Karthikeyan K

Karthikeyan K, Shaily Bhatt, Pankaj Singh et al. · cmu

Multilingual evaluation benchmarks usually contain limited high-resource languages and do not test models for specific linguistic capabilities. CheckList is a template-based evaluation approach that tests models for specific capabilities. The CheckList template creation process requires native speakers, posing a challenge in scaling to hundreds of languages. In this work, we explore multiple approaches to generate Multilingual CheckLists. We device an algorithm - Template Extraction Algorithm (TEA) for automatically extracting target language CheckList templates from machine translated instances of a source language templates. We compare the TEA CheckLists with CheckLists created with different levels of human intervention. We further introduce metrics along the dimensions of cost, diversity, utility, and correctness to compare the CheckLists. We thoroughly analyze different approaches to creating CheckLists in Hindi. Furthermore, we experiment with 9 more different languages. We find that TEA followed by human verification is ideal for scaling Checklist-based evaluation to multiple languages while TEA gives a good estimates of model performance.

Karthikeyan K, Raghuveer Thirukovalluru, Bhuwan Dhingra et al.

Large language model agents for data analysis typically generate and execute code directly on databases. However, when applied to sensitive data, this approach poses significant security risks. To address this issue, we propose a security-motivated alternative: restrict LLMs from direct code generation and data access, and require them to interact with data exclusively through a predefined set of secure, verified tools. Although recent tool-use benchmarks exist, they primarily target tool selection and simple execution rather than the compositional, multi-step reasoning needed for complex data analysis. To reduce this gap, we introduce Indirect Data Engagement (InData), a dataset designed to assess LLMs' multi-step tool-based reasoning ability. InData includes data analysis questions at three difficulty levels--Easy, Medium, and Hard--capturing increasing reasoning complexity. We benchmark 15 open-source LLMs on InData and find that while large models (e.g., gpt-oss-120b) achieve high accuracy on Easy tasks (97.3%), performance drops sharply on Hard tasks (69.6%). These results show that current LLMs still lack robust multi-step tool-based reasoning ability. With InData, we take a step toward enabling the development and evaluation of LLMs with stronger multi-step tool-use capabilities. We will publicly release the dataset and code.

Karthikeyan K, Raghuveer Thirukovalluru, David Carlson

Clinical notes contain valuable, context-rich information, but their unstructured format introduces several challenges, including unintended biases (e.g., gender or racial bias), and poor generalization across clinical settings (e.g., models trained on one EHR system may perform poorly on another due to format differences) and poor interpretability. To address these issues, we present ClinStructor, a pipeline that leverages large language models (LLMs) to convert clinical free-text into structured, task-specific question-answer pairs prior to predictive modeling. Our method substantially enhances transparency and controllability and only leads to a modest reduction in predictive performance (a 2-3% drop in AUC), compared to direct fine-tuning, on the ICU mortality prediction task. ClinStructor lays a strong foundation for building reliable, interpretable, and generalizable machine learning models in clinical environments.

Karthikeyan K, Raghuveer Thirukovalluru, David Carlson

Large Language Models have advanced clinical text classification, but their opaque predictions remain a critical barrier to practical adoption in research and clinical settings where investigators and physicians need to understand which parts of a patient's record drive risk signals. To address this challenge, we introduce \textbf{CALM}, short for \textbf{Classification with Additive Large Language Models}, an interpretable framework for semi-structured text where inputs are composed of semantically meaningful components, such as sections of an admission note or question-answer fields from an intake form. CALM predicts outcomes as the additive sum of each component's contribution, making these contributions part of the forward computation itself and enabling faithful explanations at both the patient and population level. The additive structure also enables clear visualizations, such as component-level risk curves similar to those used in generalized additive models, making the learned relationships easier to inspect and communicate. Although CALM expects semi-structured inputs, many clinical documents already have this form, and similar structure can often be automatically extracted from free-text notes. CALM achieves performance comparable to conventional LLM classifiers while improving trust, supporting quality-assurance checks, and revealing clinically meaningful patterns during model development and auditing.

Karthikeyan K, Philip Wu, Xin Tang et al.

The Science Consultant Agent is a web-based Artificial Intelligence (AI) tool that helps practitioners select and implement the most effective modeling strategy for AI-based solutions. It operates through four core components: Questionnaire, Smart Fill, Research-Guided Recommendation, and Prototype Builder. By combining structured questionnaires, literature-backed solution recommendations, and prototype generation, the Science Consultant Agent accelerates development for everyone from Product Managers and Software Developers to Researchers. The full pipeline is illustrated in Figure 1.

Karthikeyan K, Michelle Yuan, Elman Mansimov et al.

Recent works have demonstrated that incorporating search during inference can significantly improve reasoning capabilities of language agents. Some approaches may make use of the ground truth or rely on model's own generated feedback. The search algorithm uses this feedback to then produce values that will update its criterion for exploring and exploiting various reasoning paths. In this study, we investigate how search and model's self-feedback can be leveraged for reasoning tasks. First, we explore differences in ground-truth feedback and self-feedback during search for math reasoning. Second, we observe limitations in applying search techniques to more complex tasks like tool-calling and design domain-specific approaches to address these gaps. Our experiments reveal challenges related to generalization when solely relying on self-feedback during search. For search to work effectively, either access to the ground-truth is needed or feedback mechanisms need to be carefully designed for the specific task.

Karthikeyan K, Yogarshi Vyas, Jie Ma et al.

Training a Named Entity Recognition (NER) model often involves fixing a taxonomy of entity types. However, requirements evolve and we might need the NER model to recognize additional entity types. A simple approach is to re-annotate entire dataset with both existing and additional entity types and then train the model on the re-annotated dataset. However, this is an extremely laborious task. To remedy this, we propose a novel approach called Partial Label Model (PLM) that uses only partially annotated datasets. We experiment with 6 diverse datasets and show that PLM consistently performs better than most other approaches (0.5 - 2.5 F1), including in novel settings for taxonomy expansion not considered in prior work. The gap between PLM and all other approaches is especially large in settings where there is limited data available for the additional entity types (as much as 11 F1), thus suggesting a more cost effective approaches to taxonomy expansion.

Karthikeyan K, Anders Søgaard

Several instance-based explainability methods for finding influential training examples for test-time decisions have been proposed recently, including Influence Functions, TraceIn, Representer Point Selection, Grad-Dot, and Grad-Cos. Typically these methods are evaluated using LOO influence (Cook's distance) as a gold standard, or using various heuristics. In this paper, we show that all of the above methods are unstable, i.e., extremely sensitive to initialization, ordering of the training data, and batch size. We suggest that this is a natural consequence of how in the literature, the influence of examples is assumed to be independent of model state and other examples -- and argue it is not. We show that LOO influence and heuristics are, as a result, poor metrics to measure the quality of instance-based explanations, and instead propose to evaluate such explanations by their ability to detect poisoning attacks. Further, we provide a simple, yet effective baseline to improve all of the above methods and show how it leads to very significant improvements on downstream tasks.

Karthikeyan K, Aalok Sathe, Somak Aditya et al.

Multilingual language models achieve impressive zero-shot accuracies in many languages in complex tasks such as Natural Language Inference (NLI). Examples in NLI (and equivalent complex tasks) often pertain to various types of sub-tasks, requiring different kinds of reasoning. Certain types of reasoning have proven to be more difficult to learn in a monolingual context, and in the crosslingual context, similar observations may shed light on zero-shot transfer efficiency and few-shot sample selection. Hence, to investigate the effects of types of reasoning on transfer performance, we propose a category-annotated multilingual NLI dataset and discuss the challenges to scale monolingual annotations to multiple languages. We statistically observe interesting effects that the confluence of reasoning types and language similarities have on transfer performance.

Zihan Wang, Karthikeyan K, Stephen Mayhew et al.

Multilingual BERT (M-BERT) has been a huge success in both supervised and zero-shot cross-lingual transfer learning. However, this success has focused only on the top 104 languages in Wikipedia that it was trained on. In this paper, we propose a simple but effective approach to extend M-BERT (E-BERT) so that it can benefit any new language, and show that our approach benefits languages that are already in M-BERT as well. We perform an extensive set of experiments with Named Entity Recognition (NER) on 27 languages, only 16 of which are in M-BERT, and show an average increase of about 6% F1 on languages that are already in M-BERT and 23% F1 increase on new languages.

Karthikeyan K, Zihan Wang, Stephen Mayhew et al.

Recent work has exhibited the surprising cross-lingual abilities of multilingual BERT (M-BERT) -- surprising since it is trained without any cross-lingual objective and with no aligned data. In this work, we provide a comprehensive study of the contribution of different components in M-BERT to its cross-lingual ability. We study the impact of linguistic properties of the languages, the architecture of the model, and the learning objectives. The experimental study is done in the context of three typologically different languages -- Spanish, Hindi, and Russian -- and using two conceptually different NLP tasks, textual entailment and named entity recognition. Among our key conclusions is the fact that the lexical overlap between languages plays a negligible role in the cross-lingual success, while the depth of the network is an integral part of it. All our models and implementations can be found on our project page: http://cogcomp.org/page/publication_view/900 .

Karthikeyan K, Shubham Kumar Bharti, Piyush Rai

Despite the effectiveness of multitask deep neural network (MTDNN), there is a limited theoretical understanding on how the information is shared across different tasks in MTDNN. In this work, we establish a formal connection between MTDNN with infinitely-wide hidden layers and multitask Gaussian Process (GP). We derive multitask GP kernels corresponding to both single-layer and deep multitask Bayesian neural networks (MTBNN) and show that information among different tasks is shared primarily due to correlation across last layer weights of MTBNN and shared hyper-parameters, which is contrary to the popular hypothesis that information is shared because of shared intermediate layer weights. Our construction enables using multitask GP to perform efficient Bayesian inference for the equivalent MTDNN with infinitely-wide hidden layers. Prior work on the connection between deep neural networks and GP for single task settings can be seen as special cases of our construction. We also present an adaptive multitask neural network architecture that corresponds to a multitask GP with more flexible kernels, such as Linear Model of Coregionalization (LMC) and Cross-Coregionalization (CC) kernels. We provide experimental results to further illustrate these ideas on synthetic and real datasets.