Allen Nie

Aarohi Srivastava, Abhinav Rastogi, Abhishek Rao et al. · allen-ai, amazon-science

Language models demonstrate both quantitative improvement and new qualitative capabilities with increasing scale. Despite their potentially transformative impact, these new capabilities are as yet poorly characterized. In order to inform future research, prepare for disruptive new model capabilities, and ameliorate socially harmful effects, it is vital that we understand the present and near-future capabilities and limitations of language models. To address this challenge, we introduce the Beyond the Imitation Game benchmark (BIG-bench). BIG-bench currently consists of 204 tasks, contributed by 450 authors across 132 institutions. Task topics are diverse, drawing problems from linguistics, childhood development, math, common-sense reasoning, biology, physics, social bias, software development, and beyond. BIG-bench focuses on tasks that are believed to be beyond the capabilities of current language models. We evaluate the behavior of OpenAI's GPT models, Google-internal dense transformer architectures, and Switch-style sparse transformers on BIG-bench, across model sizes spanning millions to hundreds of billions of parameters. In addition, a team of human expert raters performed all tasks in order to provide a strong baseline. Findings include: model performance and calibration both improve with scale, but are poor in absolute terms (and when compared with rater performance); performance is remarkably similar across model classes, though with benefits from sparsity; tasks that improve gradually and predictably commonly involve a large knowledge or memorization component, whereas tasks that exhibit "breakthrough" behavior at a critical scale often involve multiple steps or components, or brittle metrics; social bias typically increases with scale in settings with ambiguous context, but this can be improved with prompting.

Sherry Ruan, Allen Nie, William Steenbergen et al. · stanford

Resource limitations make it hard to provide all students with one of the most effective educational interventions: personalized instruction. Reinforcement learning could be a key tool to reduce the development cost and improve the effectiveness of intelligent tutoring software that aims to provide the right support, at the right time, to a student. Here we illustrate that deep reinforcement learning can be used to provide adaptive pedagogical support to students learning about the concept of volume in a narrative storyline software. Using explainable artificial intelligence tools, we extracted interpretable insights about the pedagogical policy learned and demonstrated that the resulting policy had similar performance in a different student population. Most importantly, in both studies, the reinforcement-learning narrative system had the largest benefit for those students with the lowest initial pretest scores, suggesting the opportunity for AI to adapt and provide support for those most in need.

Anirudhan Badrinath, Yannis Flet-Berliac, Allen Nie et al. · stanford

Despite the recent advancements in offline reinforcement learning via supervised learning (RvS) and the success of the decision transformer (DT) architecture in various domains, DTs have fallen short in several challenging benchmarks. The root cause of this underperformance lies in their inability to seamlessly connect segments of suboptimal trajectories. To overcome this limitation, we present a novel approach to enhance RvS methods by integrating intermediate targets. We introduce the Waypoint Transformer (WT), using an architecture that builds upon the DT framework and conditioned on automatically-generated waypoints. The results show a significant increase in the final return compared to existing RvS methods, with performance on par or greater than existing state-of-the-art temporal difference learning-based methods. Additionally, the performance and stability improvements are largest in the most challenging environments and data configurations, including AntMaze Large Play/Diverse and Kitchen Mixed/Partial.

Allen Nie, Yannis Flet-Berliac, Deon R. Jordan et al. · stanford

Offline reinforcement learning (RL) can be used to improve future performance by leveraging historical data. There exist many different algorithms for offline RL, and it is well recognized that these algorithms, and their hyperparameter settings, can lead to decision policies with substantially differing performance. This prompts the need for pipelines that allow practitioners to systematically perform algorithm-hyperparameter selection for their setting. Critically, in most real-world settings, this pipeline must only involve the use of historical data. Inspired by statistical model selection methods for supervised learning, we introduce a task- and method-agnostic pipeline for automatically training, comparing, selecting, and deploying the best policy when the provided dataset is limited in size. In particular, our work highlights the importance of performing multiple data splits to produce more reliable algorithm-hyperparameter selection. While this is a common approach in supervised learning, to our knowledge, this has not been discussed in detail in the offline RL setting. We show it can have substantial impacts when the dataset is small. Compared to alternate approaches, our proposed pipeline outputs higher-performing deployed policies from a broad range of offline policy learning algorithms and across various simulation domains in healthcare, education, and robotics. This work contributes toward the development of a general-purpose meta-algorithm for automatic algorithm-hyperparameter selection for offline RL.

Evan Zheran Liu, Moritz Stephan, Allen Nie et al. · stanford

Developing interactive software, such as websites or games, is a particularly engaging way to learn computer science. However, teaching and giving feedback on such software is time-consuming -- standard approaches require instructors to manually grade student-implemented interactive programs. As a result, online platforms that serve millions, like Code.org, are unable to provide any feedback on assignments for implementing interactive programs, which critically hinders students' ability to learn. One approach toward automatic grading is to learn an agent that interacts with a student's program and explores states indicative of errors via reinforcement learning. However, existing work on this approach only provides binary feedback of whether a program is correct or not, while students require finer-grained feedback on the specific errors in their programs to understand their mistakes. In this work, we show that exploring to discover errors can be cast as a meta-exploration problem. This enables us to construct a principled objective for discovering errors and an algorithm for optimizing this objective, which provides fine-grained feedback. We evaluate our approach on a set of over 700K real anonymized student programs from a Code.org interactive assignment. Our approach provides feedback with 94.3% accuracy, improving over existing approaches by 17.7% and coming within 1.5% of human-level accuracy. Project web page: https://ezliu.github.io/dreamgrader.

Kefan Dong, Yannis Flet-Berliac, Allen Nie et al. · stanford

We present a model-based offline reinforcement learning policy performance lower bound that explicitly captures dynamics model misspecification and distribution mismatch and we propose an empirical algorithm for optimal offline policy selection. Theoretically, we prove a novel safe policy improvement theorem by establishing pessimism approximations to the value function. Our key insight is to jointly consider selecting over dynamics models and policies: as long as a dynamics model can accurately represent the dynamics of the state-action pairs visited by a given policy, it is possible to approximate the value of that particular policy. We analyze our lower bound in the LQR setting and also show competitive performance to previous lower bounds on policy selection across a set of D4RL tasks.

Allen Nie, Yuhui Zhang, Atharva Amdekar et al. · stanford

Human commonsense understanding of the physical and social world is organized around intuitive theories. These theories support making causal and moral judgments. When something bad happens, we naturally ask: who did what, and why? A rich literature in cognitive science has studied people's causal and moral intuitions. This work has revealed a number of factors that systematically influence people's judgments, such as the violation of norms and whether the harm is avoidable or inevitable. We collected a dataset of stories from 24 cognitive science papers and developed a system to annotate each story with the factors they investigated. Using this dataset, we test whether large language models (LLMs) make causal and moral judgments about text-based scenarios that align with those of human participants. On the aggregate level, alignment has improved with more recent LLMs. However, using statistical analyses, we find that LLMs weigh the different factors quite differently from human participants. These results show how curated, challenge datasets combined with insights from cognitive science can help us go beyond comparisons based merely on aggregate metrics: we uncover LLMs implicit tendencies and show to what extent these align with human intuitions.

Xuanfei Ren, Allen Nie, Tengyang Xie et al.

Optimizing complex systems, ranging from LLM prompts to multi-turn agents, traditionally requires labor-intensive manual iteration. We formalize this challenge as a stochastic generative optimization problem where a generative language model acts as the optimizer, guided by numerical rewards and text feedback to discover the best system. We introduce Prioritized Optimization with Local Contextual Aggregation (POLCA), a scalable framework designed to handle stochasticity in optimization -- such as noisy feedback, sampling minibatches, and stochastic system behaviors -- while effectively managing the unconstrained expansion of solution space. POLCA maintains a priority queue to manage the exploration-exploitation tradeoff, systematically tracking candidate solutions and their evaluation histories. To enhance efficiency, we integrate an $\varepsilon$-Net mechanism to maintain parameter diversity and an LLM Summarizer to perform meta-learning across historical trials. We theoretically prove that POLCA converges to near-optimal candidate solutions under stochasticity. We evaluate our framework on diverse benchmarks, including $τ$-bench, HotpotQA (agent optimization), VeriBench (code translation) and KernelBench (CUDA kernel generation). Experimental results demonstrate that POLCA achieves robust, sample and time-efficient performance, consistently outperforming state-of-the-art algorithms in both deterministic and stochastic problems. The codebase for this work is publicly available at https://github.com/rlx-lab/POLCA.

Allen Nie, Xavier Daull, Zhiyi Kuang et al.

Generative optimization uses large language models (LLMs) to iteratively improve artifacts (such as code, workflows or prompts) using execution feedback. It is a promising approach to building self-improving agents, yet in practice remains brittle: despite active research, only 9% of surveyed agents used any automated optimization. We argue that this brittleness arises because, to set up a learning loop, an engineer must make ``hidden'' design choices: What can the optimizer edit and what is the "right" learning evidence to provide at each update? We investigate three factors that affect most applications: the starting artifact, the credit horizon for execution traces, and batching trials and errors into learning evidence. Through case studies in MLAgentBench, Atari, and BigBench Extra Hard, we find that these design decisions can determine whether generative optimization succeeds, yet they are rarely made explicit in prior work. Different starting artifacts determine which solutions are reachable in MLAgentBench, truncated traces can still improve Atari agents, and larger minibatches do not monotonically improve generalization on BBEH. We conclude that the lack of a simple, universal way to set up learning loops across domains is a major hurdle for productionization and adoption. We provide practical guidance for making these choices.

Meghana Kshirsagar, Allen Nie, Ching-An Cheng et al.

We introduce RosettaSearch, an inference-time multi-objective optimization approach for protein sequence optimization. We use large language models (LLMs) as a generative optimizer within a search algorithm capable of controlled exploration and exploitation, using rewards computed from RosettaFold3, a structure prediction model. In a large-scale evaluation, we apply RosettaSearch to 400 suboptimal sequences generated by LigandMPNN (a state-of-the-art model trained for protein sequence design), recovering high-fidelity designs that LigandMPNN's single-pass decoding fails to produce. RosettaSearch's designs show improvements in structural fidelity metrics ranging between 18\% to 68\%, translating to a 2.5$\times$ improvement in design success rate. We observe that these gains in success rate are robust when RosettaSearch-designed sequences are evaluated with an independent structure prediction oracle (Chai-1) and generalize across two distinct LLM families (o4-mini and Gemini-3), with performance scaling consistently with reasoning capability. We further demonstrate that RosettaSearch improves sequence fidelity for ProteinMPNN-designed sequences on \textit{de novo} backbones from the Dayhoff atlas, showing that the approach generalizes beyond native protein structures to computationally generated backbones. We also demonstrate a multi-modal extension of RosettaSearch with vision-language models, where images of predicted protein structures are used as feedback to incorporate structural context to guide protein sequence generation. The sequence trajectories generated by our approach can be used as training data in sequence design models or in post-training and will be released along with the code and datasets upon publication.

Andre He, Nathaniel Weir, Kaj Bostrom et al.

Reinforcement learning with verifiable rewards (RLVR) has emerged as a promising approach for training reasoning language models (RLMs) by leveraging supervision from verifiers. Although verifier implementation is easier than solution annotation for many tasks, existing synthetic data generation methods remain largely solution-centric, while verifier-based methods rely on a few hand-crafted procedural environments. In this work, we scale RLVR by introducing ReSyn, a pipeline that generates diverse reasoning environments equipped with instance generators and verifiers, covering tasks such as constraint satisfaction, algorithmic puzzles, and spatial reasoning. A Qwen2.5-7B-Instruct model trained with RL on ReSyn data achieves consistent gains across reasoning benchmarks and out-of-domain math benchmarks, including a 27\% relative improvement on the challenging BBEH benchmark. Ablations show that verifier-based supervision and increased task diversity both contribute significantly, providing empirical evidence that generating reasoning environments at scale can enhance reasoning abilities in RLMs

Genghan Zhang, Shaowei Zhu, Anjiang Wei et al. · stanford

We present AccelOpt, a self-improving large language model (LLM) agentic system that autonomously optimizes kernels for emerging AI acclerators, eliminating the need for expert-provided hardware-specific optimization knowledge. AccelOpt explores the kernel optimization space through iterative generation, informed by an optimization memory that curates experiences and insights from previously encountered slow-fast kernel pairs. We build NKIBench, a new benchmark suite of AWS Trainium accelerator kernels with varying complexity extracted from real-world LLM workloads to evaluate the effectiveness of AccelOpt. Our evaluation confirms that AccelOpt's capability improves over time, boosting the average percentage of peak throughput from $49\%$ to $61\%$ on Trainium 1 and from $45\%$ to $59\%$ on Trainium 2 for NKIBench kernels. Moreover, AccelOpt is highly cost-effective: using open-source models, it matches the kernel improvements of Claude Sonnet 4 while being $26\times$ cheaper.

Allen Nie, Yash Chandak, Miroslav Suzara et al.

Large language models (LLMs) are quickly being adopted in a wide range of learning experiences, especially via ubiquitous and broadly accessible chat interfaces like ChatGPT and Copilot. This type of interface is readily available to students and teachers around the world, yet relatively little research has been done to assess the impact of such generic tools on student learning. Coding education is an interesting test case, both because LLMs have strong performance on coding tasks, and because LLM-powered support tools are rapidly becoming part of the workflow of professional software engineers. To help understand the impact of generic LLM use on coding education, we conducted a large-scale randomized control trial with 5,831 students from 146 countries in an online coding class in which we provided some students with access to a chat interface with GPT-4. We estimate positive benefits on exam performance for adopters, the students who used the tool, but over all students, the advertisement of GPT-4 led to a significant average decrease in exam participation. We observe similar decreases in other forms of course engagement. However, this decrease is modulated by the student's country of origin. Offering access to LLMs to students from low human development index countries increased their exam participation rate on average. Our results suggest there may be promising benefits to using LLMs in an introductory coding class, but also potential harms for engagement, which makes their longer term impact on student success unclear. Our work highlights the need for additional investigations to help understand the potential impact of future adoption and integration of LLMs into classrooms.

Ching-An Cheng, Andrey Kolobov, Dipendra Misra et al.

We introduce a new benchmark, LLF-Bench (Learning from Language Feedback Benchmark; pronounced as "elf-bench"), to evaluate the ability of AI agents to interactively learn from natural language feedback and instructions. Learning from language feedback (LLF) is essential for people, largely because the rich information this feedback provides can help a learner avoid much of trial and error and thereby speed up the learning process. Large Language Models (LLMs) have recently enabled AI agents to comprehend natural language -- and hence AI agents can potentially benefit from language feedback during learning like humans do. But existing interactive benchmarks do not assess this crucial capability: they either use numeric reward feedback or require no learning at all (only planning or information retrieval). LLF-Bench is designed to fill this omission. LLF-Bench is a diverse collection of sequential decision-making tasks that includes user recommendation, poem writing, navigation, and robot control. The objective of an agent is to interactively solve these tasks based on their natural-language instructions and the feedback received after taking actions. Crucially, to ensure that the agent actually "learns" from the feedback, LLF-Bench implements several randomization techniques (such as paraphrasing and environment randomization) to ensure that the task isn't familiar to the agent and that the agent is robust to various verbalizations. In addition, LLF-Bench provides a unified OpenAI Gym interface for all its tasks and allows the users to easily configure the information the feedback conveys (among suggestion, explanation, and instantaneous performance) to study how agents respond to different types of feedback. Together, these features make LLF-Bench a unique research platform for developing and testing LLF agents.

Anjiang Wei, Allen Nie, Thiago S. F. X. Teixeira et al. · stanford

Modern scientific discovery increasingly relies on high-performance computing for complex modeling and simulation. A key challenge in improving parallel program performance is efficiently mapping tasks to processors and data to memory, a process dictated by intricate, low-level system code known as mappers. Developing high-performance mappers demands days of manual tuning, posing a significant barrier for domain scientists without systems expertise. We introduce a framework that automates mapper development with generative optimization, leveraging richer feedback beyond scalar performance metrics. Our approach features the Agent-System Interface, which includes a Domain-Specific Language (DSL) to abstract away the low-level complexity of system code and define a structured search space, as well as AutoGuide, a mechanism that interprets raw execution output into actionable feedback. Unlike traditional reinforcement learning methods such as OpenTuner, which rely solely on scalar feedback, our method finds superior mappers in far fewer iterations. With just 10 iterations, it outperforms OpenTuner even after 1000 iterations, achieving 3.8X faster performance. Our approach finds mappers that surpass expert-written mappers by up to 1.34X speedup across nine benchmarks while reducing tuning time from days to minutes.

Jiin Woo, Shaowei Zhu, Allen Nie et al.

With the rapid evolution of large language models (LLMs), the demand for automated, high-performance system kernels has emerged as a key enabler for accelerating development and deployment. We introduce TritonRL, a domain-specialized LLM for Triton kernel generation, trained with a novel training framework that enables robust and automated kernel synthesis. Unlike general-purpose programming languages, Triton kernel generation faces unique challenges due to data scarcity and incomplete evaluation criteria, vulnerable to reward hacking. Our approach addresses these challenges end-to-end by distilling Triton-specific knowledge through supervised fine-tuning on curated datasets, and further improving code quality via reinforcement learning (RL) with robust, verifiable rewards and hierarchical reward assignment. Our RL framework robustly detects reward hacking and guides both reasoning traces and code tokens through fine-grained verification and hierarchical reward decomposition, enabling the model to generate high-quality Triton kernels that can truly replace existing modules. With robust and fine-grained evaluation, our experiments on KernelBench demonstrate that TritonRL achieves state-of-the-art correctness and speedup, surpassing all other Triton-specific models and underscoring the effectiveness of our RL-based training paradigm.

Calvin Isley, Joshua Gilbert, Evangelos Kassos et al.

While large language models (LLMs) challenge conventional methods of teaching and learning, they present an exciting opportunity to improve efficiency and scale high-quality instruction. One promising application is the generation of customized exams, tailored to specific course content. There has been significant recent excitement on automatically generating questions using artificial intelligence, but also comparatively little work evaluating the psychometric quality of these items in real-world educational settings. Filling this gap is an important step toward understanding generative AI's role in effective test design. In this study, we introduce and evaluate an iterative refinement strategy for question generation, repeatedly producing, assessing, and improving questions through cycles of LLM-generated critique and revision. We evaluate the quality of these AI-generated questions in a large-scale field study involving 91 classes -- covering computer science, mathematics, chemistry, and more -- in dozens of colleges across the United States, comprising nearly 1700 students. Our analysis, based on item response theory (IRT), suggests that for students in our sample the AI-generated questions performed comparably to expert-created questions designed for standardized exams. Our results illustrate the power of AI to make high-quality assessments more readily available, benefiting both teachers and students.

Wanqiao Xu, Allen Nie, Ruijie Zheng et al.

Interactively learning from observation and language feedback is an increasingly studied area driven by the emergence of large language model (LLM) agents. While impressive empirical demonstrations have been shown, so far a principled framing of these decision problems remains lacking. In this paper, we formalize the Learning from Language Feedback (LLF) problem, assert sufficient assumptions to enable learning despite latent rewards, and introduce $\textit{transfer eluder dimension}$ as a complexity measure to characterize the hardness of LLF problems. We show that transfer eluder dimension captures the intuition that information in the feedback changes the learning complexity of the LLF problem. We demonstrate cases where learning from rich language feedback can be exponentially faster than learning from reward. We develop a no-regret algorithm, called $\texttt{HELiX}$, that provably solves LLF problems through sequential interactions, with performance guarantees that scale with the transfer eluder dimension of the problem. Across several empirical domains, we show that $\texttt{HELiX}$ performs well even when repeatedly prompting LLMs does not work reliably. Our contributions mark a first step towards designing principled interactive learning algorithms from generic language feedback.

Zhiyi Kuang, Ryan Rong, YuCheng Yuan et al.

We present a generative optimization approach for learning game-playing agents, where policies are represented as Python programs and refined using large language models (LLMs). Our method treats decision-making policies as self-evolving code, with current observation as input and an in-game action as output, enabling agents to self-improve through execution traces and natural language feedback with minimal human intervention. Applied to Atari games, our game-playing Python program achieves performance competitive with deep reinforcement learning (RL) baselines while using significantly less training time and much fewer environment interactions. This work highlights the promise of programmatic policy representations for building efficient, adaptable agents capable of complex, long-horizon reasoning.

Tianwei Ni, Allen Nie, Sapana Chaudhary et al.

Leveraging inference-time search in large language models has proven effective in further enhancing a trained model's capability to solve complex mathematical and reasoning problems. However, this approach significantly increases computational costs and inference time, as the model must generate and evaluate multiple candidate solutions to identify a viable reasoning path. To address this, we propose an effective approach that integrates search capabilities directly into the model by fine-tuning it on unpaired successful (learning) and failed reasoning paths (forgetting) derived from diverse search methods. A key challenge we identify is that naive fine-tuning can degrade the model's search capability; we show this can be mitigated with a smaller learning rate. Extensive experiments on the challenging Game-of-24 and Countdown arithmetic puzzles show that, replacing CoT-generated data with search-generated data for offline fine-tuning improves success rates by around 23% over inference-time search baselines, while reducing inference time by 180$\times$. On top of this, our learning and forgetting objective consistently outperforms both supervised fine-tuning and preference-based methods.

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.

Ching-An Cheng, Allen Nie, Adith Swaminathan

We study a class of optimization problems motivated by automating the design and update of AI systems like coding assistants, robots, and copilots. AutoDiff frameworks, like PyTorch, enable efficient end-to-end optimization of differentiable systems. However, general computational workflows can be non-differentiable and involve rich feedback (e.g. console output or user's responses), heterogeneous parameters (e.g. prompts, codes), and intricate objectives (beyond maximizing a score). We investigate end-to-end generative optimization -- using generative models such as LLMs within the optimizer for automatic updating of general computational workflows. We discover that workflow execution traces are akin to back-propagated gradients in AutoDiff and can provide key information to interpret feedback for efficient optimization. Formally, we frame a new mathematical setup, Optimization with Trace Oracle (OPTO). In OPTO, an optimizer receives an execution trace along with feedback on the computed output and updates parameters iteratively. We provide a Python library, Trace, that efficiently converts a workflow optimization problem into an OPTO instance using PyTorch-like syntax. Using Trace, we develop a general LLM-based generative optimizer called OptoPrime. In empirical studies, we find that OptoPrime is capable of first-order numerical optimization, prompt optimization, hyper-parameter tuning, robot controller design, code debugging, etc., and is often competitive with specialized optimizers for each domain. We envision Trace as an open research platform for devising novel generative optimizers and developing the next generation of interactive learning agents. Website: https://microsoft.github.io/Trace/.

Allen Nie, Emma Brunskill, Chris Piech

Contemporary coding education often presents students with the task of developing programs that have user interaction and complex dynamic systems, such as mouse based games. While pedagogically compelling, there are no contemporary autonomous methods for providing feedback. Notably, interactive programs are impossible to grade by traditional unit tests. In this paper we formalize the challenge of providing feedback to interactive programs as a task of classifying Markov Decision Processes (MDPs). Each student's program fully specifies an MDP where the agent needs to operate and decide, under reasonable generalization, if the dynamics and reward model of the input MDP should be categorized as correct or broken. We demonstrate that by designing a cooperative objective between an agent and an autoregressive model, we can use the agent to sample differential trajectories from the input MDP that allows a classifier to determine membership: Play to Grade. Our method enables an automatic feedback system for interactive code assignments. We release a dataset of 711,274 anonymized student submissions to a single assignment with hand-coded bug labels to support future research.

Rishi Bommasani, Drew A. Hudson, Ehsan Adeli et al.

AI is undergoing a paradigm shift with the rise of models (e.g., BERT, DALL-E, GPT-3) that are trained on broad data at scale and are adaptable to a wide range of downstream tasks. We call these models foundation models to underscore their critically central yet incomplete character. This report provides a thorough account of the opportunities and risks of foundation models, ranging from their capabilities (e.g., language, vision, robotics, reasoning, human interaction) and technical principles(e.g., model architectures, training procedures, data, systems, security, evaluation, theory) to their applications (e.g., law, healthcare, education) and societal impact (e.g., inequity, misuse, economic and environmental impact, legal and ethical considerations). Though foundation models are based on standard deep learning and transfer learning, their scale results in new emergent capabilities,and their effectiveness across so many tasks incentivizes homogenization. Homogenization provides powerful leverage but demands caution, as the defects of the foundation model are inherited by all the adapted models downstream. Despite the impending widespread deployment of foundation models, we currently lack a clear understanding of how they work, when they fail, and what they are even capable of due to their emergent properties. To tackle these questions, we believe much of the critical research on foundation models will require deep interdisciplinary collaboration commensurate with their fundamentally sociotechnical nature.

Allen Nie, Reuben Cohn-Gordon, Christopher Potts

Image captioning systems have recently improved dramatically, but they still tend to produce captions that are insensitive to the communicative goals that captions should meet. To address this, we propose Issue-Sensitive Image Captioning (ISIC). In ISIC, a captioning system is given a target image and an issue, which is a set of images partitioned in a way that specifies what information is relevant. The goal of the captioner is to produce a caption that resolves this issue. To model this task, we use an extension of the Rational Speech Acts model of pragmatic language use. Our extension is built on top of state-of-the-art pretrained neural image captioners and explicitly reasons about issues in our sense. We establish experimentally that these models generate captions that are both highly descriptive and issue-sensitive, and we show how ISIC can complement and enrich the related task of Visual Question Answering.

Allen Nie, Arturo L. Pineda, Matt W. Wright Hannah Wand et al.

As genetic sequencing costs decrease, the lack of clinical interpretation of variants has become the bottleneck in using genetics data. A major rate limiting step in clinical interpretation is the manual curation of evidence in the genetic literature by highly trained biocurators. What makes curation particularly time-consuming is that the curator needs to identify papers that study variant pathogenicity using different types of approaches and evidences---e.g. biochemical assays or case control analysis. In collaboration with the Clinical Genomic Resource (ClinGen)---the flagship NIH program for clinical curation---we propose the first machine learning system, LitGen, that can retrieve papers for a particular variant and filter them by specific evidence types used by curators to assess for pathogenicity. LitGen uses semi-supervised deep learning to predict the type of evidence provided by each paper. It is trained on papers annotated by ClinGen curators and systematically evaluated on new test data collected by ClinGen. LitGen further leverages rich human explanations and unlabeled data to gain 7.9%-12.6% relative performance improvement over models learned only on the annotated papers. It is a useful framework to improve clinical variant curation.

Allen Nie, Erin D. Bennett, Noah D. Goodman

Providing plausible responses to why questions is a challenging but critical goal for language based human-machine interaction. Explanations are challenging in that they require many different forms of abstract knowledge and reasoning. Previous work has either relied on human-curated structured knowledge bases or detailed domain representation to generate satisfactory explanations. They are also often limited to ranking pre-existing explanation choices. In our work, we contribute to the under-explored area of generating natural language explanations for general phenomena. We automatically collect large datasets of explanation-phenomenon pairs which allow us to train sequence-to-sequence models to generate natural language explanations. We compare different training strategies and evaluate their performance using both automatic scores and human ratings. We demonstrate that our strategy is sufficient to generate highly plausible explanations for general open-domain phenomena compared to other models trained on different datasets.

Yuhui Zhang, Allen Nie, James Zou

Supervised learning is limited both by the quantity and quality of the labeled data. In the field of medical record tagging, writing styles between hospitals vary drastically. The knowledge learned from one hospital might not transfer well to another. This problem is amplified in veterinary medicine domain because veterinary clinics rarely apply medical codes to their records. We proposed and trained the first large-scale generative modeling algorithm in automated disease coding. We demonstrate that generative modeling can learn discriminative features when additionally trained with supervised fine-tuning. We systematically ablate and evaluate the effect of generative modeling on the final system's performance. We compare the performance of our model with several baselines in a challenging cross-hospital setting with substantial domain shift. We outperform competitive baselines by a large margin. In addition, we provide interpretation for what is learned by our model.

Allen Nie, Ashley Zehnder, Rodney L. Page et al.

Large scale veterinary clinical records can become a powerful resource for patient care and research. However, clinicians lack the time and resource to annotate patient records with standard medical diagnostic codes and most veterinary visits are captured in free text notes. The lack of standard coding makes it challenging to use the clinical data to improve patient care. It is also a major impediment to cross-species translational research, which relies on the ability to accurately identify patient cohorts with specific diagnostic criteria in humans and animals. In order to reduce the coding burden for veterinary clinical practice and aid translational research, we have developed a deep learning algorithm, DeepTag, which automatically infers diagnostic codes from veterinary free text notes. DeepTag is trained on a newly curated dataset of 112,558 veterinary notes manually annotated by experts. DeepTag extends multi-task LSTM with an improved hierarchical objective that captures the semantic structures between diseases. To foster human-machine collaboration, DeepTag also learns to abstain in examples when it is uncertain and defers them to human experts, resulting in improved performance. DeepTag accurately infers disease codes from free text even in challenging cross-hospital settings where the text comes from different clinical settings than the ones used for training. It enables automated disease annotation across a broad range of clinical diagnoses with minimal pre-processing. The technical framework in this work can be applied in other medical domains that currently lack medical coding resources.

Allen Nie, Erin D. Bennett, Noah D. Goodman

Learning effective representations of sentences is one of the core missions of natural language understanding. Existing models either train on a vast amount of text, or require costly, manually curated sentence relation datasets. We show that with dependency parsing and rule-based rubrics, we can curate a high quality sentence relation task by leveraging explicit discourse relations. We show that our curated dataset provides an excellent signal for learning vector representations of sentence meaning, representing relations that can only be determined when the meanings of two sentences are combined. We demonstrate that the automatically curated corpus allows a bidirectional LSTM sentence encoder to yield high quality sentence embeddings and can serve as a supervised fine-tuning dataset for larger models such as BERT. Our fixed sentence embeddings achieve high performance on a variety of transfer tasks, including SentEval, and we achieve state-of-the-art results on Penn Discourse Treebank's implicit relation prediction task.