No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
ReX: A Framework for Incorporating Temporal Information in Model-Agnostic Local Explanation Techniques
Abstract
Existing local model-agnostic explanation techniques are ineffective for machine learning models that consider inputs of variable lengths, as they do not consider temporal information embedded in these models. To address this limitation, we propose ReX, a general framework for incorporating temporal information in these techniques. Our key insight is that these techniques typically learn a model surrogate by sampling model inputs and outputs, and we can incorporate temporal information in a uniform way by only changing the sampling process and the surrogate features. We instantiate our approach on three popular explanation techniques: Anchors, LIME, and Kernel SHAP. To evaluate the effectiveness of ReX, we apply our approach to six models in three different tasks. Our evaluation results demonstrate that our approach 1) significantly improves the fidelity of explanations, making model-agnostic techniques outperform a state-of-the-art model-specific technique on its target model, and 2) helps end users better understand the models' behaviors.
Large Language Models (LLMs) have grown in popularity in recent years and are now employed in a variety of software engineering domains thanks to their Natural Language Processing (NLP) capabilities, which include source code generation, understanding, and documentation. Selecting the appropriate model for source code generation presents a problem to developers as more and more powerful LLMs become available. While some studies have evaluated Copilot or ChatGPT, there is a lack of research on how developers can choose from available LLMs, which is a key factor in the growing set of available models and services. It is crucial to know if a model is capable of generating useful source code that meets the quality requirements and if the developers will be able to use the generated code. Regarding these factors, one has to decide which model to utilize during everyday tasks. This paper shows a methodology to compare such models by demonstrating an actual comparison of two models. Subsequently, we investigated the functional and non-functional qualities of the code synthesized by the models on a program synthesis benchmark containing 25 tasks. On average, the functional testing shows that ChatGPT generated 17 perfect solutions, while Copilot could only solve 13. The non-functional analysis reflected that both models generated good quality code, however, both have characteristic code smells. Our evaluation shows that ChatGPT performs better using this methodology, which is supported by human reviewers who evaluated the generated code by hand.
We present the Natural Questions corpus, a question answering data set. Questions consist of real anonymized, aggregated queries issued to the Google search engine. An annotator is presented with a question along with a Wikipedia page from the top 5 search results, and annotates a long answer (typically a paragraph) and a short answer (one or more entities) if present on the page, or marks null if no long/short answer is present. The public release consists of 307,373 training examples with single annotations; 7,830 examples with 5-way annotations for development data; and a further 7,842 examples with 5-way annotated sequestered as test data. We present experiments validating quality of the data. We also describe analysis of 25-way annotations on 302 examples, giving insights into human variability on the annotation task. We introduce robust metrics for the purposes of evaluating question answering systems; demonstrate high human upper bounds on these
metrics; and establish baseline results using competitive methods drawn from related literature.
This paper is a contribution towards interpretability of the deep learning models in different applications of time-series. We propose a temporal attention layer that is capable of selecting the relevant information to perform various tasks, including data completion, key-frame detection and classification. The method uses the whole input sequence to calculate an attention value for each time step. This results in more focused attention values and more plausible visualisation than previous methods. We apply the proposed method to three different tasks. Experimental results show that the proposed network produces comparable results to a state of the art. In addition, the network provides better interpretability of the decision, that is, it generates more significant attention weight to related frames compared to similar techniques attempted in the past.
The ubiquity of systems using artificial intelligence or "AI" has brought increasing attention to how those systems should be regulated. The choice of how to regulate AI systems will require care. AI systems have the potential to synthesize large amounts of data, allowing for greater levels of personalization and precision than ever before---applications range from clinical decision support to autonomous driving and predictive policing. That said, there exist legitimate concerns about the intentional and unintentional negative consequences of AI systems. There are many ways to hold AI systems accountable. In this work, we focus on one: explanation. Questions about a legal right to explanation from AI systems was recently debated in the EU General Data Protection Regulation, and thus thinking carefully about when and how explanation from AI systems might improve accountability is timely. In this work, we review contexts in which explanation is currently required under the law, and then list the technical considerations that must be considered if we desired AI systems that could provide kinds of explanations that are currently required of humans.
There has been much discussion of the right to explanation in the EU General Data Protection Regulation, and its existence, merits, and disadvantages. Implementing a right to explanation that opens the black box of algorithmic decision-making faces major legal and technical barriers. Explaining the functionality of complex algorithmic decision-making systems and their rationale in specific cases is a technically challenging problem. Some explanations may offer little meaningful information to data subjects, raising questions around their value. Explanations of automated decisions need not hinge on the general public understanding how algorithmic systems function. Even though such interpretability is of great importance and should be pursued, explanations can, in principle, be offered without opening the black box. Looking at explanations as a means to help a data subject act rather than merely understand, one could gauge the scope and content of explanations according to the specific goal or action they are intended to support. From the perspective of individuals affected by automated decision-making, we propose three aims for explanations: (1) to inform and help the individual understand why a particular decision was reached, (2) to provide grounds to contest the decision if the outcome is undesired, and (3) to understand what would need to change in order to receive a desired result in the future, based on the current decision-making model. We assess how each of these goals finds support in the GDPR. We suggest data controllers should offer a particular type of explanation, unconditional counterfactual explanations, to support these three aims. These counterfactual explanations describe the smallest change to the world that can be made to obtain a desirable outcome, or to arrive at the closest possible world, without needing to explain the internal logic of the system.
Rule extraction from black box models is critical in domains that require model validation before implementation, as can be the case in credit scoring and medical diagnosis. Though already a challenging problem in statistical learning in general, the difficulty is even greater when highly nonlinear, recursive models, such as recurrent neural networks (RNNs), are fit to data. Here, we study the extraction of rules from second-order RNNs trained to recognize the Tomita grammars. We show that production rules can be stably extracted from trained RNNs and that in certain cases, the rules outperform the trained RNNs.
We present a hierarchical convolutional document model with an architecture
designed to support introspection of the document structure. Using this model,
we show how to use visualisation techniques from the computer vision literature
to identify and extract topic-relevant sentences.
We also introduce a new scalable evaluation technique for automatic sentence
extraction systems that avoids the need for time consuming human annotation of
validation data.
Rule extraction (RE) from recurrent neural networks (RNNs) refers to finding models of the underlying RNN, typically in the form of finite state machines, that mimic the network to a satisfactory degree while having the advantage of being more transparent. RE from RNNs can be argued to allow a deeper and more profound form of analysis of RNNs than other, more or less ad hoc methods. RE may give us understanding of RNNs in the intermediate levels between quite abstract theoretical knowledge of RNNs as a class of computing devices and quantitative performance evaluations of RNN instantiations. The development of techniques for extraction of rules from RNNs has been an active field since the early 1990s. This article reviews the progress of this development and analyzes it in detail. In order to structure the survey and evaluate the techniques, a taxonomy specifically designed for this purpose has been developed. Moreover, important open research issues are identified that, if addressed properly, possibly can give the field a significant push forward.
This article presents Individual Conditional Expectation (ICE) plots, a tool
for visualizing the model estimated by any supervised learning algorithm.
Classical partial dependence plots (PDPs) help visualize the average partial
relationship between the predicted response and one or more features. In the
presence of substantial interaction effects, the partial response relationship
can be heterogeneous and an average curve, such as the PDP, can obfuscate the
complexity of the modeled relationship. Accordingly, ICE plots refine the
partial dependence plot by graphing the functional relationship between the
predicted response and the feature for individual observations. Specifically,
ICE plots highlight the variation in the fitted values across the range of a
covariate, suggesting where and to what extent they may exist. In addition to
providing a plotting suite for exploratory analysis, we include a visual test
for additive structure in the data generating model. Through simulated examples
and real data sets, we demonstrate how ICE plots can shed light on estimated
models in ways PDPs cannot. Procedures outlined are available in the R package
ICEbox.
Unpacking and comprehending how black-box machine learning algorithms (such as deep learning models) make decisions has been a persistent challenge for researchers and end-users. Explaining time-series predictive models is useful for clinical applications with high stakes to understand the behavior of prediction models, e.g., to determine how different variables and time points influence the clinical outcome. However, existing approaches to explain such models are frequently unique to architectures and data where the features do not have a time-varying component. In this paper, we introduce WindowSHAP, a model-agnostic framework for explaining time-series classifiers using Shapley values. We intend for WindowSHAP to mitigate the computational complexity of calculating Shapley values for long time-series data as well as improve the quality of explanations. WindowSHAP is based on partitioning a sequence into time windows. Under this framework, we present three distinct algorithms of Stationary, Sliding and Dynamic WindowSHAP, each evaluated against baseline approaches, KernelSHAP and TimeSHAP, using perturbation and sequence analyses metrics. We applied our framework to clinical time-series data from both a specialized clinical domain (Traumatic Brain Injury - TBI) as well as a broad clinical domain (critical care medicine). The experimental results demonstrate that, based on the two quantitative metrics, our framework is superior at explaining clinical time-series classifiers, while also reducing the complexity of computations. We show that for time-series data with 120 time steps (hours), merging 10 adjacent time points can reduce the CPU time of WindowSHAP by 80% compared to KernelSHAP. We also show that our Dynamic WindowSHAP algorithm focuses more on the most important time steps and provides more understandable explanations. As a result, WindowSHAP not only accelerates the calculation of Shapley values for time-series data, but also delivers more understandable explanations with higher quality.
As our dependence on intelligent machines continues to grow, so does the demand for more transparent and interpretable models. In addition, the ability to explain the model generally is now the gold standard for building trust and deployment of Artificial Intelligence (AI) systems in critical domains. Explainable Artificial Intelligence (XAI) aims to provide a suite of machine learning (ML) techniques that enable human users to understand, appropriately trust, and produce more explainable models. Selecting an appropriate approach for building an XAI-enabled application requires a clear understanding of the core ideas within XAI and the associated programming frameworks. We survey state-of-the-art programming techniques for XAI and present the different phases of XAI in a typical ML development process. We classify the various XAI approaches and using this taxonomy, discuss the key differences among the existing XAI techniques. Furthermore, concrete examples are used to describe these techniques that are mapped to programming frameworks and software toolkits. It is the intention that this survey will help stakeholders in selecting the appropriate approaches, programming frameworks, and software toolkits by comparing them through the lens of the presented taxonomy.
Along with the great success of deep neural networks, there is also growing concern about their black-box nature. The interpretability issue affects people’s trust on deep learning systems. It is also related to many ethical problems, e.g., algorithmic discrimination. Moreover, interpretability is a desired property for deep networks to become powerful tools in other research fields, e.g., drug discovery and genomics. In this survey, we conduct a comprehensive review of the neural network interpretability research. We first clarify the definition of interpretability as it has been used in many different contexts. Then we elaborate on the importance of interpretability and propose a novel taxonomy organized along three dimensions: type of engagement (passive vs. active interpretation approaches), the type of explanation, and the focus (from local to global interpretability). This taxonomy provides a meaningful 3D view of distribution of papers from the relevant literature as two of the dimensions are not simply categorical but allow ordinal subcategories. Finally, we summarize the existing interpretability evaluation methods and suggest possible research directions inspired by our new taxonomy.
Big data, high-performance computing, and (deep) machine learning are increasingly becoming key to precision medicine—from identifying disease risks and taking preventive measures, to making diagnoses and personalizing treatment for individuals. Precision medicine, however, is not only about predicting risks and outcomes, but also about weighing interventions. Interventional clinical predictive models require the correct specification of cause and effect, and the calculation of so-called counterfactuals, that is, alternative scenarios. In biomedical research, observational studies are commonly affected by confounding and selection bias. Without robust assumptions, often requiring a priori domain knowledge, causal inference is not feasible. Data-driven prediction models are often mistakenly used to draw causal effects, but neither their parameters nor their predictions necessarily have a causal interpretation. Therefore, the premise that data-driven prediction models lead to trustable decisions/interventions for precision medicine is questionable. When pursuing intervention modelling, the bio-health informatics community needs to employ causal approaches and learn causal structures. Here we discuss how target trials (algorithmic emulation of randomized studies), transportability (the licence to transfer causal effects from one population to another) and prediction invariance (where a true causal model is contained in the set of all prediction models whose accuracy does not vary across different settings) are linchpins to developing and testing intervention models. Machine learning models are commonly used to predict risks and outcomes in biomedical research. But healthcare often requires information about cause–effect relations and alternative scenarios, that is, counterfactuals. Prosperi et al. discuss the importance of interventional and counterfactual models, as opposed to purely predictive models, in the context of precision medicine.
Deep Learning (DL) has achieved tremendous success in many cutting-edge applications. However, the state-of-the-art DL systems still suffer from quality issues. While some recent progress has been made on the analysis of feed-forward DL systems, little study has been done on the Recurrent Neural Network (RNN)-based stateful DL systems, which are widely used in audio, natural languages and video processing, etc. In this paper, we initiate the very first step towards the quantitative analysis of RNN-based DL systems. We model RNN as an abstract state transition system to characterize its internal behaviors. Based on the abstract model, we design two trace similarity metrics and five coverage criteria which enable the quantitative analysis of RNNs. We further propose two algorithms powered by the quantitative measures for adversarial sample detection and coverage-guided test generation. We evaluate DeepStellar on four RNN-based systems covering image classification and automated speech recognition. The results demonstrate that the abstract model is useful in capturing the internal behaviors of RNNs, and confirm that (1) the similarity metrics could effectively capture the differences between samples even with very small perturbations (achieving 97% accuracy for detecting adversarial samples) and (2) the coverage criteria are useful in revealing erroneous behaviors (generating three times more adversarial samples than random testing and hundreds times more than the unrolling approach).
Recently, a technique called Layer-wise Relevance Propagation (LRP) was shown to deliver insightful explanations in the form of input space relevances for understanding feed-forward neural network classification decisions. In the present work, we extend the usage of LRP to recurrent neural networks. We propose a specific propagation rule applicable to multiplicative connections as they arise in recurrent network architectures such as LSTMs and GRUs. We apply our technique to a word-based bi-directional LSTM model on a five-class sentiment prediction task, and evaluate the resulting LRP relevances both qualitatively and quantitatively, obtaining better results than a gradient-based related method which was used in previous work.
The dominant sequence transduction models are based on complex recurrent or convolutional neural networks in an encoder-decoder configuration. The best performing models also connect the encoder and decoder through an attention mechanism. We propose a new simple network architecture, the Transformer, based solely on attention mechanisms, dispensing with recurrence and convolutions entirely. Experiments on two machine translation tasks show these models to be superior in quality while being more parallelizable and requiring significantly less time to train. Our model achieves 28.4 BLEU on the WMT 2014 English-to-German translation task, improving over the existing best results, including ensembles by over 2 BLEU. On the WMT 2014 English-to-French translation task, our model establishes a new single-model state-of-the-art BLEU score of 41.0 after training for 3.5 days on eight GPUs, a small fraction of the training costs of the best models from the literature. We show that the Transformer generalizes well to other tasks by applying it successfully to English constituency parsing both with large and limited training data.
One of the most important obstacles to deploying predictive models is the fact that humans do not understand and trust them. Knowing which variables are important in a model's prediction and how they are combined can be very powerful in helping people understand and trust automatic decision making systems.
Here we propose interpretable decision sets, a framework for building predictive models that are highly accurate, yet also highly interpretable. Decision sets are sets of independent if-then rules. Because each rule can be applied independently, decision sets are simple, concise, and easily interpretable. We formalize decision set learning through an objective function that simultaneously optimizes accuracy and interpretability of the rules. In particular, our approach learns short, accurate, and non-overlapping rules that cover the whole feature space and pay attention to small but important classes. Moreover, we prove that our objective is a non-monotone submodular function, which we efficiently optimize to find a near-optimal set of rules.
Experiments show that interpretable decision sets are as accurate at classification as state-of-the-art machine learning techniques. They are also three times smaller on average than rule-based models learned by other methods. Finally, results of a user study show that people are able to answer multiple-choice questions about the decision boundaries of interpretable decision sets and write descriptions of classes based on them faster and more accurately than with other rule-based models that were designed for interpretability. Overall, our framework provides a new approach to interpretable machine learning that balances accuracy, interpretability, and computational efficiency.
We present a sensitivity analysis-based method for explaining prediction models that can be applied to any type of classification or regression model. Its advantage over existing general methods is that all subsets of input features are perturbed, so interactions and redundancies between features are taken into account. Furthermore, when explaining an additive model, the method is equivalent to commonly used additive model-specific methods. We illustrate the method's usefulness with examples from artificial and real-world data sets and an empirical analysis of running times. Results from a controlled experiment with 122 participants suggest that the method's explanations improved the participants' understanding of the model.
The extraction of symbolic knowledge from trained neural networks and the direct encoding of (partial) knowledge into networks prior to training are important issues. They allow the exchange of information between symbolic and connectionist knowledge representations. The focus of this paper is on the quality of the rules that are extracted from recurrent neural networks. Discrete-time recurrent neural networks can be trained to correctly classify strings of a regular language. Rules defining the learned grammar can be extracted from networks in the form of deterministic finite-state automata (DFAs) by applying clustering algorithms in the output space of recurrent state neurons. Our algorithm can extract different finite-state automata that are consistent with a training set from the same network. We compare the generalization performances of these different models and the trained network and we introduce a heuristic that permits us to choose among the consistent DFAs the model which best approximates the learned regular grammar.
A significant limitation of neural networks is that the representations they learn are usually incomprehensible to humans. We present a novel algorithm, Trepan, for extracting comprehensible, symbolic representations from trained neural networks. Our algorithm uses queries to induce a decision tree that approximates the concept represented by a given network. Our experiments demonstrate that Trepan is able to produce decision trees that maintain a high level of fidelity to their respective networks while being comprehensible and accurate. Unlike previous work in this area, our algorithm is general in its applicability and scales well to large networks and problems with high-dimensional input spaces.
- J Achiam
Achiam, J.; et al. 2023. Gpt-4 technical report. arXiv preprint
arXiv:2303.08774.
Evaluating Recurrent Neural Network Explanations
- L Arras
- A Osman
- K Müller
- W Samek
- T Linzen
- G Chrupala
- Y Belinkov
- D Hupkes
Arras, L.; Osman, A.; Müller, K.; and Samek, W. 2019. Evaluating
Recurrent Neural Network Explanations. In Linzen, T.; Chrupala,
G.; Belinkov, Y.; and Hupkes, D., eds., Proceedings of the 2019
ACL Workshop BlackboxNLP: Analyzing and Interpreting Neural
Networks for NLP, BlackboxNLP@ACL 2019, Florence, Italy, August 1, 2019, 113-126. Association for Computational Linguistics.
Explaining Black Boxes on Sequential Data using Weighted Automata
- S Ayache
- R Eyraud
- N Goudian
Ayache, S.; Eyraud, R.; and Goudian, N. 2018. Explaining Black
Boxes on Sequential Data using Weighted Automata. In Unold,
O.; Dyrka, W.; and Wieczorek, W., eds., Proceedings of the 14th
International Conference on Grammatical Inference, ICGI 2018,
Wrocław, Poland, September 5-7, 2018, volume 93 of Proceedings
of Machine Learning Research, 81-103. PMLR.
Multi-Objective Counterfactual Explanations
- S Dandl
- C Molnar
- M Binder
- B Bischl
- T Bäck
- M Preuss
- A H Deutz
- H Wang
- C Doerr
- M T Emmerich
Dandl, S.; Molnar, C.; Binder, M.; and Bischl, B. 2020. Multi-Objective Counterfactual Explanations. In Bäck, T.; Preuss, M.;
Deutz, A. H.; Wang, H.; Doerr, C.; Emmerich, M. T. M.; and Trautmann, H., eds., Parallel Problem Solving from Nature -PPSN XVI
-16th International Conference, PPSN 2020, Leiden, The Netherlands, September 5-9, 2020, Proceedings, Part I, volume 12269 of
Lecture Notes in Computer Science, 448-469. Springer.
- J Devlin
- M.-W Chang
- K Lee
- K Toutanova
Devlin, J.; Chang, M.-W.; Lee, K.; and Toutanova, K. 2018. Bert:
Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805.
Local Rule-Based Explanations of Black Box Decision Systems
- R Guidotti
- A Monreale
- S Ruggieri
- D Pedreschi
- F Turini
- F Giannotti
Guidotti, R.; Monreale, A.; Ruggieri, S.; Pedreschi, D.; Turini, F.;
and Giannotti, F. 2018. Local Rule-Based Explanations of Black
Box Decision Systems. CoRR, abs/1805.10820.
Improving deep learning interpretability by saliency guided training
- A A Ismail
- H Bravo
- S Feizi
Ismail, A. A.; Corrada Bravo, H.; and Feizi, S. 2021. Improving deep learning interpretability by saliency guided training. Advances in Neural Information Processing Systems, 34: 26726-26739.
Information complexity in bandit subset selection
- E Kaufmann
- S Kalyanakrishnan
Kaufmann, E.; and Kalyanakrishnan, S. 2013. Information complexity in bandit subset selection. In Conference on Learning Theory, 228-251. PMLR.
BERT meets Shapley: Extending SHAP Explanations to Transformer-based Classifiers
- E Kokalj
- B Škrlj
- N Lavrač
- S Pollak
- M Robnik-Šikonja
Kokalj, E.;Škrlj, B.; Lavrač, N.; Pollak, S.; and Robnik-Šikonja,
M. 2021. BERT meets Shapley: Extending SHAP Explanations to
Transformer-based Classifiers. In Toivonen, H.; and Boggia, M.,
eds., Proceedings of the EACL Hackashop on News Media Content
Analysis and Automated Report Generation, 16-21. Online: Association for Computational Linguistics.
A Unified Approach to Interpreting Model Predictions
- S M Lundberg
- S Lee
- I Guyon
- U Von Luxburg
- S Bengio
- H M Wallach
- R Fergus
- S V N Vishwanathan
- R Garnett
Lundberg, S. M.; and Lee, S. 2017. A Unified Approach to Interpreting Model Predictions. In Guyon, I.; von Luxburg, U.; Bengio,
S.; Wallach, H. M.; Fergus, R.; Vishwanathan, S. V. N.; and Garnett, R., eds., Advances in Neural Information Processing Systems
30: Annual Conference on Neural Information Processing Systems
2017, December 4-9, 2017, Long Beach, CA, USA, 4765-4774.
Beyond Word Importance: Contextual Decomposition to Extract Interactions from LSTMs
- W J Murdoch
- P J Liu
- B Yu
Murdoch, W. J.; Liu, P. J.; and Yu, B. 2018. Beyond Word Importance: Contextual Decomposition to Extract Interactions from
LSTMs. In 6th International Conference on Learning Representations, ICLR 2018, Vancouver, BC, Canada, April 30 -May 3, 2018,
Conference Track Proceedings. OpenReview.net.
Multi-Level Explanations for Generative Language Models
- L M Paes
Paes, L. M.; et al. 2024. Multi-Level Explanations for Generative
Language Models. arXiv preprint arXiv:2403.14459.
RNN based Time-series Anomaly Detector Model Implemented in Pytorch
- J Park
Park, J. 2018. RNN based Time-series Anomaly Detector Model
Implemented in Pytorch. https://github.com/chickenbestlover/
RNN-Time-series-Anomaly-Detection. Accessed: 2022-06-01.
Language models are unsupervised multitask learners
- A Radford
Radford, A.; et al. 2019. Language models are unsupervised multitask learners. OpenAI blog, 1(8): 9.
Exploring the limits of transfer learning with a unified text-to-text transformer
- C Raffel
Raffel, C.; et al. 2020. Exploring the limits of transfer learning
with a unified text-to-text transformer. The Journal of Machine
Learning Research, 21(1): 5485-5551.
Anchors: High-Precision Model-Agnostic Explanations
- M T Ribeiro
- S Singh
- C Guestrin
Ribeiro, M. T.; Singh, S.; and Guestrin, C. 2018. Anchors: High-Precision Model-Agnostic Explanations. In McIlraith, S. A.; and
Weinberger, K. Q., eds., Proceedings of the Thirty-Second AAAI
Conference on Artificial Intelligence, (AAAI-18), the 30th innovative Applications of Artificial Intelligence (IAAI-18), and the 8th
AAAI Symposium on Educational Advances in Artificial Intelligence (EAAI-18), New Orleans, Louisiana, USA, February 2-7,
2018, 1527-1535. AAAI Press.
Hierarchical interpretations for neural network predictions
- C Singh
- W J Murdoch
- B Yu
Singh, C.; Murdoch, W. J.; and Yu, B. 2018. Hierarchical interpretations for neural network predictions. In International Conference
on Learning Representations.
LIMESegment: Meaningful, Realistic Time Series Explanations
- T Sivill
- P Flach
Sivill, T.; and Flach, P. 2022. LIMESegment: Meaningful, Realistic
Time Series Explanations. In International Conference on Artificial Intelligence and Statistics, 3418-3433. PMLR.
- Z Tan
- Y Tian
- J Li
Tan, Z.; Tian, Y.; and Li, J. 2023. GLIME: General, Stable and
Local LIME Explanation. arXiv:2311.15722.
Llama 2: Open Foundation and Fine-Tuned Chat Models
- H Touvron
Touvron, H.; et al. 2023. Llama 2: Open Foundation and Fine-Tuned Chat Models. arXiv:2307.09288.