arXiv subjects

Besmira Nushi

Publications and source records attributed to Besmira Nushi.

At least 37 records · Page 2Linked to original sources

arXiv2024

Eureka: Evaluating and Understanding Large Foundation Models

Rigorous and reproducible evaluation is critical for assessing the state of the art and for guiding scientific advances in Artificial Intelligence. Evaluation is challenging in practice due to several reasons, including benchmark saturation, lack of transparency in methods used for measurement, development challenges in extracting measurements for generative tasks, and, more generally, the extensive number of capabilities required for a well-rounded comparison across models. We make three contributions to alleviate the above challenges. First, we present Eureka, an open-source framework for standardizing evaluations of large foundation models beyond single-score reporting and rankings. Second, we introduce Eureka-Bench as an extensible collection of benchmarks testing capabilities that (i) are still challenging for state-of-the-art models and (ii) represent fundamental but overlooked language and multimodal capabilities. The inherent space for improvement in non-saturated benchmarks enables us to discover meaningful differences between models at a capability level. Third, using Eureka, we conduct an analysis of 12 state-of-the-art models, providing in-depth insights into failure understanding and model comparison, which can be leveraged to plan targeted improvements. In contrast to recent trends in reports and leaderboards showing absolute rankings and claims for one model or another to be the best, our analysis shows that there is no such best model. Different models have different strengths, but there are models that appear more often than others as best performers for some capabilities. Despite the recent improvements, current models still struggle with several fundamental capabilities including detailed image understanding, benefiting from multimodal input when available rather than fully relying on language, factuality and grounding for information retrieval, and over refusals.

cs.LG↗

arXiv2024

Understanding Information Storage and Transfer in Multi-modal Large Language Models

Understanding the mechanisms of information storage and transfer in Transformer-based models is important for driving model understanding progress. Recent work has studied these mechanisms for Large Language Models (LLMs), revealing insights on how information is stored in a model's parameters and how information flows to and from these parameters in response to specific prompts. However, these studies have not yet been extended to Multi-modal Large Language Models (MLLMs). Given their expanding capabilities and real-world use, we start by studying one aspect of these models -- how MLLMs process information in a factual visual question answering task. We use a constraint-based formulation which views a visual question as having a set of visual or textual constraints that the model's generated answer must satisfy to be correct (e.g. What movie directed by the director in this photo has won a Golden Globe?). Under this setting, we contribute i) a method that extends causal information tracing from pure language to the multi-modal setting, and ii) VQA-Constraints, a test-bed of 9.7K visual questions annotated with constraints. We use these tools to study two open-source MLLMs, LLaVa and multi-modal Phi-2. Our key findings show that these MLLMs rely on MLP and self-attention blocks in much earlier layers for information storage, compared to LLMs whose mid-layer MLPs are more important. We also show that a consistent small subset of visual tokens output by the vision encoder are responsible for transferring information from the image to these causal blocks. We validate these mechanisms by introducing MultEdit, a model-editing algorithm that can correct errors and insert new long-tailed information into MLLMs by targeting these causal blocks.

cs.CV↗

arXiv2024

Introducing v0.5 of the AI Safety Benchmark from MLCommons

This paper introduces v0.5 of the AI Safety Benchmark, which has been created by the MLCommons AI Safety Working Group. The AI Safety Benchmark has been designed to assess the safety risks of AI systems that use chat-tuned language models. We introduce a principled approach to specifying and constructing the benchmark, which for v0.5 covers only a single use case (an adult chatting to a general-purpose assistant in English), and a limited set of personas (i.e., typical users, malicious users, and vulnerable users). We created a new taxonomy of 13 hazard categories, of which 7 have tests in the v0.5 benchmark. We plan to release version 1.0 of the AI Safety Benchmark by the end of 2024. The v1.0 benchmark will provide meaningful insights into the safety of AI systems. However, the v0.5 benchmark should not be used to assess the safety of AI systems. We have sought to fully document the limitations, flaws, and challenges of v0.5. This release of v0.5 of the AI Safety Benchmark includes (1) a principled approach to specifying and constructing the benchmark, which comprises use cases, types of systems under test (SUTs), language and context, personas, tests, and test items; (2) a taxonomy of 13 hazard categories with definitions and subcategories; (3) tests for seven of the hazard categories, each comprising a unique set of test items, i.e., prompts. There are 43,090 test items in total, which we created with templates; (4) a grading system for AI systems against the benchmark; (5) an openly available platform, and downloadable tool, called ModelBench that can be used to evaluate the safety of AI systems on the benchmark; (6) an example evaluation report which benchmarks the performance of over a dozen openly available chat-tuned language models; (7) a test specification for the benchmark.

cs.CL↗

arXiv2024

Attention Satisfies: A Constraint-Satisfaction Lens on Factual Errors of Language Models

We investigate the internal behavior of Transformer-based Large Language Models (LLMs) when they generate factually incorrect text. We propose modeling factual queries as constraint satisfaction problems and use this framework to investigate how the LLM interacts internally with factual constraints. We find a strong positive relationship between the LLM's attention to constraint tokens and the factual accuracy of generations. We curate a suite of 10 datasets containing over 40,000 prompts to study the task of predicting factual errors with the Llama-2 family across all scales (7B, 13B, 70B). We propose SAT Probe, a method probing attention patterns, that can predict factual errors and fine-grained constraint satisfaction, and allow early error identification. The approach and findings take another step towards using the mechanistic understanding of LLMs to enhance their reliability.

cs.CL↗

arXiv2024

Diversity of Thought Improves Reasoning Abilities of LLMs

Large language models (LLMs) are documented to struggle in settings that require complex reasoning. Nevertheless, instructing the model to break down the problem into smaller reasoning steps, or ensembling various generations through modifying decoding steps boosts performance. However, these methods assume that the input prompt is fixed and expect the decoding strategies to introduce the diversity needed for ensembling. In this work, we discuss how one can create and leverage variations of the input prompt as a means of diversity of thought. We propose a method that automatically improves prompt diversity by soliciting feedback from the LLM to ideate approaches that are apt for the problem. We then ensemble the diverse prompts in our method DIVSE (DIVerse reasoning path Self-Ensemble) across multiple inference calls, or use diverse approaches within a single inference call; we call the latter IDIV-SE (In-call DIVerse reasoning path Self-Ensemble). Apart from our approaches outperforming prior work, DIV-SE(in particular) advances state-of-the-art performance on the challenging planning and graph coloring benchmarks. Our results improve the Pareto frontier of the accuracy-cost trade-off.

cs.CL↗

arXiv2023

Benchmarking Spatial Relationships in Text-to-Image Generation

Spatial understanding is a fundamental aspect of computer vision and integral for human-level reasoning about images, making it an important component for grounded language understanding. While recent text-to-image synthesis (T2I) models have shown unprecedented improvements in photorealism, it is unclear whether they have reliable spatial understanding capabilities. We investigate the ability of T2I models to generate correct spatial relationships among objects and present VISOR, an evaluation metric that captures how accurately the spatial relationship described in text is generated in the image. To benchmark existing models, we introduce a dataset, $\mathrm{SR}_{2D}$, that contains sentences describing two or more objects and the spatial relationships between them. We construct an automated evaluation pipeline to recognize objects and their spatial relationships, and employ it in a large-scale evaluation of T2I models. Our experiments reveal a surprising finding that, although state-of-the-art T2I models exhibit high image quality, they are severely limited in their ability to generate multiple objects or the specified spatial relations between them. Our analyses demonstrate several biases and artifacts of T2I models such as the difficulty with generating multiple objects, a bias towards generating the first object mentioned, spatially inconsistent outputs for equivalent relationships, and a correlation between object co-occurrence and spatial understanding capabilities. We conduct a human study that shows the alignment between VISOR and human judgement about spatial understanding. We offer the $\mathrm{SR}_{2D}$ dataset and the VISOR metric to the community in support of T2I reasoning research.

cs.CV↗

arXiv2023

KITAB: Evaluating LLMs on Constraint Satisfaction for Information Retrieval

We study the ability of state-of-the art models to answer constraint satisfaction queries for information retrieval (e.g., 'a list of ice cream shops in San Diego'). In the past, such queries were considered to be tasks that could only be solved via web-search or knowledge bases. More recently, large language models (LLMs) have demonstrated initial emergent abilities in this task. However, many current retrieval benchmarks are either saturated or do not measure constraint satisfaction. Motivated by rising concerns around factual incorrectness and hallucinations of LLMs, we present KITAB, a new dataset for measuring constraint satisfaction abilities of language models. KITAB consists of book-related data across more than 600 authors and 13,000 queries, and also offers an associated dynamic data collection and constraint verification approach for acquiring similar test data for other authors. Our extended experiments on GPT4 and GPT3.5 characterize and decouple common failure modes across dimensions such as information popularity, constraint types, and context availability. Results show that in the absence of context, models exhibit severe limitations as measured by irrelevant information, factual errors, and incompleteness, many of which exacerbate as information popularity decreases. While context availability mitigates irrelevant information, it is not helpful for satisfying constraints, identifying fundamental barriers to constraint satisfaction. We open source our contributions to foster further research on improving constraint satisfaction abilities of future models.

cs.LG↗

arXiv2023

Mitigating Spurious Correlations in Multi-modal Models during Fine-tuning

Spurious correlations that degrade model generalization or lead the model to be right for the wrong reasons are one of the main robustness concerns for real-world deployments. However, mitigating these correlations during pre-training for large-scale models can be costly and impractical, particularly for those without access to high-performance computing resources. This paper proposes a novel approach to address spurious correlations during fine-tuning for a given domain of interest. With a focus on multi-modal models (e.g., CLIP), the proposed method leverages different modalities in these models to detect and explicitly set apart spurious attributes from the affected class, achieved through a multi-modal contrastive loss function that expresses spurious relationships through language. Our experimental results and in-depth visualizations on CLIP show that such an intervention can effectively i) improve the model's accuracy when spurious attributes are not present, and ii) directs the model's activation maps towards the actual class rather than the spurious attribute when present. In particular, on the Waterbirds dataset, our algorithm achieved a worst-group accuracy 23% higher than ERM on CLIP with a ResNet-50 backbone, and 32% higher on CLIP with a ViT backbone, while maintaining the same average accuracy as ERM.

cs.LG↗

arXiv2023

Social Biases through the Text-to-Image Generation Lens

Text-to-Image (T2I) generation is enabling new applications that support creators, designers, and general end users of productivity software by generating illustrative content with high photorealism starting from a given descriptive text as a prompt. Such models are however trained on massive amounts of web data, which surfaces the peril of potential harmful biases that may leak in the generation process itself. In this paper, we take a multi-dimensional approach to studying and quantifying common social biases as reflected in the generated images, by focusing on how occupations, personality traits, and everyday situations are depicted across representations of (perceived) gender, age, race, and geographical location. Through an extensive set of both automated and human evaluation experiments we present findings for two popular T2I models: DALLE-v2 and Stable Diffusion. Our results reveal that there exist severe occupational biases of neutral prompts majorly excluding groups of people from results for both models. Such biases can get mitigated by increasing the amount of specification in the prompt itself, although the prompting mitigation will not address discrepancies in image quality or other usages of the model or its representations in other scenarios. Further, we observe personality traits being associated with only a limited set of people at the intersection of race, gender, and age. Finally, an analysis of geographical location representations on everyday situations (e.g., park, food, weddings) shows that for most situations, images generated through default location-neutral prompts are closer and more similar to images generated for locations of United States and Germany.

cs.CY↗

arXiv2022

Advancing Human-AI Complementarity: The Impact of User Expertise and Algorithmic Tuning on Joint Decision Making

Human-AI collaboration for decision-making strives to achieve team performance that exceeds the performance of humans or AI alone. However, many factors can impact success of Human-AI teams, including a user's domain expertise, mental models of an AI system, trust in recommendations, and more. This work examines users' interaction with three simulated algorithmic models, all with similar accuracy but different tuning on their true positive and true negative rates. Our study examined user performance in a non-trivial blood vessel labeling task where participants indicated whether a given blood vessel was flowing or stalled. Our results show that while recommendations from an AI-Assistant can aid user decision making, factors such as users' baseline performance relative to the AI and complementary tuning of AI error types significantly impact overall team performance. Novice users improved, but not to the accuracy level of the AI. Highly proficient users were generally able to discern when they should follow the AI recommendation and typically maintained or improved their performance. Mid-performers, who had a similar level of accuracy to the AI, were most variable in terms of whether the AI recommendations helped or hurt their performance. In addition, we found that users' perception of the AI's performance relative on their own also had a significant impact on whether their accuracy improved when given AI recommendations. This work provides insights on the complexity of factors related to Human-AI collaboration and provides recommendations on how to develop human-centered AI algorithms to complement users in decision-making tasks.

cs.HC↗

arXiv2022

Who Goes First? Influences of Human-AI Workflow on Decision Making in Clinical Imaging

Details of the designs and mechanisms in support of human-AI collaboration must be considered in the real-world fielding of AI technologies. A critical aspect of interaction design for AI-assisted human decision making are policies about the display and sequencing of AI inferences within larger decision-making workflows. We have a poor understanding of the influences of making AI inferences available before versus after human review of a diagnostic task at hand. We explore the effects of providing AI assistance at the start of a diagnostic session in radiology versus after the radiologist has made a provisional decision. We conducted a user study where 19 veterinary radiologists identified radiographic findings present in patients' X-ray images, with the aid of an AI tool. We employed two workflow configurations to analyze (i) anchoring effects, (ii) human-AI team diagnostic performance and agreement, (iii) time spent and confidence in decision making, and (iv) perceived usefulness of the AI. We found that participants who are asked to register provisional responses in advance of reviewing AI inferences are less likely to agree with the AI regardless of whether the advice is accurate and, in instances of disagreement with the AI, are less likely to seek the second opinion of a colleague. These participants also reported the AI advice to be less useful. Surprisingly, requiring provisional decisions on cases in advance of the display of AI inferences did not lengthen the time participants spent on the task. The study provides generalizable and actionable insights for the deployment of clinical AI tools in human-in-the-loop systems and introduces a methodology for studying alternative designs for human-AI collaboration. We make our experimental platform available as open source to facilitate future research on the influence of alternate designs on human-AI workflows.

cs.HC↗

arXiv2022

Investigations of Performance and Bias in Human-AI Teamwork in Hiring

In AI-assisted decision-making, effective hybrid (human-AI) teamwork is not solely dependent on AI performance alone, but also on its impact on human decision-making. While prior work studies the effects of model accuracy on humans, we endeavour here to investigate the complex dynamics of how both a model's predictive performance and bias may transfer to humans in a recommendation-aided decision task. We consider the domain of ML-assisted hiring, where humans -- operating in a constrained selection setting -- can choose whether they wish to utilize a trained model's inferences to help select candidates from written biographies. We conduct a large-scale user study leveraging a re-created dataset of real bios from prior work, where humans predict the ground truth occupation of given candidates with and without the help of three different NLP classifiers (random, bag-of-words, and deep neural network). Our results demonstrate that while high-performance models significantly improve human performance in a hybrid setting, some models mitigate hybrid bias while others accentuate it. We examine these findings through the lens of decision conformity and observe that our model architecture choices have an impact on human-AI conformity and bias, motivating the explicit need to assess these complex dynamics prior to deployment.

cs.HC↗

arXiv2021

Hierarchical Analysis of Visual COVID-19 Features from Chest Radiographs

Chest radiography has been a recommended procedure for patient triaging and resource management in intensive care units (ICUs) throughout the COVID-19 pandemic. The machine learning efforts to augment this workflow have been long challenged due to deficiencies in reporting, model evaluation, and failure mode analysis. To address some of those shortcomings, we model radiological features with a human-interpretable class hierarchy that aligns with the radiological decision process. Also, we propose the use of a data-driven error analysis methodology to uncover the blind spots of our model, providing further transparency on its clinical utility. For example, our experiments show that model failures highly correlate with ICU imaging conditions and with the inherent difficulty in distinguishing certain types of radiological features. Also, our hierarchical interpretation and analysis facilitates the comparison with respect to radiologists' findings and inter-variability, which in return helps us to better assess the clinical applicability of models.

eess.IV↗

arXiv2021

Understanding Failures of Deep Networks via Robust Feature Extraction

Traditional evaluation metrics for learned models that report aggregate scores over a test set are insufficient for surfacing important and informative patterns of failure over features and instances. We introduce and study a method aimed at characterizing and explaining failures by identifying visual attributes whose presence or absence results in poor performance. In distinction to previous work that relies upon crowdsourced labels for visual attributes, we leverage the representation of a separate robust model to extract interpretable features and then harness these features to identify failure modes. We further propose a visualization method aimed at enabling humans to understand the meaning encoded in such features and we test the comprehensibility of the features. An evaluation of the methods on the ImageNet dataset demonstrates that: (i) the proposed workflow is effective for discovering important failure modes, (ii) the visualization techniques help humans to understand the extracted features, and (iii) the extracted insights can assist engineers with error analysis and debugging.

cs.CV↗

arXiv2021

Is the Most Accurate AI the Best Teammate? Optimizing AI for Teamwork

AI practitioners typically strive to develop the most accurate systems, making an implicit assumption that the AI system will function autonomously. However, in practice, AI systems often are used to provide advice to people in domains ranging from criminal justice and finance to healthcare. In such AI-advised decision making, humans and machines form a team, where the human is responsible for making final decisions. But is the most accurate AI the best teammate? We argue "No" -- predictable performance may be worth a slight sacrifice in AI accuracy. Instead, we argue that AI systems should be trained in a human-centered manner, directly optimized for team performance. We study this proposal for a specific type of human-AI teaming, where the human overseer chooses to either accept the AI recommendation or solve the task themselves. To optimize the team performance for this setting we maximize the team's expected utility, expressed in terms of the quality of the final decision, cost of verifying, and individual accuracies of people and machines. Our experiments with linear and non-linear models on real-world, high-stakes datasets show that the most accuracy AI may not lead to highest team performance and show the benefit of modeling teamwork during training through improvements in expected team utility across datasets, considering parameters such as human skill and the cost of mistakes. We discuss the shortcoming of current optimization approaches beyond well-studied loss functions such as log-loss, and encourage future work on AI optimization problems motivated by human-AI collaboration.

cs.AI↗

arXiv2021

Does the Whole Exceed its Parts? The Effect of AI Explanations on Complementary Team Performance

Many researchers motivate explainable AI with studies showing that human-AI team performance on decision-making tasks improves when the AI explains its recommendations. However, prior studies observed improvements from explanations only when the AI, alone, outperformed both the human and the best team. Can explanations help lead to complementary performance, where team accuracy is higher than either the human or the AI working solo? We conduct mixed-method user studies on three datasets, where an AI with accuracy comparable to humans helps participants solve a task (explaining itself in some conditions). While we observed complementary improvements from AI augmentation, they were not increased by explanations. Rather, explanations increased the chance that humans will accept the AI's recommendation, regardless of its correctness. Our result poses new challenges for human-centered AI: Can we develop explanatory approaches that encourage appropriate trust in AI, and therefore help generate (or improve) complementary performance?

cs.AI↗

arXiv2020

An Empirical Analysis of Backward Compatibility in Machine Learning Systems

In many applications of machine learning (ML), updates are performed with the goal of enhancing model performance. However, current practices for updating models rely solely on isolated, aggregate performance analyses, overlooking important dependencies, expectations, and needs in real-world deployments. We consider how updates, intended to improve ML models, can introduce new errors that can significantly affect downstream systems and users. For example, updates in models used in cloud-based classification services, such as image recognition, can cause unexpected erroneous behavior in systems that make calls to the services. Prior work has shown the importance of "backward compatibility" for maintaining human trust. We study challenges with backward compatibility across different ML architectures and datasets, focusing on common settings including data shifts with structured noise and ML employed in inferential pipelines. Our results show that (i) compatibility issues arise even without data shift due to optimization stochasticity, (ii) training on large-scale noisy datasets often results in significant decreases in backward compatibility even when model accuracy increases, and (iii) distributions of incompatible points align with noise bias, motivating the need for compatibility aware de-noising and robustness methods.

cs.LG↗

arXiv2020

SQuINTing at VQA Models: Introspecting VQA Models with Sub-Questions

Existing VQA datasets contain questions with varying levels of complexity. While the majority of questions in these datasets require perception for recognizing existence, properties, and spatial relationships of entities, a significant portion of questions pose challenges that correspond to reasoning tasks - tasks that can only be answered through a synthesis of perception and knowledge about the world, logic and / or reasoning. Analyzing performance across this distinction allows us to notice when existing VQA models have consistency issues; they answer the reasoning questions correctly but fail on associated low-level perception questions. For example, in Figure 1, models answer the complex reasoning question "Is the banana ripe enough to eat?" correctly, but fail on the associated perception question "Are the bananas mostly green or yellow?" indicating that the model likely answered the reasoning question correctly but for the wrong reason. We quantify the extent to which this phenomenon occurs by creating a new Reasoning split of the VQA dataset and collecting VQA-introspect, a new dataset1 which consists of 238K new perception questions which serve as sub questions corresponding to the set of perceptual tasks needed to effectively answer the complex reasoning questions in the Reasoning split. Our evaluation shows that state-of-the-art VQA models have comparable performance in answering perception and reasoning questions, but suffer from consistency problems. To address this shortcoming, we propose an approach called Sub-Question Importance-aware Network Tuning (SQuINT), which encourages the model to attend to the same parts of the image when answering the reasoning question and the perception sub question. We show that SQuINT improves model consistency by ~5%, also marginally improving performance on the Reasoning questions in VQA, while also displaying better attention maps.

cs.CV↗