SimulST), which deals with pre-segmented speech, StreamST
faces the challenges of handling continuous and unbounded audio streams. This
requires additional decisions about what to retain of the previous history,
which is impractical to keep entirely due to latency and computational
constraints. Despite the real-world demand for real-time ST, research on
streaming translation remains limited, with existing works solely focusing on
SimulST. To fill this gap, we introduce StreamAtt, the first StreamST policy,
and propose StreamLAAL, the first StreamST latency metric designed to be
comparable with existing metrics for SimulST. Extensive experiments across all
8 languages of MuST-C v1.0 show the effectiveness of StreamAtt compared to a
naive streaming baseline and the related state-of-the-art SimulST policy,
providing a first step in StreamST research. | cs.SD | StreamAtt: Direct Streaming Speech-to-Text Translation with Attention-based Audio History Selection |
2024-06-10T08:23:52Z | http://arxiv.org/pdf/2406.10251v1 | http://arxiv.org/abs/2406.10251v1 | Mert Yazan, Suzan Verberne, Frederik Situmeang | 2024-06-10T08:23:52Z | cs.CL, cs.AI | Post-training quantization reduces the computational demand of Large Language
Models (LLMs) but can weaken some of their capabilities. Since LLM abilities
emerge with scale, smaller LLMs are more sensitive to quantization. In this
paper, we explore how quantization affects smaller LLMs' ability to perform
retrieval-augmented generation (RAG), specifically in longer contexts. We chose
personalization for evaluation because it is a challenging domain to perform
using RAG as it requires long-context reasoning over multiple documents. We
compare the original FP16 and the quantized INT4 performance of multiple 7B and
8B LLMs on two tasks while progressively increasing the number of retrieved
documents to test how quantized models fare against longer contexts. To better
understand the effect of retrieval, we evaluate three retrieval models in our
experiments. Our findings reveal that if a 7B LLM performs the task well,
quantization does not impair its performance and long-context reasoning
capabilities. We conclude that it is possible to utilize RAG with quantized
smaller LLMs. | cs.CL | The Impact of Quantization on Retrieval-Augmented Generation: An Analysis of Small LLMs |
2024-06-10T07:36:55Z | http://arxiv.org/pdf/2406.06073v1 | http://arxiv.org/abs/2406.06073v1 | Yan Gao, Zhiwei Cao, Zhongjian Miao, Baosong Yang, Shiyu Liu, Min Zhang, Jinsong Su | 2024-06-10T07:36:55Z | cs.CL | To achieve non-parametric NMT domain adaptation, $k$-Nearest-Neighbor Machine
Translation ($k$NN-MT) constructs an external datastore to store
domain-specific translation knowledge, which derives a $k$NN distribution to
interpolate the prediction distribution of the NMT model via a linear
interpolation coefficient $\lambda$. Despite its success, $k$NN retrieval at
each timestep leads to substantial time overhead. To address this issue,
dominant studies resort to $k$NN-MT with adaptive retrieval ($k$NN-MT-AR),
which dynamically estimates $\lambda$ and skips $k$NN retrieval if $\lambda$ is
less than a fixed threshold. Unfortunately, $k$NN-MT-AR does not yield
satisfactory results. In this paper, we first conduct a preliminary study to
reveal two key limitations of $k$NN-MT-AR: 1) the optimization gap leads to
inaccurate estimation of $\lambda$ for determining $k$NN retrieval skipping,
and 2) using a fixed threshold fails to accommodate the dynamic demands for
$k$NN retrieval at different timesteps. To mitigate these limitations, we then
propose $k$NN-MT with dynamic retrieval ($k$NN-MT-DR) that significantly
extends vanilla $k$NN-MT in two aspects. Firstly, we equip $k$NN-MT with a
MLP-based classifier for determining whether to skip $k$NN retrieval at each
timestep. Particularly, we explore several carefully-designed scalar features
to fully exert the potential of the classifier. Secondly, we propose a
timestep-aware threshold adjustment method to dynamically generate the
threshold, which further improves the efficiency of our model. Experimental
results on the widely-used datasets demonstrate the effectiveness and
generality of our model.\footnote{Our code is available at
\url{https://github.com/DeepLearnXMU/knn-mt-dr}. | cs.CL | Efficient k-Nearest-Neighbor Machine Translation with Dynamic Retrieval |
2024-06-10T07:03:36Z | http://arxiv.org/pdf/2406.06056v1 | http://arxiv.org/abs/2406.06056v1 | Avijit Mitra, Emily Druhl, Raelene Goodwin, Hong Yu | 2024-06-10T07:03:36Z | cs.CL | Social and behavioral determinants of health (SBDH) play a crucial role in
health outcomes and are frequently documented in clinical text. Automatically
extracting SBDH information from clinical text relies on publicly available
good-quality datasets. However, existing SBDH datasets exhibit substantial
limitations in their availability and coverage. In this study, we introduce
Synth-SBDH, a novel synthetic dataset with detailed SBDH annotations,
encompassing status, temporal information, and rationale across 15 SBDH
categories. We showcase the utility of Synth-SBDH on three tasks using
real-world clinical datasets from two distinct hospital settings, highlighting
its versatility, generalizability, and distillation capabilities. Models
trained on Synth-SBDH consistently outperform counterparts with no Synth-SBDH
training, achieving up to 62.5% macro-F improvements. Additionally, Synth-SBDH
proves effective for rare SBDH categories and under-resource constraints. Human
evaluation demonstrates a Human-LLM alignment of 71.06% and uncovers areas for
future refinements. | cs.CL | Synth-SBDH: A Synthetic Dataset of Social and Behavioral Determinants of Health for Clinical Text |
2024-06-10T06:46:09Z | http://arxiv.org/pdf/2406.06052v1 | http://arxiv.org/abs/2406.06052v1 | Naomi Baes, Nick Haslam, Ekaterina Vylomova | 2024-06-10T06:46:09Z | cs.CL | Historical linguists have identified multiple forms of lexical semantic
change. We present a three-dimensional framework for integrating these forms
and a unified computational methodology for evaluating them concurrently. The
dimensions represent increases or decreases in semantic 1) sentiment, 2)
breadth, and 3) intensity. These dimensions can be complemented by the
evaluation of shifts in the frequency of the target words and the thematic
content of its collocates. This framework enables lexical semantic change to be
mapped economically and systematically and has applications in computational
social science. We present an illustrative analysis of semantic shifts in
mental health and mental illness in two corpora, demonstrating patterns of
semantic change that illuminate contemporary concerns about pathologization,
stigma, and concept creep. | cs.CL | A Multidimensional Framework for Evaluating Lexical Semantic Change with Social Science Applications |
2024-06-10T06:27:42Z | http://arxiv.org/pdf/2406.06046v1 | http://arxiv.org/abs/2406.06046v1 | Zichun Yu, Spandan Das, Chenyan Xiong | 2024-06-10T06:27:42Z | cs.CL, cs.LG | Pretraining data selection has the potential to improve language model
pretraining efficiency by utilizing higher-quality data from massive web data
corpora. Current data selection methods, which rely on either hand-crafted
rules or larger reference models, are conducted statically and do not capture
the evolving data preferences during pretraining. In this paper, we introduce
model-aware data selection with data influence models (MATES), where a data
influence model continuously adapts to the evolving data preferences of the
pretraining model and then selects the data most effective for the current
pretraining progress. Specifically, we fine-tune a small data influence model
to approximate oracle data preference signals collected by locally probing the
pretraining model and to select data accordingly for the next pretraining
stage. Experiments on Pythia and the C4 dataset demonstrate that MATES
significantly outperforms random data selection on extensive downstream tasks
in both zero- and few-shot settings. It doubles the gains achieved by recent
data selection approaches that leverage larger reference models and reduces the
total FLOPs required to reach certain performances by half. Further analysis
validates the ever-changing data preferences of pretraining models and the
effectiveness of our data influence models to capture them. Our code is
open-sourced at https://github.com/cxcscmu/MATES. | cs.CL | MATES: Model-Aware Data Selection for Efficient Pretraining with Data Influence Models |
2024-06-10T05:50:23Z | http://arxiv.org/pdf/2406.06032v1 | http://arxiv.org/abs/2406.06032v1 | Ryosuke Takahashi, Go Kamoda, Benjamin Heinzerling, Keisuke Sakaguchi, Kentaro Inui | 2024-06-10T05:50:23Z | cs.CL | Language models (LMs) encode world knowledge in their internal parameters
through training. However, LMs may learn personal and confidential information
from the training data, leading to privacy concerns such as data leakage.
Therefore, research on knowledge deletion from LMs is essential. This study
focuses on the knowledge stored in LMs and analyzes the relationship between
the side effects of knowledge deletion and the entities related to the
knowledge. Our findings reveal that deleting knowledge related to popular
entities can have catastrophic side effects. Furthermore, this research is the
first to analyze knowledge deletion in models trained on synthetic knowledge
graphs, indicating a new direction for controlled experiments. | cs.CL | The Curse of Popularity: Popular Entities have Catastrophic Side Effects when Deleting Knowledge from Language Models |
2024-06-10T05:22:49Z | http://arxiv.org/pdf/2406.06027v1 | http://arxiv.org/abs/2406.06027v1 | Pranoy Panda, Ankush Agarwal, Chaitanya Devaguptapu, Manohar Kaul, Prathosh A P | 2024-06-10T05:22:49Z | cs.CL | Given unstructured text, Large Language Models (LLMs) are adept at answering
simple (single-hop) questions. However, as the complexity of the questions
increase, the performance of LLMs degrade. We believe this is due to the
overhead associated with understanding the complex question followed by
filtering and aggregating unstructured information in the raw text. Recent
methods try to reduce this burden by integrating structured knowledge triples
into the raw text, aiming to provide a structured overview that simplifies
information processing. However, this simplistic approach is query-agnostic and
the extracted facts are ambiguous as they lack context. To address these
drawbacks and to enable LLMs to answer complex (multi-hop) questions with ease,
we propose to use a knowledge graph (KG) that is context-aware and is distilled
to contain query-relevant information. The use of our compressed distilled KG
as input to the LLM results in our method utilizing up to $67\%$ fewer tokens
to represent the query relevant information present in the supporting
documents, compared to the state-of-the-art (SoTA) method. Our experiments show
consistent improvements over the SoTA across several metrics (EM, F1,
BERTScore, and Human Eval) on two popular benchmark datasets (HotpotQA and
MuSiQue). | cs.CL | HOLMES: Hyper-Relational Knowledge Graphs for Multi-hop Question Answering using LLMs |
2024-06-10T05:15:30Z | http://arxiv.org/pdf/2406.06025v1 | http://arxiv.org/abs/2406.06025v1 | Jiawei Liu, Jia Le Tian, Vijay Daita, Yuxiang Wei, Yifeng Ding, Yuhan Katherine Wang, Jun Yang, Lingming Zhang | 2024-06-10T05:15:30Z | cs.SE, cs.CL, cs.LG | Recent advances have been improving the context windows of Large Language
Models (LLMs). To quantify the real long-context capabilities of LLMs,
evaluators such as the popular Needle in a Haystack have been developed to test
LLMs over a large chunk of raw texts. While effective, current evaluations
overlook the insight of how LLMs work with long-context code, i.e.,
repositories. To this end, we initiate the RepoQA benchmark to evaluate LLMs on
long-context code understanding. Traditional needle testers ask LLMs to
directly retrieve the answer from the context without necessary deep
understanding. In RepoQA, we built our initial task, namely Searching Needle
Function (SNF), which exercises LLMs to search functions given their
natural-language description, i.e., LLMs cannot find the desired function if
they cannot understand the description and code. RepoQA is multilingual and
comprehensive: it includes 500 code search tasks gathered from 50 popular
repositories across 5 modern programming languages. By evaluating 26 general
and code-specific LLMs on RepoQA, we show (i) there is still a small gap
between the best open and proprietary models; (ii) different models are good at
different languages; and (iii) models may understand code better without
comments. | cs.SE | RepoQA: Evaluating Long Context Code Understanding |
2024-06-10T05:01:26Z | http://arxiv.org/pdf/2406.06648v1 | http://arxiv.org/abs/2406.06648v1 | Jung-Ho Kim, Mathew Huerta-Enochian, Changyong Ko, Du Hui Lee | 2024-06-10T05:01:26Z | cs.CL, cs.AI, cs.LG | Sign languages are multi-channel languages that communicate information
through not just the hands (manual signals) but also facial expressions and
upper body movements (non-manual signals). However, since automatic sign
language translation is usually performed by generating a single sequence of
glosses, researchers eschew non-manual and co-occurring manual signals in favor
of a simplified list of manual glosses. This can lead to significant
information loss and ambiguity. In this paper, we introduce a new task named
multi-channel sign language translation (MCSLT) and present a novel metric,
SignBLEU, designed to capture multiple signal channels. We validated SignBLEU
on a system-level task using three sign language corpora with varied linguistic
structures and transcription methodologies and examined its correlation with
human judgment through two segment-level tasks. We found that SignBLEU
consistently correlates better with human judgment than competing metrics. To
facilitate further MCSLT research, we report benchmark scores for the three
sign language corpora and release the source code for SignBLEU at
https://github.com/eq4all-projects/SignBLEU. | cs.CL | SignBLEU: Automatic Evaluation of Multi-channel Sign Language Translation |
2024-06-10T04:47:27Z | http://arxiv.org/pdf/2406.06021v1 | http://arxiv.org/abs/2406.06021v1 | Surangika Ranathunga, Nisansa de Silva, Dilith Jayakody, Aloka Fernando | 2024-06-10T04:47:27Z | cs.CL | We analysed a sample of NLP research papers archived in ACL Anthology as an
attempt to quantify the degree of openness and the benefit of such an open
culture in the NLP community. We observe that papers published in different NLP
venues show different patterns related to artefact reuse. We also note that
more than 30% of the papers we analysed do not release their artefacts
publicly, despite promising to do so. Further, we observe a wide language-wise
disparity in publicly available NLP-related artefacts. | cs.CL | Shoulders of Giants: A Look at the Degree and Utility of Openness in NLP Research |
2024-06-10T04:07:09Z | http://arxiv.org/pdf/2406.06007v1 | http://arxiv.org/abs/2406.06007v1 | Peng Xia, Ze Chen, Juanxi Tian, Yangrui Gong, Ruibo Hou, Yue Xu, Zhenbang Wu, Zhiyuan Fan, Yiyang Zhou, Kangyu Zhu, Wenhao Zheng, Zhaoyang Wang, Xiao Wang, Xuchao Zhang, Chetan Bansal, Marc Niethammer, Junzhou Huang, Hongtu Zhu, Yun Li, Jimeng Sun, Zongyuan Ge, Gang Li, James Zou, Huaxiu Yao | 2024-06-10T04:07:09Z | cs.LG, cs.CL, cs.CV, cs.CY | Artificial intelligence has significantly impacted medical applications,
particularly with the advent of Medical Large Vision Language Models
(Med-LVLMs), sparking optimism for the future of automated and personalized
healthcare. However, the trustworthiness of Med-LVLMs remains unverified,
posing significant risks for future model deployment. In this paper, we
introduce CARES and aim to comprehensively evaluate the Trustworthiness of
Med-LVLMs across the medical domain. We assess the trustworthiness of Med-LVLMs
across five dimensions, including trustfulness, fairness, safety, privacy, and
robustness. CARES comprises about 41K question-answer pairs in both closed and
open-ended formats, covering 16 medical image modalities and 27 anatomical
regions. Our analysis reveals that the models consistently exhibit concerns
regarding trustworthiness, often displaying factual inaccuracies and failing to
maintain fairness across different demographic groups. Furthermore, they are
vulnerable to attacks and demonstrate a lack of privacy awareness. We publicly
release our benchmark and code in https://github.com/richard-peng-xia/CARES. | cs.LG | CARES: A Comprehensive Benchmark of Trustworthiness in Medical Vision Language Models |
2024-06-10T03:57:39Z | http://arxiv.org/pdf/2406.06004v1 | http://arxiv.org/abs/2406.06004v1 | Yebin Lee, Imseong Park, Myungjoo Kang | 2024-06-10T03:57:39Z | cs.CV, cs.AI, cs.CL | Most existing image captioning evaluation metrics focus on assigning a single
numerical score to a caption by comparing it with reference captions. However,
these methods do not provide an explanation for the assigned score. Moreover,
reference captions are expensive to acquire. In this paper, we propose FLEUR,
an explainable reference-free metric to introduce explainability into image
captioning evaluation metrics. By leveraging a large multimodal model, FLEUR
can evaluate the caption against the image without the need for reference
captions, and provide the explanation for the assigned score. We introduce
score smoothing to align as closely as possible with human judgment and to be
robust to user-defined grading criteria. FLEUR achieves high correlations with
human judgment across various image captioning evaluation benchmarks and
reaches state-of-the-art results on Flickr8k-CF, COMPOSITE, and Pascal-50S
within the domain of reference-free evaluation metrics. Our source code and
results are publicly available at: https://github.com/Yebin46/FLEUR. | cs.CV | FLEUR: An Explainable Reference-Free Evaluation Metric for Image Captioning Using a Large Multimodal Model |
2024-06-10T03:47:24Z | http://arxiv.org/pdf/2406.06000v1 | http://arxiv.org/abs/2406.06000v1 | Pontakorn Trakuekul, Wei Qi Leong, Charin Polpanumas, Jitkapat Sawatphol, William Chandra Tjhi, Attapol T. Rutherford | 2024-06-10T03:47:24Z | cs.CL | While coreference resolution is a well-established research area in Natural
Language Processing (NLP), research focusing on Thai language remains limited
due to the lack of large annotated corpora. In this work, we introduce
ThaiCoref, a dataset for Thai coreference resolution. Our dataset comprises
777,271 tokens, 44,082 mentions and 10,429 entities across four text genres:
university essays, newspapers, speeches, and Wikipedia. Our annotation scheme
is built upon the OntoNotes benchmark with adjustments to address Thai-specific
phenomena. Utilizing ThaiCoref, we train models employing a multilingual
encoder and cross-lingual transfer techniques, achieving a best F1 score of
67.88\% on the test set. Error analysis reveals challenges posed by Thai's
unique linguistic features. To benefit the NLP community, we make the dataset
and the model publicly available at http://www.github.com/nlp-chula/thai-coref . | cs.CL | ThaiCoref: Thai Coreference Resolution Dataset |
2024-06-10T03:29:23Z | http://arxiv.org/pdf/2406.05995v1 | http://arxiv.org/abs/2406.05995v1 | Yutong Han, Yan Yuan, Lili Mou | 2024-06-10T03:29:23Z | cs.CL, cs.AI, cs.LG | Radiology report analysis provides valuable information that can aid with
public health initiatives, and has been attracting increasing attention from
the research community. In this work, we present a novel insight that the
structure of a radiology report (namely, the Findings and Impression sections)
offers different views of a radiology scan. Based on this intuition, we further
propose a co-training approach, where two machine learning models are built
upon the Findings and Impression sections, respectively, and use each other's
information to boost performance with massive unlabeled data in a
semi-supervised manner. We conducted experiments in a public health
surveillance study, and results show that our co-training approach is able to
improve performance using the dual views and surpass competing supervised and
semi-supervised methods. | cs.CL | A Dual-View Approach to Classifying Radiology Reports by Co-Training |
2024-06-10T02:47:55Z | http://arxiv.org/pdf/2406.05981v2 | http://arxiv.org/abs/2406.05981v2 | Haoran You, Yipin Guo, Yichao Fu, Wei Zhou, Huihong Shi, Xiaofan Zhang, Souvik Kundu, Amir Yazdanbakhsh, Yingyan, Lin | 2024-06-11T15:14:30Z | cs.LG, cs.AI, cs.CL | Large language models (LLMs) have shown impressive performance on language
tasks but face challenges when deployed on resource-constrained devices due to
their extensive parameters and reliance on dense multiplications, resulting in
high memory demands and latency bottlenecks. Shift-and-add reparameterization
offers a promising solution by replacing costly multiplications with
hardware-friendly primitives in both the attention and multi-layer perceptron
(MLP) layers of an LLM. However, current reparameterization techniques require
training from scratch or full parameter fine-tuning to restore accuracy, which
is resource-intensive for LLMs. To address this, we propose accelerating
pretrained LLMs through post-training shift-and-add reparameterization,
creating efficient multiplication-free models, dubbed ShiftAddLLM.
Specifically, we quantize each weight matrix into binary matrices paired with
group-wise scaling factors. The associated multiplications are reparameterized
into (1) shifts between activations and scaling factors and (2) queries and
adds according to the binary matrices. To reduce accuracy loss, we present a
multi-objective optimization method to minimize both weight and output
activation reparameterization errors. Additionally, based on varying
sensitivity across layers to reparameterization, we develop an automated bit
allocation strategy to further reduce memory usage and latency. Experiments on
five LLM families and eight tasks consistently validate the effectiveness of
ShiftAddLLM, achieving average perplexity improvements of 5.6 and 22.7 points
at comparable or lower latency compared to the most competitive quantized LLMs
at 3 and 2 bits, respectively, and more than 80% memory and energy reductions
over the original LLMs. Codes and models are available at
https://github.com/GATECH-EIC/ShiftAddLLM. | cs.LG | ShiftAddLLM: Accelerating Pretrained LLMs via Post-Training Multiplication-Less Reparameterization |
2024-06-10T02:04:28Z | http://arxiv.org/pdf/2406.05968v1 | http://arxiv.org/abs/2406.05968v1 | Wonjune Kang, Deb Roy | 2024-06-10T02:04:28Z | eess.AS, cs.CL | In this work, we introduce a framework for speech summarization that
leverages the processing and reasoning capabilities of large language models
(LLMs). We propose an end-to-end system that combines an instruction-tuned LLM
with an audio encoder that converts speech into token representations that the
LLM can interpret. Using a dataset with paired speech-text data, the overall
system is trained to generate consistent responses to prompts with the same
semantic information regardless of the input modality. The resulting framework
allows the LLM to process speech inputs in the same way as text, enabling
speech summarization by simply prompting the LLM. Unlike prior approaches, our
method is able to summarize spoken content from any arbitrary domain, and it
can produce summaries in different styles by varying the LLM prompting
strategy. Experiments demonstrate that our approach outperforms a cascade
baseline of speech recognition followed by LLM text processing. | eess.AS | Prompting Large Language Models with Audio for General-Purpose Speech Summarization |
2024-06-10T01:59:00Z | http://arxiv.org/pdf/2406.05967v1 | http://arxiv.org/abs/2406.05967v1 | David Romero, Chenyang Lyu, Haryo Akbarianto Wibowo, Teresa Lynn, Injy Hamed, Aditya Nanda Kishore, Aishik Mandal, Alina Dragonetti, Artem Abzaliev, Atnafu Lambebo Tonja, Bontu Fufa Balcha, Chenxi Whitehouse, Christian Salamea, Dan John Velasco, David Ifeoluwa Adelani, David Le Meur, Emilio Villa-Cueva, Fajri Koto, Fauzan Farooqui, Frederico Belcavello, Ganzorig Batnasan, Gisela Vallejo, Grainne Caulfield, Guido Ivetta, Haiyue Song, Henok Biadglign Ademtew, Hernán Maina, Holy Lovenia, Israel Abebe Azime, Jan Christian Blaise Cruz, Jay Gala, Jiahui Geng, Jesus-German Ortiz-Barajas, Jinheon Baek, Jocelyn Dunstan, Laura Alonso Alemany, Kumaranage Ravindu Yasas Nagasinghe, Luciana Benotti, Luis Fernando D'Haro, Marcelo Viridiano, Marcos Estecha-Garitagoitia, Maria Camila Buitrago Cabrera, Mario Rodríguez-Cantelar, Mélanie Jouitteau, Mihail Mihaylov, Mohamed Fazli Mohamed Imam, Muhammad Farid Adilazuarda, Munkhjargal Gochoo, Munkh-Erdene Otgonbold, Naome Etori, Olivier Niyomugisha, Paula Mónica Silva, Pranjal Chitale, Raj Dabre, Rendi Chevi, Ruochen Zhang, Ryandito Diandaru, Samuel Cahyawijaya, Santiago Góngora, Soyeong Jeong, Sukannya Purkayastha, Tatsuki Kuribayashi, Thanmay Jayakumar, Tiago Timponi Torrent, Toqeer Ehsan, Vladimir Araujo, Yova Kementchedjhieva, Zara Burzo, Zheng Wei Lim, Zheng Xin Yong, Oana Ignat, Joan Nwatu, Rada Mihalcea, Thamar Solorio, Alham Fikri Aji | 2024-06-10T01:59:00Z | cs.CV, cs.AI, cs.CL, cs.LG | Visual Question Answering (VQA) is an important task in multimodal AI, and it
is often used to test the ability of vision-language models to understand and
reason on knowledge present in both visual and textual data. However, most of
the current VQA models use datasets that are primarily focused on English and a
few major world languages, with images that are typically Western-centric.
While recent efforts have tried to increase the number of languages covered on
VQA datasets, they still lack diversity in low-resource languages. More
importantly, although these datasets often extend their linguistic range via
translation or some other approaches, they usually keep images the same,
resulting in narrow cultural representation. To address these limitations, we
construct CVQA, a new Culturally-diverse multilingual Visual Question Answering
benchmark, designed to cover a rich set of languages and cultures, where we
engage native speakers and cultural experts in the data collection process. As
a result, CVQA includes culturally-driven images and questions from across 28
countries on four continents, covering 26 languages with 11 scripts, providing
a total of 9k questions. We then benchmark several Multimodal Large Language
Models (MLLMs) on CVQA, and show that the dataset is challenging for the
current state-of-the-art models. This benchmark can serve as a probing
evaluation suite for assessing the cultural capability and bias of multimodal
models and hopefully encourage more research efforts toward increasing cultural
awareness and linguistic diversity in this field. | cs.CV | CVQA: Culturally-diverse Multilingual Visual Question Answering Benchmark |
2024-06-10T01:21:59Z | http://arxiv.org/pdf/2406.05955v2 | http://arxiv.org/abs/2406.05955v2 | Yixin Song, Haotong Xie, Zhengyan Zhang, Bo Wen, Li Ma, Zeyu Mi, Haibo Chen | 2024-06-11T02:15:47Z | cs.LG, cs.CL | Exploiting activation sparsity is a promising approach to significantly
accelerating the inference process of large language models (LLMs) without
compromising performance. However, activation sparsity is determined by
activation functions, and commonly used ones like SwiGLU and GeGLU exhibit
limited sparsity. Simply replacing these functions with ReLU fails to achieve
sufficient sparsity. Moreover, inadequate training data can further increase
the risk of performance degradation. To address these challenges, we propose a
novel dReLU function, which is designed to improve LLM activation sparsity,
along with a high-quality training data mixture ratio to facilitate effective
sparsification. Additionally, we leverage sparse activation patterns within the
Feed-Forward Network (FFN) experts of Mixture-of-Experts (MoE) models to
further boost efficiency. By applying our neuron sparsification method to the
Mistral and Mixtral models, only 2.5 billion and 4.3 billion parameters are
activated per inference iteration, respectively, while achieving even more
powerful model performance. Evaluation results demonstrate that this sparsity
achieves a 2-5x decoding speedup. Remarkably, on mobile phones, our
TurboSparse-Mixtral-47B achieves an inference speed of 11 tokens per second.
Our models are available at \url{https://huggingface.co/PowerInfer} | cs.LG | Turbo Sparse: Achieving LLM SOTA Performance with Minimal Activated Parameters |
2024-06-09T22:46:41Z | http://arxiv.org/pdf/2406.05930v1 | http://arxiv.org/abs/2406.05930v1 | Liang Lu, Peirong Xie, David R. Mortensen | 2024-06-09T22:46:41Z | cs.CL | Existing work implementing comparative reconstruction of ancestral languages
(proto-languages) has usually required full supervision. However, historical
reconstruction models are only of practical value if they can be trained with a
limited amount of labeled data. We propose a semisupervised historical
reconstruction task in which the model is trained on only a small amount of
labeled data (cognate sets with proto-forms) and a large amount of unlabeled
data (cognate sets without proto-forms). We propose a neural architecture for
comparative reconstruction (DPD-BiReconstructor) incorporating an essential
insight from linguists' comparative method: that reconstructed words should not
only be reconstructable from their daughter words, but also deterministically
transformable back into their daughter words. We show that this architecture is
able to leverage unlabeled cognate sets to outperform strong semisupervised
baselines on this novel task. | cs.CL | Semisupervised Neural Proto-Language Reconstruction |
2024-06-09T21:58:32Z | http://arxiv.org/pdf/2406.05925v1 | http://arxiv.org/abs/2406.05925v1 | Hao Li, Chenghao Yang, An Zhang, Yang Deng, Xiang Wang, Tat-Seng Chua | 2024-06-09T21:58:32Z | cs.CL, cs.AI | Open-domain dialogue systems have seen remarkable advancements with the
development of large language models (LLMs). Nonetheless, most existing
dialogue systems predominantly focus on brief single-session interactions,
neglecting the real-world demands for long-term companionship and personalized
interactions with chatbots. Crucial to addressing this real-world need are
event summary and persona management, which enable reasoning for appropriate
long-term dialogue responses. Recent progress in the human-like cognitive and
reasoning capabilities of LLMs suggests that LLM-based agents could
significantly enhance automated perception, decision-making, and
problem-solving. In response to this potential, we introduce a model-agnostic
framework, the Long-term Dialogue Agent (LD-Agent), which incorporates three
independently tunable modules dedicated to event perception, persona
extraction, and response generation. For the event memory module, long and
short-term memory banks are employed to separately focus on historical and
ongoing sessions, while a topic-based retrieval mechanism is introduced to
enhance the accuracy of memory retrieval. Furthermore, the persona module
conducts dynamic persona modeling for both users and agents. The integration of
retrieved memories and extracted personas is subsequently fed into the
generator to induce appropriate responses. The effectiveness, generality, and
cross-domain capabilities of LD-Agent are empirically demonstrated across
various illustrative benchmarks, models, and tasks. The code is released at
https://github.com/leolee99/LD-Agent. | cs.CL | Hello Again! LLM-powered Personalized Agent for Long-term Dialogue |
2024-06-09T21:12:15Z | http://arxiv.org/pdf/2406.05918v1 | http://arxiv.org/abs/2406.05918v1 | Abdelrahman Zayed, Goncalo Mordido, Ioana Baldini, Sarath Chandar | 2024-06-09T21:12:15Z | cs.CL, cs.AI, cs.CY, cs.LG | The widespread use of large language models has brought up essential
questions about the potential biases these models might learn. This led to the
development of several metrics aimed at evaluating and mitigating these biases.
In this paper, we first demonstrate that prompt-based fairness metrics exhibit
poor agreement, as measured by correlation, raising important questions about
the reliability of fairness assessment using prompts. Then, we outline six
relevant reasons why such a low correlation is observed across existing
metrics. Based on these insights, we propose a method called Correlated
Fairness Output (CAIRO) to enhance the correlation between fairness metrics.
CAIRO augments the original prompts of a given fairness metric by using several
pre-trained language models and then selects the combination of the augmented
prompts that achieves the highest correlation across metrics. We show a
significant improvement in Pearson correlation from 0.3 and 0.18 to 0.90 and
0.98 across metrics for gender and religion biases, respectively. Our code is
available at https://github.com/chandar-lab/CAIRO. | cs.CL | Why Don't Prompt-Based Fairness Metrics Correlate? |
2024-06-09T20:18:58Z | http://arxiv.org/pdf/2406.05906v1 | http://arxiv.org/abs/2406.05906v1 | Chufan Gao, Xuan Wang, Jimeng Sun | 2024-06-09T20:18:58Z | cs.CL, cs.AI | Document-level relation extraction aims to categorize the association between
any two entities within a document. We find that previous methods for
document-level relation extraction are ineffective in exploiting the full
potential of large amounts of training data with varied noise levels. For
example, in the ReDocRED benchmark dataset, state-of-the-art methods trained on
the large-scale, lower-quality, distantly supervised training data generally do
not perform better than those trained solely on the smaller, high-quality,
human-annotated training data. To unlock the full potential of large-scale
noisy training data for document-level relation extraction, we propose TTM-RE,
a novel approach that integrates a trainable memory module, known as the Token
Turing Machine, with a noisy-robust loss function that accounts for the
positive-unlabeled setting. Extensive experiments on ReDocRED, a benchmark
dataset for document-level relation extraction, reveal that TTM-RE achieves
state-of-the-art performance (with an absolute F1 score improvement of over
3%). Ablation studies further illustrate the superiority of TTM-RE in other
domains (the ChemDisGene dataset in the biomedical domain) and under highly
unlabeled settings. | cs.CL | TTM-RE: Memory-Augmented Document-Level Relation Extraction |
2024-06-09T19:42:25Z | http://arxiv.org/pdf/2406.05902v1 | http://arxiv.org/abs/2406.05902v1 | Emilia Agis Lerner, Florian E. Dorner, Elliott Ash, Naman Goel | 2024-06-09T19:42:25Z | cs.LG, cs.AI, cs.CL, cs.CY | There is a growing body of work on learning from human feedback to align
various aspects of machine learning systems with human values and preferences.
We consider the setting of fairness in content moderation, in which human
feedback is used to determine how two comments -- referencing different
sensitive attribute groups -- should be treated in comparison to one another.
With a novel dataset collected from Prolific and MTurk, we find significant
gaps in fairness preferences depending on the race, age, political stance,
educational level, and LGBTQ+ identity of annotators. We also demonstrate that
demographics mentioned in text have a strong influence on how users perceive
individual fairness in moderation. Further, we find that differences also exist
in downstream classifiers trained to predict human preferences. Finally, we
observe that an ensemble, giving equal weight to classifiers trained on
annotations from different demographics, performs better for different
demographic intersections; compared to a single classifier that gives equal
weight to each annotation. | cs.LG | Whose Preferences? Differences in Fairness Preferences and Their Impact on the Fairness of AI Utilizing Human Feedback |
2024-06-09T19:08:33Z | http://arxiv.org/pdf/2406.05888v2 | http://arxiv.org/abs/2406.05888v2 | Mamadou K. Keita, Elysabhete Amadou Ibrahim, Habibatou Abdoulaye Alfari, Christopher Homan | 2024-06-17T22:36:18Z | cs.CL | Machine translation (MT) is a rapidly expanding field that has experienced
significant advancements in recent years with the development of models capable
of translating multiple languages with remarkable accuracy. However, the
representation of African languages in this field still needs to improve due to
linguistic complexities and limited resources. This applies to the Zarma
language, a dialect of Songhay (of the Nilo-Saharan language family) spoken by
over 5 million people across Niger and neighboring countries
\cite{lewis2016ethnologue}. This paper introduces Feriji, the first robust
French-Zarma parallel corpus and glossary designed for MT. The corpus,
containing 61,085 sentences in Zarma and 42,789 in French, and a glossary of
4,062 words represent a significant step in addressing the need for more
resources for Zarma. We fine-tune three large language models on our dataset,
obtaining a BLEU score of 30.06 on the best-performing model. We further
evaluate the models on human judgments of fluency, comprehension, and
readability and the importance and impact of the corpus and models. Our
contributions help to bridge a significant language gap and promote an
essential and overlooked indigenous African language. | cs.CL | Feriji: A French-Zarma Parallel Corpus, Glossary & Translator |
2024-06-09T18:45:41Z | http://arxiv.org/pdf/2406.05885v1 | http://arxiv.org/abs/2406.05885v1 | Sourabrata Mukherjee, Atul Kr. Ojha, Ondřej Dušek | 2024-06-09T18:45:41Z | cs.CL | We analyze the performance of large language models (LLMs) on Text Style
Transfer (TST), specifically focusing on sentiment transfer and text
detoxification across three languages: English, Hindi, and Bengali. Text Style
Transfer involves modifying the linguistic style of a text while preserving its
core content. We evaluate the capabilities of pre-trained LLMs using zero-shot
and few-shot prompting as well as parameter-efficient finetuning on publicly
available datasets. Our evaluation using automatic metrics, GPT-4 and human
evaluations reveals that while some prompted LLMs perform well in English,
their performance in on other languages (Hindi, Bengali) remains average.
However, finetuning significantly improves results compared to zero-shot and
few-shot prompting, making them comparable to previous state-of-the-art. This
underscores the necessity of dedicated datasets and specialized models for
effective TST. | cs.CL | Are Large Language Models Actually Good at Text Style Transfer? |
2024-06-09T18:40:24Z | http://arxiv.org/pdf/2406.05881v2 | http://arxiv.org/abs/2406.05881v2 | Utsav Singh, Pramit Bhattacharyya, Vinay P. Namboodiri | 2024-06-16T10:28:45Z | cs.LG, cs.CL, cs.RO | Developing interactive systems that leverage natural language instructions to
solve complex robotic control tasks has been a long-desired goal in the
robotics community. Large Language Models (LLMs) have demonstrated exceptional
abilities in handling complex tasks, including logical reasoning, in-context
learning, and code generation. However, predicting low-level robotic actions
using LLMs poses significant challenges. Additionally, the complexity of such
tasks usually demands the acquisition of policies to execute diverse subtasks
and combine them to attain the ultimate objective. Hierarchical Reinforcement
Learning (HRL) is an elegant approach for solving such tasks, which provides
the intuitive benefits of temporal abstraction and improved exploration.
However, HRL faces the recurring issue of non-stationarity due to unstable
lower primitive behaviour. In this work, we propose LGR2, a novel HRL framework
that leverages language instructions to generate a stationary reward function
for the higher-level policy. Since the language-guided reward is unaffected by
the lower primitive behaviour, LGR2 mitigates non-stationarity and is thus an
elegant method for leveraging language instructions to solve robotic control
tasks. To analyze the efficacy of our approach, we perform empirical analysis
and demonstrate that LGR2 effectively alleviates non-stationarity in HRL. Our
approach attains success rates exceeding 70$\%$ in challenging, sparse-reward
robotic navigation and manipulation environments where the baselines fail to
achieve any significant progress. Additionally, we conduct real-world robotic
manipulation experiments and demonstrate that CRISP shows impressive
generalization in real-world scenarios. | cs.LG | LGR2: Language Guided Reward Relabeling for Accelerating Hierarchical Reinforcement Learning |
2024-06-09T18:13:36Z | http://arxiv.org/pdf/2406.05876v1 | http://arxiv.org/abs/2406.05876v1 | Yanis Labrak, Adel Moumen, Richard Dufour, Mickael Rouvier | 2024-06-09T18:13:36Z | cs.CL, cs.LG, cs.SD, eess.AS | In the rapidly evolving landscape of spoken question-answering (SQA), the
integration of large language models (LLMs) has emerged as a transformative
development. Conventional approaches often entail the use of separate models
for question audio transcription and answer selection, resulting in significant
resource utilization and error accumulation. To tackle these challenges, we
explore the effectiveness of end-to-end (E2E) methodologies for SQA in the
medical domain. Our study introduces a novel zero-shot SQA approach, compared
to traditional cascade systems. Through a comprehensive evaluation conducted on
a new open benchmark of 8 medical tasks and 48 hours of synthetic audio, we
demonstrate that our approach requires up to 14.7 times fewer resources than a
combined 1.3B parameters LLM with a 1.55B parameters ASR model while improving
average accuracy by 0.5\%. These findings underscore the potential of E2E
methodologies for SQA in resource-constrained contexts. | cs.CL | Zero-Shot End-To-End Spoken Question Answering In Medical Domain |
2024-06-09T18:07:47Z | http://arxiv.org/pdf/2406.05872v1 | http://arxiv.org/abs/2406.05872v1 | Shreyas Basavatia, Keerthiram Murugesan, Shivam Ratnakar | 2024-06-09T18:07:47Z | cs.LG, cs.AI, cs.CL | Interactive fiction games have emerged as an important application to improve
the generalization capabilities of language-based reinforcement learning (RL)
agents. Existing environments for interactive fiction games are domain-specific
or time-consuming to generate and do not train the RL agents to master a
specific set of skills. In this work, we introduce an interactive environment
for self-supervised RL, STARLING, for text-based games that bootstraps the
text-based RL agents with automatically generated games (based on the seed set
of game ideas) to boost the performance and generalization capabilities to
reach a goal of the target environment. These games let the agent hone their
skills on a predefined set of tasks. We create and test an environment with 100
games, generated using this automated framework that uses large language models
(GPT-3) and an interactive fiction game engine (based on Inform7) to provide
the user with the ability to generate more games under minimal human
supervision. Experimental results based on both the human participants and
baseline text-based RL agents reveal that current state-of-the-art text-based
RL agents cannot use previously learned skills in new situations at the level
humans can. These results enforce STARLING's potential to serve as a sandbox
environment for further research in self-supervised text-based RL. | cs.LG | STARLING: Self-supervised Training of Text-based Reinforcement Learning Agent with Large Language Models |
2024-06-09T17:55:55Z | http://arxiv.org/pdf/2406.05870v1 | http://arxiv.org/abs/2406.05870v1 | Avital Shafran, Roei Schuster, Vitaly Shmatikov | 2024-06-09T17:55:55Z | cs.CR, cs.CL, cs.LG | Retrieval-augmented generation (RAG) systems respond to queries by retrieving
relevant documents from a knowledge database, then generating an answer by
applying an LLM to the retrieved documents.
We demonstrate that RAG systems that operate on databases with potentially
untrusted content are vulnerable to a new class of denial-of-service attacks we
call jamming. An adversary can add a single ``blocker'' document to the
database that will be retrieved in response to a specific query and,
furthermore, result in the RAG system not answering the query - ostensibly
because it lacks the information or because the answer is unsafe.
We describe and analyze several methods for generating blocker documents,
including a new method based on black-box optimization that does not require
the adversary to know the embedding or LLM used by the target RAG system, nor
access to an auxiliary LLM to generate blocker documents. We measure the
efficacy of the considered methods against several LLMs and embeddings, and
demonstrate that the existing safety metrics for LLMs do not capture their
vulnerability to jamming. We then discuss defenses against blocker documents. | cs.CR | Machine Against the RAG: Jamming Retrieval-Augmented Generation with Blocker Documents |
2024-06-09T17:25:47Z | http://arxiv.org/pdf/2406.05862v2 | http://arxiv.org/abs/2406.05862v2 | Ziqiang Liu, Feiteng Fang, Xi Feng, Xinrun Du, Chenhao Zhang, Zekun Wang, Yuelin Bai, Qixuan Zhao, Liyang Fan, Chengguang Gan, Hongquan Lin, Jiaming Li, Yuansheng Ni, Haihong Wu, Yaswanth Narsupalli, Zhigang Zheng, Chengming Li, Xiping Hu, Ruifeng Xu, Xiaojun Chen, Min Yang, Jiaheng Liu, Ruibo Liu, Wenhao Huang, Ge Zhang, Shiwen Ni | 2024-06-11T12:33:42Z | cs.CL, cs.AI, cs.CV | The rapid advancements in the development of multimodal large language models
(MLLMs) have consistently led to new breakthroughs on various benchmarks. In
response, numerous challenging and comprehensive benchmarks have been proposed
to more accurately assess the capabilities of MLLMs. However, there is a dearth
of exploration of the higher-order perceptual capabilities of MLLMs. To fill
this gap, we propose the Image Implication understanding Benchmark, II-Bench,
which aims to evaluate the model's higher-order perception of images. Through
extensive experiments on II-Bench across multiple MLLMs, we have made
significant findings. Initially, a substantial gap is observed between the
performance of MLLMs and humans on II-Bench. The pinnacle accuracy of MLLMs
attains 74.8%, whereas human accuracy averages 90%, peaking at an impressive
98%. Subsequently, MLLMs perform worse on abstract and complex images,
suggesting limitations in their ability to understand high-level semantics and
capture image details. Finally, it is observed that most models exhibit
enhanced accuracy when image sentiment polarity hints are incorporated into the
prompts. This observation underscores a notable deficiency in their inherent
understanding of image sentiment. We believe that II-Bench will inspire the
community to develop the next generation of MLLMs, advancing the journey
towards expert artificial general intelligence (AGI). II-Bench is publicly
available at https://huggingface.co/datasets/m-a-p/II-Bench. | cs.CL | II-Bench: An Image Implication Understanding Benchmark for Multimodal Large Language Models |
2024-06-09T16:33:28Z | http://arxiv.org/pdf/2406.05845v1 | http://arxiv.org/abs/2406.05845v1 | Juraj Vladika, Phillip Schneider, Florian Matthes | 2024-06-09T16:33:28Z | cs.CL | In recent years, Large Language Models (LLMs) have demonstrated an impressive
ability to encode knowledge during pre-training on large text corpora. They can
leverage this knowledge for downstream tasks like question answering (QA), even
in complex areas involving health topics. Considering their high potential for
facilitating clinical work in the future, understanding the quality of encoded
medical knowledge and its recall in LLMs is an important step forward. In this
study, we examine the capability of LLMs to exhibit medical knowledge recall by
constructing a novel dataset derived from systematic reviews -- studies
synthesizing evidence-based answers for specific medical questions. Through
experiments on the new MedREQAL dataset, comprising question-answer pairs
extracted from rigorous systematic reviews, we assess six LLMs, such as GPT and
Mixtral, analyzing their classification and generation performance. Our
experimental insights into LLM performance on the novel biomedical QA dataset
reveal the still challenging nature of this task. | cs.CL | MedREQAL: Examining Medical Knowledge Recall of Large Language Models via Question Answering |
2024-06-09T16:25:14Z | http://arxiv.org/pdf/2406.06641v1 | http://arxiv.org/abs/2406.06641v1 | Yun Bai, Simon Camal, Andrea Michiorri | 2024-06-09T16:25:14Z | cs.CL, cs.LG | The relationship between energy demand and variables such as economic
activity and weather is well established. However, this paper aims to explore
the connection between energy demand and other social aspects, which receive
little attention. Through the use of natural language processing on a large
news corpus, we shed light on this important link. This study was carried out
in five regions of the UK and Ireland and considers multiple horizons from 1 to
30 days. It also considers economic variables such as GDP, unemployment and
inflation. We found that: 1) News about military conflicts, transportation, the
global pandemic, regional economics, and the international energy market are
related to electricity demand. 2) Economic indicators are more important in the
East Midlands and Northern Ireland, while social indicators are more useful in
the West Midlands and the South West of England. 3) The use of these indices
improved forecasting performance by up to 9%. | cs.CL | Investigation of the Impact of Economic and Social Factors on Energy Demand through Natural Language Processing |
2024-06-09T15:00:28Z | http://arxiv.org/pdf/2406.05814v1 | http://arxiv.org/abs/2406.05814v1 | Leigang Qu, Haochuan Li, Tan Wang, Wenjie Wang, Yongqi Li, Liqiang Nie, Tat-Seng Chua | 2024-06-09T15:00:28Z | cs.CV, cs.AI, cs.CL, cs.LG, cs.MM | How humans can efficiently and effectively acquire images has always been a
perennial question. A typical solution is text-to-image retrieval from an
existing database given the text query; however, the limited database typically
lacks creativity. By contrast, recent breakthroughs in text-to-image generation
have made it possible to produce fancy and diverse visual content, but it faces
challenges in synthesizing knowledge-intensive images. In this work, we rethink
the relationship between text-to-image generation and retrieval and propose a
unified framework in the context of Multimodal Large Language Models (MLLMs).
Specifically, we first explore the intrinsic discriminative abilities of MLLMs
and introduce a generative retrieval method to perform retrieval in a
training-free manner. Subsequently, we unify generation and retrieval in an
autoregressive generation way and propose an autonomous decision module to
choose the best-matched one between generated and retrieved images as the
response to the text query. Additionally, we construct a benchmark called
TIGeR-Bench, including creative and knowledge-intensive domains, to standardize
the evaluation of unified text-to-image generation and retrieval. Extensive
experimental results on TIGeR-Bench and two retrieval benchmarks, i.e.,
Flickr30K and MS-COCO, demonstrate the superiority and effectiveness of our
proposed method. | cs.CV | Unified Text-to-Image Generation and Retrieval |
2024-06-09T14:54:22Z | http://arxiv.org/pdf/2406.05812v1 | http://arxiv.org/abs/2406.05812v1 | Shraboni Sarker, Ahmad Tamim Hamad, Hulayyil Alshammari, Viviana Grieco, Praveen Rao | 2024-06-09T14:54:22Z | cs.CL, cs.AI | Large language models have gained tremendous popularity in domains such as
e-commerce, finance, healthcare, and education. Fine-tuning is a common
approach to customize an LLM on a domain-specific dataset for a desired
downstream task. In this paper, we present a valuable resource for fine-tuning
LLMs developed for the Spanish language to perform a variety of tasks such as
classification, masked language modeling, clustering, and others. Our resource
is a collection of handwritten notary records from the seventeenth century
obtained from the National Archives of Argentina. This collection contains a
combination of original images and transcribed text (and metadata) of 160+
pages that were handwritten by two notaries, namely, Estenban Agreda de Vergara
and Nicolas de Valdivia y Brisuela nearly 400 years ago. Through empirical
evaluation, we demonstrate that our collection can be used to fine-tune Spanish
LLMs for tasks such as classification and masked language modeling, and can
outperform pre-trained Spanish models and ChatGPT-3.5/ChatGPT-4o. Our resource
will be an invaluable resource for historical text analysis and is publicly
available on GitHub. | cs.CL | Seventeenth-Century Spanish American Notary Records for Fine-Tuning Spanish Large Language Models |
2024-06-09T14:44:59Z | http://arxiv.org/pdf/2406.05806v1 | http://arxiv.org/abs/2406.05806v1 | Chih-Kai Yang, Kuan-Po Huang, Hung-yi Lee | 2024-06-09T14:44:59Z | cs.CL, cs.SD, eess.AS | This research explores the interaction between Whisper, a high-performing
speech recognition model, and information in prompts. Our results unexpectedly
show that Whisper may not fully grasp textual prompts as anticipated.
Additionally, we find that performance improvement is not guaranteed even with
stronger adherence to the topic information in textual prompts. It is also
noted that English prompts generally outperform Mandarin ones on datasets of
both languages, likely due to differences in training data distributions for
these languages. Conversely, we discover that Whisper exhibits awareness of
misleading information in language tokens by effectively ignoring incorrect
language tokens and focusing on the correct ones. In summary, this work raises
questions about Whisper's prompt understanding capability and encourages
further studies. | cs.CL | Do Prompts Really Prompt? Exploring the Prompt Understanding Capability of Whisper |
2024-06-09T14:42:55Z | http://arxiv.org/pdf/2406.05804v2 | http://arxiv.org/abs/2406.05804v2 | Xinzhe Li | 2024-06-16T00:59:27Z | cs.AI, cs.CL, cs.SE | Recent advancements in Large Language Models (LLMs) have catalyzed the
development of sophisticated frameworks for developing LLM-based agents.
However, the complexity of these frameworks r poses a hurdle for nuanced
differentiation at a granular level, a critical aspect for enabling efficient
implementations across different frameworks and fostering future research.
Hence, the primary purpose of this survey is to facilitate a cohesive
understanding of diverse recently proposed frameworks by identifying common
workflows and reusable LLM-Profiled Components (LMPCs). | cs.AI | A Survey on LLM-Based Agents: Common Workflows and Reusable LLM-Profiled Components |
2024-06-09T14:25:09Z | http://arxiv.org/pdf/2406.05798v1 | http://arxiv.org/abs/2406.05798v1 | Stephen Fitz, Peter Romero, Jiyan Jonas Schneider | 2024-06-09T14:25:09Z | cs.CL, cs.AI, cs.NE | We explore the topology of representation manifolds arising in autoregressive
neural language models trained on raw text data. In order to study their
properties, we introduce tools from computational algebraic topology, which we
use as a basis for a measure of topological complexity, that we call
perforation.
Using this measure, we study the evolution of topological structure in GPT
based large language models across depth and time during training. We then
compare these to gated recurrent models, and show that the latter exhibit more
topological complexity, with a distinct pattern of changes common to all
natural languages but absent from synthetically generated data. The paper
presents a detailed analysis of the representation manifolds derived by these
models based on studying the shapes of vector clouds induced by them as they
are conditioned on sentences from corpora of natural language text.
The methods developed in this paper are novel in the field and based on
mathematical apparatus that might be unfamiliar to the target audience. To help
with that we introduce the minimum necessary theory, and provide additional
visualizations in the appendices.
The main contribution of the paper is a striking observation about the
topological structure of the transformer as compared to LSTM based neural
architectures. It suggests that further research into mathematical properties
of these neural networks is necessary to understand the operation of large
transformer language models. We hope this work inspires further explorations in
this direction within the NLP community. | cs.CL | Hidden Holes: topological aspects of language models |
2024-06-09T14:20:55Z | http://arxiv.org/pdf/2406.05797v1 | http://arxiv.org/abs/2406.05797v1 | Qizhi Pei, Lijun Wu, Kaiyuan Gao, Jinhua Zhu, Rui Yan | 2024-06-09T14:20:55Z | q-bio.BM, cs.AI, cs.CE, cs.CL, cs.LG | The integration of molecule and language has garnered increasing attention in
molecular science. Recent advancements in Language Models (LMs) have
demonstrated potential for the comprehensive modeling of molecule and language.
However, existing works exhibit notable limitations. Most existing works
overlook the modeling of 3D information, which is crucial for understanding
molecular structures and also functions. While some attempts have been made to
leverage external structure encoding modules to inject the 3D molecular
information into LMs, there exist obvious difficulties that hinder the
integration of molecular structure and language text, such as modality
alignment and separate tuning. To bridge this gap, we propose 3D-MolT5, a
unified framework designed to model both 1D molecular sequence and 3D molecular
structure. The key innovation lies in our methodology for mapping fine-grained
3D substructure representations (based on 3D molecular fingerprints) to a
specialized 3D token vocabulary for 3D-MolT5. This 3D structure token
vocabulary enables the seamless combination of 1D sequence and 3D structure
representations in a tokenized format, allowing 3D-MolT5 to encode molecular
sequence (SELFIES), molecular structure, and text sequences within a unified
architecture. Alongside, we further introduce 1D and 3D joint pre-training to
enhance the model's comprehension of these diverse modalities in a joint
representation space and better generalize to various tasks for our foundation
model. Through instruction tuning on multiple downstream datasets, our proposed
3D-MolT5 shows superior performance than existing methods in molecular property
prediction, molecule captioning, and text-based molecule generation tasks. Our
code will be available on GitHub soon. | q-bio.BM | 3D-MolT5: Towards Unified 3D Molecule-Text Modeling with 3D Molecular Tokenization |
2024-06-09T14:11:19Z | http://arxiv.org/pdf/2406.05794v2 | http://arxiv.org/abs/2406.05794v2 | Kiseung Kim, Jay-Yoon Lee | 2024-06-16T13:28:24Z | cs.CL, cs.AI | The Retrieval Augmented Generation (RAG) framework utilizes a combination of
parametric knowledge and external knowledge to demonstrate state-of-the-art
performance on open-domain question answering tasks. However, the RAG framework
suffers from performance degradation when the query is accompanied by
irrelevant contexts. In this work, we propose the RE-RAG framework, which
introduces a relevance estimator (RE) that not only provides relative relevance
between contexts as previous rerankers did, but also provides confidence, which
can be used to classify whether given context is useful for answering the given
question. We propose a weakly supervised method for training the RE simply
utilizing question-answer data without any labels for correct contexts. We show
that RE trained with a small generator (sLM) can not only improve the sLM
fine-tuned together with RE but also improve previously unreferenced large
language models (LLMs). Furthermore, we investigate new decoding strategies
that utilize the proposed confidence measured by RE such as choosing to let the
user know that it is "unanswerable" to answer the question given the retrieved
contexts or choosing to rely on LLM's parametric knowledge rather than
unrelated contexts. | cs.CL | RE-RAG: Improving Open-Domain QA Performance and Interpretability with Relevance Estimator in Retrieval-Augmented Generation |
2024-06-09T12:41:14Z | http://arxiv.org/pdf/2406.05766v1 | http://arxiv.org/abs/2406.05766v1 | Zijia Song, Zelin Zang, Yelin Wang, Guozheng Yang, Jiangbin Zheng, Kaicheng yu, Wanyu Chen, Stan Z. Li | 2024-06-09T12:41:14Z | cs.LG, cs.AI, cs.CL, cs.CV | Multimodal fusion breaks through the barriers between diverse modalities and
has already yielded numerous impressive performances. However, in various
specialized fields, it is struggling to obtain sufficient alignment data for
the training process, which seriously limits the use of previously elegant
models. Thus, semi-supervised learning attempts to achieve multimodal alignment
with fewer matched pairs but traditional methods like pseudo-labeling are
difficult to apply in domains with no label information. To address these
problems, we transform semi-supervised multimodal alignment into a manifold
matching problem and propose a new method based on CLIP, named Gentle-CLIP.
Specifically, we design a novel semantic density distribution loss to explore
implicit semantic alignment information from unpaired multimodal data by
constraining the latent representation distribution with fine granularity, thus
eliminating the need for numerous strictly matched pairs. Meanwhile, we
introduce multi-kernel maximum mean discrepancy as well as self-supervised
contrastive loss to pull separate modality distributions closer and enhance the
stability of the representation distribution. In addition, the contrastive loss
used in CLIP is employed on the supervised matched data to prevent negative
optimization. Extensive experiments conducted on a range of tasks in various
fields, including protein, remote sensing, and the general vision-language
field, demonstrate the effectiveness of our proposed Gentle-CLIP. | cs.LG | Gentle-CLIP: Exploring Aligned Semantic In Low-Quality Multimodal Data With Soft Alignment |
2024-06-09T12:30:30Z | http://arxiv.org/pdf/2406.05761v1 | http://arxiv.org/abs/2406.05761v1 | Seungone Kim, Juyoung Suk, Ji Yong Cho, Shayne Longpre, Chaeeun Kim, Dongkeun Yoon, Guijin Son, Yejin Cho, Sheikh Shafayat, Jinheon Baek, Sue Hyun Park, Hyeonbin Hwang, Jinkyung Jo, Hyowon Cho, Haebin Shin, Seongyun Lee, Hanseok Oh, Noah Lee, Namgyu Ho, Se June Joo, Miyoung Ko, Yoonjoo Lee, Hyungjoo Chae, Jamin Shin, Joel Jang, Seonghyeon Ye, Bill Yuchen Lin, Sean Welleck, Graham Neubig, Moontae Lee, Kyungjae Lee, Minjoon Seo | 2024-06-09T12:30:30Z | cs.CL | As language models (LMs) become capable of handling a wide range of tasks,
their evaluation is becoming as challenging as their development. Most
generation benchmarks currently assess LMs using abstract evaluation criteria
like helpfulness and harmlessness, which often lack the flexibility and
granularity of human assessment. Additionally, these benchmarks tend to focus
disproportionately on specific capabilities such as instruction following,
leading to coverage bias. To overcome these limitations, we introduce the
BiGGen Bench, a principled generation benchmark designed to thoroughly evaluate
nine distinct capabilities of LMs across 77 diverse tasks. A key feature of the
BiGGen Bench is its use of instance-specific evaluation criteria, closely
mirroring the nuanced discernment of human evaluation. We apply this benchmark
to assess 103 frontier LMs using five evaluator LMs. Our code, data, and
evaluation results are all publicly available at
https://github.com/prometheus-eval/prometheus-eval/tree/main/BiGGen-Bench. | cs.CL | The BiGGen Bench: A Principled Benchmark for Fine-grained Evaluation of Language Models with Language Models |
2024-06-09T12:29:55Z | http://arxiv.org/pdf/2406.05760v1 | http://arxiv.org/abs/2406.05760v1 | Salman Elgamal, Ossama Obeid, Tameem Kabbani, Go Inoue, Nizar Habash | 2024-06-09T12:29:55Z | cs.CL | The widespread absence of diacritical marks in Arabic text poses a
significant challenge for Arabic natural language processing (NLP). This paper
explores instances of naturally occurring diacritics, referred to as
"diacritics in the wild," to unveil patterns and latent information across six
diverse genres: news articles, novels, children's books, poetry, political
documents, and ChatGPT outputs. We present a new annotated dataset that maps
real-world partially diacritized words to their maximal full diacritization in
context. Additionally, we propose extensions to the analyze-and-disambiguate
approach in Arabic NLP to leverage these diacritics, resulting in notable
improvements. Our contributions encompass a thorough analysis, valuable
datasets, and an extended diacritization algorithm. We release our code and
datasets as open source. | cs.CL | Arabic Diacritics in the Wild: Exploiting Opportunities for Improved Diacritization |
2024-06-09T12:23:14Z | http://arxiv.org/pdf/2406.05756v1 | http://arxiv.org/abs/2406.05756v1 | Mengfei Du, Binhao Wu, Zejun Li, Xuanjing Huang, Zhongyu Wei | 2024-06-09T12:23:14Z | cs.AI, cs.CL, cs.CV, cs.MM | The recent rapid development of Large Vision-Language Models (LVLMs) has
indicated their potential for embodied tasks.However, the critical skill of
spatial understanding in embodied environments has not been thoroughly
evaluated, leaving the gap between current LVLMs and qualified embodied
intelligence unknown. Therefore, we construct EmbSpatial-Bench, a benchmark for
evaluating embodied spatial understanding of LVLMs.The benchmark is
automatically derived from embodied scenes and covers 6 spatial relationships
from an egocentric perspective.Experiments expose the insufficient capacity of
current LVLMs (even GPT-4V). We further present EmbSpatial-SFT, an
instruction-tuning dataset designed to improve LVLMs' embodied spatial
understanding. | cs.AI | EmbSpatial-Bench: Benchmarking Spatial Understanding for Embodied Tasks with Large Vision-Language Models |
2024-06-09T11:00:01Z | http://arxiv.org/pdf/2406.05733v1 | http://arxiv.org/abs/2406.05733v1 | Danupat Khamnuansin, Tawunrat Chalothorn, Ekapol Chuangsuwanich | 2024-06-09T11:00:01Z | cs.CL | Large Language Models (LLMs) often struggle with hallucinations and outdated
information. To address this, Information Retrieval (IR) systems can be
employed to augment LLMs with up-to-date knowledge. However, existing IR
techniques contain deficiencies, posing a performance bottleneck. Given the
extensive array of IR systems, combining diverse approaches presents a viable
strategy. Nevertheless, prior attempts have yielded restricted efficacy. In
this work, we propose an approach that leverages learning-to-rank techniques to
combine heterogeneous IR systems. We demonstrate the method on two Retrieval
Question Answering (ReQA) tasks. Our empirical findings exhibit a significant
performance enhancement, outperforming previous approaches and achieving
state-of-the-art results on ReQA SQuAD. | cs.CL | MrRank: Improving Question Answering Retrieval System through Multi-Result Ranking Model |
2024-06-09T09:51:55Z | http://arxiv.org/pdf/2406.05707v1 | http://arxiv.org/abs/2406.05707v1 | Weiping Fu, Bifan Wei, Jianxiang Hu, Zhongmin Cai, Jun Liu | 2024-06-09T09:51:55Z | cs.CL, cs.AI | Automatically generated questions often suffer from problems such as unclear
expression or factual inaccuracies, requiring a reliable and comprehensive
evaluation of their quality. Human evaluation is frequently used in the field
of question generation (QG) and is one of the most accurate evaluation methods.
It also serves as the standard for automatic metrics. However, there is a lack
of unified evaluation criteria, which hampers the development of both QG
technologies and automatic evaluation methods. To address this, we propose
QGEval, a multi-dimensional Evaluation benchmark for Question Generation, which
evaluates both generated questions and existing automatic metrics across 7
dimensions: fluency, clarity, conciseness, relevance, consistency,
answerability, and answer consistency. We demonstrate the appropriateness of
these dimensions by examining their correlations and distinctions. Analysis
with QGEval reveals that 1) most QG models perform unsatisfactorily in terms of
answerability and answer consistency, and 2) existing metrics fail to align
well with human assessments when evaluating generated questions across the 7
dimensions. We expect this work to foster the development of both QG
technologies and automatic metrics for QG. | cs.CL | QGEval: A Benchmark for Question Generation Evaluation |
2024-06-09T09:03:11Z | http://arxiv.org/pdf/2406.06636v1 | http://arxiv.org/abs/2406.06636v1 | Gony Rosenman, Lior Wolf, Talma Hendler | 2024-06-09T09:03:11Z | cs.CL, cs.LG, 68T50, I.2.7 | We employ a Large Language Model (LLM) to convert unstructured psychological
interviews into structured questionnaires spanning various psychiatric and
personality domains. The LLM is prompted to answer these questionnaires by
impersonating the interviewee. The obtained answers are coded as features,
which are used to predict standardized psychiatric measures of depression
(PHQ-8) and PTSD (PCL-C), using a Random Forest regressor. Our approach is
shown to enhance diagnostic accuracy compared to multiple baselines. It thus
establishes a novel framework for interpreting unstructured psychological
interviews, bridging the gap between narrative-driven and data-driven
approaches for mental health assessment. | cs.CL | LLM Questionnaire Completion for Automatic Psychiatric Assessment |
2024-06-09T08:31:14Z | http://arxiv.org/pdf/2406.05690v1 | http://arxiv.org/abs/2406.05690v1 | Yan Ma, Yu Qiao, Pengfei Liu | 2024-06-09T08:31:14Z | cs.CL | A story premise succinctly defines a story's main idea, foundation, and
trajectory. It serves as the initial trigger in automatic story generation.
Existing sources of story premises are limited by a lack of diversity, uneven
quality, and high costs that make them difficult to scale. In response, we
introduce Modular Story Premise Synthesis (MoPS) which breaks down story
premises into modules like background and persona for automated design and
generation. MoPS consists of three phases: (1) Precollect a consistent set of
candidates for each module to form a nested dictionary. (2) Extract a key path
from the nested dictionary as the premise design. (3) Instruct an LLM to
integrate the design into a coherent premise sentence. Thorough evaluations
demonstrate that our synthesized premises excel in diversity, fascination,
completeness, and originality compared to those induced from large language
models and captured from public story datasets. Similarly, the extended novels
and scripts generated from our premises also exhibit higher quality. In
supplementary materials, we provide the MoPS code suite, along with 7.6k
generated premises and 1k extended stories. Code:
https://github.com/GAIR-NLP/MoPS. | cs.CL | MoPS: Modular Story Premise Synthesis for Open-Ended Automatic Story Generation |
2024-06-09T08:24:17Z | http://arxiv.org/pdf/2406.05688v1 | http://arxiv.org/abs/2406.05688v1 | Cheng Tan, Dongxin Lyu, Siyuan Li, Zhangyang Gao, Jingxuan Wei, Siqi Ma, Zicheng Liu, Stan Z. Li | 2024-06-09T08:24:17Z | cs.CL, cs.AI, cs.LG | Large Language Models (LLMs) have demonstrated wide-ranging applications
across various fields and have shown significant potential in the academic
peer-review process. However, existing applications are primarily limited to
static review generation based on submitted papers, which fail to capture the
dynamic and iterative nature of real-world peer reviews. In this paper, we
reformulate the peer-review process as a multi-turn, long-context dialogue,
incorporating distinct roles for authors, reviewers, and decision makers. We
construct a comprehensive dataset containing over 26,841 papers with 92,017
reviews collected from multiple sources, including the top-tier conference and
prestigious journal. This dataset is meticulously designed to facilitate the
applications of LLMs for multi-turn dialogues, effectively simulating the
complete peer-review process. Furthermore, we propose a series of metrics to
evaluate the performance of LLMs for each role under this reformulated
peer-review setting, ensuring fair and comprehensive evaluations. We believe
this work provides a promising perspective on enhancing the LLM-driven
peer-review process by incorporating dynamic, role-based interactions. It
aligns closely with the iterative and interactive nature of real-world academic
peer review, offering a robust foundation for future research and development
in this area. We open-source the dataset at
https://github.com/chengtan9907/ReviewMT. | cs.CL | Peer Review as A Multi-Turn and Long-Context Dialogue with Role-Based Interactions |
2024-06-09T07:23:34Z | http://arxiv.org/pdf/2406.05678v1 | http://arxiv.org/abs/2406.05678v1 | Hengyu Zhang | 2024-06-09T07:23:34Z | cs.CL | Extending the functionality of the Transformer model to accommodate longer
sequence lengths has become a critical challenge. This extension is crucial not
only for improving tasks such as language translation and long-context
processing but also for enabling novel applications like chatbots, code
generation, and multimedia content creation. The primary obstacle is the
self-attention mechanism, which scales quadratically with sequence length in
terms of computation time and memory requirements. LongLoRA proposed shifted
sparse attention (S\(^2\)-Attn), effectively enabling context extension and
leading to non-trivial computation savings with similar performance to
fine-tuning with vanilla attention. However, LongLoRA is still not as efficient
as vanilla attention, reaching only 39\% of the perplexity improvement compared
to full attention. This inefficiency is due to the cyclic shift applied within
different attention head patterns, causing either chaos in the attention head
structure or unnecessary information exchange between token groups. To address
these issues, We propose \textbf{SinkLoRA}, which features better work
partitioning. Specifically, (1) we developed SF-Attn with a segmentation and
reassembly algorithm to proportionally return cyclically shifted groups of
attention heads to their un-shifted state together with global attention of
"sink attention tokens", achieving 92\% of the perplexity improvement compared
to full attention after fine tuning, and (2) applied a SOTA KV cache
compression algorithm H$_2$O to accelerate inference. Furthermore, We conducted
supervised fine-tuning with SinkLoRA using a self collected LongAlpaca-plus
dataset. All our code, models, datasets, and demos are available at
\url{https://github.com/Dexter-GT-86/SinkLoRA}. | cs.CL | SinkLoRA: Enhanced Efficiency and Chat Capabilities for Long-Context Large Language Models |
2024-06-09T07:06:58Z | http://arxiv.org/pdf/2406.05673v2 | http://arxiv.org/abs/2406.05673v2 | Fangxu Yu, Lai Jiang, Haoqiang Kang, Shibo Hao, Lianhui Qin | 2024-06-24T15:49:09Z | cs.AI, cs.CL | Divergent thinking, the cognitive process of generating diverse solutions, is
a hallmark of human creativity and problem-solving. For machines, sampling
diverse solution trajectories in complex reasoning problems is crucial for
robust outcomes, data augmentation, and enhanced model generalization. Large
language models (LLMs) often struggle with generating high-quality, diverse
reasoning. While supervised fine-tuning helps with quality, it requires
extensive supervision data to capture the full diversity of solutions.
Alternatively, reinforcement learning methods like PPO aim to find limited
highest-reward solutions while neglecting the solution diversity, akin to
convergent thinking. To address these limitations, we propose Flow of Reasoning
(FoR) -- an efficient LLM training approach enabling diverse reasoning with
minimal data. FoR formulates multi-step LLM reasoning as a Markovian flow from
an initial state to terminal states. The formulation allows to adapt principled
GFlowNet approaches to train the LLM as a policy, which is able to sample
multiple reasoning paths with probabilities proportional to the unnormalized
reward. Empirical results show that, with limited training data (e.g., 15
examples), FoR can discover diverse high-quality solutions that excel greatly
beyond current state-of-the-art methods across three tasks, including embodied
reasoning (BlocksWorld), math puzzle solving (Game24), and logical reasoning
(PrOntoQA). Code is available at https://github.com/Yu-Fangxu/FoR. | cs.AI | Flow of Reasoning: Efficient Training of LLM Policy with Divergent Thinking |
2024-06-09T06:30:28Z | http://arxiv.org/pdf/2406.05661v1 | http://arxiv.org/abs/2406.05661v1 | Hemant Yadav, Sunayana Sitaram, Rajiv Ratn Shah | 2024-06-09T06:30:28Z | cs.CL | In recent years, self-supervised pre-training methods have gained significant
traction in learning high-level information from raw speech. Among these
methods, HuBERT has demonstrated SOTA performance in automatic speech
recognition (ASR). However, HuBERT's performance lags behind data2vec due to
disparities in pre-training strategies. In this paper, we propose (i) a Swap
method to address pre-training and inference mismatch observed in HuBERT and
(ii) incorporates Multicluster masked prediction loss for more effective
utilization of the models capacity. The resulting method is, MS-HuBERT, an
end-to-end self-supervised pre-training method for learning robust speech
representations. It beats vanilla HuBERT on the ASR Librispeech benchmark on
average by a 5% margin when evaluated on different finetuning splits.
Additionally, we demonstrate that the learned embeddings obtained during
pre-training encode essential information for improving performance of content
based tasks such as ASR. | cs.CL | MS-HuBERT: Mitigating Pre-training and Inference Mismatch in Masked Language Modelling methods for learning Speech Representations |
2024-06-09T05:57:59Z | http://arxiv.org/pdf/2406.05659v1 | http://arxiv.org/abs/2406.05659v1 | Maryam Amirizaniani, Elias Martin, Maryna Sivachenko, Afra Mashhadi, Chirag Shah | 2024-06-09T05:57:59Z | cs.CL, cs.AI | Theory of Mind (ToM) reasoning entails recognizing that other individuals
possess their own intentions, emotions, and thoughts, which is vital for
guiding one's own thought processes. Although large language models (LLMs)
excel in tasks such as summarization, question answering, and translation, they
still face challenges with ToM reasoning, especially in open-ended questions.
Despite advancements, the extent to which LLMs truly understand ToM reasoning
and how closely it aligns with human ToM reasoning remains inadequately
explored in open-ended scenarios. Motivated by this gap, we assess the
abilities of LLMs to perceive and integrate human intentions and emotions into
their ToM reasoning processes within open-ended questions. Our study utilizes
posts from Reddit's ChangeMyView platform, which demands nuanced social
reasoning to craft persuasive responses. Our analysis, comparing semantic
similarity and lexical overlap metrics between responses generated by humans
and LLMs, reveals clear disparities in ToM reasoning capabilities in open-ended
questions, with even the most advanced models showing notable limitations. To
enhance LLM capabilities, we implement a prompt tuning method that incorporates
human intentions and emotions, resulting in improvements in ToM reasoning
performance. However, despite these improvements, the enhancement still falls
short of fully achieving human-like reasoning. This research highlights the
deficiencies in LLMs' social reasoning and demonstrates how integrating human
intentions and emotions can boost their effectiveness. | cs.CL | Do LLMs Exhibit Human-Like Reasoning? Evaluating Theory of Mind in LLMs for Open-Ended Responses |
2024-06-09T05:33:51Z | http://arxiv.org/pdf/2406.05654v2 | http://arxiv.org/abs/2406.05654v2 | Shuting Wang, Jiongnan Liu, Shiren Song, Jiehan Cheng, Yuqi Fu, Peidong Guo, Kun Fang, Yutao Zhu, Zhicheng Dou | 2024-06-17T03:01:39Z | cs.CL, cs.IR | Retrieval-Augmented Generation (RAG) offers a promising solution to address
various limitations of Large Language Models (LLMs), such as hallucination and
difficulties in keeping up with real-time updates. This approach is
particularly critical in expert and domain-specific applications where LLMs
struggle to cover expert knowledge. Therefore, evaluating RAG models in such
scenarios is crucial, yet current studies often rely on general knowledge
sources like Wikipedia to assess the models' abilities in solving common-sense
problems. In this paper, we evaluated LLMs by RAG settings in a domain-specific
context, college enrollment. We identified six required abilities for RAG
models, including the ability in conversational RAG, analyzing structural
information, faithfulness to external knowledge, denoising, solving
time-sensitive problems, and understanding multi-document interactions. Each
ability has an associated dataset with shared corpora to evaluate the RAG
models' performance. We evaluated popular LLMs such as Llama, Baichuan,
ChatGLM, and GPT models. Experimental results indicate that existing
closed-book LLMs struggle with domain-specific questions, highlighting the need
for RAG models to solve expert problems. Moreover, there is room for RAG models
to improve their abilities in comprehending conversational history, analyzing
structural information, denoising, processing multi-document interactions, and
faithfulness in expert knowledge. We expect future studies could solve these
problems better. | cs.CL | DomainRAG: A Chinese Benchmark for Evaluating Domain-specific Retrieval-Augmented Generation |
2024-06-09T05:26:38Z | http://arxiv.org/pdf/2406.05651v1 | http://arxiv.org/abs/2406.05651v1 | Xiangrui Kong, Thomas Braunl, Marco Fahmi, Yue Wang | 2024-06-09T05:26:38Z | cs.RO, cs.CL, cs.CV | Over the last year, significant advancements have been made in the realms of
large language models (LLMs) and multi-modal large language models (MLLMs),
particularly in their application to autonomous driving. These models have
showcased remarkable abilities in processing and interacting with complex
information. In autonomous driving, LLMs and MLLMs are extensively used,
requiring access to sensitive vehicle data such as precise locations, images,
and road conditions. These data are transmitted to an LLM-based inference cloud
for advanced analysis. However, concerns arise regarding data security, as the
protection against data and privacy breaches primarily depends on the LLM's
inherent security measures, without additional scrutiny or evaluation of the
LLM's inference outputs. Despite its importance, the security aspect of LLMs in
autonomous driving remains underexplored. Addressing this gap, our research
introduces a novel security framework for autonomous vehicles, utilizing a
multi-agent LLM approach. This framework is designed to safeguard sensitive
information associated with autonomous vehicles from potential leaks, while
also ensuring that LLM outputs adhere to driving regulations and align with
human values. It includes mechanisms to filter out irrelevant queries and
verify the safety and reliability of LLM outputs. Utilizing this framework, we
evaluated the security, privacy, and cost aspects of eleven large language
model-driven autonomous driving cues. Additionally, we performed QA tests on
these driving prompts, which successfully demonstrated the framework's
efficacy. | cs.RO | A Superalignment Framework in Autonomous Driving with Large Language Models |
2024-06-09T05:04:37Z | http://arxiv.org/pdf/2406.05644v2 | http://arxiv.org/abs/2406.05644v2 | Zhenhong Zhou, Haiyang Yu, Xinghua Zhang, Rongwu Xu, Fei Huang, Yongbin Li | 2024-06-13T05:39:31Z | cs.CL, cs.AI, cs.CR, cs.CY | Large language models (LLMs) rely on safety alignment to avoid responding to
malicious user inputs. Unfortunately, jailbreak can circumvent safety
guardrails, resulting in LLMs generating harmful content and raising concerns
about LLM safety. Due to language models with intensive parameters often
regarded as black boxes, the mechanisms of alignment and jailbreak are
challenging to elucidate. In this paper, we employ weak classifiers to explain
LLM safety through the intermediate hidden states. We first confirm that LLMs
learn ethical concepts during pre-training rather than alignment and can
identify malicious and normal inputs in the early layers. Alignment actually
associates the early concepts with emotion guesses in the middle layers and
then refines them to the specific reject tokens for safe generations. Jailbreak
disturbs the transformation of early unethical classification into negative
emotions. We conduct experiments on models from 7B to 70B across various model
families to prove our conclusion. Overall, our paper indicates the intrinsical
mechanism of LLM safety and how jailbreaks circumvent safety guardrails,
offering a new perspective on LLM safety and reducing concerns. Our code is
available at https://github.com/ydyjya/LLM-IHS-Explanation. | cs.CL | How Alignment and Jailbreak Work: Explain LLM Safety through Intermediate Hidden States |
2024-06-09T03:38:21Z | http://arxiv.org/pdf/2406.05629v1 | http://arxiv.org/abs/2406.05629v1 | Mark Hamilton, Andrew Zisserman, John R. Hershey, William T. Freeman | 2024-06-09T03:38:21Z | cs.CV, cs.CL, cs.IR, cs.LG, cs.SD, eess.AS | We present DenseAV, a novel dual encoder grounding architecture that learns
high-resolution, semantically meaningful, and audio-visually aligned features
solely through watching videos. We show that DenseAV can discover the
``meaning'' of words and the ``location'' of sounds without explicit
localization supervision. Furthermore, it automatically discovers and
distinguishes between these two types of associations without supervision. We
show that DenseAV's localization abilities arise from a new multi-head feature
aggregation operator that directly compares dense image and audio
representations for contrastive learning. In contrast, many other systems that
learn ``global'' audio and video representations cannot localize words and
sound. Finally, we contribute two new datasets to improve the evaluation of AV
representations through speech and sound prompted semantic segmentation. On
these and other datasets we show DenseAV dramatically outperforms the prior art
on speech and sound prompted semantic segmentation. DenseAV outperforms the
previous state-of-the-art, ImageBind, on cross-modal retrieval using fewer than
half of the parameters. Project Page:
\href{https://aka.ms/denseav}{https://aka.ms/denseav} | cs.CV | Separating the "Chirp" from the "Chat": Self-supervised Visual Grounding of Sound and Language |
2024-06-09T03:22:55Z | http://arxiv.org/pdf/2406.05625v1 | http://arxiv.org/abs/2406.05625v1 | Zhihao Zhang, Tomas Goldsack, Carolina Scarton, Chenghua Lin | 2024-06-09T03:22:55Z | cs.CL | Lay summarisation aims to produce summaries of scientific articles that are
comprehensible to non-expert audiences. However, previous work assumes a
one-size-fits-all approach, where the content and style of the produced summary
are entirely dependent on the data used to train the model. In practice,
audiences with different levels of expertise will have specific needs,
impacting what content should appear in a lay summary and how it should be
presented. Aiming to address this, we propose ATLAS, a novel abstractive
summarisation approach that can control various properties that contribute to
the overall "layness" of the generated summary using targeted control
attributes. We evaluate ATLAS on a combination of biomedical lay summarisation
datasets, where it outperforms state-of-the-art baselines using mainstream
summarisation metrics. Additional analyses provided on the discriminatory power
and emergent influence of our selected controllable attributes further attest
to the effectiveness of our approach. | cs.CL | ATLAS: Improving Lay Summarisation with Attribute-based Control |
2024-06-09T02:36:28Z | http://arxiv.org/pdf/2406.05615v2 | http://arxiv.org/abs/2406.05615v2 | Thong Nguyen, Yi Bin, Junbin Xiao, Leigang Qu, Yicong Li, Jay Zhangjie Wu, Cong-Duy Nguyen, See-Kiong Ng, Luu Anh Tuan | 2024-07-01T16:05:01Z | cs.CL | Humans use multiple senses to comprehend the environment. Vision and language
are two of the most vital senses since they allow us to easily communicate our
thoughts and perceive the world around us. There has been a lot of interest in
creating video-language understanding systems with human-like senses since a
video-language pair can mimic both our linguistic medium and visual environment
with temporal dynamics. In this survey, we review the key tasks of these
systems and highlight the associated challenges. Based on the challenges, we
summarize their methods from model architecture, model training, and data
perspectives. We also conduct performance comparison among the methods, and
discuss promising directions for future research. | cs.CL | Video-Language Understanding: A Survey from Model Architecture, Model Training, and Data Perspectives |
2024-06-09T02:36:00Z | http://arxiv.org/pdf/2406.10249v1 | http://arxiv.org/abs/2406.10249v1 | Ming Cheung | 2024-06-09T02:36:00Z | cs.AI, cs.CL | Large language models (LLMs) have achieved remarkable performance in language
understanding and generation tasks by leveraging vast amounts of online texts.
Unlike conventional models, LLMs can adapt to new domains through prompt
engineering without the need for retraining, making them suitable for various
business functions, such as strategic planning, project implementation, and
data-driven decision-making. However, their limitations in terms of bias,
contextual understanding, and sensitivity to prompts raise concerns about their
readiness for real-world applications. This paper thoroughly examines the
usefulness and readiness of LLMs for business processes. The limitations and
capacities of LLMs are evaluated through experiments conducted on four
accessible LLMs using real-world data. The findings have significant
implications for organizations seeking to leverage generative AI and provide
valuable insights into future research directions. To the best of our
knowledge, this represents the first quantified study of LLMs applied to core
business operations and challenges. | cs.AI | A Reality check of the benefits of LLM in business |
2024-06-09T01:16:04Z | http://arxiv.org/pdf/2406.05606v1 | http://arxiv.org/abs/2406.05606v1 | Dayoon Ko, Jinyoung Kim, Hahyeon Choi, Gunhee Kim | 2024-06-09T01:16:04Z | cs.CL | In the real world, knowledge is constantly evolving, which can render
existing knowledge-based datasets outdated. This unreliability highlights the
critical need for continuous updates to ensure both accuracy and relevance in
knowledge-intensive tasks. To address this, we propose GrowOVER-QA and
GrowOVER-Dialogue, dynamic open-domain QA and dialogue benchmarks that undergo
a continuous cycle of updates, keeping pace with the rapid evolution of
knowledge. Our research indicates that retrieval-augmented language models
(RaLMs) struggle with knowledge that has not been trained on or recently
updated. Consequently, we introduce a novel retrieval-interactive language
model framework, where the language model evaluates and reflects on its answers
for further re-retrieval. Our exhaustive experiments demonstrate that our
training-free framework significantly improves upon existing methods,
performing comparably to or even surpassing continuously trained language
models. | cs.CL | GrowOVER: How Can LLMs Adapt to Growing Real-World Knowledge? |
2024-06-09T00:54:57Z | http://arxiv.org/pdf/2406.05602v1 | http://arxiv.org/abs/2406.05602v1 | Philip Wootaek Shin, Jihyun Janice Ahn, Wenpeng Yin, Jack Sampson, Vijaykrishnan Narayanan | 2024-06-09T00:54:57Z | cs.CV, cs.CL | It has been shown that many generative models inherit and amplify societal
biases. To date, there is no uniform/systematic agreed standard to
control/adjust for these biases. This study examines the presence and
manipulation of societal biases in leading text-to-image models: Stable
Diffusion, DALL-E 3, and Adobe Firefly. Through a comprehensive analysis
combining base prompts with modifiers and their sequencing, we uncover the
nuanced ways these AI technologies encode biases across gender, race,
geography, and region/culture. Our findings reveal the challenges and potential
of prompt engineering in controlling biases, highlighting the critical need for
ethical AI development promoting diversity and inclusivity.
This work advances AI ethics by not only revealing the nuanced dynamics of
bias in text-to-image generation models but also by offering a novel framework
for future research in controlling bias. Our contributions-panning comparative
analyses, the strategic use of prompt modifiers, the exploration of prompt
sequencing effects, and the introduction of a bias sensitivity taxonomy-lay the
groundwork for the development of common metrics and standard analyses for
evaluating whether and how future AI models exhibit and respond to requests to
adjust for inherent biases. | cs.CV | Can Prompt Modifiers Control Bias? A Comparative Analysis of Text-to-Image Generative Models |
2024-06-08T22:17:52Z | http://arxiv.org/pdf/2406.05588v1 | http://arxiv.org/abs/2406.05588v1 | Jason Cai, Hang Su, Monica Sunkara, Igor Shalyminov, Saab Mansour | 2024-06-08T22:17:52Z | cs.CL, cs.AI, cs.LG | Large Language Models (LLMs) are powerful models for generation tasks, but
they may not generate good quality outputs in their first attempt. Apart from
model fine-tuning, existing approaches to improve prediction accuracy and
quality typically involve LLM self-improvement / self-reflection that
incorporate feedback from models themselves. Despite their effectiveness, these
methods are hindered by their high computational cost and lack of scalability.
In this work, we propose CERET, a method for refining text generations by
considering semantic stability, entailment and inter-sample uncertainty
measures. Experimental results show that CERET outperforms Self-consistency and
Self-rerank baselines consistently under various task setups, by ~1.6% in
Rouge-1 for abstractive summarization and ~3.5% in hit rate for question
answering. Compared to LLM Self-rerank method, our approach only requires 9.4%
of its latency and is more cost-effective. | cs.CL | CERET: Cost-Effective Extrinsic Refinement for Text Generation |
2024-06-08T22:14:51Z | http://arxiv.org/pdf/2406.05587v1 | http://arxiv.org/abs/2406.05587v1 | Behnam Mohammadi | 2024-06-08T22:14:51Z | cs.CL, cs.AI | Large Language Models (LLMs) have revolutionized natural language processing
but can exhibit biases and may generate toxic content. While alignment
techniques like Reinforcement Learning from Human Feedback (RLHF) reduce these
issues, their impact on creativity, defined as syntactic and semantic
diversity, remains unexplored. We investigate the unintended consequences of
RLHF on the creativity of LLMs through three experiments focusing on the
Llama-2 series. Our findings reveal that aligned models exhibit lower entropy
in token predictions, form distinct clusters in the embedding space, and
gravitate towards "attractor states", indicating limited output diversity. Our
findings have significant implications for marketers who rely on LLMs for
creative tasks such as copywriting, ad creation, and customer persona
generation. The trade-off between consistency and creativity in aligned models
should be carefully considered when selecting the appropriate model for a given
application. We also discuss the importance of prompt engineering in harnessing
the creative potential of base models. | cs.CL | Creativity Has Left the Chat: The Price of Debiasing Language Models |
2024-06-08T20:30:53Z | http://arxiv.org/pdf/2406.05569v1 | http://arxiv.org/abs/2406.05569v1 | Metehan Oğuz, Yusuf Umut Ciftci, Yavuz Faruk Bakman | 2024-06-08T20:30:53Z | cs.CL | Large language models (LLMs) have shown impressive capabilities in tasks such
as machine translation, text summarization, question answering, and solving
complex mathematical problems. However, their primary training on data-rich
languages like English limits their performance in low-resource languages. This
study addresses this gap by focusing on the Indexical Shift problem in Turkish.
The Indexical Shift problem involves resolving pronouns in indexical shift
contexts, a grammatical challenge not present in high-resource languages like
English. We present the first study examining indexical shift in any language,
releasing a Turkish dataset specifically designed for this purpose. Our
Indexical Shift Dataset consists of 156 multiple-choice questions, each
annotated with necessary linguistic details, to evaluate LLMs in a few-shot
setting. We evaluate recent multilingual LLMs, including GPT-4, GPT-3.5,
Cohere-AYA, Trendyol-LLM, and Turkcell-LLM, using this dataset. Our analysis
reveals that even advanced models like GPT-4 struggle with the grammatical
nuances of indexical shift in Turkish, achieving only moderate performance.
These findings underscore the need for focused research on the grammatical
challenges posed by low-resource languages. We released the dataset and code
\href{https://anonymous.4open.science/r/indexical_shift_llm-E1B4} {here}. | cs.CL | Do LLMs Recognize me, When I is not me: Assessment of LLMs Understanding of Turkish Indexical Pronouns in Indexical Shift Contexts |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.