PyTorchKR Papers

최신 논문

총 4,056편
태그·기간 등 상세 필터

카테고리 탐색

로그인 후 이 태그 구독

arXiv:2404.03163공식 코드 · Papers with Code

Uncertainty in Language Models: Assessment through Rank-Calibration

Language Models (LMs) have shown promising performance in natural language generation. However, as LMs often generate incorrect or hallucinated responses, it is crucial to correctly quantify their uncertainty in responding to given inputs. In addition to verbalized confidence elicited via prompting, many uncertainty measures ($e.g.$, semantic entropy and affinity-graph-based measures) have been proposed. However, these measures can differ greatly, and it is unclear how to compare them, partly because they take values over different ranges ($e.g.$, $[0,\infty)$ or $[0,1]$). In this work, we address this issue by developing a novel and practical framework, termed $Rank$-$Calibration$, to assess uncertainty and confidence measures for LMs. Our key tenet is that higher uncertainty (or lower confidence) should imply lower generation quality, on average. Rank-calibration quantifies deviations from this ideal relationship in a principled manner, without requiring ad hoc binary thresholding of the correctness score ($e.g.$, ROUGE or METEOR). The broad applicability and the granular interpretability of our methods are demonstrated empirically.

arXiv:2404.02431공식 코드 · Papers with Code

On the Multilingual Ability of Decoder-based Pre-trained Language Models: Finding and Controlling Language-Specific Neurons

Current decoder-based pre-trained language models (PLMs) successfully demonstrate multilingual capabilities. However, it is unclear how these models handle multilingualism. We analyze the neuron-level internal behavior of multilingual decoder-based PLMs, Specifically examining the existence of neurons that fire ``uniquely for each language'' within decoder-only multilingual PLMs. We analyze six languages: English, German, French, Spanish, Chinese, and Japanese, and show that language-specific neurons are unique, with a slight overlap (< 5%) between languages. These neurons are mainly distributed in the models' first and last few layers. This trend remains consistent across languages and models. Additionally, we tamper with less than 1% of the total neurons in each model during inference and demonstrate that tampering with a few language-specific neurons drastically changes the probability of target language occurrence in text generation.

arXiv:2404.02254

On Stronger Computational Separations Between Multimodal and Unimodal Machine Learning

Recently, multimodal machine learning has enjoyed huge empirical success (e.g. GPT-4). Motivated to develop theoretical justification for this empirical success, Lu (NeurIPS '23, ALT '24) introduces a theory of multimodal learning, and considers possible \textit{separations} between theoretical models of multimodal and unimodal learning. In particular, Lu (ALT '24) shows a computational separation, which is relevant to \textit{worst-case} instances of the learning task. In this paper, we give a stronger \textit{average-case} computational separation, where for ``typical'' instances of the learning task, unimodal learning is computationally hard, but multimodal learning is easy. We then question how ``natural'' the average-case separation is. Would it be encountered in practice? To this end, we prove that under basic conditions, any given computational separation between average-case unimodal and multimodal learning tasks implies a corresponding cryptographic key agreement protocol. We suggest to interpret this as evidence that very strong \textit{computational} advantages of multimodal learning may arise \textit{infrequently} in practice, since they exist only for the ``pathological'' case of inherently cryptographic distributions. However, this does not apply to possible (super-polynomial) \textit{statistical} advantages.

arXiv:2404.02466공식 코드 · Papers with Code

Prompting for Numerical Sequences: A Case Study on Market Comment Generation

Large language models (LLMs) have been applied to a wide range of data-to-text generation tasks, including tables, graphs, and time-series numerical data-to-text settings. While research on generating prompts for structured data such as tables and graphs is gaining momentum, in-depth investigations into prompting for time-series numerical data are lacking. Therefore, this study explores various input representations, including sequences of tokens and structured formats such as HTML, LaTeX, and Python-style codes. In our experiments, we focus on the task of Market Comment Generation, which involves taking a numerical sequence of stock prices as input and generating a corresponding market comment. Contrary to our expectations, the results show that prompts resembling programming languages yield better outcomes, whereas those similar to natural languages and longer formats, such as HTML and LaTeX, are less effective. Our findings offer insights into creating effective prompts for tasks that generate text from numerical sequences.

arXiv:2404.01327

Entertainment chatbot for the digital inclusion of elderly people without abstraction capabilities

Current language processing technologies allow the creation of conversational chatbot platforms. Even though artificial intelligence is still too immature to support satisfactory user experience in many mass market domains, conversational interfaces have found their way into ad hoc applications such as call centres and online shopping assistants. However, they have not been applied so far to social inclusion of elderly people, who are particularly vulnerable to the digital divide. Many of them relieve their loneliness with traditional media such as TV and radio, which are known to create a feeling of companionship. In this paper we present the EBER chatbot, designed to reduce the digital gap for the elderly. EBER reads news in the background and adapts its responses to the user's mood. Its novelty lies in the concept of "intelligent radio", according to which, instead of simplifying a digital information system to make it accessible to the elderly, a traditional channel they find familiar -- background news -- is augmented with interactions via voice dialogues. We make it possible by combining Artificial Intelligence Modelling Language, automatic Natural Language Generation and Sentiment Analysis. The system allows accessing digital content of interest by combining words extracted from user answers to chatbot questions with keywords extracted from the news items. This approach permits defining metrics of the abstraction capabilities of the users depending on a spatial representation of the word space. To prove the suitability of the proposed solution we present results of real experiments conducted with elderly people that provided valuable insights. Our approach was considered satisfactory during the tests and improved the information search capabilities of the participants.

arXiv:2404.01532공식 코드 · Papers with Code

Set-Aligning Framework for Auto-Regressive Event Temporal Graph Generation

Event temporal graphs have been shown as convenient and effective representations of complex temporal relations between events in text. Recent studies, which employ pre-trained language models to auto-regressively generate linearised graphs for constructing event temporal graphs, have shown promising results. However, these methods have often led to suboptimal graph generation as the linearised graphs exhibit set characteristics which are instead treated sequentially by language models. This discrepancy stems from the conventional text generation objectives, leading to erroneous penalisation of correct predictions caused by the misalignment of elements in target sequences. To address these challenges, we reframe the task as a conditional set generation problem, proposing a Set-aligning Framework tailored for the effective utilisation of Large Language Models (LLMs). The framework incorporates data augmentations and set-property regularisations designed to alleviate text generation loss penalties associated with the linearised graph edge sequences, thus encouraging the generation of more relation edges. Experimental results show that our framework surpasses existing baselines for event temporal graph generation. Furthermore, under zero-shot settings, the structural knowledge introduced through our framework notably improves model generalisation, particularly when the training examples available are limited.

arXiv:2404.01474

Finding Replicable Human Evaluations via Stable Ranking Probability

Reliable human evaluation is critical to the development of successful natural language generation models, but achieving it is notoriously difficult. Stability is a crucial requirement when ranking systems by quality: consistent ranking of systems across repeated evaluations is not just desirable, but essential. Without it, there is no reliable foundation for hill-climbing or product launch decisions. In this paper, we use machine translation and its state-of-the-art human evaluation framework, MQM, as a case study to understand how to set up reliable human evaluations that yield stable conclusions. We investigate the optimal configurations for item allocation to raters, number of ratings per item, and score normalization. Our study on two language pairs provides concrete recommendations for designing replicable human evaluation studies. We also collect and release the largest publicly available dataset of multi-segment translations rated by multiple professional translators, consisting of nearly 140,000 segment annotations across two language pairs.

arXiv:2404.01399공식 코드 · Papers with Code

Developing Safe and Responsible Large Language Model : Can We Balance Bias Reduction and Language Understanding in Large Language Models?

Large Language Models (LLMs) have advanced various Natural Language Processing (NLP) tasks, such as text generation and translation, among others. However, these models often generate texts that can perpetuate biases. Existing approaches to mitigate these biases usually compromise knowledge retention. This study explores whether LLMs can produce safe, unbiased outputs without sacrificing knowledge or comprehension. We introduce the Safe and Responsible Large Language Model (\textbf{SR}$_{\text{LLM}}$), which has been instruction fine-tuned atop of a safe fine-tuned auto-regressive decoder-only LLM to reduce biases in generated texts. We developed a specialized dataset with examples of unsafe and corresponding safe variations to train \textbf{SR}$_{\text{LLM}}$ to identify and correct biased text. Experiments on our specialized dataset and out-of-distribution test sets reveal that \textbf{SR}$_{\text{LLM}}$ effectively reduces biases while preserving knowledge integrity. This performance surpasses that of traditional fine-tuning of smaller language models and base LLMs that merely reply on prompting techniques. Our findings demonstrate that instruction fine-tuning on custom datasets tailored for tasks such as debiasing is a highly effective strategy for minimizing bias in LLM while preserving their inherent knowledge and capabilities. The code and dataset are accessible at \href{https://github.com/shainarazavi/Safe-Responsible-LLM}{SR-LLM}

arXiv:2404.01786

Generative AI-Based Text Generation Methods Using Pre-Trained GPT-2 Model

This work delved into the realm of automatic text generation, exploring a variety of techniques ranging from traditional deterministic approaches to more modern stochastic methods. Through analysis of greedy search, beam search, top-k sampling, top-p sampling, contrastive searching, and locally typical searching, this work has provided valuable insights into the strengths, weaknesses, and potential applications of each method. Each text-generating method is evaluated using several standard metrics and a comparative study has been made on the performance of the approaches. Finally, some future directions of research in the field of automatic text generation are also identified.

arXiv:2404.02138

Topic-Based Watermarks for Large Language Models

The indistinguishability of Large Language Model (LLM) output from human-authored content poses significant challenges, raising concerns about potential misuse of AI-generated text and its influence on future AI model training. Watermarking algorithms offer a viable solution by embedding detectable signatures into generated text. However, existing watermarking methods often entail trade-offs among attack robustness, generation quality, and additional overhead such as specialized frameworks or complex integrations. We propose a lightweight, topic-guided watermarking scheme for LLMs that partitions the vocabulary into topic-aligned token subsets. Given an input prompt, the scheme selects a relevant topic-specific token list, effectively "green-listing" semantically aligned tokens to embed robust marks while preserving the text's fluency and coherence. Experimental results across multiple LLMs and state-of-the-art benchmarks demonstrate that our method achieves comparable perplexity to industry-leading systems, including Google's SynthID-Text, yet enhances watermark robustness against paraphrasing and lexical perturbation attacks while introducing minimal performance overhead. Our approach avoids reliance on additional mechanisms beyond standard text generation pipelines, facilitating straightforward adoption, suggesting a practical path toward globally consistent watermarking of AI-generated content.

arXiv:2404.00909공식 코드 · Papers with Code

Learning by Correction: Efficient Tuning Task for Zero-Shot Generative Vision-Language Reasoning

Generative vision-language models (VLMs) have shown impressive performance in zero-shot vision-language tasks like image captioning and visual question answering. However, improving their zero-shot reasoning typically requires second-stage instruction tuning, which relies heavily on human-labeled or large language model-generated annotation, incurring high labeling costs. To tackle this challenge, we introduce Image-Conditioned Caption Correction (ICCC), a novel pre-training task designed to enhance VLMs' zero-shot performance without the need for labeled task-aware data. The ICCC task compels VLMs to rectify mismatches between visual and language concepts, thereby enhancing instruction following and text generation conditioned on visual inputs. Leveraging language structure and a lightweight dependency parser, we construct data samples of ICCC task from image-text datasets with low labeling and computation costs. Experimental results on BLIP-2 and InstructBLIP demonstrate significant improvements in zero-shot image-text generation-based VL tasks through ICCC instruction tuning.

arXiv:2404.00906공식 코드 · Papers with Code

From Pixels to Graphs: Open-Vocabulary Scene Graph Generation with Vision-Language Models

Scene graph generation (SGG) aims to parse a visual scene into an intermediate graph representation for downstream reasoning tasks. Despite recent advancements, existing methods struggle to generate scene graphs with novel visual relation concepts. To address this challenge, we introduce a new open-vocabulary SGG framework based on sequence generation. Our framework leverages vision-language pre-trained models (VLM) by incorporating an image-to-graph generation paradigm. Specifically, we generate scene graph sequences via image-to-text generation with VLM and then construct scene graphs from these sequences. By doing so, we harness the strong capabilities of VLM for open-vocabulary SGG and seamlessly integrate explicit relational modeling for enhancing the VL tasks. Experimental results demonstrate that our design not only achieves superior performance with an open vocabulary but also enhances downstream vision-language task performance through explicit relation modeling knowledge.

arXiv:2404.01339공식 코드 · Papers with Code

Humane Speech Synthesis through Zero-Shot Emotion and Disfluency Generation

Contemporary conversational systems often present a significant limitation: their responses lack the emotional depth and disfluent characteristic of human interactions. This absence becomes particularly noticeable when users seek more personalized and empathetic interactions. Consequently, this makes them seem mechanical and less relatable to human users. Recognizing this gap, we embarked on a journey to humanize machine communication, to ensure AI systems not only comprehend but also resonate. To address this shortcoming, we have designed an innovative speech synthesis pipeline. Within this framework, a cutting-edge language model introduces both human-like emotion and disfluencies in a zero-shot setting. These intricacies are seamlessly integrated into the generated text by the language model during text generation, allowing the system to mirror human speech patterns better, promoting more intuitive and natural user interactions. These generated elements are then adeptly transformed into corresponding speech patterns and emotive sounds using a rule-based approach during the text-to-speech phase. Based on our experiments, our novel system produces synthesized speech that's almost indistinguishable from genuine human communication, making each interaction feel more personal and authentic.

arXiv:2404.00971

Exploring and Evaluating Hallucinations in LLM-Powered Code Generation

The rise of Large Language Models (LLMs) has significantly advanced many applications on software engineering tasks, particularly in code generation. Despite the promising performance, LLMs are prone to generate hallucinations, which means LLMs might produce outputs that deviate from users' intent, exhibit internal inconsistencies, or misalign with the factual knowledge, making the deployment of LLMs potentially risky in a wide range of applications. Existing work mainly focuses on investing the hallucination in the domain of natural language generation (NLG), leaving a gap in understanding the types and extent of hallucinations in the context of code generation. To bridge the gap, we conducted a thematic analysis of the LLM-generated code to summarize and categorize the hallucinations present in it. Our study established a comprehensive taxonomy of hallucinations in LLM-generated code, encompassing 5 primary categories of hallucinations depending on the conflicting objectives and varying degrees of deviation observed in code generation. Furthermore, we systematically analyzed the distribution of hallucinations, exploring variations among different LLMs and their correlation with code correctness. Based on the results, we proposed HalluCode, a benchmark for evaluating the performance of code LLMs in recognizing hallucinations. Hallucination recognition and mitigation experiments with HalluCode and HumanEval show existing LLMs face great challenges in recognizing hallucinations, particularly in identifying their types, and are hardly able to mitigate hallucinations. We believe our findings will shed light on future research about hallucination evaluation, detection, and mitigation, ultimately paving the way for building more effective and reliable code LLMs in the future.

arXiv:2404.00752공식 코드 · Papers with Code

On the True Distribution Approximation of Minimum Bayes-Risk Decoding

Minimum Bayes-risk (MBR) decoding has recently gained renewed attention in text generation. MBR decoding considers texts sampled from a model as pseudo-references and selects the text with the highest similarity to the others. Therefore, sampling is one of the key elements of MBR decoding, and previous studies reported that the performance varies by sampling methods. From a theoretical standpoint, this performance variation is likely tied to how closely the samples approximate the true distribution of references. However, this approximation has not been the subject of in-depth study. In this study, we propose using anomaly detection to measure the degree of approximation. We first closely examine the performance variation and then show that previous hypotheses about samples do not correlate well with the variation, but our introduced anomaly scores do. The results are the first to empirically support the link between the performance and the core assumption of MBR decoding.

arXiv:2404.01768

Stereotype Detection in LLMs: A Multiclass, Explainable, and Benchmark-Driven Approach

Stereotype detection is a challenging and subjective task, as certain statements, such as "Black people like to play basketball," may not appear overtly toxic but still reinforce racial stereotypes. With the increasing prevalence of large language models (LLMs) in human-facing artificial intelligence (AI) applications, detecting these types of biases is essential. However, LLMs risk perpetuating and amplifying stereotypical outputs derived from their training data. A reliable stereotype detector is crucial for benchmarking bias, monitoring model input and output, filtering training data, and ensuring fairer model behavior in downstream applications. This paper introduces the Multi-Grain Stereotype (MGS) dataset, consisting of 51,867 instances across gender, race, profession, religion, and other stereotypes, curated from multiple existing datasets. We evaluate various machine learning approaches to establish baselines and fine-tune language models of different architectures and sizes, presenting a suite of stereotype multiclass classifiers trained on the MGS dataset. Given the subjectivity of stereotypes, explainability is essential to align model learning with human understanding of stereotypes. We employ explainable AI (XAI) tools, including SHAP, LIME, and BertViz, to assess whether the model's learned patterns align with human intuitions about stereotypes.Additionally, we develop stereotype elicitation prompts and benchmark the presence of stereotypes in text generation tasks using popular LLMs, employing the best-performing stereotype classifiers.

arXiv:2404.01361공식 코드 · Papers with Code

LLM Attributor: Interactive Visual Attribution for LLM Generation

While large language models (LLMs) have shown remarkable capability to generate convincing text across diverse domains, concerns around its potential risks have highlighted the importance of understanding the rationale behind text generation. We present LLM Attributor, a Python library that provides interactive visualizations for training data attribution of an LLM's text generation. Our library offers a new way to quickly attribute an LLM's text generation to training data points to inspect model behaviors, enhance its trustworthiness, and compare model-generated text with user-provided text. We describe the visual and interactive design of our tool and highlight usage scenarios for LLaMA2 models fine-tuned with two different datasets: online articles about recent disasters and finance-related question-answer pairs. Thanks to LLM Attributor's broad support for computational notebooks, users can easily integrate it into their workflow to interactively visualize attributions of their models. For easier access and extensibility, we open-source LLM Attributor at https://github.com/poloclub/ LLM-Attribution. The video demo is available at https://youtu.be/mIG2MDQKQxM.

arXiv:2404.01291공식 코드 · Papers with Code

Evaluating Text-to-Visual Generation with Image-to-Text Generation

Despite significant progress in generative AI, comprehensive evaluation remains challenging because of the lack of effective metrics and standardized benchmarks. For instance, the widely-used CLIPScore measures the alignment between a (generated) image and text prompt, but it fails to produce reliable scores for complex prompts involving compositions of objects, attributes, and relations. One reason is that text encoders of CLIP can notoriously act as a "bag of words", conflating prompts such as "the horse is eating the grass" with "the grass is eating the horse". To address this, we introduce the VQAScore, which uses a visual-question-answering (VQA) model to produce an alignment score by computing the probability of a "Yes" answer to a simple "Does this figure show '{text}'?" question. Though simpler than prior art, VQAScore computed with off-the-shelf models produces state-of-the-art results across many (8) image-text alignment benchmarks. We also compute VQAScore with an in-house model that follows best practices in the literature. For example, we use a bidirectional image-question encoder that allows image embeddings to depend on the question being asked (and vice versa). Our in-house model, CLIP-FlanT5, outperforms even the strongest baselines that make use of the proprietary GPT-4V. Interestingly, although we train with only images, VQAScore can also align text with video and 3D models. VQAScore allows researchers to benchmark text-to-visual generation using complex texts that capture the compositional structure of real-world prompts. We introduce GenAI-Bench, a more challenging benchmark with 1,600 compositional text prompts that require parsing scenes, objects, attributes, relationships, and high-order reasoning like comparison and logic. GenAI-Bench also offers over 15,000 human ratings for leading image and video generation models such as Stable Diffusion, DALL-E 3, and Gen2.

arXiv:2403.20279공식 코드 · Papers with Code

LUQ: Long-text Uncertainty Quantification for LLMs

Large Language Models (LLMs) have demonstrated remarkable capability in a variety of NLP tasks. However, LLMs are also prone to generate nonfactual content. Uncertainty Quantification (UQ) is pivotal in enhancing our understanding of a model's confidence on its generation, thereby aiding in the mitigation of nonfactual outputs. Existing research on UQ predominantly targets short text generation, typically yielding brief, word-limited responses. However, real-world applications frequently necessitate much longer responses. Our study first highlights the limitations of current UQ methods in handling long text generation. We then introduce \textsc{Luq} and its two variations, a series of novel sampling-based UQ approaches specifically designed for long text. Our findings reveal that \textsc{Luq} outperforms existing baseline methods in correlating with the model's factuality scores (negative coefficient of -0.85 observed for Gemini Pro). To further improve the factuality of LLM responses, we propose \textsc{Luq-Ensemble}, a method that ensembles responses from multiple models and selects the response with the lowest uncertainty. The ensembling method greatly improves the response factuality upon the best standalone LLM.

arXiv:2403.20127공식 코드 · Papers with Code

The Impact of Prompts on Zero-Shot Detection of AI-Generated Text

In recent years, there have been significant advancements in the development of Large Language Models (LLMs). While their practical applications are now widespread, their potential for misuse, such as generating fake news and committing plagiarism, has posed significant concerns. To address this issue, detectors have been developed to evaluate whether a given text is human-generated or AI-generated. Among others, zero-shot detectors stand out as effective approaches that do not require additional training data and are often likelihood-based. In chat-based applications, users commonly input prompts and utilize the AI-generated texts. However, zero-shot detectors typically analyze these texts in isolation, neglecting the impact of the original prompts. It is conceivable that this approach may lead to a discrepancy in likelihood assessments between the text generation phase and the detection phase. So far, there remains an unverified gap concerning how the presence or absence of prompts impacts detection accuracy for zero-shot detectors. In this paper, we introduce an evaluative framework to empirically analyze the impact of prompts on the detection accuracy of AI-generated text. We assess various zero-shot detectors using both white-box detection, which leverages the prompt, and black-box detection, which operates without prompt information. Our experiments reveal the significant influence of prompts on detection accuracy. Remarkably, compared with black-box detection without prompts, the white-box methods using prompts demonstrate an increase in AUC of at least $0.1$ across all zero-shot detectors tested. Code is available: \url{https://github.com/kaito25atugich/Detector}.