Multimodal large language models (MLLMs) have achieved remarkable success across various tasks. However, separate training of visual and textual encoders often results in a misalignment of the modality. Such misalignment may lead models to generate content that is absent from the input image, a phenomenon referred to as hallucination. These inaccuracies severely undermine the trustworthiness of MLLMs in real-world applications. Despite attempts to optimize text preferences to mitigate this issue, our initial investigation indicates that the trustworthiness of MLLMs remains inadequate. Specifically, these models tend to provide preferred answers even when the input image is heavily distorted. Analysis of visual token attention also indicates that the model focuses primarily on the surrounding context rather than the key object referenced in the question. These findings highlight a misalignment between the modalities, where answers inadequately leverage input images. Motivated by our findings, we propose Modality-Fair Preference Optimization (MFPO), which comprises three components: the construction of a multimodal preference dataset in which dispreferred images differ from originals solely in key regions; an image reward loss function encouraging the model to generate answers better aligned with the input images; and an easy-to-hard iterative alignment strategy to stabilize joint modality training. Extensive experiments on three trustworthiness benchmarks demonstrate that MFPO significantly enhances the trustworthiness of MLLMs. In particular, it enables the 7B models to attain trustworthiness levels on par with, or even surpass, those of the 13B, 34B, and larger models.
论文检索
输入标题、作者或关键词，从 12,319 篇学术成果中精准定位
Random Forests (RFs) are one of the most popular classifiers in machine learning. RF is an ensemble learning method that combines multiple Decision Trees (DTs), providing a more robust and accurate model than a single DT. However, one of the main step of RFs is the random selection of many different features during the construction phase of DTs, resulting in a forest with various features, which makes it difficult to extract short and concise explanations. In this paper, we propose integrating Logical Analysis of Data (LAD) into RFs. LAD is a pattern learning framework that combines optimization, Boolean functions, and combinatorial theory. One of its main goals is to generate minimal support sets (MSSes) that discriminate between different groups of data. More precisely, we show how to enhance the classical RF algorithm by randomly choosing MSSes rather than randomly choosing feature subsets that potentially contain irrelevant features for constructing DTs. Experiments on benchmark datasets reveal that integrating LAD into classical RFs using MSSes can maintain similar performance in terms of accuracy, produce forests of similar size, reduce the set of used features, and enable the extraction of significantly shorter explanations compared to classical RFs.
Recent studies have delved into constructing generalist agents for open-world environments like Minecraft. Despite the encouraging results, existing efforts mainly focus on solving basic programmatic tasks, e.g., material collection and tool-crafting following the Minecraft tech-tree, treating the ObtainDiamond task as the ultimate goal. This limitation stems from the narrowly defined set of actions available to agents, requiring them to learn effective long-horizon strategies from scratch. Consequently, discovering diverse gameplay opportunities in the open world becomes challenging. In this work, we introduce Odyssey, a new framework that empowers Large Language Model (LLM)-based agents with open-world skills to explore the vast Minecraft world. Odyssey comprises three key parts: (1) An interactive agent with an open-world skill library that consists of 40 primitive skills and 183 compositional skills. (2) A fine-tuned LLaMA-3 model trained on a large question-answering dataset with 390k+ instruction entries derived from the Minecraft Wiki. (3) A new agent capability benchmark includes the long-term planning task, the dynamic-immediate planning task, and the autonomous exploration task. Extensive experiments demonstrate that the proposed Odyssey framework can effectively evaluate different capabilities of LLM-based agents. All datasets, model weights, and code are publicly available to motivate future research on more advanced autonomous agent solutions.
We present QGen Studio: an adaptive question-answer generation, training, and evaluation platform. QGen Studio enables users to leverage large language models (LLMs) to create custom question-answer datasets and fine-tune models on this synthetic data. It features a dataset viewer and model explorer to streamline this process. The dataset viewer provides key metrics and visualizes the context from which the QA pairs are generated, offering insights into data quality. The model explorer supports model comparison, allowing users to contrast the performance of their trained LLMs against other models, supporting performance benchmarking and refinement. QGen Studio delivers an interactive, end-to-end solution for generating QA datasets and training scalable, domain-adaptable models. The studio will be open-sourced soon, allowing users to deploy it locally.
Comparative Analysis of Demonstration Selection Algorithms for In-Context Learning in Large Language Models (Student Abstract)
PDF ↗Demonstration selection algorithms play a crucial role in optimizing Large Language Models' (LLMs) in-context learning performance. Despite numerous proposed algorithms, their comparative effectiveness remains understudied. We present a comprehensive evaluation of six state-of-the-art demonstration selection algorithms across five datasets, examining both their effectiveness and computational efficiency. Our findings reveal significant trade-offs: while some demonstration selection algorithms achieve superior accuracy, they incur substantial computational costs. We also discover that increasing demonstration examples doesn't consistently improve performance, and some sophisticated algorithms struggle to outperform random selection in certain scenarios. These insights provide valuable benchmarks for future algorithm development and practical implementation. Our code is available at https://github.com/Tizzzzy/Demonstration_Selection_Overview.
This paper investigates the application of Generative Flow Networks (GFlowNets) to lead optimization in drug discovery. GFlowNets provide a novel framework for generating diverse molecular structures while optimizing for desired properties, addressing the limitations of traditional methods in exploring vast chemical spaces. We adapt GFlowNets to incrementally modify lead compounds, integrating domain-specific heuristics to guide the generation process. Our method employs the trajectory balance objective on a graph neural network (GNN), to learn a policy that samples fragments based on a multi-objective reward. The reward function ensures increase in cell permeability and similarity to the starting molecule. The results on benchmark datasets of activity cliffs demonstrate that GFlowNets can generate diverse modifications, producing optimized candidate molecules with improvement in cell permeability. This work can be extended with other pharmacokinetic properties for lead optimization in early-stage drug development, potentially accelerating the discovery of novel therapeutics.
QAagent: A Multiagent System for Unit Test Generation via Natural Language Pseudocode (Student Abstract)
PDF ↗Unit testing is essential for ensuring software quality, but it is often time-consuming and prone to developer oversight. With the rise of large language models (LLMs) in code generation, there is an increasing need for reliable and automated test generation systems. This work presents QAagent, a multi-agent system designed to generate unit tests using natural language pseudocode. QAagent leverages LLMs to create a detailed natural language plan of a function's implementation and then generates a comprehensive suite of test cases covering both base and edge scenarios. Experiments conducted on two widely-used benchmarks, HumanEval and MBPP, show that QAagent consistently outperforms existing frameworks in terms of code coverage, although its accuracy varies across datasets, demonstrating the potential for utilizing natural language pseudocode to to enhance automated test generation in LLM-driven coding environments.
Predicting the next activity in an ongoing process is one of the most common tasks in the business process management (BPM) domain. It allows businesses to optimize resource allocation, enhance operational efficiency, and aid both in risk mitigation and strategic decision-making. Existing state-of-the-art AI models for BPM do not fully capitalize on available semantic information within process event logs. As current advanced AI-BPM systems provide semantically richer textual data, the need for new adequate models grows. To address this gap, we develop SNAP—a novel system that utilizes LLMs by constructing narratives and semantic contextual stories for historical event logs, which are then used to generate precise and actionable predictions for the ongoing process. SNAP was evaluated on six benchmark datasets, where it demonstrated significant performance improvements over eleven SOTA models, particularly on datasets with high levels of semantic content. This work showcases the potential of integrating LLMs in BPM and outlines a clear path toward future deployment, emphasizing the relevance and innovation of our approach within the broader AI application landscape.
Drought is one of the most destructive and expensive natural disasters, severely impacting natural resources and risks by depleting water resources and diminishing agricultural yields. Under climate change, accurately predicting drought is critical for mitigating drought-induced risks. However, the intricate interplay among the physical and biological drivers that regulate droughts limits the predictability and understanding of drought, particularly at a subseasonal to seasonal (S2S) time scale. While deep learning has demonstrated the potential to address climate forecasting challenges, its application to drought prediction has received relatively less attention. In this work, we propose a new dataset, DroughtSet, which integrates relevant predictive features and three drought indices from multiple remote sensing and reanalysis datasets across the contiguous United States (CONUS). DroughtSet specifically provides the machine learning community with a new real-world dataset to benchmark drought prediction models and more generally, time-series forecasting methods. Furthermore, we propose a spatial-temporal model SPDrought to predict and interpret S2S droughts. Our model learns from the spatial and temporal information of physical and biological features to predict three types of droughts simultaneously. Multiple strategies are employed to quantify the importance of physical and biological features for drought prediction. Our results provide insights for researchers to better understand the predictability and sensitivity of drought to biological and physical conditions. We aim to contribute to the climate field by proposing a new tool to predict and understand the occurrence of droughts and provide the AI community with a new benchmark to study deep learning applications in climate science.
Medical benchmark datasets significantly contribute to developing Large Language Models (LLMs) for medical knowledge extraction, diagnosis, summarization, and other uses. Yet, current benchmarks are mainly derived from exam questions given to medical students or cases described in the medical literature, lacking the complexity of real-world patient cases that deviate from classic textbook abstractions. These include rare diseases, uncommon presentations of common diseases, and unexpected treatment responses. Here, we construct Clinically Uncommon Patient Cases and Diagnosis Dataset (CUPCase) based on 3,563 real-world case reports from BMC, which we formulate into diagnoses in open-ended textual format and as multiple-choice options with distractors. Using this dataset, we evaluate the ability of state-of-the-art LLMs, including both general-purpose and Clinical LLMs, to identify and correctly diagnose a patient case, and test models' performance when only partial information about cases is available. Our findings show that general-purpose GPT-4o attains the best performance in both the multiple-choice task (average accuracy of 87.9%) and the open-ended task (BERTScore F1 of 0.764), outperforming several LLMs with a focus on the medical domain such as Meditron-70B and MedLM-Large. Moreover, GPT-4o was able to maintain 87% and 88% of its performance with only the first 20% of tokens of the case presentation in multiple-choice and free text, respectively, highlighting the potential of LLMs to aid in early diagnosis in real-world cases. An error analysis demonstrates the complexity of the task, and attempts to hypothesise about the models' reasoning. CUPCase expands our ability to evaluate LLMs for clinical decision support in an open and reproducible manner.
Fields of The World: A Machine Learning Benchmark Dataset for Global Agricultural Field Boundary Segmentation
PDF ↗Crop field boundaries are foundational datasets for agricultural monitoring and assessments but are expensive to collect manually. Machine learning (ML) methods for automatically extracting field boundaries from remotely sensed images could help realize the demand for these datasets at a global scale. However, current ML methods for field instance segmentation lack sufficient geographic coverage, accuracy, and generalization capabilities. Further, research on improving ML methods is restricted by the lack of labeled datasets representing the diversity of global agricultural fields. We present Fields of The World (FTW)---a novel ML benchmark dataset for agricultural field instance segmentation spanning 24 countries on four continents (Europe, Africa, Asia, and South America). FTW is an order of magnitude larger than previous datasets with 70,462 samples, each containing instance and semantic segmentation masks paired with multi-date, multi-spectral Sentinel-2 satellite images. We provide results from baseline models for the new FTW benchmark, show that models trained on FTW have better zero-shot and fine-tuning performance in held-out countries than models that aren't pre-trained with diverse datasets, and show positive qualitative zero-shot results of FTW models in a real-world scenario -- running on Sentinel-2 scenes over Ethiopia.
PersonalizedUS: Interpretable Breast Cancer Risk Assessment with Local Coverage Uncertainty Quantification
PDF ↗Correctly assessing the malignancy of breast lesions identified during ultrasound examinations is crucial for effective clinical decision-making. However, the current "gold standard" relies on manual BI-RADS scoring by clinicians, often leading to unnecessary biopsies and a significant mental health burden on patients and their families. In this paper, we introduce PersonalizedUS, an interpretable machine learning system that leverages recent advances in conformal prediction to provide precise and personalized risk estimates with local coverage guarantees and sensitivity, specificity, and predictive values above 0.9 across various threshold levels. In particular, we identify meaningful lesion subgroups where distribution-free, model-agnostic conditional coverage holds, with approximately 90% of our prediction sets containing only the ground truth in most lesion subgroups, thus explicitly characterizing for which patients the model is most suitably applied. Moreover, we make available a curated tabular dataset of 1936 biopsied breast lesions from a recent observational multicenter study and benchmark the performance of several state-of-the-art learning algorithms. We also report a successful case study of the deployed system in the same multicenter context. Concrete clinical benefits include up to a 65% reduction in requested biopsies among BI-RADS 4a and 4b lesions, with minimal to no missed cancer cases.
FoMo: Multi-Modal, Multi-Scale and Multi-Task Remote Sensing Foundation Models for Forest Monitoring
PDF ↗Forests are vital to ecosystems, supporting biodiversity and essential services, but are rapidly changing due to land use and climate change. Understanding and mitigating negative effects requires parsing data on forests at global scale from a broad array of sensory modalities, and using them in diverse forest monitoring applications. Such diversity in data and applications can be effectively addressed through the development of a large, pre-trained foundation model that serves as a versatile base for various downstream tasks. However, remote sensing modalities, which are an excellent fit for several forest management tasks, are particularly challenging considering the variation in environmental conditions, object scales, image acquisition modes and spatio-temporal resolutions, etc. With that in mind, we present the first unified Forest Monitoring Benchmark (FoMo-Bench), carefully constructed to evaluate foundation models with such flexibility. FoMo-Bench consists of 15 diverse datasets encompassing satellite, aerial, and inventory data, covering a variety of geographical regions, and including multispectral, red-green-blue, synthetic aperture radar and LiDAR data with various temporal, spatial and spectral resolutions. FoMo-Bench includes multiple types of forest-monitoring tasks, spanning classification, segmentation, and object detection. To enhance task and geographic diversity in FoMo-Bench, we introduce TalloS, a global dataset combining satellite imagery with ground-based annotations for tree species classification across 1,000+ categories and hierarchical taxonomic levels. Finally, we propose FoMo-Net, a pre-training framework to develop foundation models with the capacity to process any combination of commonly used modalities and spectral bands in remote sensing. This work aims to inspire research collaborations between machine learning and forest biology researchers in exploring scalable multi-modal and multi-task models for forest monitoring and beyond. All code, data and appendices are published in the repository and on ArXiv.
Bridging the Gap: Enhancing LLM Performance for Low-Resource African Languages with New Benchmarks, Fine-Tuning, and Cultural Adjustments
PDF ↗Large Language Models (LLMs) have shown remarkable performance across various tasks, yet significant disparities remain for non-English languages, and especially native African languages. This paper addresses these disparities by creating approximately 1 million human-translated words of new benchmark data in 8 low-resource African languages, covering a population of over 160 million speakers of: Amharic, Bambara, Igbo, Sepedi (Northern Sotho), Shona, Sesotho (Southern Sotho), Setswana, and Tsonga. Our benchmarks are translations of Winogrande and three sections of MMLU: college medicine, clinical knowledge, and virology. Using the translated benchmarks, we report previously unknown performance gaps between state-of-the-art (SOTA) LLMs in English and African languages. Finally, using results from over 400 fine-tuned models, we explore several methods to reduce the LLM performance gap, including high-quality dataset fine-tuning (using an LLM-as-an-Annotator), cross-lingual transfer, and cultural appropriateness adjustments. Key findings include average mono-lingual improvements of 5.6% with fine-tuning (with 5.4% average mono-lingual improvements when using high-quality data over low-quality data), 2.9% average gains from cross-lingual transfer, and a 3.0% out-of-the-box performance boost on culturally appropriate questions. The publicly available benchmarks, translations, and code from this study support further research and development aimed at creating more inclusive and effective language technologies.
The increasing frequency and intensity of natural disasters call for rapid and accurate damage assessment. In response, disaster benchmark datasets from high-resolution satellite imagery have been constructed to develop methods for detecting damaged areas. However, these methods face significant challenges when applied to previously unseen regions due to the limited geographical and disaster-type diversity in the existing datasets. We introduce DAVI (Disaster Assessment with VIsion foundation model), a novel approach that addresses domain disparities and detects structural damage at the building level without requiring ground-truth labels for target regions. DAVI combines task-specific knowledge from a model trained on source regions with task-agnostic knowledge from an image segmentation model to generate pseudo labels indicating potential damage in target regions. It then utilizes a two-stage refinement process, which operate at both pixel and image levels, to accurately identify changes in disaster-affected areas. Our evaluation, including a case study on the 2023 Türkiye earthquake, demonstrates that our model achieves exceptional performance across diverse terrains (e.g., North America, Asia, and the Middle East) and disaster types (e.g., wildfires, hurricanes, and tsunamis). This confirms its robustness in disaster assessment without dependence on ground-truth labels and highlights its practical applicability.
Large Language Models (LLMs) have demonstrated impressive capabilities in complex reasoning tasks. However, they can be easily misled by unfaithful arguments during conversations, even when their original statements are correct. To this end, we investigate the problem of maintaining faithful integrity in LLMs. This involves ensuring that LLMs adhere to their faithful statements in the face of opposing arguments and are able to correct their incorrect statements when presented with faithful arguments. In this work, we propose a novel framework, named Alignment for Faithful Integrity with Confidence Estimation (AFICE), which aims to align the LLM responses with faithful integrity. Specifically, AFICE first designs a Bilateral Confidence Estimation (BCE) approach for estimating the uncertainty of each response generated by the LLM given a specific context, which simultaneously estimate the model's confidence to the question based on the internal states during decoding as well as to the answer based on cumulative probability ratios. With the BCE, we construct a conversational preference dataset composed of context, original statement, and argument, which is adopted for aligning the LLM for faithful integrity using Direct Preference Optimization (DPO). Extensive experimental results on a wide range of benchmarks demonstrate significant improvements in the LLM's ability to maintain faithful responses when encountering opposing arguments, ensuring both the practical utility and trustworthiness of LLMs in complex interactive settings.
As deep learning advances, Large Language Models (LLMs) and their multimodal counterparts, Vision-Language Models (VLMs), have shown exceptional performance in many real-world tasks. However, VLMs face significant security challenges, such as jailbreak attacks, where attackers attempt to bypass the model’s safety alignment to elicit harmful responses. The threat of jailbreak attacks on VLMs arises from both the inherent vulnerabilities of LLMs and the multiple information channels that VLMs process. While various attacks and defenses have been proposed, there is a notable gap in unified and comprehensive evaluations, as each method is evaluated on different dataset and metrics, making it impossible to compare the effectiveness of each method. To address this gap, we introduce MMJ-Bench, a unified pipeline for evaluating jailbreak attacks and defense techniques for VLMs. Through extensive experiments, we assess the effectiveness of various attack methods against SoTA VLMs and evaluate the impact of defense mechanisms on both defense effectiveness and model utility for normal tasks. Our comprehensive evaluation contribute to the field by offering a unified and systematic evaluation framework and the first public-available benchmark for VLM jailbreak research. We also demonstrate several insightful findings that highlights directions for future studies.
Reinforcement Learning Platform for Adversarial Black-box Attacks with Custom Distortion Filters
PDF ↗We present a Reinforcement Learning Platform for Adversarial Black-box untargeted and targeted attacks, RLAB, that allows users to select from various distortion filters to create adversarial examples. The platform uses a Reinforcement Learning agent to add minimum distortion to input images while still causing misclassification by the target model. The agent uses a novel dual-action method to explore the input image at each step to identify sensitive regions for adding distortions while removing noises that have less impact on the target model. This dual action leads to faster and more efficient convergence of the attack. The platform can also be used to measure the robustness of image classification models against specific distortion types. Also, retraining the model with adversarial samples significantly improved robustness when evaluated on benchmark datasets. The proposed platform outperforms state-of-the-art methods in terms of the average number of queries required to cause misclassification. This advances trustworthiness with a positive social impact.
SafetyPrompts: A Systematic Review of Open Datasets for Evaluating and Improving Large Language Model Safety
PDF ↗The last two years have seen a rapid growth in concerns around the safety of large language models (LLMs). Researchers and practitioners have met these concerns by creating an abundance of datasets for evaluating and improving LLM safety. However, much of this work has happened in parallel, and with very different goals in mind, ranging from the mitigation of near-term risks around bias and toxic content generation to the assessment of longer-term catastrophic risk potential. This makes it difficult for researchers and practitioners to find the most relevant datasets for their use case, and to identify gaps in dataset coverage that future work may fill. To remedy these issues, we conduct a first systematic review of open datasets for evaluating and improving LLM safety. We review 144 datasets, which we identified through an iterative and community-driven process over the course of several months. We highlight patterns and trends, such as a trend towards fully synthetic datasets, as well as gaps in dataset coverage, such as a clear lack of non-English and naturalistic datasets. We also examine how LLM safety datasets are used in practice -- in LLM release publications and popular LLM benchmarks -- finding that current evaluation practices are highly idiosyncratic and make use of only a small fraction of available datasets. Our contributions are based on SafetyPrompts.com, a living catalogue of open datasets for LLM safety, which we plan to update continuously as the field of LLM safety develops.
The increasing parameters and expansive dataset of large lan- guage models (LLMs) highlight the urgent demand for a technical solution to audit the underlying privacy risks and copyright issues associated with LLMs. Existing studies have partially addressed this need through an exploration of the pre-training data detection problem, which is an instance of a membership inference attack (MIA). This problem involves determining whether a given piece of text has been used during the pre-training phase of the target LLM. Although existing methods have designed various sophisticated MIA score functions to achieve considerable detection performance in pre-trained LLMs, how to achieve high-confidence detection and how to perform MIA on aligned LLMs remain challenging. In this paper, we propose MIA-Tuner, a novel instruction-based MIA method, which instructs LLMs themselves to serve as a more precise pre-training data detector internally, rather than design an external MIA score function. Furthermore, we design two instruction-based safeguards to respectively mitigate the privacy risks brought by the existing methods and MIA-Tuner. To comprehensively evaluate the most recent state-of-the-art LLMs, we collect a more up-to-date MIA benchmark dataset, named WIKIMIA-24, to replace the widely adopted benchmark WIKIMIA. We conduct extensive experiments across various aligned and unaligned LLMs over the two benchmark datasets. The results demonstrate that MIA-Tuner increases the AUC of MIAs from 0.7 to a significantly high level of 0.9.