Jialin Song — accepted AI-conference papers

2026

FluidGaussian: Propagating Simulation-Based Uncertainty Toward Functionally-Intelligent 3D Reconstruction

Yuqiu Liu, Jialin Song, Marissa Ramirez de Chanlatte, Rochishnu Chowdhury, Rushil Paresh Desai, Wuyang Chen +2

CVPR 2026

Real objects that inhabit the physical world follow physical laws and thus behave plausibly during interaction with other physical objects. However, current methods that perform 3D reconstructions of real-world scenes from multi-view 2D images optimize primarily for visual fidelity, i.e., they train

Cited by 0 Source ↗code ↗Scholar ↗

2026

MultiBreak: A Scalable and Diverse Multi-turn Jailbreak Benchmark for Evaluating LLM Safety

Jialin Song, Xiaodong Liu, Weiwei Yang, Wuyang Chen, Mingqian Feng, Xuekai Zhu +1

ICML 2026poster

We present MultiBreak, a scalable and diverse multi-turn jailbreak benchmark to evaluate large language model (LLM) safety. Multi-turn jailbreaks mimic natural conversational settings, making them easier to bypass safety-aligned LLM than single-turn jailbreaks. Existing multi-turn benchmarks are lim…

Cited by 0 Source ↗Scholar ↗

2026

SEMA: Simple yet Effective Learning for Multi-Turn Jailbreak Attacks

Mingqian Feng, Xiaodong Liu, Weiwei Yang, Jialin Song, Xuekai Zhu, Chenliang Xu +1

ICLR 2026poster

Multi-turn jailbreaks capture the real threat model for safety-aligned chatbots, where single-turn attacks are merely a special case. Yet existing approaches break under exploration complexity and intent drift. We propose SEMA, a simple yet effective framework that trains a multi-turn attacker witho…

Cited by 0 Source ↗code ↗Scholar ↗

2025

Beyond Single Frames: Can LMMs Comprehend Implicit Narratives in Comic Strip?

Xiaochen Wang, Heming Xia, Jialin Song, Longyu Guan, Qingxiu Dong, Rui Li +7

EMNLP 2025

Large Multimodal Models (LMMs) have demonstrated strong performance on vision-language benchmarks, yet current evaluations predominantly focus on single-image reasoning. In contrast, real-world scenarios always involve understanding sequences of images. A typical scenario is comic strips understandi

Cited by 0 Source ↗PDF ↗Scholar ↗

2025

Nemotron-CORTEXA: Enhancing LLM Agents for Software Engineering Tasks via Improved Localization and Solution Diversity

Atefeh Sohrabizadeh, Jialin Song, Mingjie Liu, Rajarshi Roy, Chankyu Lee, Jonathan Raiman +1

ICML 2025poster

Large Language Models (LLMs) have demonstrated significant potential in code generation by following natural language instructions. Unfortunately, crucial real-world software engineering tasks, such as debugging or repository-level feature implementation, involve processing extensive contexts beyon…

Cited by 0 Source ↗PDF ↗Scholar ↗

2025

PDE-Controller: LLMs for Autoformalization and Reasoning of PDEs

Mauricio Soroco, Jialin Song, Mengzhou Xia, Kye Emond, Weiran Sun, Wuyang Chen

ICML 2025poster

We present PDE-Controller, a framework that enables large language models (LLMs) to control systems governed by partial differential equations (PDEs). Traditional LLMs have excelled in commonsense reasoning but fall short in rigorous logical reasoning. While recent AI-for-math has made strides in pu…

Cited by 1 Source ↗PDF ↗Scholar ↗

2024

Data-Efficient Operator Learning via Unsupervised Pretraining and In-Context Learning

Wuyang Chen, Jialin Song, Pu Ren, Shashank Subramanian, Dmitriy Morozov, Michael W. Mahoney

NeurIPS 2024poster

Recent years have witnessed the promise of coupling machine learning methods and physical domain-specific insights for solving scientific problems based on partial differential equations (PDEs). However, being data-intensive, these methods still require a large amount of PDE data. This reintroduces…

Cited by 12 Source ↗PDF ↗code ↗Scholar ↗

2023

Learning Regions of Interest for Bayesian Optimization with Adaptive Level-Set Estimation

Fengxue Zhang, Jialin Song, James C Bowden, Alexander Ladd, Yisong Yue, Thomas Desautels +1

ICML 2023poster

We study Bayesian optimization (BO) in high-dimensional and non-stationary scenarios. Existing algorithms for such scenarios typically require extensive hyperparameter tuning, which limits their practical effectiveness. We propose a framework, called BALLET, which adaptively filters for a high-confi…

Cited by 8 Source ↗PDF ↗Scholar ↗

2022

MLNav: Learning to Safely Navigate on Martian Terrains

Shreyansh Daftry, Neil Abcouwer, Tyler del Sesto, Siddarth Venkatraman, Jialin Song, Lucas Igel +4

RA-L 2022

We present <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">MLNav</i> , a learning-enhanced path planning framework for safety-critical and resource-limited systems operating in complex environments, such as rovers navigating on Mars. MLNav makes judi

Cited by 23 Source ↗Scholar ↗

2021

Learning to Make Decisions via Submodular Regularization

Ayya Alieva, Aiden Aceves, Jialin Song, Stephen Mayo, Yisong Yue, Yuxin Chen

ICLR 2021poster

Many sequential decision making tasks can be viewed as combinatorial optimization problems over a large number of actions. When the cost of evaluating an action is high, even a greedy algorithm, which iteratively picks the best action given the history, is prohibitive to run. In this paper, we aim t…

Cited by 14 Source ↗PDF ↗Scholar ↗

2020

A General Large Neighborhood Search Framework for Solving Integer Linear Programs

Jialin Song, ravi lanka, Yisong Yue, Bistra Dilkina

NeurIPS 2020poster

This paper studies how to design abstractions of large-scale combinatorial optimization problems that can leverage existing state-of-the-art solvers in general-purpose ways, and that are amenable to data-driven design. The goal is to arrive at new approaches that can reliably outperform existing so…

Cited by 91 Source ↗PDF ↗Scholar ↗

2019

A General Framework for Multi-fidelity Bayesian Optimization with Gaussian Processes

Jialin Song, Yuxin Chen, Yisong Yue

AISTATS 2019poster

How can we efficiently gather information to optimize an unknown function, when presented with multiple, mutually dependent information sources with different costs? For example, when optimizing a physical system, intelligently trading off computer simulations and real-world tests can lead to signif…

Cited by 149 Source ↗PDF ↗Scholar ↗

2019

Co-training for Policy Learning

Jialin Song, Ravi Lanka, Yisong Yue, Masahiro Ono

UAI 2019poster

We study the problem of learning sequential decision-making policies in settings with multiple state-action representations. Such settings naturally arise in many domains, such as planning (e.g., multiple integer programming formulations) and various combinatorial optimization problems (e.g., those…

Cited by 20 Source ↗PDF ↗Scholar ↗