Hyunji Nam

Hyunji Alex Nam, Yash Chandak, Emma Brunskill

From incorporating LLMs in education, to identifying new drugs and improving ways to charge batteries, innovators constantly try new strategies in search of better long-term outcomes for students, patients and consumers. One major bottleneck in this innovation cycle is the amount of time it takes to observe the downstream effects of a decision policy that incorporates new interventions. The key question is whether we can quickly evaluate long-term outcomes of a new decision policy without making long-term observations. Organizations often have access to prior data about past decision policies and their outcomes, evaluated over the full horizon of interest. Motivated by this, we introduce a new setting for short-long policy evaluation for sequential decision making tasks. Our proposed methods significantly outperform prior results on simulators of HIV treatment, kidney dialysis and battery charging. We also demonstrate that our methods can be useful for applications in AI safety by quickly identifying when a new decision policy is likely to have substantially lower performance than past policies.

Hyunji Nam, Lucia Langlois, James Malamut et al.

Large language models (LLMs) are increasingly applied to open-ended, interpretive annotation tasks, such as thematic analysis by researchers or generating feedback on student work by teachers. These tasks involve free-text annotations requiring expert-level judgments grounded in specific objectives (e.g., research questions or instructional goals). Evaluating whether LLM-generated annotations align with those generated by expert humans is challenging to do at scale, and currently, no validated, scalable measure of similarity in ideas exists. In this paper, we (i) introduce the scalable evaluation of interpretive annotation by LLMs as a critical and understudied task, (ii) propose IDEAlgin, an intuitive benchmarking paradigm for capturing expert similarity ratings via a "pick-the-odd-one-out" triplet judgment task, and (iii) evaluate various similarity metrics, including vector-based ones (topic models, embeddings) and LLM-as-a-judge via IDEAlgin, against these human benchmarks. Applying this approach to two real-world educational datasets (interpretive analysis and feedback generation), we find that vector-based metrics largely fail to capture the nuanced dimensions of similarity meaningful to experts. Prompting LLMs via IDEAlgin significantly improves alignment with expert judgments (9-30% increase) compared to traditional lexical and vector-based metrics. These results establish IDEAlgin as a promising paradigm for evaluating LLMs against open-ended expert annotations at scale, informing responsible deployment of LLMs in education and beyond.

Hyunji Nam, Omer Gottesman, Amy Zhang et al.

Large language models (LLMs) built on existing reinforcement learning with human feedback (RLHF) frameworks typically optimize responses based on immediate turn-level human preferences. However, this approach falls short in multi-turn dialogue settings, such as online math tutoring. We propose a method to enhance LLM-based tutors by representing the dialogue history with a lower-dimensional latent state representation of a student and optimizing a long-term policy to determine high-level actions based on the latent state. The goal is to better align the tutor's behavior with the long-term objective of guiding the student towards solving a target math problem on their own. Our model is lightweight, requiring less computational resources than prior work of training the tutor policy end-to-end to directly output the tutor's next utterance. Our experiment results demonstrate that these modifications lead to improved long-term outcomes compared to prompting in LLM-simulated tutoring tasks.

Hyunji Nam, Yanming Wan, Mickel Liu et al.

As everyday use cases of large language model (LLM) AI assistants have expanded, it is becoming increasingly important to personalize responses to align to different users' preferences and goals. While reinforcement learning from human feedback (RLHF) is effective at improving LLMs to be generally more helpful and fluent, it does not account for variability across users, as it models the entire user population with a single reward model, meaning it assumes that everyone's preferences are the same. We present a novel framework, Preference Learning Using Summarization (PLUS), that uses reinforcement learning (RL) to learn to produce text-based summaries of each user's preferences, characteristics, and past conversations. These summaries condition the reward model, enabling it to make personalized predictions about the types of responses valued by each user. Both the user-summarization model and reward model are trained simultaneously, creating an online co-adaptation loop. We show that in contrast to the standard Bradley-Terry model, summaries produced by PLUS capture diverse aspects of user preferences, achieving a 11-77% improvement in reward model accuracy. Key strengths of PLUS are: (1) robust performance with new users and conversation topics, achieving a 25% improvement over the best personalized RLHF technique; (2) zero-shot personalization with state-of-the-art proprietary models like GPT-4 (e.g., PLUS-summary-conditioned responses achieved a 72% win rate compared to 28% for default GPT-4o); (3) learning from flexible user contexts beyond preference labels, and (4) interpretable representation of users, enabling greater transparency and user control in pluralistic LLM alignment.

Hyunji Nam, Allen Nie, Ge Gao et al.

Off-policy policy evaluation (OPE) estimates the outcome of a new policy using historical data collected from a different policy. However, existing OPE methods cannot handle cases when the new policy introduces novel actions. This issue commonly occurs in real-world domains, like healthcare, as new drugs and treatments are continuously developed. Novel actions necessitate on-policy data collection, which can be burdensome and expensive if the outcome of interest takes a substantial amount of time to observe--for example, in multi-year clinical trials. This raises a key question of how to predict the long-term outcome of a policy after only observing its short-term effects? Though in general this problem is intractable, under some surrogacy conditions, the short-term on-policy data can be combined with the long-term historical data to make accurate predictions about the new policy's long-term value. In two simulated healthcare examples--HIV and sepsis management--we show that our estimators can provide accurate predictions about the policy value only after observing 10\% of the full horizon data. We also provide finite sample analysis of our doubly robust estimators.

Rui Aguiar, Nikka Mofid, Hyunji Alex Nam

There has been a current trend in reinforcement learning for healthcare literature, where in order to prepare clinical datasets, researchers will carry forward the last results of the non-administered test known as the last-observation-carried-forward (LOCF) value to fill in gaps, assuming that it is still an accurate indicator of the patient's current state. These values are carried forward without maintaining information about exactly how these values were imputed, leading to ambiguity. Our approach models this problem using OpenAI Gym's Mountain Car and aims to address when to observe the patient's physiological state and partly how to intervene, as we have assumed we can only act after following an observation. So far, we have found that for a last-observation-carried-forward implementation of the state space, augmenting the state with counters for each state variable tracking the time since last observation was made, improves the predictive performance of an agent, supporting the notion of "informative missingness", and using a neural network based Dynamics Model to predict the most probable next state value of non-observed state variables instead of carrying forward the last observed value through LOCF further improves the agent's performance, leading to faster convergence and reduced variance.