16 citations · 41 across the 21 of their papers we have counts for
23 papers
Learning Strategies to Break Judges
Guruprerana Shabadi, Aaditya Naik, Rajeev Alur +1
As AI agents surpass human performance, it becomes exceedingly hard for system designers to evaluate them directly and understand their failure modes. Consequently, agents themselv…
The Output-Space Hypothesis: Enumerative Equivalence Checking for Tensor Programs
Paul Biberstein, Joseph Devietti, Mayur Naik
Tensor programs, as used in deep learning models, are a prime target for optimization, as small performance improvements can have a large impact across training or inference worklo…
Detecting Safety Violations Across Many Agent Traces
Adam Stein, Davis Brown, Hamed Hassani +2
To identify safety violations, auditors often search over large sets of agent traces. This search is difficult because failures are often rare, complex, and sometimes even adversar…
Do We Need Frontier Models to Verify Mathematical Proofs?
Aaditya Naik, Guruprerana Shabadi, Rajeev Alur +1
Advances in training, post-training, and inference-time methods have enabled frontier reasoning models to win gold medals in math competitions and settle challenging open problems.…
MathDuels: A Self-Play Benchmark That Grows
Zhiqiu Xu, Shibo Jin, Shreya Arya +1
As frontier language models attain near-ceiling performance on static mathematical benchmarks, existing evaluations are increasingly unable to differentiate model capabilities, lar…
CAMEL: An ECG Language Model for Forecasting Cardiac Events
Neelay Velingker, Alaia Solko-Breslin, Mayank Keoliya +9
Electrocardiograms (ECG) are electrical recordings of the heart that are critical for diagnosing cardiovascular conditions. ECG language models (ELMs) have recently emerged as a pr…
On Improving Neurosymbolic Learning by Exploiting the Representation Space
Aaditya Naik, Efthymia Tsamoura, Shibo Jin +2
We study the problem of learning neural classifiers in a neurosymbolic setting where the hidden gold labels of input instances must satisfy a logical formula. Learning in this sett…
QLCoder: A Query Synthesizer For Static Analysis of Security Vulnerabilities
Claire Wang, Ziyang Li, Saikat Dutta +1
Static analysis tools provide a powerful means to detect security vulnerabilities by specifying queries that encode vulnerable code patterns. However, writing such queries is chall…
ESCA: Contextualizing Embodied Agents via Scene-Graph Generation
Jiani Huang, Amish Sethi, Matthew Kuo +6
Multi-modal large language models (MLLMs) are making rapid progress toward general-purpose embodied agents. However, existing MLLMs do not reliably capture fine-grained links betwe…
Once Upon an Input: Reasoning via Per-Instance Program Synthesis
Adam Stein, Neelay Velingker, Mayur Naik +1
Large language models (LLMs) excel at zero-shot inference but continue to struggle with complex, multi-step reasoning. Recent methods that augment LLMs with intermediate reasoning…
Stable Prediction of Adverse Events in Medical Time-Series Data
Mayank Keoliya, Seewon Choi, Rajeev Alur +2
Early event prediction (EEP) systems continuously estimate a patient's imminent risk to support clinical decision-making. For bedside trust, risk trajectories must be accurate and…
The Road to Generalizable Neuro-Symbolic Learning Should be Paved with Foundation Models
Adam Stein, Aaditya Naik, Neelay Velingker +2
Neuro-symbolic learning was proposed to address challenges with training neural networks for complex reasoning tasks with the added benefits of interpretability, reliability, and e…
Lobster: A GPU-Accelerated Framework for Neurosymbolic Programming
Paul Biberstein, Ziyang Li, Joseph Devietti +1
Neurosymbolic programs combine deep learning with symbolic reasoning to achieve better data efficiency, interpretability, and generalizability compared to standalone deep learning…
Where's the Bug? Attention Probing for Scalable Fault Localization
Adam Stein, Arthur Wayne, Aaditya Naik +2
Ensuring code correctness remains a challenging problem even as large language models (LLMs) become increasingly capable at code-related tasks. While LLM-based program repair syste…
Advances in RNA secondary structure prediction and RNA modifications: Methods, data, and applications
Shu Yang, Nhat Truong Pham, Ziyang Li +10
Due to the hierarchical organization of RNA structures and their pivotal roles in fulfilling RNA functions, the formation of RNA secondary structure critically influences many biol…
Relational Programming with Foundation Models
Ziyang Li, Jiani Huang, Jason Liu +6
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 w…
Dolphin: A Programmable Framework for Scalable Neurosymbolic Learning
Aaditya Naik, Jason Liu, Claire Wang +4
Neurosymbolic learning enables the integration of symbolic reasoning with deep learning but faces significant challenges in scaling to complex symbolic programs, large datasets, or…
Towards Compositionality in Concept Learning
Adam Stein, Aaditya Naik, Yinjun Wu +2
Concept-based interpretability methods offer a lens into the internals of foundation models by decomposing their embeddings into high-level concepts. These concept representations…
DISCRET: Synthesizing Faithful Explanations For Treatment Effect Estimation
Yinjun Wu, Mayank Keoliya, Kan Chen +7
Designing faithful yet accurate AI models is challenging, particularly in the field of individual treatment effect estimation (ITE). ITE prediction models deployed in critical sett…
Data-Efficient Learning with Neural Programs
Alaia Solko-Breslin, Seewon Choi, Ziyang Li +4
Many computational tasks can be naturally expressed as a composition of a DNN followed by a program written in a traditional programming language or an API call to an LLM. We call…
IRIS: LLM-Assisted Static Analysis for Detecting Security Vulnerabilities
Ziyang Li, Saikat Dutta, Mayur Naik
Software is prone to security vulnerabilities. Program analysis tools to detect them have limited effectiveness in practice due to their reliance on human labeled specifications. L…
Understanding the Effectiveness of Large Language Models in Detecting Security Vulnerabilities
Avishree Khare, Saikat Dutta, Ziyang Li +3
While automated vulnerability detection techniques have made promising progress in detecting security vulnerabilities, their scalability and applicability remain challenging. The r…
TorchQL: A Programming Framework for Integrity Constraints in Machine Learning
Aaditya Naik, Adam Stein, Yinjun Wu +2
Finding errors in machine learning applications requires a thorough exploration of their behavior over data. Existing approaches used by practitioners are often ad-hoc and lack the…