Searcharxiv⌕ Search

arXiv subjects

Haonan Li

Publications and source records attributed to Haonan Li.

At least 73 records · Page 4Linked to original sources
arXiv2024

A Chinese Dataset for Evaluating the Safeguards in Large Language Models

Many studies have demonstrated that large language models (LLMs) can produce harmful responses, exposing users to unexpected risks when LLMs are deployed. Previous studies have proposed comprehensive taxonomies of the risks posed by LLMs, as well as corresponding prompts that can be used to examine the safety mechanisms of LLMs. However, the focus has been almost exclusively on English, and little has been explored for other languages. Here we aim to bridge this gap. We first introduce a dataset for the safety evaluation of Chinese LLMs, and then extend it to two other scenarios that can be used to better identify false negative and false positive examples in terms of risky prompt rejections. We further present a set of fine-grained safety assessment criteria for each risk type, facilitating both manual annotation and automatic evaluation in terms of LLM response harmfulness. Our experiments on five LLMs show that region-specific risks are the prevalent type of risk, presenting the major issue with all Chinese LLMs we experimented with. Our data is available at https://github.com/Libr-AI/do-not-answer. Warning: this paper contains example data that may be offensive, harmful, or biased.

cs.CL↗
arXiv2024

ArabicMMLU: Assessing Massive Multitask Language Understanding in Arabic

The focus of language model evaluation has transitioned towards reasoning and knowledge-intensive tasks, driven by advancements in pretraining large models. While state-of-the-art models are partially trained on large Arabic texts, evaluating their performance in Arabic remains challenging due to the limited availability of relevant datasets. To bridge this gap, we present \datasetname{}, the first multi-task language understanding benchmark for the Arabic language, sourced from school exams across diverse educational levels in different countries spanning North Africa, the Levant, and the Gulf regions. Our data comprises 40 tasks and 14,575 multiple-choice questions in Modern Standard Arabic (MSA) and is carefully constructed by collaborating with native speakers in the region. Our comprehensive evaluations of 35 models reveal substantial room for improvement, particularly among the best open-source models. Notably, BLOOMZ, mT0, LLaMA2, and Falcon struggle to achieve a score of 50%, while even the top-performing Arabic-centric model only achieves a score of 62.3%.

cs.CL↗
arXiv2024

Fact-Checking the Output of Large Language Models via Token-Level Uncertainty Quantification

Large language models (LLMs) are notorious for hallucinating, i.e., producing erroneous claims in their output. Such hallucinations can be dangerous, as occasional factual inaccuracies in the generated text might be obscured by the rest of the output being generally factually correct, making it extremely hard for the users to spot them. Current services that leverage LLMs usually do not provide any means for detecting unreliable generations. Here, we aim to bridge this gap. In particular, we propose a novel fact-checking and hallucination detection pipeline based on token-level uncertainty quantification. Uncertainty scores leverage information encapsulated in the output of a neural network or its layers to detect unreliable predictions, and we show that they can be used to fact-check the atomic claims in the LLM output. Moreover, we present a novel token-level uncertainty quantification method that removes the impact of uncertainty about what claim to generate on the current step and what surface form to use. Our method Claim Conditioned Probability (CCP) measures only the uncertainty of a particular claim value expressed by the model. Experiments on the task of biography generation demonstrate strong improvements for CCP compared to the baselines for seven LLMs and four languages. Human evaluation reveals that the fact-checking pipeline based on uncertainty quantification is competitive with a fact-checking tool that leverages external knowledge.

cs.CL↗
arXiv2024

3D Hand Mesh Recovery from Monocular RGB in Camera Space

With the rapid advancement of technologies such as virtual reality, augmented reality, and gesture control, users expect interactions with computer interfaces to be more natural and intuitive. Existing visual algorithms often struggle to accomplish advanced human-computer interaction tasks, necessitating accurate and reliable absolute spatial prediction methods. Moreover, dealing with complex scenes and occlusions in monocular images poses entirely new challenges. This study proposes a network model that performs parallel processing of root-relative grids and root recovery tasks. The model enables the recovery of 3D hand meshes in camera space from monocular RGB images. To facilitate end-to-end training, we utilize an implicit learning approach for 2D heatmaps, enhancing the compatibility of 2D cues across different subtasks. Incorporate the Inception concept into spectral graph convolutional network to explore relative mesh of root, and integrate it with the locally detailed and globally attentive method designed for root recovery exploration. This approach improves the model's predictive performance in complex environments and self-occluded scenes. Through evaluation on the large-scale hand dataset FreiHAND, we have demonstrated that our proposed model is comparable with state-of-the-art models. This study contributes to the advancement of techniques for accurate and reliable absolute spatial prediction in various human-computer interaction applications.

cs.CV↗
arXiv2024

Learning From Failure: Integrating Negative Examples when Fine-tuning Large Language Models as Agents

Large language models (LLMs) have achieved success in acting as agents, which interact with environments through tools such as search engines. However, LLMs are optimized for language generation instead of tool use during training or alignment, limiting their effectiveness as agents. To resolve this problem, previous work has first collected interaction trajectories between LLMs and environments, using only trajectories that successfully finished the task to fine-tune smaller models, making fine-tuning data scarce and acquiring it both difficult and costly. Discarding failed trajectories also leads to significant wastage of data and resources and limits the possible optimization paths during fine-tuning. In this paper, we argue that unsuccessful trajectories offer valuable insights, and LLMs can learn from these trajectories through appropriate quality control and fine-tuning strategies. By simply adding a prefix or suffix that tells the model whether to generate a successful trajectory during training, we improve model performance by a large margin on mathematical reasoning, multi-hop question answering, and strategic question answering tasks. We further analyze the inference results and find that our method provides a better trade-off between valuable information and errors in unsuccessful trajectories. To our knowledge, we are the first to demonstrate the value of negative trajectories and their application in agent-tunning scenarios. Our findings offer guidance for developing better agent-tuning methods and low-resource data usage techniques.

cs.CL↗
arXiv2024

On the number of prime numbers between $n^2$ and ${(n+1)}^2$

Let $p_{r+1}-1>n \geq p_r-1$, based on a sequence $\{1,2,3\cdots\ M_r(M_r=p_1p_2\cdots p_r)\}$, we compare the density of coprime numbers and establish a correlation between the proportions of coprime numbers in the ranges from 1 to consecutive square numbers. Then, we derive the relationship between the number of coprimes in the interval of $n^2 \sim {(n+1)}^2$ and the proportion of coprimes in the interval of $1 \sim n^2$, proving that there is at least one prime number between any $n^2$ and ${(n+1)}^2$. By extending our research to the range of $1 \sim M_r^2$, we establish the relationship between the proportions of backwards coprime numbers in the ranges from ${M_r}^2$ to consecutive square numbers; furthermore, we establish a relationship between the proportions of coprimes in small interval and the whole interval. Then, in conclusion, the number of coprimes between $n^2$ and ${(n+1)}^2$ is greater than $n\prod_{i=1}^{r}{(1-\frac{1}{p_i}})$, thus proving that there are at least 2 prime numbers between $n^2$ and ${(n+1)}^2$.

math.NT↗
arXiv2024

EXAMS-V: A Multi-Discipline Multilingual Multimodal Exam Benchmark for Evaluating Vision Language Models

We introduce EXAMS-V, a new challenging multi-discipline multimodal multilingual exam benchmark for evaluating vision language models. It consists of 20,932 multiple-choice questions across 20 school disciplines covering natural science, social science, and other miscellaneous studies, e.g., religion, fine arts, business, etc. EXAMS-V includes a variety of multimodal features such as text, images, tables, figures, diagrams, maps, scientific symbols, and equations. The questions come in 11 languages from 7 language families. Unlike existing benchmarks, EXAMS-V is uniquely curated by gathering school exam questions from various countries, with a variety of education systems. This distinctive approach calls for intricate reasoning across diverse languages and relies on region-specific knowledge. Solving the problems in the dataset requires advanced perception and joint reasoning over the text and the visual content of the image. Our evaluation results demonstrate that this is a challenging dataset, which is difficult even for advanced vision-text models such as GPT-4V and Gemini; this underscores the inherent complexity of the dataset and its significance as a future benchmark.

cs.CL↗
arXiv2024

Demystifying Instruction Mixing for Fine-tuning Large Language Models

Instruction tuning significantly enhances the performance of large language models (LLMs) across various tasks. However, the procedure to optimizing the mixing of instruction datasets for LLM fine-tuning is still poorly understood. This study categorizes instructions into three primary types: NLP downstream tasks, coding, and general chat. We explore the effects of instruction tuning on different combinations of datasets on LLM performance, and find that certain instruction types are more advantageous for specific applications but can negatively impact other areas. This work provides insights into instruction mixtures, laying the foundations for future research.

cs.CL↗
arXiv2024

CMMLU: Measuring massive multitask language understanding in Chinese

As the capabilities of large language models (LLMs) continue to advance, evaluating their performance becomes increasingly crucial and challenging. This paper aims to bridge this gap by introducing CMMLU, a comprehensive Chinese benchmark that covers various subjects, including natural science, social sciences, engineering, and humanities. We conduct a thorough evaluation of 18 advanced multilingual- and Chinese-oriented LLMs, assessing their performance across different subjects and settings. The results reveal that most existing LLMs struggle to achieve an average accuracy of 50%, even when provided with in-context examples and chain-of-thought prompts, whereas the random baseline stands at 25%. This highlights significant room for improvement in LLMs. Additionally, we conduct extensive experiments to identify factors impacting the models' performance and propose directions for enhancing LLMs. CMMLU fills the gap in evaluating the knowledge and reasoning capabilities of large language models within the Chinese context.

cs.CL↗
arXiv2024

Location Aware Modular Biencoder for Tourism Question Answering

Answering real-world tourism questions that seek Point-of-Interest (POI) recommendations is challenging, as it requires both spatial and non-spatial reasoning, over a large candidate pool. The traditional method of encoding each pair of question and POI becomes inefficient when the number of candidates increases, making it infeasible for real-world applications. To overcome this, we propose treating the QA task as a dense vector retrieval problem, where we encode questions and POIs separately and retrieve the most relevant POIs for a question by utilizing embedding space similarity. We use pretrained language models (PLMs) to encode textual information, and train a location encoder to capture spatial information of POIs. Experiments on a real-world tourism QA dataset demonstrate that our approach is effective, efficient, and outperforms previous methods across all metrics. Enabled by the dense retrieval architecture, we further build a global evaluation baseline, expanding the search space by 20 times compared to previous work. We also explore several factors that impact on the model's performance through follow-up experiments. Our code and model are publicly available at https://github.com/haonan-li/LAMB.

cs.CL↗
arXiv2023

LLM360: Towards Fully Transparent Open-Source LLMs

The recent surge in open-source Large Language Models (LLMs), such as LLaMA, Falcon, and Mistral, provides diverse options for AI practitioners and researchers. However, most LLMs have only released partial artifacts, such as the final model weights or inference code, and technical reports increasingly limit their scope to high-level design choices and surface statistics. These choices hinder progress in the field by degrading transparency into the training of LLMs and forcing teams to rediscover many details in the training process. We present LLM360, an initiative to fully open-source LLMs, which advocates for all training code and data, model checkpoints, and intermediate results to be made available to the community. The goal of LLM360 is to support open and collaborative AI research by making the end-to-end LLM training process transparent and reproducible by everyone. As a first step of LLM360, we release two 7B parameter LLMs pre-trained from scratch, Amber and CrystalCoder, including their training code, data, intermediate checkpoints, and analyses (at https://www.llm360.ai). We are committed to continually pushing the boundaries of LLMs through this open-source effort. More large-scale and stronger models are underway and will be released in the future.

cs.CL↗
arXiv2023

The Hitchhiker's Guide to Program Analysis: A Journey with Large Language Models

Static analysis is a widely used technique in software engineering for identifying and mitigating bugs. However, a significant hurdle lies in achieving a delicate balance between precision and scalability. Large Language Models (LLMs) offer a promising alternative, as recent advances demonstrate remarkable capabilities in comprehending, generating, and even debugging code. Yet, the logic of bugs can be complex and require sophisticated reasoning and a large analysis scope spanning multiple functions. Therefore, at this point, LLMs are better used in an assistive role to complement static analysis. In this paper, we take a deep dive into the open space of LLM-assisted static analysis, using use-before-initialization (UBI) bugs as a case study. To this end, we develop LLift, a fully automated framework that interfaces with both a static analysis tool and an LLM. By carefully designing the framework and the prompts, we are able to overcome a number of challenges, including bug-specific modeling, the large problem scope, the non-deterministic nature of LLMs, etc. Tested in a real-world scenario analyzing nearly a thousand potential UBI bugs produced by static analysis, LLift demonstrates a potent capability, showcasing a reasonable precision (50%) and appearing to have no missing bugs. It even identified 13 previously unknown UBI bugs in the Linux kernel. This research paves the way for new opportunities and methodologies in using LLMs for bug discovery in extensive, real-world datasets.

cs.SE↗
arXiv2023

Large Language Models Only Pass Primary School Exams in Indonesia: A Comprehensive Test on IndoMMLU

Although large language models (LLMs) are often pre-trained on large-scale multilingual texts, their reasoning abilities and real-world knowledge are mainly evaluated based on English datasets. Assessing LLM capabilities beyond English is increasingly vital but hindered due to the lack of suitable datasets. In this work, we introduce IndoMMLU, the first multi-task language understanding benchmark for Indonesian culture and languages, which consists of questions from primary school to university entrance exams in Indonesia. By employing professional teachers, we obtain 14,981 questions across 64 tasks and education levels, with 46% of the questions focusing on assessing proficiency in the Indonesian language and knowledge of nine local languages and cultures in Indonesia. Our empirical evaluations show that GPT-3.5 only manages to pass the Indonesian primary school level, with limited knowledge of local Indonesian languages and culture. Other smaller models such as BLOOMZ and Falcon perform at even lower levels.

cs.CL↗
arXiv2023

Can Large Language Model Comprehend Ancient Chinese? A Preliminary Test on ACLUE

Large language models (LLMs) have showcased remarkable capabilities in understanding and generating language. However, their ability in comprehending ancient languages, particularly ancient Chinese, remains largely unexplored. To bridge this gap, we present ACLUE, an evaluation benchmark designed to assess the capability of language models in comprehending ancient Chinese. ACLUE consists of 15 tasks cover a range of skills, spanning phonetic, lexical, syntactic, semantic, inference and knowledge. Through the evaluation of eight state-of-the-art LLMs, we observed a noticeable disparity in their performance between modern Chinese and ancient Chinese. Among the assessed models, ChatGLM2 demonstrates the most remarkable performance, achieving an average score of 37.4%. We have made our code and data public available.

cs.CL↗
arXiv2023

Bactrian-X: Multilingual Replicable Instruction-Following Models with Low-Rank Adaptation

Instruction tuning has shown great promise in improving the performance of large language models. However, research on multilingual instruction tuning has been limited due to the scarcity of high-quality instruction-response datasets across different languages. To bridge this gap, we present Bactrian-X, a comprehensive multilingual parallel dataset of 3.4 million instruction-response pairs across 52 languages. Leveraging this dataset, we train a set of adapters using low-rank adaptation (LoRA), which are lightweight components that seamlessly integrate with large language models. These adapters have a substantially lower parameter count than the base model, making them easily replaceable and usable as plug-ins for different languages or language groups. Extensive experiments in various multilingual evaluation settings demonstrate that models derived from LoRA-based training over Bactrian-X outperform both the vanilla models and existing instruction-tuned models. The code and models are publicly available at https://github.com/mbzuai-nlp/bactrian-x

cs.CL↗
arXiv2023

Jais and Jais-chat: Arabic-Centric Foundation and Instruction-Tuned Open Generative Large Language Models

We introduce Jais and Jais-chat, new state-of-the-art Arabic-centric foundation and instruction-tuned open generative large language models (LLMs). The models are based on the GPT-3 decoder-only architecture and are pretrained on a mixture of Arabic and English texts, including source code in various programming languages. With 13 billion parameters, they demonstrate better knowledge and reasoning capabilities in Arabic than any existing open Arabic and multilingual models by a sizable margin, based on extensive evaluation. Moreover, the models are competitive in English compared to English-centric open models of similar size, despite being trained on much less English data. We provide a detailed description of the training, the tuning, the safety alignment, and the evaluation of the models. We release two open versions of the model -- the foundation Jais model, and an instruction-tuned Jais-chat variant -- with the aim of promoting research on Arabic LLMs. Available at https://huggingface.co/inception-mbzuai/jais-13b-chat

cs.CL↗
arXiv2023

Do-Not-Answer: A Dataset for Evaluating Safeguards in LLMs

With the rapid evolution of large language models (LLMs), new and hard-to-predict harmful capabilities are emerging. This requires developers to be able to identify risks through the evaluation of "dangerous capabilities" in order to responsibly deploy LLMs. In this work, we collect the first open-source dataset to evaluate safeguards in LLMs, and deploy safer open-source LLMs at a low cost. Our dataset is curated and filtered to consist only of instructions that responsible language models should not follow. We annotate and assess the responses of six popular LLMs to these instructions. Based on our annotation, we proceed to train several BERT-like classifiers, and find that these small classifiers can achieve results that are comparable with GPT-4 on automatic safety evaluation. Warning: this paper contains example data that may be offensive, harmful, or biased.

cs.CL↗
arXiv2022

Generalized Koszul Algebra and Koszul Duality

We define generalized Koszul modules and rings and develop a generalized Koszul theory for $\mathbb{N}$-graded rings with the degree zero part noetherian semiperfect. This theory specializes to the classical Koszul theory for graded rings with degree zero part artinian semisimple developed by Beilinson-Ginzburg-Soergel and the ungraded Koszul theory for noetherian semiperfect rings developed by Green and Martin{é}z-Villa. Let $A$ be a left finite $\mathbb{N}$-graded ring generated in degree $1$ with $A_0$ noetherian semiperfect, $J$ be its graded Jacobson radical and $S=A/J$. By the Koszul dual of $A$ we mean the Yoneda Ext ring $\underline{\text{Ext}}_A^\bullet(S,S)$. If $A$ is a generalized Koszul ring and $M$ is a generalized Koszul module, then it is proved that the Koszul dual of the Koszul dual of $A$ is $\text{Gr}_J A$ and the Koszul dual of the Koszul dual of $M$ is $\text{Gr}_J M$. If $A$ is a locally finite algebra, then the following statements are proved to be equivalent: $A$ is generalized Koszul; the Koszul dual $\underline{\text{Ext}}_A^\bullet(S,S)$ of $A$ is (classically) Koszul; $\text{Gr}_J A$ is (classically) Koszul; the opposite ring $A^{op}$ of $A$ is generalized Koszul. It is also proved that if $A$ is generalized Koszul with finite global dimension then $A$ is generalized AS regular if and only if the Koszul dual of $A$ is self-injective.

math.RA↗