Jason Liu

Marc Walden, Jason Liu, Shaashwath Sivakumar et al.

We investigate multi-agent deep reinforcement learning and propose two enhancements to the Multi-Agent Deep Deterministic Policy Gradient (MADDPG) algorithm. First, we introduce a novel Action Inference mechanism that enables each agent to predict other agents' intended actions, thereby improving the accuracy and stability of its own policy. Second, we apply an importance sampling strategy, using geometric distribution, in the replay buffer to prioritize more recent and informative experiences, which helps mitigate the non-stationarity inherent in multi-agent environments. We evaluate both modifications on the discrete-action Predator-Prey task provided by the PettingZoo library, a flexible Python interface for general multi-agent reinforcement learning benchmarks. Our results indicate that Action Inference is effective in improving learning stability and inter-agent cooperation and that importance sampling using geometric distribution can lead to significant improvements in exploration efficiency over standard MADDPG. Code available at https://github.com/shaashwathsivakumar/MARL_Proj

Boyuan Guan, Jason Liu, Yanzhao Wu et al.

Embedding LLM-driven agents into environmental FAIR data management is compelling - they can externalize operational knowledge and scale curation across heterogeneous data and evolving conventions. However, replacing deterministic components with probabilistic workflows changes the failure mode: LLM pipelines may generate plausible but incorrect outputs that pass superficial checks and propagate into irreversible actions such as DOI minting and public release. We introduce EnviSmart, a production data management system deployed on campus-wide storage infrastructure for environmental research. EnviSmart treats reliability as an architectural property through two mechanisms: a three-track knowledge architecture that externalizes behaviors (governance constraints), domain knowledge (retrievable context), and skills (tool-using procedures) as persistent, interlocking artifacts; and a role-separated multi-agent design where deterministic validators and audited handoffs restore fail-stop semantics at trust boundaries before irreversible steps. We compare two production deployments. The University's GIS Center Ecological Archive (849 curated datasets) serves as a single-agent baseline. SF2Bench, a compound flooding benchmark comprising 2,452 monitoring stations and 8,557 published files spanning 39 years, validates the multi-agent workflow. The multi-agent approach improved both efficiency - completed by a single operator in two days with repeated artifact reuse across deployments - and reliability: audited handoffs detected and blocked a coordinate transformation error affecting all 2,452 stations before publication. A representative incident (ISS-004) demonstrated boundary-based containment with 10-minute detection latency, zero user exposure, and 80-minute resolution. This paper has been accepted at PEARC 2026.

Hamed Najafi, Gareth Lagerwall, Jayantha Obeysekera et al.

Deep super-resolution networks for precipitation downscaling achieve strong bulk skill yet systematically under-predict the heavy-tail events that drive flood risk. We demonstrate that the primary obstacle is the loss function, not the data: under intensity-weighted MAE, real and synthetic labels at the same input are simply averaged, meaning data augmentation shifts the predicted mean rather than the conditional distribution. We resolve this with Q-SRDRN, a multi-quantile super-resolution network trained with pinball loss at tau in 0.50, 0.95, 0.99, 0.999. Two CNN-specific design choices make this practical: IncrementBound enforces monotonicity while preserving each quantile channel's gradient identity, and separate per-quantile output heads provide independent filter banks for bulk and tail detection. Under this design, data augmentation via cVAE becomes complementary: the median head absorbs synthetic patterns without contaminating upper quantiles. Empirically, on Florida (convective/tropical-cyclone dominated), the un-augmented Q-SRDRN P999 head detects 1,598 of 2,111 events at 200 mm/day versus 88 for the deterministic baseline--an 18x detection-rate gain (4.2% to 75.7%)--with 63% lower KL divergence and 3.9% lower RMSE. Adding cVAE-generated samples lifts the P50 channel from 14 to 1,038 hits at 200 mm/day. On California (atmospheric-river dominated), the architecture reaches near-perfect detection (P999 SEDI >= 0.996 through 300 mm/day). On Texas, the baseline catches only 2 of 10,720 events at 200 mm/day while the P999 head catches 8,776 (81.9%). While the cVAE does not transfer across regions, multi-quantile regression captures extremes wherever the large-scale signal is strong, while augmentation rescues the median where it is not.

Michael Pradel, Georgios Gousios, Jason Liu et al.

Maintaining large code bases written in dynamically typed languages, such as JavaScript or Python, can be challenging due to the absence of type annotations: simple data compatibility errors proliferate, IDE support is limited, and APIs are hard to comprehend. Recent work attempts to address those issues through either static type inference or probabilistic type prediction. Unfortunately, static type inference for dynamic languages is inherently limited, while probabilistic approaches suffer from imprecision. This paper presents TypeWriter, the first combination of probabilistic type prediction with search-based refinement of predicted types. TypeWriter's predictor learns to infer the return and argument types for functions from partially annotated code bases by combining the natural language properties of code with programming language-level information. To validate predicted types, TypeWriter invokes a gradual type checker with different combinations of the predicted types, while navigating the space of possible type combinations in a feedback-directed manner. We implement the TypeWriter approach for Python and evaluate it on two code corpora: a multi-million line code base at Facebook and a collection of 1,137 popular open-source projects. We show that TypeWriter's type predictor achieves an F1 score of 0.64 (0.79) in the top-1 (top-5) predictions for return types, and 0.57 (0.80) for argument types, which clearly outperforms prior type prediction models. By combining predictions with search-based validation, TypeWriter can fully annotate between 14% to 44% of the files in a randomly selected corpus, while ensuring type correctness. A comparison with a static type inference tool shows that TypeWriter adds many more non-trivial types. TypeWriter currently suggests types to developers at Facebook and several thousands of types have already been accepted with minimal changes.

Hamed Najafi, Dongsheng Luo, Jason Liu

Explainable AI (XAI) is critical for building trust in complex machine learning models, yet mainstream attribution methods often provide an incomplete, static picture of a model's final state. By collapsing a feature's role into a single score, they are confounded by non-linearity and interactions. To address this, we introduce Feature-Function Curvature Analysis (FFCA), a novel framework that analyzes the geometry of a model's learned function. FFCA produces a 4-dimensional signature for each feature, quantifying its: (1) Impact, (2) Volatility, (3) Non-linearity, and (4) Interaction. Crucially, we extend this framework into Dynamic Archetype Analysis, which tracks the evolution of these signatures throughout the training process. This temporal view moves beyond explaining what a model learned to revealing how it learns. We provide the first direct, empirical evidence of hierarchical learning, showing that models consistently learn simple linear effects before complex interactions. Furthermore, this dynamic analysis provides novel, practical diagnostics for identifying insufficient model capacity and predicting the onset of overfitting. Our comprehensive experiments demonstrate that FFCA, through its static and dynamic components, provides the essential geometric context that transforms model explanation from simple quantification to a nuanced, trustworthy analysis of the entire learning process.

Ziyang Li, Jiani Huang, Jason Liu et al.

Foundation models have vast potential to enable diverse AI applications. The powerful yet incomplete nature of these models has spurred a wide range of mechanisms to augment them with capabilities such as in-context learning, information retrieval, and code interpreting. We propose Vieira, a declarative framework that unifies these mechanisms in a general solution for programming with foundation models. Vieira follows a probabilistic relational paradigm and treats foundation models as stateless functions with relational inputs and outputs. It supports neuro-symbolic applications by enabling the seamless combination of such models with logic programs, as well as complex, multi-modal applications by streamlining the composition of diverse sub-models. We implement Vieira by extending the Scallop compiler with a foreign interface that supports foundation models as plugins. We implement plugins for 12 foundation models including GPT, CLIP, and SAM. We evaluate Vieira on 9 challenging tasks that span language, vision, and structured and vector databases. Our evaluation shows that programs in Vieira are concise, can incorporate modern foundation models, and have comparable or better accuracy than competitive baselines.

Caojin Zhang, Qiang Zhang, Ke Li et al.

Large decoder-only language models (LLMs) have achieved remarkable success in generation and reasoning tasks, where they generate text responses given instructions. However, many applications, e.g., retrieval augmented generation (RAG), still rely on separate embedding models to generate text embeddings, which can complicate the system and introduce discrepancies in understanding of the query between the embedding model and LLMs. To address this limitation, we propose a simple self-supervised approach, Generative Embedding large language Model (GEM), that enables any large decoder-only LLM to generate high-quality text embeddings while maintaining its original text generation and reasoning capabilities. Our method inserts new special token(s) into a text body, and generates summarization embedding of the text by manipulating the attention mask. This method could be easily integrated into post-training or fine tuning stages of any existing LLMs. We demonstrate the effectiveness of our approach by applying it to two popular LLM families, ranging from 1B to 8B parameters, and evaluating the transformed models on both text embedding benchmarks (MTEB) and NLP benchmarks (MMLU). The results show that our proposed method significantly improves the original LLMs on MTEB while having a minimal impact on MMLU. Our strong results indicate that our approach can empower LLMs with state-of-the-art text embedding capabilities while maintaining their original NLP performance

Xu Zheng, Chaohao Lin, Sipeng Chen et al.

Forecasting compound floods presents a significant challenge due to the intricate interplay of meteorological, hydrological, and oceanographic factors. Analyzing compound floods has become more critical as the global climate increases flood risks. Traditional physics-based methods, such as the Hydrologic Engineering Center's River Analysis System, are often time-inefficient. Machine learning has recently demonstrated promise in both modeling accuracy and computational efficiency. However, the scarcity of comprehensive datasets currently hinders systematic analysis. Existing water-related datasets are often limited by a sparse network of monitoring stations and incomplete coverage of relevant factors. To address this challenge, we introduce SF2Bench, a comprehensive time series collection on compound floods in South Florida, which integrates four key factors: tide, rainfall, groundwater, and human management activities (gate and pump controlling). This integration allows for a more detailed analysis of the individual contributions of these drivers to compound flooding and informs the development of improved flood forecasting approaches. To comprehensively evaluate the potential of various modeling paradigms, we assess the performance of six categories of methods, encompassing Multilayer Perceptrons, Convolutional Neural Networks, Recurrent Neural Networks, Graph Neural Networks, Transformers, and Large Language Models. We verified the impact of different key features on flood forecasting through experiments. Our analysis examines temporal and spatial aspects, providing insights into the influence of historical data and spatial dependencies. The varying performance across these approaches underscores the diverse capabilities of each in capturing complex temporal and spatial dependencies inherent in compound floods.

Fabrice Y Harel-Canada, Boran Erol, Connor Choi et al.

Watermarking AI-generated text is critical for combating misuse. Yet recent theoretical work argues that any watermark can be erased via random walk attacks that perturb text while preserving quality. However, such attacks rely on two key assumptions: (1) rapid mixing (watermarks dissolve quickly under perturbations) and (2) reliable quality preservation (automated quality oracles perfectly guide edits). Through large-scale experiments and human-validated assessments, we find mixing is slow: 100% of perturbed texts retain traces of their origin after hundreds of edits, defying rapid mixing. Oracles falter, as state-of-the-art quality detectors misjudge edits (77% accuracy), compounding errors during attacks. Ultimately, attacks underperform: automated walks remove watermarks just 26% of the time -- dropping to 10% under human quality review. These findings challenge the inevitability of watermark removal. Instead, practical barriers -- slow mixing and imperfect quality control -- reveal watermarking to be far more robust than theoretical models suggest. The gap between idealized attacks and real-world feasibility underscores the need for stronger watermarking methods and more realistic attack models.

Jason Liu

A critical component when developing question-answering AIs is an adversarial dataset that challenges models to adapt to the complex syntax and reasoning underlying our natural language. Present techniques for procedurally generating adversarial texts are not robust enough for training on complex tasks such as answering multi-sentence trivia questions. We instead turn to human-generated data by introducing an interface for collecting adversarial human-written trivia questions. Our interface is aimed towards question writers and players of Quiz Bowl, a buzzer-based trivia competition where paragraph-long questions consist of a sequence of clues of decreasing difficulty. To incentivize usage, a suite of machine learning-based tools in our interface assist humans in writing questions that are more challenging to answer for Quiz Bowl players and computers alike. Not only does our interface gather training data for the groundbreaking Quiz Bowl AI project QANTA, but it is also a proof-of-concept of future adversarial data collection for question-answering systems. The results of performance-testing our interface with ten originally-composed questions indicate that, despite some flaws, our interface's novel question-writing features as well as its real-time exposure of useful responses from our machine models could facilitate and enhance the collection of adversarial questions.

Jason Liu, Shohreh Deldari, Hao Xue et al.

In the context of mobile sensing environments, various sensors on mobile devices continually generate a vast amount of data. Analyzing this ever-increasing data presents several challenges, including limited access to annotated data and a constantly changing environment. Recent advancements in self-supervised learning have been utilized as a pre-training step to enhance the performance of conventional supervised models to address the absence of labelled datasets. This research examines the impact of using a self-supervised representation learning model for time series classification tasks in which data is incrementally available. We proposed and evaluated a workflow in which a model learns to extract informative features using a corpus of unlabeled time series data and then conducts classification on labelled data using features extracted by the model. We analyzed the effect of varying the size, distribution, and source of the unlabeled data on the final classification performance across four public datasets, including various types of sensors in diverse applications.

Monica Roy, Kaiyu Zheng, Jason Liu et al.

We envision robots that can collaborate and communicate seamlessly with humans. It is necessary for such robots to decide both what to say and how to act, while interacting with humans. To this end, we introduce a new task, dialogue object search: A robot is tasked to search for a target object (e.g. fork) in a human environment (e.g., kitchen), while engaging in a "video call" with a remote human who has additional but inexact knowledge about the target's location. That is, the robot conducts speech-based dialogue with the human, while sharing the image from its mounted camera. This task is challenging at multiple levels, from data collection, algorithm and system development,to evaluation. Despite these challenges, we believe such a task blocks the path towards more intelligent and collaborative robots. In this extended abstract, we motivate and introduce the dialogue object search task and analyze examples collected from a pilot study. We then discuss our next steps and conclude with several challenges on which we hope to receive feedback.

Zhanlin Chen, Jeremy Goldwasser, Philip Tuckman et al.

In the era of single-cell sequencing, there is a growing need to extract insights from data with clustering methods. Here, we introduce Forest Fire Clustering, an efficient and interpretable method for cell-type discovery from single-cell data. Forest Fire Clustering makes minimal prior assumptions and, different from current approaches, calculates a non-parametric posterior probability that each cell is assigned a cell-type label. These posterior distributions allow for the evaluation of a label confidence for each cell and enable the computation of "label entropies," highlighting transitions along developmental trajectories. Furthermore, we show that Forest Fire Clustering can make robust, inductive inferences in an online-learning context and can readily scale to millions of cells. Finally, we demonstrate that our method outperforms state-of-the-art clustering approaches on diverse benchmarks of simulated and experimental data. Overall, Forest Fire Clustering is a useful tool for rare cell type discovery in large-scale single-cell analysis.