arXiv subjects

Bahareh Sarrafzadeh

Publications and source records attributed to Bahareh Sarrafzadeh.

9 recordsLinked to original sources

arXiv2026

Evaluating Style-Personalized Text Generation: Challenges and Directions

With the surge of large language models (LLMs) and their ability to produce customized output, style-personalized text generation--"write like me"--has become a rapidly growing area of interest. However, style personalization is highly specific, relative to every user, and depends strongly on the pragmatic context, which makes it uniquely challenging. Although prior research has introduced benchmarks and metrics for this area, they tend to be non-standardized and have known limitations (e.g., poor correlation with human subjects). LLMs have been found to not capture author-specific style well, it follows that the metrics themselves must be scrutinized carefully. In this work we critically examine the effectiveness of the most common metrics used in the field, such as BLEU, embeddings, and LLMs-as-judges. We evaluate these metrics using our proposed style discrimination benchmark, which spans eight diverse writing tasks across three evaluation settings: domain discrimination, authorship attribution, and LLM-generated personalized vs non-personalized discrimination. We find strong evidence that employing ensembles of diverse evaluation metrics consistently outperforms single-evaluator methods, and conclude by providing guidance on how to reliably assess style-personalized text generation.

cs.CL↗

arXiv2025

Conversational User-AI Intervention: A Study on Prompt Rewriting for Improved LLM Response Generation

Human-LLM conversations are increasingly becoming more pervasive in peoples' professional and personal lives, yet many users still struggle to elicit helpful responses from LLM Chatbots. One of the reasons for this issue is users' lack of understanding in crafting effective prompts that accurately convey their information needs. Meanwhile, the existence of real-world conversational datasets on the one hand, and the text understanding faculties of LLMs on the other, present a unique opportunity to study this problem, and its potential solutions at scale. Thus, in this paper we present the first LLM-centric study of real human-AI chatbot conversations, focused on investigating aspects in which user queries fall short of expressing information needs, and the potential of using LLMs to rewrite suboptimal user prompts. Our findings demonstrate that rephrasing ineffective prompts can elicit better responses from a conversational system, while preserving the user's original intent. Notably, the performance of rewrites improves in longer conversations, where contextual inferences about user needs can be made more accurately. Additionally, we observe that LLMs often need to -- and inherently do -- make \emph{plausible} assumptions about a user's intentions and goals when interpreting prompts. Our findings largely hold true across conversational domains, user intents, and LLMs of varying sizes and families, indicating the promise of using prompt rewriting as a solution for better human-AI interactions.

cs.CL↗

arXiv2025

Prototypical Human-AI Collaboration Behaviors from LLM-Assisted Writing in the Wild

As large language models (LLMs) are used in complex writing workflows, users engage in multi-turn interactions to steer generations to better fit their needs. Rather than passively accepting output, users actively refine, explore, and co-construct text. We conduct a large-scale analysis of this collaborative behavior for users engaged in writing tasks in the wild with two popular AI assistants, Bing Copilot and WildChat. Our analysis goes beyond simple task classification or satisfaction estimation common in prior work and instead characterizes how users interact with LLMs through the course of a session. We identify prototypical behaviors in how users interact with LLMs in prompts following their original request. We refer to these as Prototypical Human-AI Collaboration Behaviors (PATHs) and find that a small group of PATHs explain a majority of the variation seen in user-LLM interaction. These PATHs span users revising intents, exploring texts, posing questions, adjusting style or injecting new content. Next, we find statistically significant correlations between specific writing intents and PATHs, revealing how users' intents shape their collaboration behaviors. We conclude by discussing the implications of our findings on LLM alignment.

cs.CL↗

arXiv2024

Pearl: Personalizing Large Language Model Writing Assistants with Generation-Calibrated Retrievers

Powerful large language models have facilitated the development of writing assistants that promise to significantly improve the quality and efficiency of composition and communication. However, a barrier to effective assistance is the lack of personalization in LLM outputs to the author's communication style, specialized knowledge, and values. In this paper, we address this challenge by proposing Pearl, a LLM writing assistant personalized with a retriever that is trained to be generation-calibrated for personalization. Generation calibration ensures that our retriever selects historic user authored documents to augment an LLM prompt such that they are likely to help an LLM generation better adhere to a users' preferences. We propose two key novelties for training such a retriever: (1) A training data selection method that identifies user requests likely to benefit from personalization and documents that provide that benefit; and (2) A scale-calibrating KL-divergence objective that ensures that our retriever scores remain proportional to the downstream generation quality from using the document for personalized generation. In a series of holistic evaluations, we demonstrate the effectiveness of Pearl in generating long-form texts on multiple social media datasets. Finally, we demonstrate how a generation-calibrated retriever can double as a performance predictor -- detecting low quality retrieval, and improving potentially under-performing outputs via revision with LLMs.

cs.CL↗

arXiv2024

"I Want It That Way": Enabling Interactive Decision Support Using Large Language Models and Constraint Programming

A critical factor in the success of decision support systems is the accurate modeling of user preferences. Psychology research has demonstrated that users often develop their preferences during the elicitation process, highlighting the pivotal role of system-user interaction in developing personalized systems. This paper introduces a novel approach, combining Large Language Models (LLMs) with Constraint Programming to facilitate interactive decision support. We study this hybrid framework through the lens of meeting scheduling, a time-consuming daily activity faced by a multitude of information workers. We conduct three studies to evaluate the novel framework, including a diary study (n=64) to characterize contextual scheduling preferences, a quantitative evaluation of the system's performance, and a user study (n=10) with a prototype system. Our work highlights the potential for a hybrid LLM and optimization approach for iterative preference elicitation and design considerations for building systems that support human-system collaborative decision-making processes.

cs.HC↗

arXiv2022

Fast Parallel Hypertree Decompositions in Logarithmic Recursion Depth

Modern trends in data collection are bringing current mainstream techniques for database query processing to their limits. Consequently, various novel approaches for efficient query processing are being actively studied. One such approach is based on hypertree decompositions (HDs), which have been shown to carry great potential to process complex queries more efficiently and with stronger theoretical guarantees. However, using HDs for query execution relies on the difficult task of computing decompositions of the query structure, which guides the efficient execution of the query. From theoretical results we know that the performance of purely sequential methods is inherently limited, yet the problem is susceptible to parallelisation. In this paper we propose the first algorithm for computing hypertree decompositions that is well-suited for parallelisation. The proposed algorithm log-k-decomp requires only a logarithmic number of recursion levels and additionally allows for highly parallelised pruning of the search space by restriction to balanced separators. We provide detailed experimental evaluation over the HyperBench benchmark and demonstrate that our approach is highly effective especially for complex queries.

cs.DB↗

arXiv2020

Characterizing Stage-Aware Writing Assistance in Collaborative Document Authoring

Writing is a complex non-linear process that begins with a mental model of intent, and progresses through an outline of ideas, to words on paper (and their subsequent refinement). Despite past research in understanding writing, Web-scale consumer and enterprise collaborative digital writing environments are yet to greatly benefit from intelligent systems that understand the stages of document evolution, providing opportune assistance based on authors' situated actions and context. In this paper, we present three studies that explore temporal stages of document authoring. We first survey information workers at a large technology company about their writing habits and preferences, concluding that writers do in fact conceptually progress through several distinct phases while authoring documents. We also explore, qualitatively, how writing stages are linked to document lifespan. We supplement these qualitative findings with an analysis of the longitudinal user interaction logs of a popular digital writing platform over several million documents. Finally, as a first step towards facilitating an intelligent digital writing assistant, we conduct a preliminary investigation into the utility of user interaction log data for predicting the temporal stage of a document. Our results support the benefit of tools tailored to writing stages, identify primary tasks associated with these stages, and show that it is possible to predict stages from anonymous interaction logs. Together, these results argue for the benefit and feasibility of more tailored digital writing assistance.

cs.HC↗

arXiv2020

Hierarchical Knowledge Graphs: A Novel Information Representation for Exploratory Search Tasks

In exploratory search tasks, alongside information retrieval, information representation is an important factor in sensemaking. In this paper, we explore a multi-layer extension to knowledge graphs, hierarchical knowledge graphs (HKGs), that combines hierarchical and network visualizations into a unified data representation asa tool to support exploratory search. We describe our algorithm to construct these visualizations, analyze interaction logs to quantitatively demonstrate performance parity with networks and performance advantages over hierarchies, and synthesize data from interaction logs, interviews, and thinkalouds on a testbed data set to demonstrate the utility of the unified hierarchy+network structure in our HKGs. Alongside the above study, we perform an additional mixed methods analysis of the effect of precision and recall on the performance of hierarchical knowledge graphs for two different exploratory search tasks. While the quantitative data shows a limited effect of precision and recall on user performance and user effort, qualitative data combined with post-hoc statistical analysis provides evidence that the type of exploratory search task (e.g., learning versus investigating) can be impacted by precision and recall. Furthermore, our qualitative analyses find that users are unable to perceive differences in the quality of extracted information. We discuss the implications of our results and analyze other factors that more significantly impact exploratory search performance in our experimental tasks.

cs.IR↗

arXiv2019

Characterizing and Predicting Email Deferral Behavior

Email triage involves going through unhandled emails and deciding what to do with them. This familiar process can become increasingly challenging as the number of unhandled email grows. During a triage session, users commonly defer handling emails that they cannot immediately deal with to later. These deferred emails, are often related to tasks that are postponed until the user has more time or the right information to deal with them. In this paper, through qualitative interviews and a large-scale log analysis, we study when and what enterprise email users tend to defer. We found that users are more likely to defer emails when handling them involves replying, reading carefully, or clicking on links and attachments. We also learned that the decision to defer emails depends on many factors such as user's workload and the importance of the sender. Our qualitative results suggested that deferring is very common, and our quantitative log analysis confirms that 12% of triage sessions and 16% of daily active users had at least one deferred email on weekdays. We also discuss several deferral strategies such as marking emails as unread and flagging that are reported by our interviewees, and illustrate how such patterns can be also observed in user logs. Inspired by the characteristics of deferred emails and contextual factors involved in deciding if an email should be deferred, we train a classifier for predicting whether a recently triaged email is actually deferred. Our experimental results suggests that deferral can be classified with modest effectiveness. Overall, our work provides novel insights about how users handle their emails and how deferral can be modeled.

cs.IR↗