Recent advances in LLMs have transformed AI coding assistants from simple autocompletion tools into conversational partners that support a wide range of development tasks through natural language interaction. While this shift promotes closer human-AI collaboration, this also places a burden on developers to craft precise prompts and integrate sufficient context to accomplish their task.
To address this, there is growing interest in proactive AI systems that can anticipate developer intent and surface timely, contextually relevant suggestions - potentially revealing insights developers might not have considered.

However, existing research on proactive coding assistance has largely been confined to controlled, experimental settings, leaving open questions about its effectiveness and integration in real-world development environments.

In this work, we design and implement a proactive AI assistant within JetBrains Fleet, a polyglot enterprise IDE. Our system signals AI activity via in-editor cues and delivers context-aware code improvement suggestions via a chat interface. We adopt heuristics based on IDE activity to determine when it is timely to intervene, during various stages across the development workflow: formulating needs, idea execution and committing changes.
Lastly, we conducted a five-day in-the-wild study with 18 professional developers to evaluate its impact on user experience and engagement patterns.

Our findings demonstrate that timing is a critical aspect influencing perceived value of proactive AI assistance.
Suggestions delivered at natural workflow boundaries (e.g. post-commit) achieved the highest engagement, while mid-task interventions were often dismissed.
The consistent engagement trends (and voluntary continued use) indicate strong potential for integrating proactive AI into everyday software engineering workflows.
Our study not only underscores the feasibility and practical value of proactive assistance in enterprise development environments, but also provides actionable design insights for future tools.
While challenges remain - such as improving suggestion relevance and supporting diverse user preferences - these highlight important directions for advancing adaptive, intent-aware proactive systems that can further enhance developer workflows.

All over the world, people plan their daily activities. These plans include a lot of different tasks and can vary widely in kinds of activities. These plans must account for uncertainties and unknowns in the world. Planning around these uncertainties is difficult and hard to accomplish with traditional means of programming. For this set of problems, probabilistic programming is proposed. Given the Minecraft planner from the Planning Domain Definition Language (PDDL) gym library, is it possible to create Robust plans that incorporate inference without changing the underlying planner? "PDDL is a human-readable format for problems in automated planning that gives a description of the possible states of the world, a description of the set of possible actions, a specific initial state of the world, and a specific set of desired goals." [6] The current approach is using heuristics to find the optimal plan for the problem. In this research paper, an alternative method is proposed; using probabilistic programming and the existing planner to create a simulated world of Minecraft. This model introduces inference without changing the already existing planner.

Planning problems are a set of problems in which an objective must be reached by a sequence of actions. Planning problems traditionally do not consider uncertainty, however for most real-world planning problems uncertainty must be considered to create effective plans. The objective of this paper is to use an existing deterministic planning algorithm in order to create robust plans that solve an uncertain version of Sokoban called Uncertain Move Sokoban. To this end a probabilistic programming language, Gen.jl, is used, which enables creating probabilistic models and inferring its parameters using code. A probabilistic model is created in Gen.jl, that generates plans for the problem as well as robustness scores using a simulator embedded in the model. Probabilistic inference techniques are then used to obtain a robust plan for the uncertain problem, namely: importance sampling and Metropolis-Hastings. We find that the technique is able to create robust plans for small to medium-sized problems and that Metropolis-Hastings is the better-performing inference technique.

Probabilistic deterministic Finite Automata (PDFA) learning is a machine learning method used for tasks requiring human understandability and more formal validation. In recent years we saw numerous applications of ensemble techniques with other machine learning models such as decision trees. Following the success of these attempts, in this paper, we aim to integrate ensemble methods into Alergia, which is a famous algorithm in the PDFA learning realm. We present a randomized variation of the Alergia algorithm and show how to build an ensemble out of it. Such an ensemble can visibly outperform a single Alergia model, which is documented by a series of experiments. Next, we present a custom distance metric measuring dissimilarity between a pair of Alergia models. We show how it can be used to build an Inter-Model Variety score quantifying the overall diversity of a group of models. Lastly, we analyze several methods that strive to select a well-performing diverse ensemble out of a big population of generated models.

Learning Deterministic Finite Automata (DFA) from given input data has been a central task in the field of Grammatical Inference, and progress in this area is of great interest from both theoretical and practical points of view. To address this challenge, several algorithms have been proposed and evaluated using established benchmarks. One such competition-winning algorithm, Evidence Driven State Merging (EDSM), uses a heuristic to learn a DFA from given data. However, improvements leveraging ensemble techniques from machine learning have yet to be explored. In this paper, we investigate ways to adapt the EDSM algorithm to fit into the ensemble learning framework and analyze the performance of such obtained models when applied to unseen data. To this end, we compare the performance of the ensembles to that of a standard EDSM-learned model, evaluating both their output quality and the diversity within each ensemble. The results indicate significant improvements in scenarios where the data is sparse.

Scheduling problems are present in many real-world situations, such as construction projects, manufacturing processes, or train timetabling. One common formalization is the Resource Constrained Project Scheduling Problem (RCPSP), where the goal is to find an optimal schedule given limited resources. Traditional algorithms optimize for project duration under deterministic assumptions, which could lead to poor performance under uncertainty. This thesis explores how to create robust schedules, meaning they can withstand uncertainty, for stochastic task durations. Robust schedules minimize delays, measured as the difference between a task's completion time and its deadline. The method proposed uses probabilistic programming as a tool to accomplish this. A robustness distribution over schedules is inferred using importance sampling, and a robust schedule can be selected from that explored distribution. Importantly, this approach presented in this thesis can be applied on top of existing scheduling and simulation algorithms without requiring any knowledge or changes to themselves. Created schedules can also be abstracted, thus do not need to be analyzed or seen. This makes the proposed method general and easy to adopt in practice.

Planning is very important in everyday life, whether it would be creating schedules for planes or plans for manufacturing. These domains contain uncertainties requiring plans that are robust. However, there is a need for an approach which creates robust plans regardless of the domain and without changing its planning agent. Here, a replanning approach is proposed akin to the Metropolis-Hastings algorithm and its performance is compared to the performance of importance sampling. Replanning works by iteratively trying to improve the previously generated plan. The performance is compared by means of the Keys and Doors problem. It is found that replanning performed better than importance sampling in the two analysed problems. Furthermore, changing the parameter, σ, used in the replanning approach showed a significant difference in its corresponding performance. While the replanning approach has only been tested on the Keys and Doors problem, the results show that replanning is a promising approach which could work irrespective of the domain at hand.

Train Unit Shunting is a complex process that directs trains through a shunting yard. In real-world railway operations, disturbances are common, requiring shunting schedules to be robust against uncertainties such as delays. Previous research has proposed algorithms for the Train Unit Shunting Problem (TUSP) and one study attempted to create robust shunting plans by defining a probabilistic model of the uncertainties involved and inferring a distribution of robust solutions for the TUSP. Following this approach, this paper investigates the use of probabilistic programming in increasing robustness of shunting plans using an advanced TUSP solver. This research develops a model for uncertain shunting scenarios, solves these scenarios, and applies importance sampling to infer the posterior distribution, producing a distribution of robust shunting plans instead of a single plan. The paper presents examples demonstrating that it is beneficial to use one of the output robust plans over the plan made for the deterministic scenario, revealing the potential of integrating probabilistic programming techniques into the planning process to improve railway efficiency and reduce delays.

Learning a Deterministic Finite Automaton (DFA) from a language sample is an essential problem in grammatical inference, with applications in various fields, such as modeling and analyzing software systems. In this work, we propose approaches to create an ensemble of DFAs learned with the Evidence Driven State Merging algorithm. To produce varying models from the given data, we introduce two algorithms for manipulating the sequence orders. Additionally, we propose a similarity metric that allows for reducing the ensemble size by discarding similar models. The proposed approaches were analyzed and empirically evaluated using the dataset used during the StaMinA competition. Experimental results demonstrate that the methods for obtaining ensembles of DFAs presented in this work provide a number of advantages over the single DFA learned from the classical prefix tree acceptor using EDSM.

Deterministic Finite Automata (DFAs) are interpretable classification models, typically learned through merging states of a large tree-like automaton, an Augmented Prefix Tree Acceptor (APTA), according to heuristic suitability metrics. This paper introduces an ensembling approach for DFAs that does not depend on such heuristics. Starting from the APTA, we construct diverse automata by applying randomized sequences of state merges, while avoiding repetition of merges whenever possible. We also propose a novel graph-connectivity-based metric for inter-model variety. Experimental results on the STAMINA competition datasets yield improved predictive performance compared to models learned using state-of-the-art heuristics on sparse datasets, as well as a tight connection between inter-model variety and performance.

Multiple benchmarks for question answering (QA) systems often under-represent questions that require lists to be answered, referred to in this work as ListQA. This type of question can provide valuable insights into the system’s ability to structure its internal knowledge. In this work, we introduce TimelineQA (TLQA), a specialized subset of ListQA that requires both list comprehension and temporal understanding to generate correct answers. We automatically curate a benchmark for TLQA and test it against generative large language models (LLMs) commonly employed in QA systems. Our findings reveal significant shortcomings in current models, particularly in their inability to provide complete answers and accurately align facts temporally. To address these issues, we explore how Retrieval Augmented Generation (RAG) can enhance performance in this task. Our results indicate that while RAG improves performance, especially in retrieving relevant contextual information, there remains considerable room for further improvement. Both the benchmark and the evaluation of the models can be publicly accessed here.

Dutch State-of-the-art Automatic Speech Recognition (ASR) systems do not perform equally well for different speaker groups. Existing metrics to quantify this bias rely on demographic metadata, which is often unavailable. Recent advances in the field use machine learning to find groups of similar speakers instead. However, its black-box nature obscures the interpretability of resulting groups. This paper proposes an interpretable approach to bias discovery by clustering speakers based on acoustic and prosodic features. Different feature subsets were compared in their ability to find performance disparities in five ASR systems for two separate speaking styles. Results show that these feature sets can uncover bias approaching known disparities between demographic groups. While the effectiveness per feature set differed between the speaking styles, the most successful ones found significant disparities between clusters with diverse demographic compositions.

Cancer poses a significant clinical, social, and economic burden, necessitating the development of effective treatments. Understanding how drugs interact with cancer cells and their downstream effects is critical for creating new therapies and overcoming drug resistance. This paper compares the predictive performance of the Geneformer model with traditional machine learning methods in predicting the response of cancer cells to perturbation combinations using the sciplex2 dataset.

The research involves preprocessing the sciplex2 dataset, training the models, and evaluating their performance in binary classification of cells as either treated or untreated, and the prediction of gene perturbation impacts. While traditional ML models demonstrated higher accuracy in binary classification tasks, Geneformer excelled in predicting the impact of gene perturbations due to its advanced architecture and extensive pre-training on single-cell transcriptomes.

Key findings reveal that highly expression-correlated gene pairs cause the largest shifts in cell classification, underscoring the importance of gene correlations in biological predictions. Geneformer showed a deeper understanding of gene network dynamics, achieving higher maximum Cosine Shifts compared to PCA embeddings and placing less emphasis on highly differentially expressed (HDE) Single Genes. Instead, it focused on HDE Gene Pairs, indicating its potential ability to capture complex downstream effects of gene perturbations.

This study highlights the potential of integrating advanced deep learning models like Geneformer into drug discovery, offering a pathway for more effective and targeted therapeutic interventions.

Automatic Speech Recognition (ASR) systems have become an integral part of daily lives. Despite their widespread use, these systems can exhibit biases that express themselves in the differences in their accuracy and performance across different demographic groups. Methods quantifying these biases have been developed. This paper investigates the relationship between bias and the acoustic characteristics of speakers. By examining various acoustic embeddings, derived from models like wav2vec 2.0 and XLSR, we aim to identify which embeddings correlate most strongly with bias. The findings offer insights into improving the fairness of ASRs by exploring how acoustic features influence bias in ASR systems. Future research directions include exploring isolated speech properties and extending the study to diverse linguistic contexts to deepen understanding in this area.

Studying the interactions of genes within a cell is an area of significant interest in the field of medicine as it can provide answers to what exactly drives the behavior of a cell under specific circumstances, such as diseases. Once understood, gene interactions can enable the synthesis of efficient, possibly personalized treatments for these diseases and other disorders. However, studying gene interactions requires a large number of samples which might be costly and laborious to obtain in the case of rare disorders for which there is not much recorded data. Geneformer, a context-aware, attention-based deep-learning model, was created specifically for solving this problem. The model makes use of transfer learning to apply any relevant knowledge gained from a larger, similar domain onto a downstream domain with limited data which can be used to further train the model. In this paper, we assessed four fine-tuning strategies, including the one used throughout the in silico experiments presented in the original Geneformer paper. We did this to assess whether the accuracy of Geneformer on the downstream task of predicting the sensitivity of cancer cells to different treatments can be improved versus the default implementation as found within the model's paper. The model was firstly fine-tuned using a training dataset compiled from the sciplex2 dataset, followed by the prediction of the dosage levels to which samples from a test set were exposed. Upon performing the experiment, we concluded that, depending on the way in which knowledge from the source domain is stored inside the pre-trained model and the similarity between the source and the target domains, different fine-tuning strategies were suitable for a given task. Hence, there is no single optimal fine-tuning method which can be used to predict the level to which cancer cells were exposed to treatments such as nutlin-3A.

Powerful new machine learning models in biomedicine are being developed constantly, further hastened by the advent of transformer-based architectures. These advanced systems can be used for various applications, from diagnostics to assessing drug effectiveness. Many of these are fundamentally cell classification problems. Models like Geneformer [1] use gene expression data to learn how to distinguish between these cell classes. This information is usually obtained through single-cell RNA sequencing. However, the alternative source, bulk RNA sequencing, offers some advantages that make exploring the feasibility of using it to train Geneformer enticing, such as its greater availability and lower cost. In this paper, pseudo-bulk datasets are created from single-cell data by aggregation of gene expressions. A method to generate synthetic single-cell-like data from a bulk dataset is used to create new datasets. Some remain purely synthetic, while others are mixed with real single-cell data. Geneformer is fine-tuned on all generated datasets separately, and its performance in a cell classification problem is measured. It is shown that the more a dataset resembles real single-cell data, the better the model’s performance. Using bulk data to fine-tune Geneformer is proven to be infeasible. The synthetic data fails to effectively fine-tune the model and is proven to not have a meaningful impact when added to a singlecell dataset. It is concluded that the generated synthetic data is of too low quality and that alternative generation methods should be explored.

Accurately predicting how cancer cells respond to drug treatment is important to advance drug development. This paper presents a comparative analysis of Geneformer, a deep-learning transformer pre-trained on transcriptomic data, and Support Vector Machine. Using the Sciplex2 dataset, which includes transcriptomic data from lung cancer cells treated with three drugs, both models were trained to predict the response of cancer cells to drug treatments.

This paper investigates how Geneformer and SVM perform in predicting the treatment label of cells across different drugs and doses, which drug doses are suitable for conducting single-gene perturbation experiments, how accurately can these experiments replicate drug effects, and what are the differences in results between Geneformer and SVM regarding their ability to identify significant genes affecting drug response.

Results indicate that while SVM generally achieves higher accuracy in predicting treatment labels of cells, Geneformer demonstrates better capability in identifying genes whose perturbations mimic drug effects. Geneformer's embeddings show significant shifts towards treated cell states after single-gene perturbations, indicating a deeper understanding of gene interactions in drug response. On the other hand, SVM's predictions rely more on differential gene expression. This comparative analysis underscores the strengths and limitations of each approach in modelling complex biological systems and predicting the drug response of cancer cells.

This paper presents a novel approach to measuring bias in Automatic Speech Recognition (ASR) systems by proposing a metric that does not use the conventional approach of a reference group. Current methods typically measure bias through comparison with a ’norm’ or minimum error group, potentially introducing additional biases. To address this issue, this study introduces a new metric: a combination of the Groupto-Average Log Ratio and the Sum of Group Error Differences. This metric aims to provide a fair comparison by measuring performance relative to the average of the groups rather than a single reference group. Results indicate that the new metric reveals different aspects of bias not captured by traditional methods. This study contributes to the ongoing research on fairness in speech technology by challenging the existing bias metrics and proposing alternatives that might offer more equitable evaluations. Future research should explore further refinements of these metrics, and apply them across more varied datasets and environments. Ultimately, this research moves towards making ASR technologies more inclusive, ensuring that they serve all user groups equitably.

The advancement of artificial intelligence (AI) has led to an increased demand for both a greater volume and quality of data. In many companies, data is dispersed across multiple tables, yet AI models typically require data in a single table format. This necessitates the merging of these tables and the selection of optimal features for the model, a process known as Tabular Data Augmentation (TDA). With the rapid growth of TDA, automated tools have been developed to streamline this process. However, these state-of-the-art tools often make assumptions about user workflows that may not align with the actual needs of data specialists, potentially making them efficient yet not fully user-friendly. Additionally, without thorough evaluation through user studies, these tools may overlook critical steps in the TDA process.

This thesis is divided into two main parts. The first part is dedicated to uncovering the assumptions and oversights within current TDA research through an exhaustive review of recent literature. This is followed by conducting interviews with 19 data specialists. These discussions aim to verify the identified assumptions and reveal any missing elements in state-of-the-art research. The second part focuses on creating a new tool to meet the requirements identified from validated assumptions and the gaps discovered. This tool is then subjected to evaluation interviews to assess its effectiveness.

The findings indicate that data specialists prefer a TDA tool that offers enhanced control and deeper insights into the data augmentation process. To meet these preferences, Human in the Loop AutoTDA was developed, embodying the desired functionalities. Feedback from the evaluation phase confirmed that data specialists find Human in the Loop AutoTDA suitable for their TDA workflows, marking a significant advancement in the field.

In Reinforcement Learning (RL), an agent learns to make decisions by interacting with an environment and receiving feedback in the form of rewards. Multi-Task Reinforcement Learning (MTRL) extends this concept by training a single agent to perform multiple tasks simultaneously, allowing for more efficient use of resources and behavior sharing between tasks. Policy Distillation (PD) is a technique commonly used in MTRL, where policies from multiple single-task agents (teachers) are distilled into a single multi-task agent (student). This is done by merging common structure across tasks, while separating task-specific properties.

However, existing PD approaches require interactions with the environment during training. In this work, we investigate the effectiveness of PD in the offline setting, where the agent has no interaction with the environment before deployment and can only learn from previously collected data. Through a series of experiments, we demonstrate that a straightforward approach yields the highest performance. This approach involves first learning teacher policies using an existing offline RL algorithm, then distilling these policies into a student by sampling states from the offline data and applying a Mean Squared Error (MSE) loss between the teachers’ and student’s best actions. Moreover, we investigate the effect of a state distribution shift—a major challenge in offline RL—on our approach. We find that such shifts impact performance only slightly in cases of relatively small neural networks or substantial distribution shifts.

We also explore how PD can be enhanced to better capture common structure across related tasks, a key to improving efficiency in MTRL. To this end, we formally define common structure at two levels: the trajectory level and the computational level. To the best of our knowledge, we present the first attempt to quantify the amount of common structure shared across tasks. This measurement reveals that task commonalities are not fully exploited automatically. At the computational level, we attempt to improve sharing of common structure by reducing the network size and adding a regularization term to the loss function. To capture more common structure at the trajectory level, we argue that multi-task exploration is required, meaning that behaviors from one task must be evaluated in the context of another task. We propose two extensions to our approach that introduce multi-task exploration: Data Sharing (DS) and Offline Q-Switch (OQS). While these extensions are capable of improving performance, they also have clear limitations.

Overall, we propose a new, high-performing offline MTRL method and provide valuable insights into the fundamental capabilities and limitations of PD in capturing common structure across tasks, specifically within the offline MTRL setting.