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
|
|---|---|---|---|---|---|---|---|---|
2024-04-23T20:26:07Z | http://arxiv.org/pdf/2404.15501v1 | http://arxiv.org/abs/2404.15501v1 | Chihiro Taguchi, Jefferson Saransig, Dayana Velásquez, David Chiang | 2024-04-23T20:26:07Z | cs.CL, cs.AI | This paper presents Killkan, the first dataset for automatic speech
recognition (ASR) in the Kichwa language, an indigenous language of Ecuador.
Kichwa is an extremely low-resource endangered language, and there have been no
resources before Killkan for Kichwa to be incorporated in applications of
natural language processing. The dataset contains approximately 4 hours of
audio with transcription, translation into Spanish, and morphosyntactic
annotation in the format of Universal Dependencies. The audio data was
retrieved from a publicly available radio program in Kichwa. This paper also
provides corpus-linguistic analyses of the dataset with a special focus on the
agglutinative morphology of Kichwa and frequent code-switching with Spanish.
The experiments show that the dataset makes it possible to develop the first
ASR system for Kichwa with reliable quality despite its small dataset size.
This dataset, the ASR model, and the code used to develop them will be publicly
available. Thus, our study positively showcases resource building and its
applications for low-resource languages and their community. | cs.CL | Killkan: The Automatic Speech Recognition Dataset for Kichwa with Morphosyntactic Information |
2024-04-23T20:23:37Z | http://arxiv.org/pdf/2404.15500v1 | http://arxiv.org/abs/2404.15500v1 | Simranjit Singh, Michael Fore, Dimitrios Stamoulis | 2024-04-23T20:23:37Z | cs.AI, cs.CL, cs.LG | Geospatial Copilots unlock unprecedented potential for performing Earth
Observation (EO) applications through natural language instructions. However,
existing agents rely on overly simplified single tasks and template-based
prompts, creating a disconnect with real-world scenarios. In this work, we
present GeoLLM-Engine, an environment for tool-augmented agents with intricate
tasks routinely executed by analysts on remote sensing platforms. We enrich our
environment with geospatial API tools, dynamic maps/UIs, and external
multimodal knowledge bases to properly gauge an agent's proficiency in
interpreting realistic high-level natural language commands and its functional
correctness in task completions. By alleviating overheads typically associated
with human-in-the-loop benchmark curation, we harness our massively parallel
engine across 100 GPT-4-Turbo nodes, scaling to over half a million diverse
multi-tool tasks and across 1.1 million satellite images. By moving beyond
traditional single-task image-caption paradigms, we investigate
state-of-the-art agents and prompting techniques against long-horizon prompts. | cs.AI | GeoLLM-Engine: A Realistic Environment for Building Geospatial Copilots |
2024-04-23T20:00:37Z | http://arxiv.org/pdf/2404.15488v1 | http://arxiv.org/abs/2404.15488v1 | Jean-Philippe Corbeil | 2024-04-23T20:00:37Z | cs.CL, cs.AI, cs.MA | In natural language processing applied to the clinical domain, utilizing
large language models has emerged as a promising avenue for error detection and
correction on clinical notes, a knowledge-intensive task for which annotated
data is scarce. This paper presents MedReAct'N'MedReFlex, which leverages a
suite of four LLM-based medical agents. The MedReAct agent initiates the
process by observing, analyzing, and taking action, generating trajectories to
guide the search to target a potential error in the clinical notes.
Subsequently, the MedEval agent employs five evaluators to assess the targeted
error and the proposed correction. In cases where MedReAct's actions prove
insufficient, the MedReFlex agent intervenes, engaging in reflective analysis
and proposing alternative strategies. Finally, the MedFinalParser agent formats
the final output, preserving the original style while ensuring the integrity of
the error correction process. One core component of our method is our RAG
pipeline based on our ClinicalCorp corpora. Among other well-known sources
containing clinical guidelines and information, we preprocess and release the
open-source MedWiki dataset for clinical RAG application. Our results
demonstrate the central role of our RAG approach with ClinicalCorp leveraged
through the MedReAct'N'MedReFlex framework. It achieved the ninth rank on the
MEDIQA-CORR 2024 final leaderboard. | cs.CL | IryoNLP at MEDIQA-CORR 2024: Tackling the Medical Error Detection & Correction Task On the Shoulders of Medical Agents |
2024-04-23T19:57:03Z | http://arxiv.org/pdf/2405.00708v1 | http://arxiv.org/abs/2405.00708v1 | Furui Cheng, Vilém Zouhar, Robin Shing Moon Chan, Daniel Fürst, Hendrik Strobelt, Mennatallah El-Assady | 2024-04-23T19:57:03Z | cs.CL, cs.AI, cs.HC, cs.LG, I.2.7; H.5.2 | Counterfactual examples are useful for exploring the decision boundaries of
machine learning models and determining feature attributions. How can we apply
counterfactual-based methods to analyze and explain LLMs? We identify the
following key challenges. First, the generated textual counterfactuals should
be meaningful and readable to users and thus can be mentally compared to draw
conclusions. Second, to make the solution scalable to long-form text, users
should be equipped with tools to create batches of counterfactuals from
perturbations at various granularity levels and interactively analyze the
results. In this paper, we tackle the above challenges and contribute 1) a
novel algorithm for generating batches of complete and meaningful textual
counterfactuals by removing and replacing text segments in different
granularities, and 2) LLM Analyzer, an interactive visualization tool to help
users understand an LLM's behaviors by interactively inspecting and aggregating
meaningful counterfactuals. We evaluate the proposed algorithm by the
grammatical correctness of its generated counterfactuals using 1,000 samples
from medical, legal, finance, education, and news datasets. In our experiments,
97.2% of the counterfactuals are grammatically correct. Through a use case,
user studies, and feedback from experts, we demonstrate the usefulness and
usability of the proposed interactive visualization tool. | cs.CL | Interactive Analysis of LLMs using Meaningful Counterfactuals |
2024-04-23T19:55:18Z | http://arxiv.org/pdf/2404.15485v3 | http://arxiv.org/abs/2404.15485v3 | Het Patel, Umair Rehman, Farkhund Iqbal | 2024-06-06T21:03:03Z | cs.CL, cs.AI | Phishing, a prevalent cybercrime tactic for decades, remains a significant
threat in today's digital world. By leveraging clever social engineering
elements and modern technology, cybercrime targets many individuals,
businesses, and organizations to exploit trust and security. These
cyber-attackers are often disguised in many trustworthy forms to appear as
legitimate sources. By cleverly using psychological elements like urgency,
fear, social proof, and other manipulative strategies, phishers can lure
individuals into revealing sensitive and personalized information. Building on
this pervasive issue within modern technology, this paper aims to analyze the
effectiveness of 15 Large Language Models (LLMs) in detecting phishing
attempts, specifically focusing on a randomized set of "419 Scam" emails. The
objective is to determine which LLMs can accurately detect phishing emails by
analyzing a text file containing email metadata based on predefined criteria.
The experiment concluded that the following models, ChatGPT 3.5,
GPT-3.5-Turbo-Instruct, and ChatGPT, were the most effective in detecting
phishing emails. | cs.CL | Evaluating the Efficacy of Large Language Models in Identifying Phishing Attempts |
2024-04-23T18:55:49Z | http://arxiv.org/pdf/2405.02326v1 | http://arxiv.org/abs/2405.02326v1 | Jason Blocklove, Siddharth Garg, Ramesh Karri, Hammond Pearce | 2024-04-23T18:55:49Z | cs.AR, cs.AI, cs.CL, cs.LG, cs.PL | Large Language Models (LLMs) have demonstrated capabilities for producing
code in Hardware Description Languages (HDLs). However, most of the focus
remains on their abilities to write functional code, not test code. The
hardware design process consists of both design and test, and so eschewing
validation and verification leaves considerable potential benefit unexplored,
given that a design and test framework may allow for progress towards full
automation of the digital design pipeline. In this work, we perform one of the
first studies exploring how a LLM can both design and test hardware modules
from provided specifications. Using a suite of 8 representative benchmarks, we
examined the capabilities and limitations of the state-of-the-art
conversational LLMs when producing Verilog for functional and verification
purposes. We taped out the benchmarks on a Skywater 130nm shuttle and received
the functional chip. | cs.AR | Evaluating LLMs for Hardware Design and Test |
2024-04-23T18:53:33Z | http://arxiv.org/pdf/2405.03695v1 | http://arxiv.org/abs/2405.03695v1 | Daniele Grandi, Yash Patawari Jain, Allin Groom, Brandon Cramer, Christopher McComb | 2024-04-23T18:53:33Z | cs.CL | Material selection is a crucial step in conceptual design due to its
significant impact on the functionality, aesthetics, manufacturability, and
sustainability impact of the final product. This study investigates the use of
Large Language Models (LLMs) for material selection in the product design
process and compares the performance of LLMs against expert choices for various
design scenarios. By collecting a dataset of expert material preferences, the
study provides a basis for evaluating how well LLMs can align with expert
recommendations through prompt engineering and hyperparameter tuning. The
divergence between LLM and expert recommendations is measured across different
model configurations, prompt strategies, and temperature settings. This
approach allows for a detailed analysis of factors influencing the LLMs'
effectiveness in recommending materials. The results from this study highlight
two failure modes, and identify parallel prompting as a useful
prompt-engineering method when using LLMs for material selection. The findings
further suggest that, while LLMs can provide valuable assistance, their
recommendations often vary significantly from those of human experts. This
discrepancy underscores the need for further research into how LLMs can be
better tailored to replicate expert decision-making in material selection. This
work contributes to the growing body of knowledge on how LLMs can be integrated
into the design process, offering insights into their current limitations and
potential for future improvements. | cs.CL | Evaluating Large Language Models for Material Selection |
2024-04-23T18:10:42Z | http://arxiv.org/pdf/2404.15420v1 | http://arxiv.org/abs/2404.15420v1 | João Monteiro, Étienne Marcotte, Pierre-André Noël, Valentina Zantedeschi, David Vázquez, Nicolas Chapados, Christopher Pal, Perouz Taslakian | 2024-04-23T18:10:42Z | cs.CL, cs.AI | In-context learning (ICL) approaches typically leverage prompting to
condition decoder-only language model generation on reference information.
Just-in-time processing of a context is inefficient due to the quadratic cost
of self-attention operations, and caching is desirable. However, caching
transformer states can easily require almost as much space as the model
parameters. When the right context isn't known in advance, caching ICL can be
challenging. This work addresses these limitations by introducing models that,
inspired by the encoder-decoder architecture, use cross-attention to condition
generation on reference text without the prompt. More precisely, we leverage
pre-trained decoder-only models and only train a small number of added layers.
We use Question-Answering (QA) as a testbed to evaluate the ability of our
models to perform conditional generation and observe that they outperform ICL,
are comparable to fine-tuned prompted LLMs, and drastically reduce the space
footprint relative to standard KV caching by two orders of magnitude. | cs.CL | XC-Cache: Cross-Attending to Cached Context for Efficient LLM Inference |
2024-04-23T18:00:09Z | http://arxiv.org/pdf/2404.15406v2 | http://arxiv.org/abs/2404.15406v2 | Davide Caffagni, Federico Cocchi, Nicholas Moratelli, Sara Sarto, Marcella Cornia, Lorenzo Baraldi, Rita Cucchiara | 2024-05-22T07:15:18Z | cs.CV, cs.AI, cs.CL, cs.MM | Multimodal LLMs are the natural evolution of LLMs, and enlarge their
capabilities so as to work beyond the pure textual modality. As research is
being carried out to design novel architectures and vision-and-language
adapters, in this paper we concentrate on endowing such models with the
capability of answering questions that require external knowledge. Our
approach, termed Wiki-LLaVA, aims at integrating an external knowledge source
of multimodal documents, which is accessed through a hierarchical retrieval
pipeline. Relevant passages, using this approach, are retrieved from the
external knowledge source and employed as additional context for the LLM,
augmenting the effectiveness and precision of generated dialogues. We conduct
extensive experiments on datasets tailored for visual question answering with
external data and demonstrate the appropriateness of our approach. | cs.CV | Wiki-LLaVA: Hierarchical Retrieval-Augmented Generation for Multimodal LLMs |
2024-04-23T17:59:01Z | http://arxiv.org/pdf/2404.15272v3 | http://arxiv.org/abs/2404.15272v3 | Jingyang Lin, Yingda Xia, Jianpeng Zhang, Ke Yan, Le Lu, Jiebo Luo, Ling Zhang | 2024-04-29T03:25:14Z | cs.CV, cs.AI, cs.CL | Medical Vision-Language Pretraining (Med-VLP) establishes a connection
between visual content from medical images and the relevant textual
descriptions. Existing Med-VLP methods primarily focus on 2D images depicting a
single body part, notably chest X-rays. In this paper, we extend the scope of
Med-VLP to encompass 3D images, specifically targeting full-body scenarios, by
using a multimodal dataset of CT images and reports. Compared with the 2D
counterpart, 3D VLP is required to effectively capture essential semantics from
significantly sparser representation in 3D imaging. In this paper, we introduce
CT-GLIP (Grounded Language-Image Pretraining with CT scans), a novel method
that constructs organ-level image-text pairs to enhance multimodal contrastive
learning, aligning grounded visual features with precise diagnostic text.
Additionally, we developed an abnormality dictionary to augment contrastive
learning with diverse contrastive pairs. Our method, trained on a multimodal CT
dataset comprising 44,011 organ-level vision-text pairs from 17,702 patients
across 104 organs, demonstrates it can identify organs and abnormalities in a
zero-shot manner using natural languages. The performance of CT-GLIP is
validated on a separate test set of 1,130 patients, focusing on the 16 most
frequent abnormalities across 7 organs. The experimental results show our
model's superior performance over the standard CLIP framework across zero-shot
and fine-tuning scenarios, using both CNN and ViT architectures. | cs.CV | CT-GLIP: 3D Grounded Language-Image Pretraining with CT Scans and Radiology Reports for Full-Body Scenarios |
2024-04-23T17:58:33Z | http://arxiv.org/pdf/2404.15271v1 | http://arxiv.org/abs/2404.15271v1 | Wanrong Zhu, Jennifer Healey, Ruiyi Zhang, William Yang Wang, Tong Sun | 2024-04-23T17:58:33Z | cs.CV, cs.AI, cs.CL | Recent advancements in instruction-following models have made user
interactions with models more user-friendly and efficient, broadening their
applicability. In graphic design, non-professional users often struggle to
create visually appealing layouts due to limited skills and resources. In this
work, we introduce a novel multimodal instruction-following framework for
layout planning, allowing users to easily arrange visual elements into tailored
layouts by specifying canvas size and design purpose, such as for book covers,
posters, brochures, or menus. We developed three layout reasoning tasks to
train the model in understanding and executing layout instructions. Experiments
on two benchmarks show that our method not only simplifies the design process
for non-professionals but also surpasses the performance of few-shot GPT-4V
models, with mIoU higher by 12% on Crello. This progress highlights the
potential of multimodal instruction-following models to automate and simplify
the design process, providing an approachable solution for a wide range of
design tasks on visually-rich documents. | cs.CV | Automatic Layout Planning for Visually-Rich Documents with Instruction-Following Models |
2024-04-23T17:57:47Z | http://arxiv.org/pdf/2404.15269v2 | http://arxiv.org/abs/2404.15269v2 | Ge Gao, Alexey Taymanov, Eduardo Salinas, Paul Mineiro, Dipendra Misra | 2024-06-09T21:45:09Z | cs.CL, cs.AI, cs.IR, cs.LG | We study interactive learning of LLM-based language agents based on user
edits made to the agent's output. In a typical setting such as writing
assistants, the user interacts with a language agent to generate a response
given a context, and may optionally edit the agent response to personalize it
based on their latent preference, in addition to improving the correctness. The
edit feedback is naturally generated, making it a suitable candidate for
improving the agent's alignment with the user's preference, and for reducing
the cost of user edits over time. We propose a learning framework, PRELUDE that
infers a description of the user's latent preference based on historic edit
data. The inferred user preference descriptions are used to define prompts for
generating responses in the future. This avoids fine-tuning the agent, which is
costly, challenging to scale with the number of users, and may even degrade its
performance on other tasks. Furthermore, learning descriptive preference
improves interpretability, allowing the user to view and modify the learned
preference. However, user preference can be complex, subtle, and vary based on
context, making it challenging to learn. To address this, we propose a simple
yet effective algorithm named CIPHER that leverages the LLM to infer the user
preference for a given context based on user edits. In the future, CIPHER
retrieves inferred preferences from the k-closest contexts in the history, and
forms an aggregate preference for response generation. We introduce two
interactive environments -- summarization and email writing, and use a GPT-4
simulated user for evaluation. On both tasks, CIPHER outperforms several
baselines by achieving the lowest edit distance cost while only having a small
overhead in LLM query cost. Our analysis reports that user preferences learned
by CIPHER show significant similarity to the ground truth latent preferences. | cs.CL | Aligning LLM Agents by Learning Latent Preference from User Edits |
2024-04-23T17:32:24Z | http://arxiv.org/pdf/2404.15247v2 | http://arxiv.org/abs/2404.15247v2 | Yifeng Ding, Jiawei Liu, Yuxiang Wei, Terry Yue Zhuo, Lingming Zhang | 2024-06-06T18:18:21Z | cs.CL, cs.AI, cs.LG, cs.SE | We introduce XFT, a simple yet powerful training scheme, by simply merging
upcycled Mixture-of-Experts (MoE) to unleash the performance limit of
instruction-tuned code Large Language Models (LLMs). While vanilla sparse
upcycling fails to improve instruction tuning, XFT introduces a shared expert
mechanism with a novel routing weight normalization strategy into sparse
upcycling, which significantly boosts instruction tuning. After fine-tuning the
upcycled MoE model, XFT introduces a learnable model merging mechanism to
compile the upcycled MoE model back to a dense model, achieving upcycled
MoE-level performance with only dense-model compute. By applying XFT to a 1.3B
model, we create a new state-of-the-art tiny code LLM (<3B) with 67.1 and 64.6
pass@1 on HumanEval and HumanEval+ respectively. With the same data and model
architecture, XFT improves supervised fine-tuning (SFT) by 13% on HumanEval+,
along with consistent improvements from 2% to 13% on MBPP+, MultiPL-E, and
DS-1000, demonstrating its generalizability. XFT is fully orthogonal to
existing techniques such as Evol-Instruct and OSS-Instruct, opening a new
dimension for improving code instruction tuning. Codes are available at
https://github.com/ise-uiuc/xft. | cs.CL | XFT: Unlocking the Power of Code Instruction Tuning by Simply Merging Upcycled Mixture-of-Experts |
2024-04-23T17:16:08Z | http://arxiv.org/pdf/2404.15238v1 | http://arxiv.org/abs/2404.15238v1 | Weiyan Shi, Ryan Li, Yutong Zhang, Caleb Ziems, Chunhua yu, Raya Horesh, Rogério Abreu de Paula, Diyi Yang | 2024-04-23T17:16:08Z | cs.CL, cs.AI | To enhance language models' cultural awareness, we design a generalizable
pipeline to construct cultural knowledge bases from different online
communities on a massive scale. With the pipeline, we construct CultureBank, a
knowledge base built upon users' self-narratives with 12K cultural descriptors
sourced from TikTok and 11K from Reddit. Unlike previous cultural knowledge
resources, CultureBank contains diverse views on cultural descriptors to allow
flexible interpretation of cultural knowledge, and contextualized cultural
scenarios to help grounded evaluation. With CultureBank, we evaluate different
LLMs' cultural awareness, and identify areas for improvement. We also fine-tune
a language model on CultureBank: experiments show that it achieves better
performances on two downstream cultural tasks in a zero-shot setting. Finally,
we offer recommendations based on our findings for future culturally aware
language technologies. The project page is https://culturebank.github.io . The
code and model is at https://github.com/SALT-NLP/CultureBank . The released
CultureBank dataset is at https://huggingface.co/datasets/SALT-NLP/CultureBank . | cs.CL | CultureBank: An Online Community-Driven Knowledge Base Towards Culturally Aware Language Technologies |
2024-04-23T17:06:24Z | http://arxiv.org/pdf/2405.01575v1 | http://arxiv.org/abs/2405.01575v1 | Thuy Nguyen Thi, Anh Nguyen Viet, Thin Dang Van, Ngan Nguyen Luu Thuy | 2024-04-23T17:06:24Z | cs.SE, cs.AI, cs.CL | This paper describes our systems for the sub-task I in the Software Mention
Detection in Scholarly Publications shared-task. We propose three approaches
leveraging different pre-trained language models (BERT, SciBERT, and XLM-R) to
tackle this challenge. Our bestperforming system addresses the named entity
recognition (NER) problem through a three-stage framework. (1) Entity Sentence
Classification - classifies sentences containing potential software mentions;
(2) Entity Extraction - detects mentions within classified sentences; (3)
Entity Type Classification - categorizes detected mentions into specific
software types. Experiments on the official dataset demonstrate that our
three-stage framework achieves competitive performance, surpassing both other
participating teams and our alternative approaches. As a result, our framework
based on the XLM-R-based model achieves a weighted F1-score of 67.80%,
delivering our team the 3rd rank in Sub-task I for the Software Mention
Recognition task. | cs.SE | Software Mention Recognition with a Three-Stage Framework Based on BERTology Models at SOMD 2024 |
2024-04-23T16:59:02Z | http://arxiv.org/pdf/2404.15228v1 | http://arxiv.org/abs/2404.15228v1 | Peter Kulits, Haiwen Feng, Weiyang Liu, Victoria Abrevaya, Michael J. Black | 2024-04-23T16:59:02Z | cs.CV, cs.CL | Inverse graphics -- the task of inverting an image into physical variables
that, when rendered, enable reproduction of the observed scene -- is a
fundamental challenge in computer vision and graphics. Disentangling an image
into its constituent elements, such as the shape, color, and material
properties of the objects of the 3D scene that produced it, requires a
comprehensive understanding of the environment. This requirement limits the
ability of existing carefully engineered approaches to generalize across
domains. Inspired by the zero-shot ability of large language models (LLMs) to
generalize to novel contexts, we investigate the possibility of leveraging the
broad world knowledge encoded in such models in solving inverse-graphics
problems. To this end, we propose the Inverse-Graphics Large Language Model
(IG-LLM), an inverse-graphics framework centered around an LLM, that
autoregressively decodes a visual embedding into a structured, compositional
3D-scene representation. We incorporate a frozen pre-trained visual encoder and
a continuous numeric head to enable end-to-end training. Through our
investigation, we demonstrate the potential of LLMs to facilitate inverse
graphics through next-token prediction, without the use of image-space
supervision. Our analysis opens up new possibilities for precise spatial
reasoning about images that exploit the visual knowledge of LLMs. We will
release our code and data to ensure the reproducibility of our investigation
and to facilitate future research at https://ig-llm.is.tue.mpg.de/ | cs.CV | Re-Thinking Inverse Graphics With Large Language Models |
2024-04-23T16:51:26Z | http://arxiv.org/pdf/2404.15219v1 | http://arxiv.org/abs/2404.15219v1 | Brendan King, Jeffrey Flanigan | 2024-04-23T16:51:26Z | cs.CL | Training task-oriented dialogue systems typically requires turn-level
annotations for interacting with their APIs: e.g. a dialogue state and the
system actions taken at each step. These annotations can be costly to produce,
error-prone, and require both domain and annotation expertise. With advances in
LLMs, we hypothesize unlabelled data and a schema definition are sufficient for
building a working task-oriented dialogue system, completely unsupervised.
Using only (1) a well-defined API schema (2) a set of unlabelled dialogues
between a user and agent, we develop a novel approach for inferring turn-level
annotations as latent variables using a noisy channel model. We iteratively
improve these pseudo-labels with expectation-maximization (EM), and use the
inferred labels to train an end-to-end dialogue agent. Evaluating our approach
on the MultiWOZ benchmark, our method more than doubles the dialogue success
rate of a strong GPT-3.5 baseline. | cs.CL | The Power of the Noisy Channel: Unsupervised End-to-End Task-Oriented Dialogue with LLMs |
2024-04-23T16:39:03Z | http://arxiv.org/pdf/2404.15206v2 | http://arxiv.org/abs/2404.15206v2 | Constanza Fierro, Jiaang Li, Anders Søgaard | 2024-04-30T09:25:40Z | cs.CL | The purpose of instruction tuning is enabling zero-shot performance, but
instruction tuning has also been shown to improve chain-of-thought reasoning
and value alignment (Si et al., 2023). Here we consider the impact on
$\textit{consistency}$, i.e., the sensitivity of language models to small
perturbations in the input. We compare 10 instruction-tuned LLaMA models to the
original LLaMA-7b model and show that almost across-the-board they become more
consistent, both in terms of their representations and their predictions in
zero-shot and downstream tasks. We explain these improvements through
mechanistic analyses of factual recall. | cs.CL | Does Instruction Tuning Make LLMs More Consistent? |
2024-04-23T16:34:34Z | http://arxiv.org/pdf/2404.15196v1 | http://arxiv.org/abs/2404.15196v1 | Yurii Paniv, Dmytro Chaplynskyi, Nikita Trynus, Volodymyr Kyrylov | 2024-04-23T16:34:34Z | cs.CL | To build large language models for Ukrainian we need to expand our corpora
with large amounts of new algorithmic tasks expressed in natural language.
Examples of task performance expressed in English are abundant, so with a
high-quality translation system our community will be enabled to curate
datasets faster. To aid this goal, we introduce a recipe to build a translation
system using supervised finetuning of a large pretrained language model with a
noisy parallel dataset of 3M pairs of Ukrainian and English sentences followed
by a second phase of training using 17K examples selected by k-fold perplexity
filtering on another dataset of higher quality. Our decoder-only model named
Dragoman beats performance of previous state of the art encoder-decoder models
on the FLORES devtest set. | cs.CL | Setting up the Data Printer with Improved English to Ukrainian Machine Translation |
2024-04-23T15:57:55Z | http://arxiv.org/pdf/2404.15156v1 | http://arxiv.org/abs/2404.15156v1 | Shashank Sonkar, Naiming Liu, Richard G. Baraniuk | 2024-04-23T15:57:55Z | cs.CL | This paper presents a novel exploration into the regressive side effects of
training Large Language Models (LLMs) to mimic student misconceptions for
personalized education. We highlight the problem that as LLMs are trained to
more accurately mimic student misconceptions, there is a compromise in the
factual integrity and reasoning ability of the models. Our work involved
training an LLM on a student-tutor dialogue dataset to predict student
responses. The results demonstrated a decrease in the model's performance
across multiple benchmark datasets, including the ARC reasoning challenge and
TruthfulQA, which evaluates the truthfulness of model's generated responses.
Furthermore, the HaluEval Dial dataset, used for hallucination detection, and
MemoTrap, a memory-based task dataset, also reported a decline in the model
accuracy. To combat these side effects, we introduced a "hallucination token"
technique. This token, appended at the beginning of each student response
during training, instructs the model to switch between mimicking student
misconceptions and providing factually accurate responses. Despite the
significant improvement across all datasets, the technique does not completely
restore the LLM's baseline performance, indicating the need for further
research in this area. This paper contributes to the ongoing discussion on the
use of LLMs for student modeling, emphasizing the need for a balance between
personalized education and factual accuracy. | cs.CL | Regressive Side Effects of Training Language Models to Mimic Student Misconceptions |
2024-04-23T15:52:52Z | http://arxiv.org/pdf/2404.15149v1 | http://arxiv.org/abs/2404.15149v1 | Raphael Poulain, Hamed Fayyaz, Rahmatollah Beheshti | 2024-04-23T15:52:52Z | cs.CL, cs.LG | Large Language Models (LLMs) have emerged as powerful candidates to inform
clinical decision-making processes. While these models play an increasingly
prominent role in shaping the digital landscape, two growing concerns emerge in
healthcare applications: 1) to what extent do LLMs exhibit social bias based on
patients' protected attributes (like race), and 2) how do design choices (like
architecture design and prompting strategies) influence the observed biases? To
answer these questions rigorously, we evaluated eight popular LLMs across three
question-answering (QA) datasets using clinical vignettes (patient
descriptions) standardized for bias evaluations. We employ red-teaming
strategies to analyze how demographics affect LLM outputs, comparing both
general-purpose and clinically-trained models. Our extensive experiments reveal
various disparities (some significant) across protected groups. We also observe
several counter-intuitive patterns such as larger models not being necessarily
less biased and fined-tuned models on medical data not being necessarily better
than the general-purpose models. Furthermore, our study demonstrates the impact
of prompt design on bias patterns and shows that specific phrasing can
influence bias patterns and reflection-type approaches (like Chain of Thought)
can reduce biased outcomes effectively. Consistent with prior studies, we call
on additional evaluations, scrutiny, and enhancement of LLMs used in clinical
decision support applications. | cs.CL | Bias patterns in the application of LLMs for clinical decision support: A comprehensive study |
2024-04-23T15:49:37Z | http://arxiv.org/pdf/2404.15146v2 | http://arxiv.org/abs/2404.15146v2 | Avi Schwarzschild, Zhili Feng, Pratyush Maini, Zachary C. Lipton, J. Zico Kolter | 2024-07-01T14:43:11Z | cs.LG, cs.CL | Large language models (LLMs) trained on web-scale datasets raise substantial
concerns regarding permissible data usage. One major question is whether these
models "memorize" all their training data or they integrate many data sources
in some way more akin to how a human would learn and synthesize information.
The answer hinges, to a large degree, on how we define memorization. In this
work, we propose the Adversarial Compression Ratio (ACR) as a metric for
assessing memorization in LLMs. A given string from the training data is
considered memorized if it can be elicited by a prompt (much) shorter than the
string itself -- in other words, if these strings can be "compressed" with the
model by computing adversarial prompts of fewer tokens. The ACR overcomes the
limitations of existing notions of memorization by (i) offering an adversarial
view of measuring memorization, especially for monitoring unlearning and
compliance; and (ii) allowing for the flexibility to measure memorization for
arbitrary strings at a reasonably low compute. Our definition serves as a
practical tool for determining when model owners may be violating terms around
data usage, providing a potential legal tool and a critical lens through which
to address such scenarios. | cs.LG | Rethinking LLM Memorization through the Lens of Adversarial Compression |
2024-04-23T15:27:19Z | http://arxiv.org/pdf/2404.15127v1 | http://arxiv.org/abs/2404.15127v1 | Sunan He, Yuxiang Nie, Zhixuan Chen, Zhiyuan Cai, Hongmei Wang, Shu Yang, Hao Chen | 2024-04-23T15:27:19Z | cs.CV, cs.CL | The rapid advancement of large-scale vision-language models has showcased
remarkable capabilities across various tasks. However, the lack of extensive
and high-quality image-text data in medicine has greatly hindered the
development of large-scale medical vision-language models. In this work, we
present a diagnosis-guided bootstrapping strategy that exploits both image and
label information to construct vision-language datasets. Based on the
constructed dataset, we developed MedDr, a generalist foundation model for
healthcare capable of handling diverse medical data modalities, including
radiology, pathology, dermatology, retinography, and endoscopy. Moreover,
during inference, we propose a simple but effective retrieval-augmented medical
diagnosis strategy, which enhances the model's generalization ability.
Extensive experiments on visual question answering, medical report generation,
and medical image diagnosis demonstrate the superiority of our method. | cs.CV | MedDr: Diagnosis-Guided Bootstrapping for Large-Scale Medical Vision-Language Learning |
2024-04-23T14:56:15Z | http://arxiv.org/pdf/2404.15104v2 | http://arxiv.org/abs/2404.15104v2 | Kevin Stowe, Benny Longwill, Alyssa Francis, Tatsuya Aoyama, Debanjan Ghosh, Swapna Somasundaran | 2024-05-01T19:45:14Z | cs.CL, I.2.7 | Natural language generation tools are powerful and effective for generating
content. However, language models are known to display bias and fairness
issues, making them impractical to deploy for many use cases. We here focus on
how fairness issues impact automatically generated test content, which can have
stringent requirements to ensure the test measures only what it was intended to
measure. Specifically, we review test content generated for a large-scale
standardized English proficiency test with the goal of identifying content that
only pertains to a certain subset of the test population as well as content
that has the potential to be upsetting or distracting to some test takers.
Issues like these could inadvertently impact a test taker's score and thus
should be avoided. This kind of content does not reflect the more
commonly-acknowledged biases, making it challenging even for modern models that
contain safeguards. We build a dataset of 601 generated texts annotated for
fairness and explore a variety of methods for classification: fine-tuning,
topic-based classification, and prompting, including few-shot and
self-correcting prompts. We find that combining prompt self-correction and
few-shot learning performs best, yielding an F1 score of 0.79 on our held-out
test set, while much smaller BERT- and topic-based models have competitive
performance on out-of-domain data. | cs.CL | Identifying Fairness Issues in Automatically Generated Testing Content |
2024-04-23T14:55:32Z | http://arxiv.org/pdf/2404.15103v1 | http://arxiv.org/abs/2404.15103v1 | Kuicai Dong, Derrick Goh Xin Deik, Yi Quan Lee, Hao Zhang, Xiangyang Li, Cong Zhang, Yong Liu | 2024-04-23T14:55:32Z | cs.CL | Long document question answering (DocQA) aims to answer questions from long
documents over 10k words. They usually contain content structures such as
sections, sub-sections, and paragraph demarcations. However, the indexing
methods of long documents remain under-explored, while existing systems
generally employ fixed-length chunking. As they do not consider content
structures, the resultant chunks can exclude vital information or include
irrelevant content. Motivated by this, we propose the Multi-view Content-aware
indexing (MC-indexing) for more effective long DocQA via (i) segment structured
document into content chunks, and (ii) represent each content chunk in
raw-text, keywords, and summary views. We highlight that MC-indexing requires
neither training nor fine-tuning. Having plug-and-play capability, it can be
seamlessly integrated with any retrievers to boost their performance. Besides,
we propose a long DocQA dataset that includes not only question-answer pair,
but also document structure and answer scope. When compared to state-of-art
chunking schemes, MC-indexing has significantly increased the recall by 42.8%,
30.0%, 23.9%, and 16.3% via top k= 1.5, 3, 5, and 10 respectively. These
improved scores are the average of 8 widely used retrievers (2 sparse and 6
dense) via extensive experiments. | cs.CL | Multi-view Content-aware Indexing for Long Document Retrieval |
2024-04-23T14:43:35Z | http://arxiv.org/pdf/2405.00706v2 | http://arxiv.org/abs/2405.00706v2 | David M. Markowitz | 2024-06-11T12:35:51Z | cs.CL, cs.AI, cs.CY | This paper evaluated the effectiveness of using generative AI to simplify
science communication and enhance the public's understanding of science. By
comparing lay summaries of journal articles from PNAS, yoked to those generated
by AI, this work first assessed linguistic simplicity across such summaries and
public perceptions. Study 1a analyzed simplicity features of PNAS abstracts
(scientific summaries) and significance statements (lay summaries), observing
that lay summaries were indeed linguistically simpler, but effect size
differences were small. Study 1b used a large language model, GPT-4, to create
significance statements based on paper abstracts and this more than doubled the
average effect size without fine-tuning. Study 2 experimentally demonstrated
that simply-written GPT summaries facilitated more favorable perceptions of
scientists (they were perceived as more credible and trustworthy, but less
intelligent) than more complexly-written human PNAS summaries. Crucially, Study
3 experimentally demonstrated that participants comprehended scientific writing
better after reading simple GPT summaries compared to complex PNAS summaries.
In their own words, participants also summarized scientific papers in a more
detailed and concrete manner after reading GPT summaries compared to PNAS
summaries of the same article. AI has the potential to engage scientific
communities and the public via a simple language heuristic, advocating for its
integration into scientific dissemination for a more informed society. | cs.CL | From Complexity to Clarity: How AI Enhances Perceptions of Scientists and the Public's Understanding of Science |
2024-04-23T14:13:53Z | http://arxiv.org/pdf/2404.15067v1 | http://arxiv.org/abs/2404.15067v1 | Haohao Zhu, Xiaokun Zhang, Junyu Lu, Youlin Wu, Zewen Bai, Changrong Min, Liang Yang, Bo Xu, Dongyu Zhang, Hongfei Lin | 2024-04-23T14:13:53Z | cs.CL | Textual personality detection aims to identify personality characteristics by
analyzing user-generated content toward social media platforms. Numerous
psychological literature highlighted that personality encompasses both
long-term stable traits and short-term dynamic states. However, existing
studies often concentrate only on either long-term or short-term personality
representations, without effectively combining both aspects. This limitation
hinders a comprehensive understanding of individuals' personalities, as both
stable traits and dynamic states are vital. To bridge this gap, we propose a
Dual Enhanced Network(DEN) to jointly model users' long-term and short-term
personality for textual personality detection. In DEN, a Long-term Personality
Encoding is devised to effectively model long-term stable personality traits.
Short-term Personality Encoding is presented to capture short-term dynamic
personality states. The Bi-directional Interaction component facilitates the
integration of both personality aspects, allowing for a comprehensive
representation of the user's personality. Experimental results on two
personality detection datasets demonstrate the effectiveness of the DEN model
and the benefits of considering both the dynamic and stable nature of
personality characteristics for textual personality detection. | cs.CL | Enhancing Textual Personality Detection toward Social Media: Integrating Long-term and Short-term Perspectives |
2024-04-23T13:47:09Z | http://arxiv.org/pdf/2404.15045v1 | http://arxiv.org/abs/2404.15045v1 | Xun Wu, Shaohan Huang, Wenhui Wang, Furu Wei | 2024-04-23T13:47:09Z | cs.CL, cs.AI, cs.LG | Sparse Mixtures of Experts (SMoE) scales model capacity without significant
increases in training and inference costs, but exhibits the following two
issues: (1) Low expert activation, where only a small subset of experts are
activated for optimization. (2) Lacking fine-grained analytical capabilities
for multiple semantic concepts within individual tokens. We propose Multi-Head
Mixture-of-Experts (MH-MoE), which employs a multi-head mechanism to split each
token into multiple sub-tokens. These sub-tokens are then assigned to and
processed by a diverse set of experts in parallel, and seamlessly reintegrated
into the original token form. The multi-head mechanism enables the model to
collectively attend to information from various representation spaces within
different experts, while significantly enhances expert activation, thus deepens
context understanding and alleviate overfitting. Moreover, our MH-MoE is
straightforward to implement and decouples from other SMoE optimization
methods, making it easy to integrate with other SMoE models for enhanced
performance. Extensive experimental results across three tasks: English-focused
language modeling, Multi-lingual language modeling and Masked multi-modality
modeling tasks, demonstrate the effectiveness of MH-MoE. | cs.CL | Multi-Head Mixture-of-Experts |
2024-04-23T13:09:11Z | http://arxiv.org/pdf/2404.15004v2 | http://arxiv.org/abs/2404.15004v2 | Derek Powell, Walter Gerych, Thomas Hartvigsen | 2024-06-06T13:46:36Z | cs.CL | Humans rarely learn one fact in isolation. Instead, learning a new fact
induces knowledge of other facts about the world. For example, in learning a
korat is a type of cat, you also infer it is a mammal and has claws, ensuring
your model of the world is consistent. Knowledge editing aims to inject new
facts into language models to improve their factuality, but current benchmarks
fail to evaluate consistency, which is critical to ensure efficient, accurate,
and generalizable edits. We manually create TAXI, a new benchmark dataset
specifically created to evaluate consistency in categorical knowledge edits.
TAXI contains 11,120 multiple-choice queries for 976 edits spanning 41
categories (e.g., Dogs), 164 subjects (e.g., Labrador), and 183 properties
(e.g., is a mammal). We then use TAXI to evaluate popular editors' categorical
consistency, measuring how often editing a subject's category appropriately
edits its properties. We find that 1) the editors achieve marginal, yet
non-random consistency, 2) their consistency far underperforms human baselines,
and 3) consistency is more achievable when editing atypical subjects Our code
and data are available at https://github.com/derekpowell/taxi. | cs.CL | TAXI: Evaluating Categorical Knowledge Editing for Language Models |
2024-04-23T13:06:32Z | http://arxiv.org/pdf/2404.15003v1 | http://arxiv.org/abs/2404.15003v1 | Aleksei Dorkin, Kairit Sirts | 2024-04-23T13:06:32Z | cs.CL | This study evaluates three different lemmatization approaches to Estonian --
Generative character-level models, Pattern-based word-level classification
models, and rule-based morphological analysis. According to our experiments, a
significantly smaller Generative model consistently outperforms the
Pattern-based classification model based on EstBERT. Additionally, we observe a
relatively small overlap in errors made by all three models, indicating that an
ensemble of different approaches could lead to improvements. | cs.CL | Comparison of Current Approaches to Lemmatization: A Case Study in Estonian |
2024-04-23T12:51:37Z | http://arxiv.org/pdf/2404.14994v3 | http://arxiv.org/abs/2404.14994v3 | Anej Svete, Ryan Cotterell | 2024-06-20T15:21:23Z | cs.CL, cs.AI, cs.CC, cs.FL, cs.LG | Existing work has analyzed the representational capacity of the transformer
architecture by means of formal models of computation. However, the focus so
far has been on analyzing the architecture in terms of language
\emph{acceptance}. We contend that this is an ill-suited problem in the study
of \emph{language models} (LMs), which are definitionally \emph{probability
distributions} over strings. In this paper, we focus on the relationship
between transformer LMs and $n$-gram LMs, a simple and historically relevant
class of language models. We show that transformer LMs using the hard or sparse
attention mechanisms can exactly represent any $n$-gram LM, giving us a
concrete lower bound on their probabilistic representational capacity. This
provides a first step towards understanding the mechanisms that transformer LMs
can use to represent probability distributions over strings. | cs.CL | Transformers Can Represent $n$-gram Language Models |
2024-04-23T12:46:53Z | http://arxiv.org/pdf/2404.14989v1 | http://arxiv.org/abs/2404.14989v1 | Sean MacAvaney, Nicola Tonellotto | 2024-04-23T12:46:53Z | cs.IR, cs.CL | The PLAID (Performance-optimized Late Interaction Driver) algorithm for
ColBERTv2 uses clustered term representations to retrieve and progressively
prune documents for final (exact) document scoring. In this paper, we reproduce
and fill in missing gaps from the original work. By studying the parameters
PLAID introduces, we find that its Pareto frontier is formed of a careful
balance among its three parameters; deviations beyond the suggested settings
can substantially increase latency without necessarily improving its
effectiveness. We then compare PLAID with an important baseline missing from
the paper: re-ranking a lexical system. We find that applying ColBERTv2 as a
re-ranker atop an initial pool of BM25 results provides better
efficiency-effectiveness trade-offs in low-latency settings. However,
re-ranking cannot reach peak effectiveness at higher latency settings due to
limitations in recall of lexical matching and provides a poor approximation of
an exhaustive ColBERTv2 search. We find that recently proposed modifications to
re-ranking that pull in the neighbors of top-scoring documents overcome this
limitation, providing a Pareto frontier across all operational points for
ColBERTv2 when evaluated using a well-annotated dataset. Curious about why
re-ranking methods are highly competitive with PLAID, we analyze the token
representation clusters PLAID uses for retrieval and find that most clusters
are predominantly aligned with a single token and vice versa. Given the
competitive trade-offs that re-ranking baselines exhibit, this work highlights
the importance of carefully selecting pertinent baselines when evaluating the
efficiency of retrieval engines. | cs.IR | A Reproducibility Study of PLAID |
2024-04-23T12:35:44Z | http://arxiv.org/pdf/2404.16880v1 | http://arxiv.org/abs/2404.16880v1 | Yikun Zhang, Geyan Ye, Chaohao Yuan, Bo Han, Long-Kai Huang, Jianhua Yao, Wei Liu, Yu Rong | 2024-04-23T12:35:44Z | q-bio.QM, cs.AI, cs.CL | Molecule-and-text cross-modal representation learning has emerged as a
promising direction for enhancing the quality of molecular representation,
thereby improving performance in various scientific fields, including drug
discovery and materials science. Existing studies adopt a global alignment
approach to learn the knowledge from different modalities. These global
alignment approaches fail to capture fine-grained information, such as
molecular fragments and their corresponding textual description, which is
crucial for downstream tasks. Furthermore, it is incapable to model such
information using a similar global alignment strategy due to data scarcity of
paired local part annotated data from existing datasets. In this paper, we
propose Atomas, a multi-modal molecular representation learning framework to
jointly learn representations from SMILES string and text. We design a
Hierarchical Adaptive Alignment model to concurrently learn the fine-grained
fragment correspondence between two modalities and align these representations
of fragments in three levels. Additionally, Atomas's end-to-end training
framework incorporates the tasks of understanding and generating molecule,
thereby supporting a wider range of downstream tasks. In the retrieval task,
Atomas exhibits robust generalization ability and outperforms the baseline by
30.8% of recall@1 on average. In the generation task, Atomas achieves
state-of-the-art results in both molecule captioning task and molecule
generation task. Moreover, the visualization of the Hierarchical Adaptive
Alignment model further confirms the chemical significance of our approach. Our
codes can be found at https://anonymous.4open.science/r/Atomas-03C3. | q-bio.QM | Atomas: Hierarchical Alignment on Molecule-Text for Unified Molecule Understanding and Generation |
2024-04-23T12:33:14Z | http://arxiv.org/pdf/2404.14977v1 | http://arxiv.org/abs/2404.14977v1 | Muhammad Asif Auyb, Muhammad Tayyab Zamir, Imran Khan, Hannia Naseem, Nasir Ahmad, Kashif Ahmad | 2024-04-23T12:33:14Z | cs.SI, cs.CL | This paper focuses on a very important societal challenge of water quality
analysis. Being one of the key factors in the economic and social development
of society, the provision of water and ensuring its quality has always remained
one of the top priorities of public authorities. To ensure the quality of
water, different methods for monitoring and assessing the water networks, such
as offline and online surveys, are used. However, these surveys have several
limitations, such as the limited number of participants and low frequency due
to the labor involved in conducting such surveys. In this paper, we propose a
Natural Language Processing (NLP) framework to automatically collect and
analyze water-related posts from social media for data-driven decisions. The
proposed framework is composed of two components, namely (i) text
classification, and (ii) topic modeling. For text classification, we propose a
merit-fusion-based framework incorporating several Large Language Models (LLMs)
where different weight selection and optimization methods are employed to
assign weights to the LLMs. In topic modeling, we employed the BERTopic library
to discover the hidden topic patterns in the water-related tweets. We also
analyzed relevant tweets originating from different regions and countries to
explore global, regional, and country-specific issues and water-related
concerns. We also collected and manually annotated a large-scale dataset, which
is expected to facilitate future research on the topic. | cs.SI | Social Media and Artificial Intelligence for Sustainable Cities and Societies: A Water Quality Analysis Use-case |
2024-04-23T12:16:05Z | http://arxiv.org/pdf/2404.14963v3 | http://arxiv.org/abs/2404.14963v3 | Qihuang Zhong, Kang Wang, Ziyang Xu, Juhua Liu, Liang Ding, Bo Du, Dacheng Tao | 2024-05-29T15:27:53Z | cs.CL, cs.AI | Chain-of-Thought (CoT) prompting has enhanced the performance of Large
Language Models (LLMs) across various reasoning tasks. However, CoT still falls
short in dealing with complex math word problems, as it usually suffers from
three pitfalls: semantic misunderstanding errors, calculation errors and
step-missing errors. Prior studies involve addressing the calculation errors
and step-missing errors, but neglect the semantic misunderstanding errors,
which is the major factor limiting the LLMs' performance. To this end, we
propose a simple-yet-effective method, namely Deeply Understanding the Problems
(DUP), to improve the LLMs' math problem-solving ability by addressing semantic
misunderstanding errors. The core of our method is to encourage the LLMs to
deeply understand the problems and extract the key problem-solving information
used for better reasoning. Extensive experiments on 10 diverse reasoning
benchmarks show that our DUP method consistently outperforms the other
counterparts by a large margin. More encouragingly, DUP achieves a new SOTA
result on the GSM8K benchmark, with an accuracy of 97.1% under zero-shot
setting. | cs.CL | Achieving >97% on GSM8K: Deeply Understanding the Problems Makes LLMs Better Solvers for Math Word Problems |
2024-04-23T11:41:35Z | http://arxiv.org/pdf/2404.14946v1 | http://arxiv.org/abs/2404.14946v1 | Sen Liu, Yiwei Guo, Xie Chen, Kai Yu | 2024-04-23T11:41:35Z | cs.SD, cs.CL, eess.AS | While acoustic expressiveness has long been studied in expressive
text-to-speech (ETTS), the inherent expressiveness in text lacks sufficient
attention, especially for ETTS of artistic works. In this paper, we introduce
StoryTTS, a highly ETTS dataset that contains rich expressiveness both in
acoustic and textual perspective, from the recording of a Mandarin storytelling
show. A systematic and comprehensive labeling framework is proposed for textual
expressiveness. We analyze and define speech-related textual expressiveness in
StoryTTS to include five distinct dimensions through linguistics, rhetoric,
etc. Then we employ large language models and prompt them with a few manual
annotation examples for batch annotation. The resulting corpus contains 61
hours of consecutive and highly prosodic speech equipped with accurate text
transcriptions and rich textual expressiveness annotations. Therefore, StoryTTS
can aid future ETTS research to fully mine the abundant intrinsic textual and
acoustic features. Experiments are conducted to validate that TTS models can
generate speech with improved expressiveness when integrating with the
annotated textual labels in StoryTTS. | cs.SD | StoryTTS: A Highly Expressive Text-to-Speech Dataset with Rich Textual Expressiveness Annotations |
2024-04-23T11:40:30Z | http://arxiv.org/pdf/2404.14943v1 | http://arxiv.org/abs/2404.14943v1 | Julien Delaunay, Luis Galárraga, Christine Largouët | 2024-04-23T11:40:30Z | cs.CL, cs.LG, 68T50 | Although counterfactual explanations are a popular approach to explain ML
black-box classifiers, they are less widespread in NLP. Most methods find those
explanations by iteratively perturbing the target document until it is
classified differently by the black box. We identify two main families of
counterfactual explanation methods in the literature, namely, (a)
\emph{transparent} methods that perturb the target by adding, removing, or
replacing words, and (b) \emph{opaque} approaches that project the target
document into a latent, non-interpretable space where the perturbation is
carried out subsequently. This article offers a comparative study of the
performance of these two families of methods on three classical NLP tasks. Our
empirical evidence shows that opaque approaches can be an overkill for
downstream applications such as fake news detection or sentiment analysis since
they add an additional level of complexity with no significant performance
gain. These observations motivate our discussion, which raises the question of
whether it makes sense to explain a black box using another black box. | cs.CL | Does It Make Sense to Explain a Black Box With Another Black Box? |
2024-04-23T11:13:39Z | http://arxiv.org/pdf/2404.14928v2 | http://arxiv.org/abs/2404.14928v2 | Wenqi Fan, Shijie Wang, Jiani Huang, Zhikai Chen, Yu Song, Wenzhuo Tang, Haitao Mao, Hui Liu, Xiaorui Liu, Dawei Yin, Qing Li | 2024-06-04T01:31:30Z | cs.LG, cs.AI, cs.CL, cs.SI | Graphs play an important role in representing complex relationships in
various domains like social networks, knowledge graphs, and molecular
discovery. With the advent of deep learning, Graph Neural Networks (GNNs) have
emerged as a cornerstone in Graph Machine Learning (Graph ML), facilitating the
representation and processing of graph structures. Recently, LLMs have
demonstrated unprecedented capabilities in language tasks and are widely
adopted in a variety of applications such as computer vision and recommender
systems. This remarkable success has also attracted interest in applying LLMs
to the graph domain. Increasing efforts have been made to explore the potential
of LLMs in advancing Graph ML's generalization, transferability, and few-shot
learning ability. Meanwhile, graphs, especially knowledge graphs, are rich in
reliable factual knowledge, which can be utilized to enhance the reasoning
capabilities of LLMs and potentially alleviate their limitations such as
hallucinations and the lack of explainability. Given the rapid progress of this
research direction, a systematic review summarizing the latest advancements for
Graph ML in the era of LLMs is necessary to provide an in-depth understanding
to researchers and practitioners. Therefore, in this survey, we first review
the recent developments in Graph ML. We then explore how LLMs can be utilized
to enhance the quality of graph features, alleviate the reliance on labeled
data, and address challenges such as graph heterogeneity and
out-of-distribution (OOD) generalization. Afterward, we delve into how graphs
can enhance LLMs, highlighting their abilities to enhance LLM pre-training and
inference. Furthermore, we investigate various applications and discuss the
potential future directions in this promising field. | cs.LG | Graph Machine Learning in the Era of Large Language Models (LLMs) |
2024-04-23T10:57:59Z | http://arxiv.org/pdf/2404.14914v1 | http://arxiv.org/abs/2404.14914v1 | Kostiantyn Omelianchuk, Andrii Liubonko, Oleksandr Skurzhanskyi, Artem Chernodub, Oleksandr Korniienko, Igor Samokhin | 2024-04-23T10:57:59Z | cs.CL | In this paper, we carry out experimental research on Grammatical Error
Correction, delving into the nuances of single-model systems, comparing the
efficiency of ensembling and ranking methods, and exploring the application of
large language models to GEC as single-model systems, as parts of ensembles,
and as ranking methods. We set new state-of-the-art performance with F_0.5
scores of 72.8 on CoNLL-2014-test and 81.4 on BEA-test, respectively. To
support further advancements in GEC and ensure the reproducibility of our
research, we make our code, trained models, and systems' outputs publicly
available. | cs.CL | Pillars of Grammatical Error Correction: Comprehensive Inspection Of Contemporary Approaches In The Era of Large Language Models |
2024-04-23T10:34:16Z | http://arxiv.org/pdf/2404.14901v2 | http://arxiv.org/abs/2404.14901v2 | Ranim Khojah, Mazen Mohamad, Philipp Leitner, Francisco Gomes de Oliveira Neto | 2024-05-21T12:53:30Z | cs.SE, cs.AI, cs.CL, cs.HC, cs.LG | Large Language Models (LLMs) are frequently discussed in academia and the
general public as support tools for virtually any use case that relies on the
production of text, including software engineering. Currently there is much
debate, but little empirical evidence, regarding the practical usefulness of
LLM-based tools such as ChatGPT for engineers in industry. We conduct an
observational study of 24 professional software engineers who have been using
ChatGPT over a period of one week in their jobs, and qualitatively analyse
their dialogues with the chatbot as well as their overall experience (as
captured by an exit survey). We find that, rather than expecting ChatGPT to
generate ready-to-use software artifacts (e.g., code), practitioners more often
use ChatGPT to receive guidance on how to solve their tasks or learn about a
topic in more abstract terms. We also propose a theoretical framework for how
(i) purpose of the interaction, (ii) internal factors (e.g., the user's
personality), and (iii) external factors (e.g., company policy) together shape
the experience (in terms of perceived usefulness and trust). We envision that
our framework can be used by future research to further the academic discussion
on LLM usage by software engineering practitioners, and to serve as a reference
point for the design of future empirical LLM research in this domain. | cs.SE | Beyond Code Generation: An Observational Study of ChatGPT Usage in Software Engineering Practice |
2024-04-23T10:25:45Z | http://arxiv.org/pdf/2404.14897v1 | http://arxiv.org/abs/2404.14897v1 | Chen Zhang, Zhuorui Liu, Dawei Song | 2024-04-23T10:25:45Z | cs.CL, cs.AI | With the increasingly giant scales of (causal) large language models (LLMs),
the inference efficiency comes as one of the core concerns along the improved
performance. In contrast to the memory footprint, the latency bottleneck seems
to be of greater importance as there can be billions of requests to a LLM
(e.g., GPT-4) per day. The bottleneck is mainly due to the autoregressive
innateness of LLMs, where tokens can only be generated sequentially during
decoding. To alleviate the bottleneck, the idea of speculative execution, which
originates from the field of computer architecture, is introduced to LLM
decoding in a \textit{draft-then-verify} style. Under this regime, a sequence
of tokens will be drafted in a fast pace by utilizing some heuristics, and then
the tokens shall be verified in parallel by the LLM. As the costly sequential
inference is parallelized, LLM decoding speed can be significantly boosted.
Driven by the success of LLMs in recent couple of years, a growing literature
in this direction has emerged. Yet, there lacks a position survey to summarize
the current landscape and draw a roadmap for future development of this
promising area. To meet this demand, we present the very first survey paper
that reviews and unifies literature of speculative execution in LLMs (e.g.,
blockwise parallel decoding, speculative decoding, etc.) in a comprehensive
framework and a systematic taxonomy. Based on the taxonomy, we present a
critical review and comparative analysis of the current arts. Finally we
highlight various key challenges and future directions to further develop the
area. | cs.CL | Beyond the Speculative Game: A Survey of Speculative Execution in Large Language Models |
2024-04-23T10:09:46Z | http://arxiv.org/pdf/2404.14883v1 | http://arxiv.org/abs/2404.14883v1 | Vittoria Dentella, Fritz Guenther, Evelina Leivada | 2024-04-23T10:09:46Z | cs.CL | Understanding the limits of language is a prerequisite for Large Language
Models (LLMs) to act as theories of natural language. LLM performance in some
language tasks presents both quantitative and qualitative differences from that
of humans, however it remains to be determined whether such differences are
amenable to model size. This work investigates the critical role of model
scaling, determining whether increases in size make up for such differences
between humans and models. We test three LLMs from different families (Bard,
137 billion parameters; ChatGPT-3.5, 175 billion; ChatGPT-4, 1.5 trillion) on a
grammaticality judgment task featuring anaphora, center embedding,
comparatives, and negative polarity. N=1,200 judgments are collected and scored
for accuracy, stability, and improvements in accuracy upon repeated
presentation of a prompt. Results of the best performing LLM, ChatGPT-4, are
compared to results of n=80 humans on the same stimuli. We find that increased
model size may lead to better performance, but LLMs are still not sensitive to
(un)grammaticality as humans are. It seems possible but unlikely that scaling
alone can fix this issue. We interpret these results by comparing language
learning in vivo and in silico, identifying three critical differences
concerning (i) the type of evidence, (ii) the poverty of the stimulus, and
(iii) the occurrence of semantic hallucinations due to impenetrable linguistic
reference. | cs.CL | Language in Vivo vs. in Silico: Size Matters but Larger Language Models Still Do Not Comprehend Language on a Par with Humans |
2024-04-23T09:05:37Z | http://arxiv.org/pdf/2404.14851v3 | http://arxiv.org/abs/2404.14851v3 | Xiaoxi Li, Jiajie Jin, Yujia Zhou, Yuyao Zhang, Peitian Zhang, Yutao Zhu, Zhicheng Dou | 2024-05-16T03:28:28Z | cs.IR, cs.AI, cs.CL | Information Retrieval (IR) systems are crucial tools for users to access
information, widely applied in scenarios like search engines, question
answering, and recommendation systems. Traditional IR methods, based on
similarity matching to return ranked lists of documents, have been reliable
means of information acquisition, dominating the IR field for years. With the
advancement of pre-trained language models, generative information retrieval
(GenIR) has emerged as a novel paradigm, gaining increasing attention in recent
years. Currently, research in GenIR can be categorized into two aspects:
generative document retrieval (GR) and reliable response generation. GR
leverages the generative model's parameters for memorizing documents, enabling
retrieval by directly generating relevant document identifiers without explicit
indexing. Reliable response generation, on the other hand, employs language
models to directly generate the information users seek, breaking the
limitations of traditional IR in terms of document granularity and relevance
matching, offering more flexibility, efficiency, and creativity, thus better
meeting practical needs. This paper aims to systematically review the latest
research progress in GenIR. We will summarize the advancements in GR regarding
model training, document identifier, incremental learning, downstream tasks
adaptation, multi-modal GR and generative recommendation, as well as progress
in reliable response generation in aspects of internal knowledge memorization,
external knowledge augmentation, generating response with citations and
personal information assistant. We also review the evaluation, challenges and
future prospects in GenIR systems. This review aims to offer a comprehensive
reference for researchers in the GenIR field, encouraging further development
in this area. | cs.IR | From Matching to Generation: A Survey on Generative Information Retrieval |
2024-04-23T09:05:09Z | http://arxiv.org/pdf/2404.14850v1 | http://arxiv.org/abs/2404.14850v1 | Yang Tan, Mingchen Li, Bingxin Zhou, Bozitao Zhong, Lirong Zheng, Pan Tan, Ziyi Zhou, Huiqun Yu, Guisheng Fan, Liang Hong | 2024-04-23T09:05:09Z | cs.CL, cs.LG, q-bio.BM | Fine-tuning Pre-trained protein language models (PLMs) has emerged as a
prominent strategy for enhancing downstream prediction tasks, often
outperforming traditional supervised learning approaches. As a widely applied
powerful technique in natural language processing, employing
Parameter-Efficient Fine-Tuning techniques could potentially enhance the
performance of PLMs. However, the direct transfer to life science tasks is
non-trivial due to the different training strategies and data forms. To address
this gap, we introduce SES-Adapter, a simple, efficient, and scalable adapter
method for enhancing the representation learning of PLMs. SES-Adapter
incorporates PLM embeddings with structural sequence embeddings to create
structure-aware representations. We show that the proposed method is compatible
with different PLM architectures and across diverse tasks. Extensive
evaluations are conducted on 2 types of folding structures with notable quality
differences, 9 state-of-the-art baselines, and 9 benchmark datasets across
distinct downstream tasks. Results show that compared to vanilla PLMs,
SES-Adapter improves downstream task performance by a maximum of 11% and an
average of 3%, with significantly accelerated training speed by a maximum of
1034% and an average of 362%, the convergence rate is also improved by
approximately 2 times. Moreover, positive optimization is observed even with
low-quality predicted structures. The source code for SES-Adapter is available
at https://github.com/tyang816/SES-Adapter. | cs.CL | Simple, Efficient and Scalable Structure-aware Adapter Boosts Protein Language Models |
2024-04-23T08:39:29Z | http://arxiv.org/pdf/2404.14831v2 | http://arxiv.org/abs/2404.14831v2 | Tianshu Wang, Hongyu Lin, Xianpei Han, Xiaoyang Chen, Boxi Cao, Le Sun | 2024-04-25T06:37:51Z | cs.DB, cs.CL, cs.IR | Blocking is a critical step in entity resolution, and the emergence of neural
network-based representation models has led to the development of dense
blocking as a promising approach for exploring deep semantics in blocking.
However, previous advanced self-supervised dense blocking approaches require
domain-specific training on the target domain, which limits the benefits and
rapid adaptation of these methods. To address this issue, we propose
UniBlocker, a dense blocker that is pre-trained on a domain-independent,
easily-obtainable tabular corpus using self-supervised contrastive learning. By
conducting domain-independent pre-training, UniBlocker can be adapted to
various downstream blocking scenarios without requiring domain-specific
fine-tuning. To evaluate the universality of our entity blocker, we also
construct a new benchmark covering a wide range of blocking tasks from multiple
domains and scenarios. Our experiments show that the proposed UniBlocker,
without any domain-specific learning, significantly outperforms previous self-
and unsupervised dense blocking methods and is comparable and complementary to
the state-of-the-art sparse blocking methods. | cs.DB | Towards Universal Dense Blocking for Entity Resolution |
2024-04-23T08:29:56Z | http://arxiv.org/pdf/2404.14827v1 | http://arxiv.org/abs/2404.14827v1 | Jingxuan Wei, Linzhuang Sun, Yichong Leng, Xu Tan, Bihui Yu, Ruifeng Guo | 2024-04-23T08:29:56Z | cs.CL | Knowledge distillation, transferring knowledge from a teacher model to a
student model, has emerged as a powerful technique in neural machine
translation for compressing models or simplifying training targets. Knowledge
distillation encompasses two primary methods: sentence-level distillation and
token-level distillation. In sentence-level distillation, the student model is
trained to align with the output of the teacher model, which can alleviate the
training difficulty and give student model a comprehensive understanding of
global structure. Differently, token-level distillation requires the student
model to learn the output distribution of the teacher model, facilitating a
more fine-grained transfer of knowledge. Studies have revealed divergent
performances between sentence-level and token-level distillation across
different scenarios, leading to the confusion on the empirical selection of
knowledge distillation methods. In this study, we argue that token-level
distillation, with its more complex objective (i.e., distribution), is better
suited for ``simple'' scenarios, while sentence-level distillation excels in
``complex'' scenarios. To substantiate our hypothesis, we systematically
analyze the performance of distillation methods by varying the model size of
student models, the complexity of text, and the difficulty of decoding
procedure. While our experimental results validate our hypothesis, defining the
complexity level of a given scenario remains a challenging task. So we further
introduce a novel hybrid method that combines token-level and sentence-level
distillation through a gating mechanism, aiming to leverage the advantages of
both individual methods. Experiments demonstrate that the hybrid method
surpasses the performance of token-level or sentence-level distillation methods
and the previous works by a margin, demonstrating the effectiveness of the
proposed hybrid method. | cs.CL | Sentence-Level or Token-Level? A Comprehensive Study on Knowledge Distillation |
2024-04-23T07:50:00Z | http://arxiv.org/pdf/2404.14812v1 | http://arxiv.org/abs/2404.14812v1 | Yufeng Zhang, Xuepeng Wang, Lingxiang Wu, Jinqiao Wang | 2024-04-23T07:50:00Z | cs.CL | Chain-of-thought (CoT) prompting can guide language models to engage in
complex multi-step reasoning. The quality of provided demonstrations
significantly impacts the success of downstream inference tasks. While existing
automated methods prioritize accuracy and semantics in these demonstrations, we
show that the underlying reasoning patterns play a more crucial role in such
tasks. In this paper, we propose Pattern-Aware CoT, a prompting method that
considers the diversity of demonstration patterns. By incorporating patterns
such as step length and reasoning process within intermediate steps, PA-CoT
effectively mitigates the issue of bias induced by demonstrations and enables
better generalization to diverse scenarios. We conduct experiments on nine
reasoning benchmark tasks using two open-source LLMs. The results show that our
method substantially enhances reasoning performance and exhibits robustness to
errors. The code will be made publicly available. | cs.CL | Pattern-Aware Chain-of-Thought Prompting in Large Language Models |
2024-04-23T07:39:24Z | http://arxiv.org/pdf/2404.14809v1 | http://arxiv.org/abs/2404.14809v1 | Wenbo Shang, Xin Huang | 2024-04-23T07:39:24Z | cs.CL, cs.AI, cs.DB | A graph is a fundamental data model to represent various entities and their
complex relationships in society and nature, such as social networks,
transportation networks, financial networks, and biomedical systems. Recently,
large language models (LLMs) have showcased a strong generalization ability to
handle various NLP and multi-mode tasks to answer users' arbitrary questions
and specific-domain content generation. Compared with graph learning models,
LLMs enjoy superior advantages in addressing the challenges of generalizing
graph tasks by eliminating the need for training graph learning models and
reducing the cost of manual annotation. In this survey, we conduct a
comprehensive investigation of existing LLM studies on graph data, which
summarizes the relevant graph analytics tasks solved by advanced LLM models and
points out the existing remaining challenges and future directions.
Specifically, we study the key problems of LLM-based generative graph analytics
(LLM-GGA) with three categories: LLM-based graph query processing (LLM-GQP),
LLM-based graph inference and learning (LLM-GIL), and graph-LLM-based
applications. LLM-GQP focuses on an integration of graph analytics techniques
and LLM prompts, including graph understanding and knowledge graph (KG) based
augmented retrieval, while LLM-GIL focuses on learning and reasoning over
graphs, including graph learning, graph-formed reasoning and graph
representation. We summarize the useful prompts incorporated into LLM to handle
different graph downstream tasks. Moreover, we give a summary of LLM model
evaluation, benchmark datasets/tasks, and a deep pro and cons analysis of LLM
models. We also explore open problems and future directions in this exciting
interdisciplinary research area of LLMs and graph analytics. | cs.CL | A Survey of Large Language Models on Generative Graph Analytics: Query, Learning, and Applications |
2024-04-23T07:19:20Z | http://arxiv.org/pdf/2404.14795v3 | http://arxiv.org/abs/2404.14795v3 | Jiaming He, Wenbo Jiang, Guanyu Hou, Wenshu Fan, Rui Zhang, Hongwei Li | 2024-05-11T10:40:58Z | cs.CL, cs.CR, cs.LG | Mainstream poisoning attacks on large language models (LLMs) typically set a
fixed trigger in the input instance and specific responses for triggered
queries. However, the fixed trigger setting (e.g., unusual words) may be easily
detected by human detection, limiting the effectiveness and practicality in
real-world scenarios. To enhance the stealthiness of the trigger, we present a
poisoning attack against LLMs that is triggered by a generation/output
condition-token limitation, which is a commonly adopted strategy by users for
reducing costs. The poisoned model performs normally for output without token
limitation, while becomes harmful for output with limited tokens. To achieve
this objective, we introduce BrieFool, an efficient attack framework. It
leverages the characteristics of generation limitation by efficient instruction
sampling and poisoning data generation, thereby influencing the behavior of
LLMs under target conditions. Our experiments demonstrate that BrieFool is
effective across safety domains and knowledge domains. For instance, with only
20 generated poisoning examples against GPT-3.5-turbo, BrieFool achieves a 100%
Attack Success Rate (ASR) and a 9.28/10 average Harmfulness Score (HS) under
token limitation conditions while maintaining the benign performance. | cs.CL | Talk Too Much: Poisoning Large Language Models under Token Limit |
2024-04-23T06:36:21Z | http://arxiv.org/pdf/2404.14779v1 | http://arxiv.org/abs/2404.14779v1 | Clément Christophe, Praveen K Kanithi, Prateek Munjal, Tathagata Raha, Nasir Hayat, Ronnie Rajan, Ahmed Al-Mahrooqi, Avani Gupta, Muhammad Umar Salman, Gurpreet Gosal, Bhargav Kanakiya, Charles Chen, Natalia Vassilieva, Boulbaba Ben Amor, Marco AF Pimentel, Shadab Khan | 2024-04-23T06:36:21Z | cs.CL | This study presents a comprehensive analysis and comparison of two
predominant fine-tuning methodologies - full-parameter fine-tuning and
parameter-efficient tuning - within the context of medical Large Language
Models (LLMs). We developed and refined a series of LLMs, based on the Llama-2
architecture, specifically designed to enhance medical knowledge retrieval,
reasoning, and question-answering capabilities. Our experiments systematically
evaluate the effectiveness of these tuning strategies across various well-known
medical benchmarks. Notably, our medical LLM Med42 showed an accuracy level of
72% on the US Medical Licensing Examination (USMLE) datasets, setting a new
standard in performance for openly available medical LLMs. Through this
comparative analysis, we aim to identify the most effective and efficient
method for fine-tuning LLMs in the medical domain, thereby contributing
significantly to the advancement of AI-driven healthcare applications. | cs.CL | Med42 -- Evaluating Fine-Tuning Strategies for Medical LLMs: Full-Parameter vs. Parameter-Efficient Approaches |
2024-04-23T06:30:53Z | http://arxiv.org/pdf/2404.14777v1 | http://arxiv.org/abs/2404.14777v1 | Ling Yue, Tianfan Fu | 2024-04-23T06:30:53Z | cs.CL, cs.LG | Large Language Models (LLMs) and multi-agent systems have shown impressive
capabilities in natural language tasks but face challenges in clinical trial
applications, primarily due to limited access to external knowledge.
Recognizing the potential of advanced clinical trial tools that aggregate and
predict based on the latest medical data, we propose an integrated solution to
enhance their accessibility and utility. We introduce Clinical Agent System
(CT-Agent), a Clinical multi-agent system designed for clinical trial tasks,
leveraging GPT-4, multi-agent architectures, LEAST-TO-MOST, and ReAct reasoning
technology. This integration not only boosts LLM performance in clinical
contexts but also introduces novel functionalities. Our system autonomously
manages the entire clinical trial process, demonstrating significant efficiency
improvements in our evaluations, which include both computational benchmarks
and expert feedback. | cs.CL | CT-Agent: Clinical Trial Multi-Agent with Large Language Model-based Reasoning |
2024-04-23T06:23:34Z | http://arxiv.org/pdf/2404.14772v1 | http://arxiv.org/abs/2404.14772v1 | Chris Samarinas, Pracha Promthaw, Atharva Nijasure, Hansi Zeng, Julian Killingback, Hamed Zamani | 2024-04-23T06:23:34Z | cs.CL | This paper explores SynTOD, a new synthetic data generation approach for
developing end-to-end Task-Oriented Dialogue (TOD) Systems capable of handling
complex tasks such as intent classification, slot filling, conversational
question-answering, and retrieval-augmented response generation, without
relying on crowdsourcing or real-world data. SynTOD utilizes a state transition
graph to define the desired behavior of a TOD system and generates diverse,
structured conversations through random walks and response simulation using
large language models (LLMs). In our experiments, using graph-guided response
simulations leads to significant improvements in intent classification, slot
filling and response relevance compared to naive single-prompt simulated
conversations. We also investigate the end-to-end TOD effectiveness of
different base and instruction-tuned LLMs, with and without the constructed
synthetic conversations. Finally, we explore how various LLMs can evaluate
responses in a TOD system and how well they are correlated with human
judgments. Our findings pave the path towards quick development and evaluation
of domain-specific TOD systems. We release our datasets, models, and code for
research purposes. | cs.CL | Simulating Task-Oriented Dialogues with State Transition Graphs and Large Language Models |
2024-04-23T05:51:45Z | http://arxiv.org/pdf/2404.14760v2 | http://arxiv.org/abs/2404.14760v2 | Sanat Sharma, David Seunghyun Yoon, Franck Dernoncourt, Dewang Sultania, Karishma Bagga, Mengjiao Zhang, Trung Bui, Varun Kotte | 2024-05-29T16:18:02Z | cs.CL, cs.AI, cs.IR, cs.LG | Question answering (QA) has become an important application in the advanced
development of large language models. General pre-trained large language models
for question-answering are not trained to properly understand the knowledge or
terminology for a specific domain, such as finance, healthcare, education, and
customer service for a product. To better cater to domain-specific
understanding, we build an in-house question-answering system for Adobe
products. We propose a novel framework to compile a large question-answer
database and develop the approach for retrieval-aware finetuning of a Large
Language model. We showcase that fine-tuning the retriever leads to major
improvements in the final generation. Our overall approach reduces
hallucinations during generation while keeping in context the latest retrieval
information for contextual grounding. | cs.CL | Retrieval Augmented Generation for Domain-specific Question Answering |
2024-04-23T05:11:08Z | http://arxiv.org/pdf/2404.14749v2 | http://arxiv.org/abs/2404.14749v2 | Yukio Ohsawa, Dingming Xue, Kaira Sekiguchi | 2024-04-26T22:16:59Z | cs.LG, cs.CL | Previous models for learning the semantic vectors of items and their groups,
such as words, sentences, nodes, and graphs, using distributed representation
have been based on the assumption that the basic sense of an item corresponds
to one vector composed of dimensions corresponding to hidden contexts in the
target real world, from which multiple senses of the item are obtained by
conforming to lexical databases or adapting to the context. However, there may
be multiple senses of an item, which are hardly assimilated and change or
evolve dynamically following the contextual shift even within a document or a
restricted period. This is a process similar to the evolution or adaptation of
a living entity with/to environmental shifts. Setting the scope of
disambiguation of items for sensemaking, the author presents a method in which
a word or item in the data embraces multiple semantic vectors that evolve via
interaction with others, similar to a cell embracing chromosomes crossing over
with each other. We obtained two preliminary results: (1) the role of a word
that evolves to acquire the largest or lower-middle variance of semantic
vectors tends to be explainable by the author of the text; (2) the epicenters
of earthquakes that acquire larger variance via crossover, corresponding to the
interaction with diverse areas of land crust, are likely to correspond to the
epicenters of forthcoming large earthquakes. | cs.LG | Semantic Cells: Evolutional Process to Acquire Sense Diversity of Items |
2024-04-23T04:56:48Z | http://arxiv.org/pdf/2405.00705v1 | http://arxiv.org/abs/2405.00705v1 | Yexiao He, Ziyao Wang, Zheyu Shen, Guoheng Sun, Yucong Dai, Yongkai Wu, Hongyi Wang, Ang Li | 2024-04-23T04:56:48Z | cs.CL, cs.LG | The pre-trained Large Language Models (LLMs) can be adapted for many
downstream tasks and tailored to align with human preferences through
fine-tuning. Recent studies have discovered that LLMs can achieve desirable
performance with only a small amount of high-quality data, suggesting that a
large amount of the data in these extensive datasets is redundant or even
harmful. Identifying high-quality data from vast datasets to curate small yet
effective datasets has emerged as a critical challenge. In this paper, we
introduce SHED, an automated dataset refinement framework based on Shapley
value for instruction fine-tuning. SHED eliminates the need for human
intervention or the use of commercial LLMs. Moreover, the datasets curated
through SHED exhibit transferability, indicating they can be reused across
different LLMs with consistently high performance. We conduct extensive
experiments to evaluate the datasets curated by SHED. The results demonstrate
SHED's superiority over state-of-the-art methods across various tasks and LLMs;
notably, datasets comprising only 10% of the original data selected by SHED
achieve performance comparable to or surpassing that of the full datasets. | cs.CL | SHED: Shapley-Based Automated Dataset Refinement for Instruction Fine-Tuning |
2024-04-23T04:47:22Z | http://arxiv.org/pdf/2404.14740v2 | http://arxiv.org/abs/2404.14740v2 | Ben Hutchinson | 2024-06-25T01:02:16Z | cs.CL | This position paper concerns the use of religious texts in Natural Language
Processing (NLP), which is of special interest to the Ethics of NLP. Religious
texts are expressions of culturally important values, and machine learned
models have a propensity to reproduce cultural values encoded in their training
data. Furthermore, translations of religious texts are frequently used by NLP
researchers when language data is scarce. This repurposes the translations from
their original uses and motivations, which often involve attracting new
followers. This paper argues that NLP's use of such texts raises considerations
that go beyond model biases, including data provenance, cultural contexts, and
their use in proselytism. We argue for more consideration of researcher
positionality, and of the perspectives of marginalized linguistic and religious
communities. | cs.CL | Modeling the Sacred: Considerations when Using Religious Texts in Natural Language Processing |
2024-04-23T04:47:22Z | http://arxiv.org/pdf/2404.14741v1 | http://arxiv.org/abs/2404.14741v1 | Yao Xu, Shizhu He, Jiabei Chen, Zihao Wang, Yangqiu Song, Hanghang Tong, Kang Liu, Jun Zhao | 2024-04-23T04:47:22Z | cs.CL, cs.AI | To address the issue of insufficient knowledge and the tendency to generate
hallucination in Large Language Models (LLMs), numerous studies have endeavored
to integrate LLMs with Knowledge Graphs (KGs). However, all these methods are
evaluated on conventional Knowledge Graph Question Answering (KGQA) with
complete KGs, where the factual triples involved in each question are entirely
covered by the given KG. In this situation, LLM mainly acts as an agent to find
answer entities by exploring the KG, rather than effectively integrating
internal and external knowledge sources. However, in real-world scenarios, KGs
are often incomplete to cover all the knowledge required to answer questions.
To simulate real-world scenarios and evaluate the ability of LLMs to integrate
internal and external knowledge, in this paper, we propose leveraging LLMs for
QA under Incomplete Knowledge Graph (IKGQA), where the given KG doesn't include
all the factual triples involved in each question. To handle IKGQA, we propose
a training-free method called Generate-on-Graph (GoG) that can generate new
factual triples while exploring on KGs. Specifically, we propose a
selecting-generating-answering framework, which not only treat the LLM as an
agent to explore on KGs, but also treat it as a KG to generate new facts based
on the explored subgraph and its inherent knowledge. Experimental results on
two datasets demonstrate that our GoG can solve IKGQA to a certain extent,
while almost all previous methods cannot perform well on IKGQA. | cs.CL | Generate-on-Graph: Treat LLM as both Agent and KG in Incomplete Knowledge Graph Question Answering |
2024-04-23T04:33:49Z | http://arxiv.org/pdf/2404.14736v1 | http://arxiv.org/abs/2404.14736v1 | Katie Seaborn, Jacqueline Urakami, Peter Pennefather, Norihisa P. Miyake | 2024-04-23T04:33:49Z | cs.HC, cs.AI, cs.CL, cs.CY, cs.SD, eess.AS | Voice is a natural mode of expression offered by modern computer-based
systems. Qualitative perspectives on voice-based user experiences (voice UX)
offer rich descriptions of complex interactions that numbers alone cannot fully
represent. We conducted a systematic review of the literature on qualitative
approaches to voice UX, capturing the nature of this body of work in a
systematic map and offering a qualitative synthesis of findings. We highlight
the benefits of qualitative methods for voice UX research, identify
opportunities for increasing rigour in methods and outcomes, and distill
patterns of experience across a diversity of devices and modes of qualitative
praxis. | cs.HC | Qualitative Approaches to Voice UX |
2024-04-23T03:55:01Z | http://arxiv.org/pdf/2404.14723v1 | http://arxiv.org/abs/2404.14723v1 | Amir Saeidi, Shivanshu Verma, Chitta Baral | 2024-04-23T03:55:01Z | cs.CL | Large Language Models (LLMs) have demonstrated remarkable performance across
a spectrum of tasks. Recently, Direct Preference Optimization (DPO) has emerged
as an RL-free approach to optimize the policy model on human preferences.
However, several limitations hinder the widespread adoption of this method. To
address these shortcomings, various versions of DPO have been introduced. Yet,
a comprehensive evaluation of these variants across diverse tasks is still
lacking. In this study, we aim to bridge this gap by investigating the
performance of alignment methods across three distinct scenarios: (1) keeping
the Supervised Fine-Tuning (SFT) part, (2) skipping the SFT part, and (3)
skipping the SFT part and utilizing an instruction-tuned model. Furthermore, we
explore the impact of different training sizes on their performance. Our
evaluation spans a range of tasks including dialogue systems, reasoning,
mathematical problem-solving, question answering, truthfulness, and multi-task
understanding, encompassing 13 benchmarks such as MT-Bench, Big Bench, and Open
LLM Leaderboard. Key observations reveal that alignment methods achieve optimal
performance with smaller training data subsets, exhibit limited effectiveness
in reasoning tasks yet significantly impact mathematical problem-solving, and
employing an instruction-tuned model notably influences truthfulness. We
anticipate that our findings will catalyze further research aimed at developing
more robust models to address alignment challenges. | cs.CL | Insights into Alignment: Evaluating DPO and its Variants Across Multiple Tasks |
2024-04-23T03:42:48Z | http://arxiv.org/pdf/2404.14716v2 | http://arxiv.org/abs/2404.14716v2 | Siyin Wang, Chao-Han Huck Yang, Ji Wu, Chao Zhang | 2024-06-16T08:49:00Z | cs.CL, cs.AI, cs.CV, cs.SD, eess.AS | Large language models (LLMs) can adapt to new tasks through in-context
learning (ICL) based on a few examples presented in dialogue history without
any model parameter update. Despite such convenience, the performance of ICL
heavily depends on the quality of the in-context examples presented, which
makes the in-context example selection approach a critical choice. This paper
proposes a novel Bayesian in-Context example Selection method (ByCS) for ICL.
Extending the inference probability conditioned on in-context examples based on
Bayes' theorem, ByCS focuses on the inverse inference conditioned on test
input. Following the assumption that accurate inverse inference probability
(likelihood) will result in accurate inference probability (posterior),
in-context examples are selected based on their inverse inference results.
Diverse and extensive cross-tasking and cross-modality experiments are
performed with speech, text, and image examples. Experimental results show the
efficacy and robustness of our ByCS method on various models, tasks and
modalities. | cs.CL | Bayesian Example Selection Improves In-Context Learning for Speech, Text, and Visual Modalities |
2024-04-23T03:42:14Z | http://arxiv.org/pdf/2404.14715v1 | http://arxiv.org/abs/2404.14715v1 | Hang Hua, Jing Shi, Kushal Kafle, Simon Jenni, Daoan Zhang, John Collomosse, Scott Cohen, Jiebo Luo | 2024-04-23T03:42:14Z | cs.CV, cs.CL | Recent progress in large-scale pre-training has led to the development of
advanced vision-language models (VLMs) with remarkable proficiency in
comprehending and generating multimodal content. Despite the impressive ability
to perform complex reasoning for VLMs, current models often struggle to
effectively and precisely capture the compositional information on both the
image and text sides. To address this, we propose FineMatch, a new aspect-based
fine-grained text and image matching benchmark, focusing on text and image
mismatch detection and correction. This benchmark introduces a novel task for
boosting and evaluating the VLMs' compositionality for aspect-based
fine-grained text and image matching. In this task, models are required to
identify mismatched aspect phrases within a caption, determine the aspect's
class, and propose corrections for an image-text pair that may contain between
0 and 3 mismatches. To evaluate the models' performance on this new task, we
propose a new evaluation metric named ITM-IoU for which our experiments show a
high correlation to human evaluation. In addition, we also provide a
comprehensive experimental analysis of existing mainstream VLMs, including
fully supervised learning and in-context learning settings. We have found that
models trained on FineMatch demonstrate enhanced proficiency in detecting
fine-grained text and image mismatches. Moreover, models (e.g., GPT-4V, Gemini
Pro Vision) with strong abilities to perform multimodal in-context learning are
not as skilled at fine-grained compositional image and text matching analysis.
With FineMatch, we are able to build a system for text-to-image generation
hallucination detection and correction. | cs.CV | FINEMATCH: Aspect-based Fine-grained Image and Text Mismatch Detection and Correction |
2024-04-23T02:57:46Z | http://arxiv.org/pdf/2404.14700v3 | http://arxiv.org/abs/2404.14700v3 | Zhen Ye, Zeqian Ju, Haohe Liu, Xu Tan, Jianyi Chen, Yiwen Lu, Peiwen Sun, Jiahao Pan, Weizhen Bian, Shulin He, Qifeng Liu, Yike Guo, Wei Xue | 2024-04-25T03:38:46Z | eess.AS, cs.AI, cs.CL, cs.LG, cs.SD | Recent progress in large-scale zero-shot speech synthesis has been
significantly advanced by language models and diffusion models. However, the
generation process of both methods is slow and computationally intensive.
Efficient speech synthesis using a lower computing budget to achieve quality on
par with previous work remains a significant challenge. In this paper, we
present FlashSpeech, a large-scale zero-shot speech synthesis system with
approximately 5\% of the inference time compared with previous work.
FlashSpeech is built on the latent consistency model and applies a novel
adversarial consistency training approach that can train from scratch without
the need for a pre-trained diffusion model as the teacher. Furthermore, a new
prosody generator module enhances the diversity of prosody, making the rhythm
of the speech sound more natural. The generation processes of FlashSpeech can
be achieved efficiently with one or two sampling steps while maintaining high
audio quality and high similarity to the audio prompt for zero-shot speech
generation. Our experimental results demonstrate the superior performance of
FlashSpeech. Notably, FlashSpeech can be about 20 times faster than other
zero-shot speech synthesis systems while maintaining comparable performance in
terms of voice quality and similarity. Furthermore, FlashSpeech demonstrates
its versatility by efficiently performing tasks like voice conversion, speech
editing, and diverse speech sampling. Audio samples can be found in
https://flashspeech.github.io/. | eess.AS | FlashSpeech: Efficient Zero-Shot Speech Synthesis |
2024-04-23T02:54:00Z | http://arxiv.org/pdf/2404.14695v1 | http://arxiv.org/abs/2404.14695v1 | Tamanna Hossain, Sunipa Dev, Sameer Singh | 2024-04-23T02:54:00Z | cs.CL | Content Warning: This paper contains examples of misgendering and erasure
that could be offensive and potentially triggering.
Misgendering, the act of incorrectly addressing someone's gender, inflicts
serious harm and is pervasive in everyday technologies, yet there is a notable
lack of research to combat it. We are the first to address this lack of
research into interventions for misgendering by conducting a survey of
gender-diverse individuals in the US to understand perspectives about automated
interventions for text-based misgendering. Based on survey insights on the
prevalence of misgendering, desired solutions, and associated concerns, we
introduce a misgendering interventions task and evaluation dataset,
MisgenderMender. We define the task with two sub-tasks: (i) detecting
misgendering, followed by (ii) correcting misgendering where misgendering is
present in domains where editing is appropriate. MisgenderMender comprises 3790
instances of social media content and LLM-generations about non-cisgender
public figures, annotated for the presence of misgendering, with additional
annotations for correcting misgendering in LLM-generated text. Using this
dataset, we set initial benchmarks by evaluating existing NLP systems and
highlighting challenges for future models to address. We release the full
dataset, code, and demo at
https://tamannahossainkay.github.io/misgendermender/. | cs.CL | MisgenderMender: A Community-Informed Approach to Interventions for Misgendering |
2024-04-23T02:32:57Z | http://arxiv.org/pdf/2404.14687v1 | http://arxiv.org/abs/2404.14687v1 | Raehyuk Jung, Hyojun Go, Jaehyuk Yi, Jiho Jang, Daniel Kim, Jay Suh, Aiden Lee, Cooper Han, Jae Lee, Jeff Kim, Jin-Young Kim, Junwan Kim, Kyle Park, Lucas Lee, Mars Ha, Minjoon Seo, Abraham Jo, Ed Park, Hassan Kianinejad, SJ Kim, Tony Moon, Wade Jeong, Andrei Popescu, Esther Kim, EK Yoon, Genie Heo, Henry Choi, Jenna Kang, Kevin Han, Noah Seo, Sunny Nguyen, Ryan Won, Yeonhoo Park, Anthony Giuliani, Dave Chung, Hans Yoon, James Le, Jenny Ahn, June Lee, Maninder Saini, Meredith Sanders, Soyoung Lee, Sue Kim, Travis Couture | 2024-04-23T02:32:57Z | cs.MM, cs.AI, cs.CL, cs.CV | This technical report introduces Pegasus-1, a multimodal language model
specialized in video content understanding and interaction through natural
language. Pegasus-1 is designed to address the unique challenges posed by video
data, such as interpreting spatiotemporal information, to offer nuanced video
content comprehension across various lengths. This technical report overviews
Pegasus-1's architecture, training strategies, and its performance in
benchmarks on video conversation, zero-shot video question answering, and video
summarization. We also explore qualitative characteristics of Pegasus-1 ,
demonstrating its capabilities as well as its limitations, in order to provide
readers a balanced view of its current state and its future direction. | cs.MM | Pegasus-v1 Technical Report |
2024-04-23T02:19:35Z | http://arxiv.org/pdf/2404.14680v1 | http://arxiv.org/abs/2404.14680v1 | Elijah Pelofske, Vincent Urias, Lorie M. Liebrock | 2024-04-23T02:19:35Z | cs.CL, cs.AI, cs.LG | The task of accurate and efficient language translation is an extremely
important information processing task. Machine learning enabled and automated
translation that is accurate and fast is often a large topic of interest in the
machine learning and data science communities. In this study, we examine using
local Generative Pretrained Transformer (GPT) models to perform automated zero
shot black-box, sentence wise, multi-natural-language translation into English
text. We benchmark 16 different open-source GPT models, with no custom
fine-tuning, from the Huggingface LLM repository for translating 50 different
non-English languages into English using translated TED Talk transcripts as the
reference dataset. These GPT model inference calls are performed strictly
locally, on single A100 Nvidia GPUs. Benchmark metrics that are reported are
language translation accuracy, using BLEU, GLEU, METEOR, and chrF text overlap
measures, and wall-clock time for each sentence translation. The best overall
performing GPT model for translating into English text for the BLEU metric is
ReMM-v2-L2-13B with a mean score across all tested languages of $0.152$, for
the GLEU metric is ReMM-v2-L2-13B with a mean score across all tested languages
of $0.256$, for the chrF metric is Llama2-chat-AYT-13B with a mean score across
all tested languages of $0.448$, and for the METEOR metric is ReMM-v2-L2-13B
with a mean score across all tested languages of $0.438$. | cs.CL | Automated Multi-Language to English Machine Translation Using Generative Pre-Trained Transformers |
2024-04-23T01:46:32Z | http://arxiv.org/pdf/2404.14662v1 | http://arxiv.org/abs/2404.14662v1 | Ansong Ni, Miltiadis Allamanis, Arman Cohan, Yinlin Deng, Kensen Shi, Charles Sutton, Pengcheng Yin | 2024-04-23T01:46:32Z | cs.LG, cs.CL, cs.PL, cs.SE | A fundamental skill among human developers is the ability to understand and
reason about program execution. As an example, a programmer can mentally
simulate code execution in natural language to debug and repair code (aka.
rubber duck debugging). However, large language models (LLMs) of code are
typically trained on the surface textual form of programs, thus may lack a
semantic understanding of how programs execute at run-time. To address this
issue, we propose NExT, a method to teach LLMs to inspect the execution traces
of programs (variable states of executed lines) and reason about their run-time
behavior through chain-of-thought (CoT) rationales. Specifically, NExT uses
self-training to bootstrap a synthetic training set of execution-aware
rationales that lead to correct task solutions (e.g., fixed programs) without
laborious manual annotation. Experiments on program repair tasks based on MBPP
and HumanEval demonstrate that NExT improves the fix rate of a PaLM 2 model, by
26.1% and 14.3% absolute, respectively, with significantly improved rationale
quality as verified by automated metrics and human raters. Our model can also
generalize to scenarios where program traces are absent at test-time. | cs.LG | NExT: Teaching Large Language Models to Reason about Code Execution |
2024-04-23T00:05:48Z | http://arxiv.org/pdf/2404.14631v1 | http://arxiv.org/abs/2404.14631v1 | Chaohao Yang | 2024-04-23T00:05:48Z | cs.CL, cs.LG | Distributed word representation (a.k.a. word embedding) is a key focus in
natural language processing (NLP). As a highly successful word embedding model,
Word2Vec offers an efficient method for learning distributed word
representations on large datasets. However, Word2Vec lacks consideration for
distances between center and context words. We propose two novel methods,
Learnable Formulated Weights (LFW) and Epoch-based Dynamic Window Size (EDWS),
to incorporate distance information into two variants of Word2Vec, the
Continuous Bag-of-Words (CBOW) model and the Continuous Skip-gram (Skip-gram)
model. For CBOW, LFW uses a formula with learnable parameters that best
reflects the relationship of influence and distance between words to calculate
distance-related weights for average pooling, providing insights for future NLP
text modeling research. For Skip-gram, we improve its dynamic window size
strategy to introduce distance information in a more balanced way. Experiments
prove the effectiveness of LFW and EDWS in enhancing Word2Vec's performance,
surpassing previous state-of-the-art methods. | cs.CL | Learning Word Embedding with Better Distance Weighting and Window Size Scheduling |
2024-04-22T23:31:28Z | http://arxiv.org/pdf/2405.00704v2 | http://arxiv.org/abs/2405.00704v2 | Ming Liu, Ran Liu, Ye Zhu, Hua Wang, Youyang Qu, Rongsheng Li, Yongpan Sheng, Wray Buntine | 2024-05-10T02:38:28Z | cs.CL, cs.AI | ChatGPT has changed the AI community and an active research line is the
performance evaluation of ChatGPT. A key challenge for the evaluation is that
ChatGPT is still closed-source and traditional benchmark datasets may have been
used by ChatGPT as the training data. In this paper, (i) we survey recent
studies which uncover the real performance levels of ChatGPT in seven
categories of NLP tasks, (ii) review the social implications and safety issues
of ChatGPT, and (iii) emphasize key challenges and opportunities for its
evaluation. We hope our survey can shed some light on its blackbox manner, so
that researchers are not misleaded by its surface generation. | cs.CL | A Survey on the Real Power of ChatGPT |
2024-04-22T23:12:03Z | http://arxiv.org/pdf/2404.14619v2 | http://arxiv.org/abs/2404.14619v2 | Sachin Mehta, Mohammad Hossein Sekhavat, Qingqing Cao, Maxwell Horton, Yanzi Jin, Chenfan Sun, Iman Mirzadeh, Mahyar Najibi, Dmitry Belenko, Peter Zatloukal, Mohammad Rastegari | 2024-05-02T00:30:57Z | cs.CL, cs.AI, cs.LG | The reproducibility and transparency of large language models are crucial for
advancing open research, ensuring the trustworthiness of results, and enabling
investigations into data and model biases, as well as potential risks. To this
end, we release OpenELM, a state-of-the-art open language model. OpenELM uses a
layer-wise scaling strategy to efficiently allocate parameters within each
layer of the transformer model, leading to enhanced accuracy. For example, with
a parameter budget of approximately one billion parameters, OpenELM exhibits a
2.36% improvement in accuracy compared to OLMo while requiring $2\times$ fewer
pre-training tokens.
Diverging from prior practices that only provide model weights and inference
code, and pre-train on private datasets, our release includes the complete
framework for training and evaluation of the language model on publicly
available datasets, including training logs, multiple checkpoints, and
pre-training configurations. We also release code to convert models to MLX
library for inference and fine-tuning on Apple devices. This comprehensive
release aims to empower and strengthen the open research community, paving the
way for future open research endeavors.
Our source code along with pre-trained model weights and training recipes is
available at \url{https://github.com/apple/corenet}. Additionally, \model
models can be found on HuggingFace at:
\url{https://huggingface.co/apple/OpenELM}. | cs.CL | OpenELM: An Efficient Language Model Family with Open Training and Inference Framework |
2024-04-22T23:06:42Z | http://arxiv.org/pdf/2404.14618v1 | http://arxiv.org/abs/2404.14618v1 | Dujian Ding, Ankur Mallick, Chi Wang, Robert Sim, Subhabrata Mukherjee, Victor Ruhle, Laks V. S. Lakshmanan, Ahmed Hassan Awadallah | 2024-04-22T23:06:42Z | cs.LG, cs.AI, cs.CL | Large language models (LLMs) excel in most NLP tasks but also require
expensive cloud servers for deployment due to their size, while smaller models
that can be deployed on lower cost (e.g., edge) devices, tend to lag behind in
terms of response quality. Therefore in this work we propose a hybrid inference
approach which combines their respective strengths to save cost and maintain
quality. Our approach uses a router that assigns queries to the small or large
model based on the predicted query difficulty and the desired quality level.
The desired quality level can be tuned dynamically at test time to seamlessly
trade quality for cost as per the scenario requirements. In experiments our
approach allows us to make up to 40% fewer calls to the large model, with no
drop in response quality. | cs.LG | Hybrid LLM: Cost-Efficient and Quality-Aware Query Routing |
2024-04-22T22:04:16Z | http://arxiv.org/pdf/2404.14607v1 | http://arxiv.org/abs/2404.14607v1 | Yanhui Guo, Shaoyuan Xu, Jinmiao Fu, Jia Liu, Chaosheng Dong, Bryan Wang | 2024-04-22T22:04:16Z | cs.CL | This paper introduces \textbf{Q-tuning}, a novel approach for continual
prompt tuning that enables the lifelong learning of a pre-trained language
model. When learning a new task, Q-tuning trains a task-specific prompt by
adding it to a prompt queue consisting of the prompts from older tasks. To
better transfer the knowledge of old tasks, we design an adaptive knowledge
aggregation technique that reweighs previous prompts in the queue with a
learnable low-rank matrix. Once the prompt queue reaches its maximum capacity,
we leverage a PCA-based eviction rule to reduce the queue's size, allowing the
newly trained prompt to be added while preserving the primary knowledge of old
tasks. In order to mitigate the accumulation of information loss caused by the
eviction, we additionally propose a globally shared prefix prompt and a memory
retention regularization based on information theory. Extensive experiments
demonstrate that our approach outperforms the state-of-the-art methods
substantially on continual prompt tuning benchmarks. Moreover, our approach
enables lifelong learning on linearly growing task sequences while requiring
constant complexity for training and inference. | cs.CL | Q-Tuning: Queue-based Prompt Tuning for Lifelong Few-shot Language Learning |
2024-04-22T21:59:35Z | http://arxiv.org/pdf/2404.14604v3 | http://arxiv.org/abs/2404.14604v3 | Mengzhao Jia, Zhihan Zhang, Wenhao Yu, Fangkai Jiao, Meng Jiang | 2024-04-26T02:34:29Z | cs.CL | Open-source multimodal large language models (MLLMs) excel in various tasks
involving textual and visual inputs but still struggle with complex multimodal
mathematical reasoning, lagging behind proprietary models like GPT-4V(ision)
and Gemini-Pro. Although fine-tuning with intermediate steps (i.e., rationales)
elicits some mathematical reasoning skills, the resulting models still fall
short in visual comprehension due to inadequate visual-centric supervision,
which leads to inaccurate interpretation of math figures. To address this
issue, we propose a two-step training pipeline VCAR, which emphasizes the
Visual Comprehension training in Addition to mathematical Reasoning learning.
It first improves the visual comprehension ability of MLLMs through the visual
description generation task, followed by another training step on generating
rationales with the assistance of descriptions. Experimental results on two
popular benchmarks demonstrate that VCAR substantially outperforms baseline
methods solely relying on rationale supervision, especially on problems with
high visual demands. | cs.CL | Describe-then-Reason: Improving Multimodal Mathematical Reasoning through Visual Comprehension Training |
2024-04-22T21:50:01Z | http://arxiv.org/pdf/2404.14600v1 | http://arxiv.org/abs/2404.14600v1 | Hansi Zeng, Chen Luo, Hamed Zamani | 2024-04-22T21:50:01Z | cs.IR, cs.CL | This paper introduces PAG-a novel optimization and decoding approach that
guides autoregressive generation of document identifiers in generative
retrieval models through simultaneous decoding. To this aim, PAG constructs a
set-based and sequential identifier for each document. Motivated by the
bag-of-words assumption in information retrieval, the set-based identifier is
built on lexical tokens. The sequential identifier, on the other hand, is
obtained via quantizing relevance-based representations of documents. Extensive
experiments on MSMARCO and TREC Deep Learning Track data reveal that PAG
outperforms the state-of-the-art generative retrieval model by a large margin
(e.g., 15.6% MRR improvements on MS MARCO), while achieving 22x speed up in
terms of query latency. | cs.IR | Planning Ahead in Generative Retrieval: Guiding Autoregressive Generation through Simultaneous Decoding |
2024-04-22T20:29:58Z | http://arxiv.org/pdf/2404.14567v1 | http://arxiv.org/abs/2404.14567v1 | Ronald Xie, Steven Palayew, Augustin Toma, Gary Bader, Bo Wang | 2024-04-22T20:29:58Z | cs.CL | This paper outlines our submission to the MEDIQA2024 Multilingual and
Multimodal Medical Answer Generation (M3G) shared task. We report results for
two standalone solutions under the English category of the task, the first
involving two consecutive API calls to the Claude 3 Opus API and the second
involving training an image-disease label joint embedding in the style of CLIP
for image classification. These two solutions scored 1st and 2nd place
respectively on the competition leaderboard, substantially outperforming the
next best solution. Additionally, we discuss insights gained from
post-competition experiments. While the performance of these two solutions have
significant room for improvement due to the difficulty of the shared task and
the challenging nature of medical visual question answering in general, we
identify the multi-stage LLM approach and the CLIP image classification
approach as promising avenues for further investigation. | cs.CL | WangLab at MEDIQA-M3G 2024: Multimodal Medical Answer Generation using Large Language Models |
2024-04-22T19:31:45Z | http://arxiv.org/pdf/2404.14544v1 | http://arxiv.org/abs/2404.14544v1 | Augustin Toma, Ronald Xie, Steven Palayew, Patrick R. Lawler, Bo Wang | 2024-04-22T19:31:45Z | cs.CL | Medical errors in clinical text pose significant risks to patient safety. The
MEDIQA-CORR 2024 shared task focuses on detecting and correcting these errors
across three subtasks: identifying the presence of an error, extracting the
erroneous sentence, and generating a corrected sentence. In this paper, we
present our approach that achieved top performance in all three subtasks. For
the MS dataset, which contains subtle errors, we developed a retrieval-based
system leveraging external medical question-answering datasets. For the UW
dataset, reflecting more realistic clinical notes, we created a pipeline of
modules to detect, localize, and correct errors. Both approaches utilized the
DSPy framework for optimizing prompts and few-shot examples in large language
model (LLM) based programs. Our results demonstrate the effectiveness of LLM
based programs for medical error correction. However, our approach has
limitations in addressing the full diversity of potential errors in medical
documentation. We discuss the implications of our work and highlight future
research directions to advance the robustness and applicability of medical
error detection and correction systems. | cs.CL | WangLab at MEDIQA-CORR 2024: Optimized LLM-based Programs for Medical Error Detection and Correction |
2024-04-22T17:59:29Z | http://arxiv.org/pdf/2404.14408v2 | http://arxiv.org/abs/2404.14408v2 | Kevin Slagle | 2024-05-23T16:41:41Z | cs.CL, cs.AI, cs.LG | Tokenization is widely used in large language models because it significantly
improves performance. However, tokenization imposes several disadvantages, such
as performance biases, increased adversarial vulnerability, decreased
character-level modeling performance, and increased modeling complexity. To
address these disadvantages without sacrificing performance, we propose
SpaceByte, a novel byte-level decoder architecture that closes the performance
gap between byte-level and subword autoregressive language modeling. SpaceByte
consists of a byte-level Transformer model, but with extra larger transformer
blocks inserted in the middle of the layers. We find that performance is
significantly improved by applying these larger blocks only after certain
bytes, such as space characters, which typically denote word boundaries. Our
experiments show that for a fixed training and inference compute budget,
SpaceByte outperforms other byte-level architectures and roughly matches the
performance of tokenized Transformer architectures. | cs.CL | SpaceByte: Towards Deleting Tokenization from Large Language Modeling |
2024-04-22T17:56:26Z | http://arxiv.org/pdf/2404.14397v1 | http://arxiv.org/abs/2404.14397v1 | Adrian de Wynter, Ishaan Watts, Nektar Ege Altıntoprak, Tua Wongsangaroonsri, Minghui Zhang, Noura Farra, Lena Baur, Samantha Claudet, Pavel Gajdusek, Can Gören, Qilong Gu, Anna Kaminska, Tomasz Kaminski, Ruby Kuo, Akiko Kyuba, Jongho Lee, Kartik Mathur, Petter Merok, Ivana Milovanović, Nani Paananen, Vesa-Matti Paananen, Anna Pavlenko, Bruno Pereira Vidal, Luciano Strika, Yueh Tsao, Davide Turcato, Oleksandr Vakhno, Judit Velcsov, Anna Vickers, Stéphanie Visser, Herdyan Widarmanto, Andrey Zaikin, Si-Qing Chen | 2024-04-22T17:56:26Z | cs.CL, cs.CY, cs.LG | Large language models (LLMs) and small language models (SLMs) are being
adopted at remarkable speed, although their safety still remains a serious
concern. With the advent of multilingual S/LLMs, the question now becomes a
matter of scale: can we expand multilingual safety evaluations of these models
with the same velocity at which they are deployed? To this end we introduce
RTP-LX, a human-transcreated and human-annotated corpus of toxic prompts and
outputs in 28 languages. RTP-LX follows participatory design practices, and a
portion of the corpus is especially designed to detect culturally-specific
toxic language. We evaluate seven S/LLMs on their ability to detect toxic
content in a culturally-sensitive, multilingual scenario. We find that,
although they typically score acceptably in terms of accuracy, they have low
agreement with human judges when judging holistically the toxicity of a prompt,
and have difficulty discerning harm in context-dependent scenarios,
particularly with subtle-yet-harmful content (e.g. microagressions, bias). We
release of this dataset to contribute to further reduce harmful uses of these
models and improve their safe deployment. | cs.CL | RTP-LX: Can LLMs Evaluate Toxicity in Multilingual Scenarios? |
2024-04-22T17:55:56Z | http://arxiv.org/pdf/2404.14395v1 | http://arxiv.org/abs/2404.14395v1 | Mitodru Niyogi, Arnab Bhattacharya | 2024-04-22T17:55:56Z | cs.CL, cs.AI, cs.LG | In this paper, we present Paramanu-Ganita, a 208 million parameter novel Auto
Regressive (AR) decoder based language model on mathematics. The model is
pretrained from scratch at context size of 4096 on our curated mixed
mathematical corpus. We evaluate our model on both perplexity metric and GSM8k
mathematical benchmark. Paramanu-Ganita despite being 35 times smaller than 7B
LLMs, outperformed generalist LLMs such as LLaMa-1 7B by 28.4% points, LLaMa-2
7B by 27.6% points, Falcon 7B by 32.6% points, PaLM 8B by 35.3% points, and
math specialised LLMs such as Minerva 8B by 23.2% points, and LLEMMA-7B by 3.0%
points in GSM8k test accuracy metric respectively. Paramanu-Ganita also
outperformed giant LLMs like PaLM 62B by 6.4% points, Falcon 40B by 19.8%
points, LLaMa-1 33B by 3.8% points and Vicuna 13B by 11.8% points respectively.
The large significant margin improvement in performance of our math model over
the existing LLMs signifies that reasoning capabilities of language model are
just not restricted to LLMs with humongous number of parameters.
Paramanu-Ganita took 146 hours of A100 training whereas math specialised LLM,
LLEMMA 7B, was trained for 23,000 A100 hours of training equivalent. Thus, our
approach of pretraining powerful domain specialised language models from
scratch for domain adaptation is much more cost-effective than performing
continual training of LLMs for domain adaptation. Hence, we conclude that for
strong mathematical reasoning abilities of language model, we do not need giant
LLMs and immense computing power to our end. In the end, we want to point out
that we have only trained Paramanu-Ganita only on a part of our entire
mathematical corpus and yet to explore the full potential of our model. | cs.CL | PARAMANU-GANITA: Language Model with Mathematical Capabilities |
2024-04-22T17:55:11Z | http://arxiv.org/pdf/2404.14394v1 | http://arxiv.org/abs/2404.14394v1 | Tamar Rott Shaham, Sarah Schwettmann, Franklin Wang, Achyuta Rajaram, Evan Hernandez, Jacob Andreas, Antonio Torralba | 2024-04-22T17:55:11Z | cs.AI, cs.CL, cs.CV | This paper describes MAIA, a Multimodal Automated Interpretability Agent.
MAIA is a system that uses neural models to automate neural model understanding
tasks like feature interpretation and failure mode discovery. It equips a
pre-trained vision-language model with a set of tools that support iterative
experimentation on subcomponents of other models to explain their behavior.
These include tools commonly used by human interpretability researchers: for
synthesizing and editing inputs, computing maximally activating exemplars from
real-world datasets, and summarizing and describing experimental results.
Interpretability experiments proposed by MAIA compose these tools to describe
and explain system behavior. We evaluate applications of MAIA to computer
vision models. We first characterize MAIA's ability to describe (neuron-level)
features in learned representations of images. Across several trained models
and a novel dataset of synthetic vision neurons with paired ground-truth
descriptions, MAIA produces descriptions comparable to those generated by
expert human experimenters. We then show that MAIA can aid in two additional
interpretability tasks: reducing sensitivity to spurious features, and
automatically identifying inputs likely to be mis-classified. | cs.AI | A Multimodal Automated Interpretability Agent |
2024-04-22T17:43:23Z | http://arxiv.org/pdf/2404.14387v2 | http://arxiv.org/abs/2404.14387v2 | Zhengwei Tao, Ting-En Lin, Xiancai Chen, Hangyu Li, Yuchuan Wu, Yongbin Li, Zhi Jin, Fei Huang, Dacheng Tao, Jingren Zhou | 2024-06-03T17:47:30Z | cs.CL, cs.AI | Large language models (LLMs) have significantly advanced in various fields
and intelligent agent applications. However, current LLMs that learn from human
or external model supervision are costly and may face performance ceilings as
task complexity and diversity increase. To address this issue, self-evolution
approaches that enable LLM to autonomously acquire, refine, and learn from
experiences generated by the model itself are rapidly growing. This new
training paradigm inspired by the human experiential learning process offers
the potential to scale LLMs towards superintelligence. In this work, we present
a comprehensive survey of self-evolution approaches in LLMs. We first propose a
conceptual framework for self-evolution and outline the evolving process as
iterative cycles composed of four phases: experience acquisition, experience
refinement, updating, and evaluation. Second, we categorize the evolution
objectives of LLMs and LLM-based agents; then, we summarize the literature and
provide taxonomy and insights for each module. Lastly, we pinpoint existing
challenges and propose future directions to improve self-evolution frameworks,
equipping researchers with critical insights to fast-track the development of
self-evolving LLMs. Our corresponding GitHub repository is available at
https://github.com/AlibabaResearch/DAMO-ConvAI/tree/main/Awesome-Self-Evolution-of-LLM | cs.CL | A Survey on Self-Evolution of Large Language Models |
2024-04-22T17:42:58Z | http://arxiv.org/pdf/2404.14469v2 | http://arxiv.org/abs/2404.14469v2 | Yuhong Li, Yingbing Huang, Bowen Yang, Bharat Venkitesh, Acyr Locatelli, Hanchen Ye, Tianle Cai, Patrick Lewis, Deming Chen | 2024-06-17T03:01:58Z | cs.CL, cs.AI | Large Language Models (LLMs) have made remarkable progress in processing
extensive contexts, with the Key-Value (KV) cache playing a vital role in
enhancing their performance. However, the growth of the KV cache in response to
increasing input length poses challenges to memory and time efficiency. To
address this problem, this paper introduces SnapKV, an innovative and
fine-tuning-free approach that efficiently minimizes KV cache size while still
delivering comparable performance in real-world applications.
We discover that each attention head in the model consistently focuses on
specific prompt attention features during generation. Meanwhile, this robust
pattern can be obtained from an 'observation' window located at the end of the
prompts. Drawing on this insight, SnapKV automatically compresses KV caches by
selecting clustered important KV positions for each attention head. Our
approach significantly reduces the growing computational overhead and memory
footprint when processing long input sequences. Specifically, SnapKV achieves a
consistent decoding speed with a 3.6x increase in generation speed and an 8.2x
enhancement in memory efficiency compared to the baseline when processing
inputs of 16K tokens. At the same time, it maintains comparable performance to
the baseline models across 16 long sequence datasets. Moreover, SnapKV can
process up to 380K context tokens on a single A100-80GB GPU using HuggingFace
implementation with minor changes, exhibiting only a negligible accuracy drop
in the Needle-in-a-Haystack test. Further comprehensive studies suggest
SnapKV's potential for practical applications. | cs.CL | SnapKV: LLM Knows What You are Looking for Before Generation |
2024-04-22T17:40:08Z | http://arxiv.org/pdf/2406.02560v2 | http://arxiv.org/abs/2406.02560v2 | Ruizhe Huang, Xiaohui Zhang, Zhaoheng Ni, Li Sun, Moto Hira, Jeff Hwang, Vimal Manohar, Vineel Pratap, Matthew Wiesner, Shinji Watanabe, Daniel Povey, Sanjeev Khudanpur | 2024-06-15T22:02:03Z | eess.AS, cs.AI, cs.CL, cs.LG | Connectionist temporal classification (CTC) models are known to have peaky
output distributions. Such behavior is not a problem for automatic speech
recognition (ASR), but it can cause inaccurate forced alignments (FA),
especially at finer granularity, e.g., phoneme level. This paper aims at
alleviating the peaky behavior for CTC and improve its suitability for forced
alignment generation, by leveraging label priors, so that the scores of
alignment paths containing fewer blanks are boosted and maximized during
training. As a result, our CTC model produces less peaky posteriors and is able
to more accurately predict the offset of the tokens besides their onset. It
outperforms the standard CTC model and a heuristics-based approach for
obtaining CTC's token offset timestamps by 12-40% in phoneme and word boundary
errors (PBE and WBE) measured on the Buckeye and TIMIT data. Compared with the
most widely used FA toolkit Montreal Forced Aligner (MFA), our method performs
similarly on PBE/WBE on Buckeye, yet falls behind MFA on TIMIT. Nevertheless,
our method has a much simpler training pipeline and better runtime efficiency.
Our training recipe and pretrained model are released in TorchAudio. | eess.AS | Less Peaky and More Accurate CTC Forced Alignment by Label Priors |
2024-04-22T17:36:33Z | http://arxiv.org/pdf/2404.15166v1 | http://arxiv.org/abs/2404.15166v1 | John R. Lawson, Montgomery L. Flora, Kevin H. Goebbert, Seth N. Lyman, Corey K. Potvin, David M. Schultz, Adam J. Stepanek, Joseph E. Trujillo-Falcón | 2024-04-22T17:36:33Z | cs.CL, cs.AI, physics.ao-ph | Generative AI, such as OpenAI's GPT-4V large-language model, has rapidly
entered mainstream discourse. Novel capabilities in image processing and
natural-language communication may augment existing forecasting methods. Large
language models further display potential to better communicate weather hazards
in a style honed for diverse communities and different languages. This study
evaluates GPT-4V's ability to interpret meteorological charts and communicate
weather hazards appropriately to the user, despite challenges of
hallucinations, where generative AI delivers coherent, confident, but incorrect
responses. We assess GPT-4V's competence via its web interface ChatGPT in two
tasks: (1) generating a severe-weather outlook from weather-chart analysis and
conducting self-evaluation, revealing an outlook that corresponds well with a
Storm Prediction Center human-issued forecast; and (2) producing hazard
summaries in Spanish and English from weather charts. Responses in Spanish,
however, resemble direct (not idiomatic) translations from English to Spanish,
yielding poorly translated summaries that lose critical idiomatic precision
required for optimal communication. Our findings advocate for cautious
integration of tools like GPT-4V in meteorology, underscoring the necessity of
human oversight and development of trustworthy, explainable AI. | cs.CL | Pixels and Predictions: Potential of GPT-4V in Meteorological Imagery Analysis and Forecast Communication |
2024-04-22T17:22:31Z | http://arxiv.org/pdf/2404.14372v1 | http://arxiv.org/abs/2404.14372v1 | Xiaochen Kev Gao, Feng Yao, Kewen Zhao, Beilei He, Animesh Kumar, Vish Krishnan, Jingbo Shang | 2024-04-22T17:22:31Z | cs.CL, cs.AI | Model scaling is becoming the default choice for many language tasks due to
the success of large language models (LLMs). However, it can fall short in
specific scenarios where simple customized methods excel. In this paper, we
delve into the patent approval pre-diction task and unveil that simple
domain-specific graph methods outperform enlarging the model, using the
intrinsic dependencies within the patent data. Specifically, we first extend
the embedding-based state-of-the-art (SOTA) by scaling up its backbone model
with various sizes of open-source LLMs, then explore prompt-based methods to
harness proprietary LLMs' potential, but find the best results close to random
guessing, underlining the ineffectiveness of model scaling-up. Hence, we
propose a novel Fine-grained cLAim depeNdency (FLAN) Graph through meticulous
patent data analyses, capturing the inherent dependencies across segments of
the patent text. As it is model-agnostic, we apply cost-effective graph models
to our FLAN Graph to obtain representations for approval prediction. Extensive
experiments and detailed analyses prove that incorporating FLAN Graph via
various graph models consistently outperforms all LLM baselines significantly.
We hope that our observations and analyses in this paper can bring more
attention to this challenging task and prompt further research into the
limitations of LLMs. Our source code and dataset can be obtained from
http://github.com/ShangDataLab/FLAN-Graph. | cs.CL | Beyond Scaling: Predicting Patent Approval with Domain-specific Fine-grained Claim Dependency Graph |
2024-04-22T17:20:38Z | http://arxiv.org/pdf/2404.14368v1 | http://arxiv.org/abs/2404.14368v1 | Yutao Cheng, Zhao Zhang, Maoke Yang, Hui Nie, Chunyuan Li, Xinglong Wu, Jie Shao | 2024-04-22T17:20:38Z | cs.CV, cs.AI, cs.CL | In the field of graphic design, automating the integration of design elements
into a cohesive multi-layered artwork not only boosts productivity but also
paves the way for the democratization of graphic design. One existing practice
is Graphic Layout Generation (GLG), which aims to layout sequential design
elements. It has been constrained by the necessity for a predefined correct
sequence of layers, thus limiting creative potential and increasing user
workload. In this paper, we present Hierarchical Layout Generation (HLG) as a
more flexible and pragmatic setup, which creates graphic composition from
unordered sets of design elements. To tackle the HLG task, we introduce
Graphist, the first layout generation model based on large multimodal models.
Graphist efficiently reframes the HLG as a sequence generation problem,
utilizing RGB-A images as input, outputs a JSON draft protocol, indicating the
coordinates, size, and order of each element. We develop new evaluation metrics
for HLG. Graphist outperforms prior arts and establishes a strong baseline for
this field. Project homepage: https://github.com/graphic-design-ai/graphist | cs.CV | Graphic Design with Large Multimodal Model |
2024-04-22T17:15:32Z | http://arxiv.org/pdf/2404.14361v3 | http://arxiv.org/abs/2404.14361v3 | Saumya Gandhi, Ritu Gala, Vijay Viswanathan, Tongshuang Wu, Graham Neubig | 2024-04-26T19:02:23Z | cs.CL | Despite recent advances in large language models, building dependable and
deployable NLP models typically requires abundant, high-quality training data.
However, task-specific data is not available for many use cases, and manually
curating task-specific data is labor-intensive. Recent work has studied
prompt-driven synthetic data generation using large language models, but these
generated datasets tend to lack complexity and diversity. To address these
limitations, we introduce a method, DataTune, to make better use of existing,
publicly available datasets to improve automatic dataset generation. DataTune
performs dataset transformation, enabling the repurposing of publicly available
datasets into a format that is directly aligned with the specific requirements
of target tasks. On a diverse set of language-based tasks from the BIG-Bench
benchmark, we find that finetuning language models via DataTune improves over a
few-shot prompting baseline by 49% and improves over existing methods that use
synthetic or retrieved training data by 34%. We find that dataset
transformation significantly increases the diversity and difficulty of
generated data on many tasks. We integrate DataTune into an open-source
repository to make this method accessible to the community:
https://github.com/neulab/prompt2model. | cs.CL | Better Synthetic Data by Retrieving and Transforming Existing Datasets |
2024-04-22T17:07:25Z | http://arxiv.org/pdf/2404.14355v3 | http://arxiv.org/abs/2404.14355v3 | Vishruth Veerendranath, Vishwa Shah, Kshitish Ghate | 2024-06-26T03:52:24Z | cs.CL, cs.AI | Quantitative and numerical comprehension in language is an important task in
many fields like education and finance, but still remains a challenging task
for language models. While tool and calculator usage has shown to be helpful to
improve mathematical reasoning in large pretrained decoder-only language
models, this remains unexplored for smaller language models with encoders. In
this paper, we propose Pre-Calc, a simple pre-finetuning objective of learning
to use the calculator for both encoder-only and encoder-decoder architectures,
formulated as a discriminative and generative task respectively. We pre-train
BERT and RoBERTa for discriminative calculator use and Flan-T5 for generative
calculator use on the MAWPS, SVAMP, and AsDiv-A datasets, which improves
performance on downstream tasks that require numerical understanding. Our code
and data are available at https://github.com/calc-cmu/pre-calc. | cs.CL | Pre-Calc: Learning to Use the Calculator Improves Numeracy in Language Models |
2024-04-22T16:56:43Z | http://arxiv.org/pdf/2404.14339v1 | http://arxiv.org/abs/2404.14339v1 | Bharathi A, Arkaitz Zubiaga | 2024-04-22T16:56:43Z | cs.CL | Stance detection has been widely studied as the task of determining if a
social media post is positive, negative or neutral towards a specific issue,
such as support towards vaccines. Research in stance detection has however
often been limited to a single language and, where more than one language has
been studied, research has focused on few-shot settings, overlooking the
challenges of developing a zero-shot cross-lingual stance detection model. This
paper makes the first such effort by introducing a novel approach to zero-shot
cross-lingual stance detection, Multilingual Translation-Augmented BERT (MTAB),
aiming to enhance the performance of a cross-lingual classifier in the absence
of explicit training data for target languages. Our technique employs
translation augmentation to improve zero-shot performance and pairs it with
adversarial learning to further boost model efficacy. Through experiments on
datasets labeled for stance towards vaccines in four languages English, German,
French, Italian. We demonstrate the effectiveness of our proposed approach,
showcasing improved results in comparison to a strong baseline model as well as
ablated versions of our model. Our experiments demonstrate the effectiveness of
model components, not least the translation-augmented data as well as the
adversarial learning component, to the improved performance of the model. We
have made our source code accessible on GitHub. | cs.CL | Zero-shot Cross-lingual Stance Detection via Adversarial Language Adaptation |
2024-04-22T16:55:44Z | http://arxiv.org/pdf/2404.14467v1 | http://arxiv.org/abs/2404.14467v1 | Hongxuan Liu, Haoyu Yin, Zhiyao Luo, Xiaonan Wang | 2024-04-22T16:55:44Z | cs.CL, cs.AI | This paper presents a study on the integration of domain-specific knowledge
in prompt engineering to enhance the performance of large language models
(LLMs) in scientific domains. A benchmark dataset is curated to encapsulate the
intricate physical-chemical properties of small molecules, their drugability
for pharmacology, alongside the functional attributes of enzymes and crystal
materials, underscoring the relevance and applicability across biological and
chemical domains.The proposed domain-knowledge embedded prompt engineering
method outperforms traditional prompt engineering strategies on various
metrics, including capability, accuracy, F1 score, and hallucination drop. The
effectiveness of the method is demonstrated through case studies on complex
materials including the MacMillan catalyst, paclitaxel, and lithium cobalt
oxide. The results suggest that domain-knowledge prompts can guide LLMs to
generate more accurate and relevant responses, highlighting the potential of
LLMs as powerful tools for scientific discovery and innovation when equipped
with domain-specific prompts. The study also discusses limitations and future
directions for domain-specific prompt engineering development. | cs.CL | Integrating Chemistry Knowledge in Large Language Models via Prompt Engineering |
2024-04-22T16:33:42Z | http://arxiv.org/pdf/2404.15153v1 | http://arxiv.org/abs/2404.15153v1 | Josef Pichlmeier, Philipp Ross, Andre Luckow | 2024-04-22T16:33:42Z | cs.CL, cs.AI, cs.PF | Large Language Models (LLMs) have experienced widespread adoption across
scientific and industrial domains due to their versatility and utility for
diverse tasks. Nevertheless, deploying and serving these models at scale with
optimal throughput and latency remains a significant challenge, primarily
because of the high computational and memory demands associated with LLMs. To
tackle this limitation, we introduce Expert Router, a system designed to
orchestrate multiple expert models efficiently, thereby enhancing scalability.
Expert Router is a parallel inference system with a central routing gateway
that distributes incoming requests using a clustering method. This approach
effectively partitions incoming requests among available LLMs, maximizing
overall throughput. Our extensive evaluations encompassed up to 1,000
concurrent users, providing comprehensive insights into the system's behavior
from user and infrastructure perspectives. The results demonstrate Expert
Router's effectiveness in handling high-load scenarios and achieving higher
throughput rates, particularly under many concurrent users. | cs.CL | Expert Router: Orchestrating Efficient Language Model Inference through Prompt Classification |
2024-04-22T16:28:09Z | http://arxiv.org/pdf/2404.14316v1 | http://arxiv.org/abs/2404.14316v1 | Shashank Sonkar, Kangqi Ni, Lesa Tran Lu, Kristi Kincaid, John S. Hutchinson, Richard G. Baraniuk | 2024-04-22T16:28:09Z | cs.CL | We introduce a new area of study in the field of educational Natural Language
Processing: Automated Long Answer Grading (ALAG). Distinguishing itself from
Automated Short Answer Grading (ASAG) and Automated Essay Grading (AEG), ALAG
presents unique challenges due to the complexity and multifaceted nature of
fact-based long answers. To study ALAG, we introduce RiceChem, a dataset
derived from a college chemistry course, featuring real student responses to
long-answer questions with an average word count notably higher than typical
ASAG datasets. We propose a novel approach to ALAG by formulating it as a
rubric entailment problem, employing natural language inference models to
verify whether each criterion, represented by a rubric item, is addressed in
the student's response. This formulation enables the effective use of MNLI for
transfer learning, significantly improving the performance of models on the
RiceChem dataset. We demonstrate the importance of rubric-based formulation in
ALAG, showcasing its superiority over traditional score-based approaches in
capturing the nuances of student responses. We also investigate the performance
of models in cold start scenarios, providing valuable insights into the
practical deployment considerations in educational settings. Lastly, we
benchmark state-of-the-art open-sourced Large Language Models (LLMs) on
RiceChem and compare their results to GPT models, highlighting the increased
complexity of ALAG compared to ASAG. Despite leveraging the benefits of a
rubric-based approach and transfer learning from MNLI, the lower performance of
LLMs on RiceChem underscores the significant difficulty posed by the ALAG task.
With this work, we offer a fresh perspective on grading long, fact-based
answers and introduce a new dataset to stimulate further research in this
important area. Code:
\url{https://github.com/luffycodes/Automated-Long-Answer-Grading}. | cs.CL | Automated Long Answer Grading with RiceChem Dataset |
2024-04-22T16:20:36Z | http://arxiv.org/pdf/2404.14313v2 | http://arxiv.org/abs/2404.14313v2 | Jan-Philipp Fränken, Eric Zelikman, Rafael Rafailov, Kanishk Gandhi, Tobias Gerstenberg, Noah D. Goodman | 2024-05-21T17:31:27Z | cs.CL | When prompting a language model (LM), users often expect the model to adhere
to a set of behavioral principles across diverse tasks, such as producing
insightful content while avoiding harmful or biased language. Instilling such
principles (i.e., a constitution) into a model is resource-intensive,
technically challenging, and generally requires human preference labels or
examples. We introduce SAMI, an iterative algorithm that finetunes a pretrained
language model (without requiring preference labels or demonstrations) to
increase the conditional mutual information between constitutions and
self-generated responses given queries from a dataset. On single-turn dialogue
and summarization, a SAMI-trained mistral-7b outperforms the initial pretrained
model, with win rates between 66% and 77%. Strikingly, it also surpasses an
instruction-finetuned baseline (mistral-7b-instruct) with win rates between 55%
and 57% on single-turn dialogue. SAMI requires a model that writes the
principles. To avoid dependence on strong models for writing principles, we
align a strong pretrained model (mixtral-8x7b) using constitutions written by a
weak instruction-finetuned model (mistral-7b-instruct), achieving a 65% win
rate on summarization. Finally, we investigate whether SAMI generalizes to
diverse summarization principles (e.g., "summaries should be scientific") and
scales to stronger models (llama3-70b), finding that it achieves win rates of
up to 68% for learned and 67% for held-out principles compared to the base
model. Our results show that a pretrained LM can learn to follow constitutions
without using preference labels, demonstrations, or human oversight. | cs.CL | Self-Supervised Alignment with Mutual Information: Learning to Follow Principles without Preference Labels |
2024-04-22T16:00:46Z | http://arxiv.org/pdf/2404.14301v1 | http://arxiv.org/abs/2404.14301v1 | Shashank Sonkar, Naiming Liu, Debshila B. Mallick, Richard G. Baraniuk | 2024-04-22T16:00:46Z | cs.CL | In this paper, we introduce "Marking", a novel grading task that enhances
automated grading systems by performing an in-depth analysis of student
responses and providing students with visual highlights. Unlike traditional
systems that provide binary scores, "marking" identifies and categorizes
segments of the student response as correct, incorrect, or irrelevant and
detects omissions from gold answers. We introduce a new dataset meticulously
curated by Subject Matter Experts specifically for this task. We frame
"Marking" as an extension of the Natural Language Inference (NLI) task, which
is extensively explored in the field of Natural Language Processing. The gold
answer and the student response play the roles of premise and hypothesis in
NLI, respectively. We subsequently train language models to identify
entailment, contradiction, and neutrality from student response, akin to NLI,
and with the added dimension of identifying omissions from gold answers. Our
experimental setup involves the use of transformer models, specifically BERT
and RoBERTa, and an intelligent training step using the e-SNLI dataset. We
present extensive baseline results highlighting the complexity of the "Marking"
task, which sets a clear trajectory for the upcoming study. Our work not only
opens up new avenues for research in AI-powered educational assessment tools,
but also provides a valuable benchmark for the AI in education community to
engage with and improve upon in the future. The code and dataset can be found
at https://github.com/luffycodes/marking. | cs.CL | Marking: Visual Grading with Highlighting Errors and Annotating Missing Bits |
2024-04-22T15:53:08Z | http://arxiv.org/pdf/2404.14294v2 | http://arxiv.org/abs/2404.14294v2 | Zixuan Zhou, Xuefei Ning, Ke Hong, Tianyu Fu, Jiaming Xu, Shiyao Li, Yuming Lou, Luning Wang, Zhihang Yuan, Xiuhong Li, Shengen Yan, Guohao Dai, Xiao-Ping Zhang, Yuhan Dong, Yu Wang | 2024-06-08T04:49:51Z | cs.CL, cs.AI | Large Language Models (LLMs) have attracted extensive attention due to their
remarkable performance across various tasks. However, the substantial
computational and memory requirements of LLM inference pose challenges for
deployment in resource-constrained scenarios. Efforts within the field have
been directed towards developing techniques aimed at enhancing the efficiency
of LLM inference. This paper presents a comprehensive survey of the existing
literature on efficient LLM inference. We start by analyzing the primary causes
of the inefficient LLM inference, i.e., the large model size, the
quadratic-complexity attention operation, and the auto-regressive decoding
approach. Then, we introduce a comprehensive taxonomy that organizes the
current literature into data-level, model-level, and system-level optimization.
Moreover, the paper includes comparative experiments on representative methods
within critical sub-fields to provide quantitative insights. Last but not
least, we provide some knowledge summary and discuss future research
directions. | cs.CL | A Survey on Efficient Inference for Large Language Models |
2024-04-22T15:15:50Z | http://arxiv.org/pdf/2404.14270v1 | http://arxiv.org/abs/2404.14270v1 | Jue Hou, Anisia Katinskaia, Lari Kotilainen, Sathianpong Trangcasanchai, Anh-Duc Vu, Roman Yangarber | 2024-04-22T15:15:50Z | cs.CL, cs.LG | This paper investigates what insights about linguistic features and what
knowledge about the structure of natural language can be obtained from the
encodings in transformer language models.In particular, we explore how BERT
encodes the government relation between constituents in a sentence. We use
several probing classifiers, and data from two morphologically rich languages.
Our experiments show that information about government is encoded across all
transformer layers, but predominantly in the early layers of the model. We find
that, for both languages, a small number of attention heads encode enough
information about the government relations to enable us to train a classifier
capable of discovering new, previously unknown types of government, never seen
in the training data. Currently, data is lacking for the research community
working on grammatical constructions, and government in particular. We release
the Government Bank -- a dataset defining the government relations for
thousands of lemmas in the languages in our experiments. | cs.CL | What do Transformers Know about Government? |
2024-04-22T14:46:10Z | http://arxiv.org/pdf/2404.14233v1 | http://arxiv.org/abs/2404.14233v1 | Wenyi Xiao, Ziwei Huang, Leilei Gan, Wanggui He, Haoyuan Li, Zhelun Yu, Hao Jiang, Fei Wu, Linchao Zhu | 2024-04-22T14:46:10Z | cs.CV, cs.AI, cs.CL, cs.LG | The rapidly developing Large Vision Language Models (LVLMs) have shown
notable capabilities on a range of multi-modal tasks, but still face the
hallucination phenomena where the generated texts do not align with the given
contexts, significantly restricting the usages of LVLMs. Most previous work
detects and mitigates hallucination at the coarse-grained level or requires
expensive annotation (e.g., labeling by proprietary models or human experts).
To address these issues, we propose detecting and mitigating hallucinations in
LVLMs via fine-grained AI feedback. The basic idea is that we generate a
small-size sentence-level hallucination annotation dataset by proprietary
models, whereby we train a hallucination detection model which can perform
sentence-level hallucination detection, covering primary hallucination types
(i.e., object, attribute, and relationship). Then, we propose a
detect-then-rewrite pipeline to automatically construct preference dataset for
training hallucination mitigating model. Furthermore, we propose
differentiating the severity of hallucinations, and introducing a Hallucination
Severity-Aware Direct Preference Optimization (HSA-DPO) for mitigating
hallucination in LVLMs by incorporating the severity of hallucinations into
preference learning. Extensive experiments demonstrate the effectiveness of our
method. | cs.CV | Detecting and Mitigating Hallucination in Large Vision Language Models via Fine-Grained AI Feedback |
2024-04-22T14:32:33Z | http://arxiv.org/pdf/2404.14219v3 | http://arxiv.org/abs/2404.14219v3 | Marah Abdin, Sam Ade Jacobs, Ammar Ahmad Awan, Jyoti Aneja, Ahmed Awadallah, Hany Awadalla, Nguyen Bach, Amit Bahree, Arash Bakhtiari, Jianmin Bao, Harkirat Behl, Alon Benhaim, Misha Bilenko, Johan Bjorck, Sébastien Bubeck, Qin Cai, Martin Cai, Caio César Teodoro Mendes, Weizhu Chen, Vishrav Chaudhary, Dong Chen, Dongdong Chen, Yen-Chun Chen, Yi-Ling Chen, Parul Chopra, Xiyang Dai, Allie Del Giorno, Gustavo de Rosa, Matthew Dixon, Ronen Eldan, Victor Fragoso, Dan Iter, Mei Gao, Min Gao, Jianfeng Gao, Amit Garg, Abhishek Goswami, Suriya Gunasekar, Emman Haider, Junheng Hao, Russell J. Hewett, Jamie Huynh, Mojan Javaheripi, Xin Jin, Piero Kauffmann, Nikos Karampatziakis, Dongwoo Kim, Mahoud Khademi, Lev Kurilenko, James R. Lee, Yin Tat Lee, Yuanzhi Li, Yunsheng Li, Chen Liang, Lars Liden, Ce Liu, Mengchen Liu, Weishung Liu, Eric Lin, Zeqi Lin, Chong Luo, Piyush Madan, Matt Mazzola, Arindam Mitra, Hardik Modi, Anh Nguyen, Brandon Norick, Barun Patra, Daniel Perez-Becker, Thomas Portet, Reid Pryzant, Heyang Qin, Marko Radmilac, Corby Rosset, Sambudha Roy, Olatunji Ruwase, Olli Saarikivi, Amin Saied, Adil Salim, Michael Santacroce, Shital Shah, Ning Shang, Hiteshi Sharma, Swadheen Shukla, Xia Song, Masahiro Tanaka, Andrea Tupini, Xin Wang, Lijuan Wang, Chunyu Wang, Yu Wang, Rachel Ward, Guanhua Wang, Philipp Witte, Haiping Wu, Michael Wyatt, Bin Xiao, Can Xu, Jiahang Xu, Weijian Xu, Sonali Yadav, Fan Yang, Jianwei Yang, Ziyi Yang, Yifan Yang, Donghan Yu, Lu Yuan, Chengruidong Zhang, Cyril Zhang, Jianwen Zhang, Li Lyna Zhang, Yi Zhang, Yue Zhang, Yunan Zhang, Xiren Zhou | 2024-05-23T22:42:40Z | cs.CL, cs.AI | We introduce phi-3-mini, a 3.8 billion parameter language model trained on
3.3 trillion tokens, whose overall performance, as measured by both academic
benchmarks and internal testing, rivals that of models such as Mixtral 8x7B and
GPT-3.5 (e.g., phi-3-mini achieves 69% on MMLU and 8.38 on MT-bench), despite
being small enough to be deployed on a phone. The innovation lies entirely in
our dataset for training, a scaled-up version of the one used for phi-2,
composed of heavily filtered publicly available web data and synthetic data.
The model is also further aligned for robustness, safety, and chat format. We
also provide some initial parameter-scaling results with a 7B and 14B models
trained for 4.8T tokens, called phi-3-small and phi-3-medium, both
significantly more capable than phi-3-mini (e.g., respectively 75% and 78% on
MMLU, and 8.7 and 8.9 on MT-bench). Moreover, we also introduce phi-3-vision, a
4.2 billion parameter model based on phi-3-mini with strong reasoning
capabilities for image and text prompts. | cs.CL | Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone |
2024-04-22T14:31:28Z | http://arxiv.org/pdf/2404.14215v1 | http://arxiv.org/abs/2404.14215v1 | Zheye Deng, Chunkit Chan, Weiqi Wang, Yuxi Sun, Wei Fan, Tianshi Zheng, Yauwai Yim, Yangqiu Song | 2024-04-22T14:31:28Z | cs.CL | The task of condensing large chunks of textual information into concise and
structured tables has gained attention recently due to the emergence of Large
Language Models (LLMs) and their potential benefit for downstream tasks, such
as text summarization and text mining. Previous approaches often generate
tables that directly replicate information from the text, limiting their
applicability in broader contexts, as text-to-table generation in real-life
scenarios necessitates information extraction, reasoning, and integration.
However, there is a lack of both datasets and methodologies towards this task.
In this paper, we introduce LiveSum, a new benchmark dataset created for
generating summary tables of competitions based on real-time commentary texts.
We evaluate the performances of state-of-the-art LLMs on this task in both
fine-tuning and zero-shot settings, and additionally propose a novel pipeline
called $T^3$(Text-Tuple-Table) to improve their performances. Extensive
experimental results demonstrate that LLMs still struggle with this task even
after fine-tuning, while our approach can offer substantial performance gains
without explicit training. Further analyses demonstrate that our method
exhibits strong generalization abilities, surpassing previous approaches on
several other text-to-table datasets. Our code and data can be found at
https://github.com/HKUST-KnowComp/LiveSum-TTT. | cs.CL | Text-Tuple-Table: Towards Information Integration in Text-to-Table Generation via Global Tuple Extraction |
2024-04-22T14:18:34Z | http://arxiv.org/pdf/2404.14209v1 | http://arxiv.org/abs/2404.14209v1 | Po-Ting Lai, Elisabeth Coudert, Lucila Aimo, Kristian Axelsen, Lionel Breuza, Edouard de Castro, Marc Feuermann, Anne Morgat, Lucille Pourcel, Ivo Pedruzzi, Sylvain Poux, Nicole Redaschi, Catherine Rivoire, Anastasia Sveshnikova, Chih-Hsuan Wei, Robert Leaman, Ling Luo, Zhiyong Lu, Alan Bridge | 2024-04-22T14:18:34Z | cs.CL | Expert curation is essential to capture knowledge of enzyme functions from
the scientific literature in FAIR open knowledgebases but cannot keep pace with
the rate of new discoveries and new publications. In this work we present
EnzChemRED, for Enzyme Chemistry Relation Extraction Dataset, a new training
and benchmarking dataset to support the development of Natural Language
Processing (NLP) methods such as (large) language models that can assist enzyme
curation. EnzChemRED consists of 1,210 expert curated PubMed abstracts in which
enzymes and the chemical reactions they catalyze are annotated using
identifiers from the UniProt Knowledgebase (UniProtKB) and the ontology of
Chemical Entities of Biological Interest (ChEBI). We show that fine-tuning
pre-trained language models with EnzChemRED can significantly boost their
ability to identify mentions of proteins and chemicals in text (Named Entity
Recognition, or NER) and to extract the chemical conversions in which they
participate (Relation Extraction, or RE), with average F1 score of 86.30% for
NER, 86.66% for RE for chemical conversion pairs, and 83.79% for RE for
chemical conversion pairs and linked enzymes. We combine the best performing
methods after fine-tuning using EnzChemRED to create an end-to-end pipeline for
knowledge extraction from text and apply this to abstracts at PubMed scale to
create a draft map of enzyme functions in literature to guide curation efforts
in UniProtKB and the reaction knowledgebase Rhea. The EnzChemRED corpus is
freely available at https://ftp.expasy.org/databases/rhea/nlp/. | cs.CL | EnzChemRED, a rich enzyme chemistry relation extraction dataset |
2024-04-22T14:01:09Z | http://arxiv.org/pdf/2404.14192v3 | http://arxiv.org/abs/2404.14192v3 | Víctor Franco-Sánchez, Arnau Martí-Llobet, Ramon Ferrer-i-Cancho | 2024-05-25T14:10:05Z | cs.CL, physics.soc-ph | Here we consider the problem of all the possible orders of a linguistic
structure formed by $n$ elements, for instance, subject, direct object and verb
($n=3$) or subject, direct object, indirect object and verb ($n=4$). We
investigate if the frequency of the $n!$ possible orders is constrained by two
principles. First, entropy minimization, a principle that has been suggested to
shape natural communication systems at distinct levels of organization. Second,
swap distance minimization, namely a preference for word orders that require
fewer swaps of adjacent elements to be produced from a source order. Here we
present average swap distance, a novel score for research on swap distance
minimization, and investigate the theoretical distribution of that score for
any $n$: its minimum and maximum values and its expected value in die rolling
experiments or when the word order frequencies are shuffled. We investigate
whether entropy and average swap distance are significantly small in distinct
linguistic structures with $n=3$ or $n=4$ in agreement with the corresponding
minimization principles. We find strong evidence of entropy minimization and
swap distance minimization with respect to a die rolling experiment. The
evidence of these two forces with respect to a Polya urn process is strong for
$n=4$ but weaker for $n=3$. We still find evidence of swap distance
minimization when word order frequencies are shuffled, indicating that swap
distance minimization effects are beyond pressure to minimize word order
entropy. | cs.CL | Swap distance minimization beyond entropy minimization in word order variation |
Subsets and Splits
No saved queries yet
Save your SQL queries to embed, download, and access them later. Queries will appear here once saved.