Using a sample of 25,000 Bing Copilot conversations, we study how the agent responds to users of varying levels of domain expertise and the resulting impact on user experience along multiple dimensions. Our findings show that across a variety of topical domains, the agent largely responds at proficient or expert levels of expertise (77% of conversations) which correlates with positive user experience regardless of the user's level of expertise. Misalignment, such that the agent responds at a level of expertise below that of the user, has a negative impact on overall user experience, with the impact more profound for more complex tasks. We also show that users engage more, as measured by the number of words in the conversation, when the agent responds at a level of expertise commensurate with that of the user. Our findings underscore the importance of alignment between user and AI when designing human-centered AI systems, to ensure satisfactory and productive interactions.

AI systems are often introduced with high expectations, yet many fail to deliver, resulting in unintended harm and missed opportunities for benefit. We frequently observe significant "AI Mismatches", where the system's actual performance falls short of what is needed to ensure safety and co-create value. These mismatches are particularly difficult to address once development is underway, highlighting the need for early-stage intervention. Navigating complex, multi-dimensional risk factors that contribute to AI Mismatches is a persistent challenge. To address it, we propose an AI Mismatch approach to anticipate and mitigate risks early on, focusing on the gap between realistic model performance and required task performance. Through an analysis of 774 AI cases, we extracted a set of critical factors, which informed the development of seven matrices that map the relationships between these factors and highlight high-risk areas. Through case studies, we demonstrate how our approach can help reduce risks in AI development.

Understanding collaborative writing dynamics between native speakers (NS) and non-native speakers (NNS) is critical for enhancing collaboration quality and team inclusivity. In this paper, we partnered with communication researchers to develop visual analytics solutions for comparing NS and NNS behaviors in 162 writing sessions across 27 teams. The primary challenges in analyzing writing behaviors are data complexity and the uncertainties introduced by automated methods. In response, we present \textsc{COALA}, a novel visual analytics tool that improves model interpretability by displaying uncertainties in author clusters, generating behavior summaries using large language models, and visualizing writing-related actions at multiple granularities. We validated the effectiveness of \textsc{COALA} through user studies with domain experts (N=2+2) and researchers with relevant experience (N=8). We present the insights discovered by participants using \textsc{COALA}, suggest features for future AI-assisted collaborative writing tools, and discuss the broader implications for analyzing collaborative processes beyond writing.

Supervised fine-tuning (SFT) followed by preference optimization (PO) denoted by SFT$\rightarrow$PO has become the standard for improving pretrained large language models (LLMs), with PO demonstrating significant performance gains. However, PO methods rely on either human-labeled preference data or a strong reward model to generate preference data. Can we fine-tune LLMs without preference data or reward models while achieving competitive performance to SFT$\rightarrow$PO? We address this question by introducing Discriminative Fine-Tuning (DFT), a novel approach that eliminates the need for preference data. Unlike SFT, which employs a generative approach and overlooks negative data, DFT adopts a discriminative paradigm that that increases the probability of positive answers while suppressing potentially negative ones, shifting from token prediction to data prediction. Our contributions include: (i) a discriminative probabilistic framework for fine-tuning LLMs by explicitly modeling the discriminative likelihood of an answer among all possible outputs given an input; (ii) efficient algorithms to optimize this discriminative likelihood; and (iii) extensive experiments demonstrating DFT's effectiveness, achieving performance better than SFT and comparable to if not better than SFT$\rightarrow$PO. The code can be found at https://github.com/PenGuln/DFT.

Learning therapeutic counseling involves significant role-play experience with mock patients, with current manual training methods providing only intermittent granular feedback. We seek to accelerate and optimize counselor training by providing frequent, detailed feedback to trainees as they interact with a simulated patient. Our first application domain involves training motivational interviewing skills for counselors. Motivational interviewing is a collaborative counseling style in which patients are guided to talk about changing their behavior, with empathetic counseling an essential ingredient. We developed and evaluated an LLM-powered training system that features a simulated patient and visualizations of turn-by-turn performance feedback tailored to the needs of counselors learning motivational interviewing. We conducted an evaluation study with professional and student counselors, demonstrating high usability and satisfaction with the system. We present design implications for the development of automated systems that train users in counseling skills and their generalizability to other types of social skills training.

Optimizing language models for use in conversational agents requires large quantities of example dialogues. Increasingly, these dialogues are synthetically generated by using powerful large language models (LLMs), especially in domains with challenges to obtain authentic human data. One such domain is human resources (HR). In this context, we compare two LLM-based dialogue generation methods for the use case of generating HR job interviews, and assess whether one method generates higher-quality dialogues that are more challenging to distinguish from genuine human discourse. The first method uses a single prompt to generate the complete interview dialog. The second method uses two agents that converse with each other. To evaluate dialogue quality under each method, we ask a judge LLM to determine whether AI was used for interview generation, using pairwise interview comparisons. We demonstrate that despite a sixfold increase in token cost, interviews generated with the dual-prompt method achieve a win rate up to ten times higher than those generated with the single-prompt method. This difference remains consistent regardless of whether GPT-4o or Llama 3.3 70B is used for either interview generation or judging quality.

Large language models (LLMs) encode a diverse range of linguistic features within their latent representations, which can be harnessed to steer their output toward specific target characteristics. In this paper, we modify the internal structure of LLMs by training sparse autoencoders to learn a sparse representation of the query embedding, allowing precise control over the model's attention distribution. We demonstrate that manipulating this sparse representation effectively transforms the output toward different stylistic and cognitive targets. Specifically, in an educational setting, we show that the cognitive complexity of LLM-generated feedback can be systematically adjusted by modifying the encoded query representation at a specific layer. To achieve this, we guide the learned sparse embedding toward the representation of samples from the desired cognitive complexity level, using gradient-based optimization in the latent space.

Generative AI significantly enhances player agency in interactive narratives (IN) by enabling just-in-time content generation that adapts to player actions. While delegating generation to AI makes IN more interactive, it becomes challenging for authors to control the space of possible narratives - within which the final story experienced by the player emerges from their interaction with AI. In this paper, we present WhatELSE, an AI-bridged IN authoring system that creates narrative possibility spaces from example stories. WhatELSE provides three views (narrative pivot, outline, and variants) to help authors understand the narrative space and corresponding tools leveraging linguistic abstraction to control the boundaries of the narrative space. Taking innovative LLM-based narrative planning approaches, WhatELSE further unfolds the narrative space into executable game events. Through a user study (N=12) and technical evaluations, we found that WhatELSE enables authors to perceive and edit the narrative space and generates engaging interactive narratives at play-time.

While Retrieval Augmented Generation (RAG) has emerged as a popular technique for improving Large Language Model (LLM) systems, it introduces a large number of choices, parameters and hyperparameters that must be made or tuned. This includes the LLM, embedding, and ranker models themselves, as well as hy-perparameters governing individual RAG components. Y et, collectively optimizing the entire configuration in a RAG or LLM system remains under-explored-- especially in multi-objective settings--due to intractably large solution spaces, noisy objective evaluations, and the high cost of evaluations. In this work, we introduce the first approach for multi-objective parameter optimization of cost, latency, safety and alignment over entire LLM and RAG systems. We find that Bayesian optimization methods significantly outperform baseline approaches, obtaining a superior Pareto front on two new RAG benchmark tasks. We conclude our work with important considerations for practitioners who are designing multi-objective RAG systems, highlighting nuances such as how optimal configurations may not generalize across tasks and objectives. Retrieval Augmented Generation (RAG) has emerged as a popular technique for improving the performance of Large Language Models (LLMs) on question-answering tasks over specific datasets. A benefit of using RAG pipelines is that they can often achieve high performance on specific tasks without the need for extensive alignment and fine-tuning (Gupta et al., 2024), a costly and time-consuming process. However, the end-to-end pipeline of a RAG system is dependent on many parameters that span different components (or modules) of the system, such as the choice of LLM, the embedding model used in retrieval, the number of chunks retrieved and hyperparameters governing a reranking model. Examples of choices, parameters, and hyperparameters that are often made or tuned when implementing a RAG pipeline are listed in Table 1.

Knowledge components (KCs) mapped to problems help model student learning, tracking their mastery levels on fine-grained skills thereby facilitating personalized learning and feedback in online learning platforms. However, crafting and tagging KCs to problems, traditionally performed by human domain experts, is highly labor-intensive. We present a fully automated, LLM-based pipeline for KC generation and tagging for open-ended programming problems. We also develop an LLM-based knowledge tracing (KT) framework to leverage these LLM-generated KCs, which we refer to as KCGen-KT. We conduct extensive quantitative and qualitative evaluations validating the effectiveness of KCGen-KT. On a real-world dataset of student code submissions to open-ended programming problems, KCGen-KT outperforms existing KT methods. We investigate the learning curves of generated KCs and show that LLM-generated KCs have a comparable level-of-fit to human-written KCs under the performance factor analysis (PFA) model. We also conduct a human evaluation to show that the KC tagging accuracy of our pipeline is reasonably accurate when compared to that by human domain experts.