Publications (58)
Study of QCD critical point with three-nucleon correlations in light nuclei yields ratios using PYTHIA8/Angantyr
Zuman Zhang, Ning Yu, Sha Li +3
This study utilizes the PYTHIA8 Angantyr model to systematically investigate the effects of three nucleons correlation on the light nuclei yield ratio in…
Experience as a Compass: Multi-agent RAG with Evolving Orchestration and Agent Prompts
Sha Li, Naren Ramakrishnan
Multi-agent Retrieval-Augmented Generation (RAG), wherein each agent takes on a specific role, supports hard queries that require multiple steps and sources, or complex reasoning.…
LLM-ACES: Closed-Loop Discovery of Dynamical Systems with LLM-Guided Adaptive Search
Nikhil Abhyankar, Sha Li, Sanchit Kabra +3
Recovering governing Ordinary Differential Equations (ODEs) from data is a central challenge in modeling dynamical systems across scientific domains. Existing approaches cast disco…
SyncMind: Measuring Agent Out-of-Sync Recovery in Collaborative Software Engineering
Xuehang Guo, Xingyao Wang, Yangyi Chen +4
Software engineering (SE) is increasingly collaborative, with developers working together on shared complex codebases. Effective collaboration in shared environments requires parti…
Learning with Rare Success but Rich Feedback via Reflection-Enhanced Self-Distillation
Yuwei Zhang, Sha Li, Changlong Yu +9
Enabling Large Language Models (LLMs) to continuously improve from environmental interactions is a central challenge in post-training. While on-policy self-distillation offers a pr…
Exploring LLMs for Scientific Information Extraction Using The SciEx Framework
Sha Li, Ayush Sadekar, Nathan Self +10
Large language models (LLMs) are increasingly touted as powerful tools for automating scientific information extraction. However, existing methods and tools often struggle with the…
The Law of Knowledge Overshadowing: Towards Understanding, Predicting, and Preventing LLM Hallucination
Yuji Zhang, Sha Li, Cheng Qian +8
Hallucination is a persistent challenge in large language models (LLMs), where even with rigorous quality control, models often generate distorted facts. This paradox, in which err…
Stacking polymorphism in PtSe drastically affects its electromechanical properties
Roman Kempt, Sebastian Lukas, Oliver Hartwig +8
PtSe is one of the most promising materials for the next generation of piezoresistive sensors. However, the large-scale synthesis of homogeneous thin films with reproducible el…
RAG without Forgetting: Continual Query-Infused Key Memory
Yuntong Hu, Sha Li, Naren Ramakrishnan +1
Retrieval-augmented generation (RAG) systems commonly improve robustness via query-time adaptations such as query expansion and iterative retrieval. While effective, these approach…
Investigating nonflow contribution subtraction in d-Au collisions with AMPT model
Zuman Zhang, Sha Li, Ning Yu +1
This paper presents research that focuses on nonflow contribution subtraction in heavy-ion collisions, using a multiphase transport model (AMPT). Specifically, the study aims to in…
Open-Domain Hierarchical Event Schema Induction by Incremental Prompting and Verification
Sha Li, Ruining Zhao, Manling Li +3
Event schemas are a form of world knowledge about the typical progression of events. Recent methods for event schema induction use information extraction systems to construct a lar…
Code4Struct: Code Generation for Few-Shot Event Structure Prediction
Xingyao Wang, Sha Li, Heng Ji
Large Language Model (LLM) trained on a mixture of text and code has demonstrated impressive capability in translating natural language (NL) into structured code. We observe that s…
Antenna system characteristic and solar radio burst observation
Sha Li, Yihua Yan, Zhijun Chen +2
Chinese Spectral Radio Heliograph (CSRH) is an advanced aperture synthesis solar radio heliograph, developed by National Astronomical Observatories, Chinese Academy of Sciences ind…
Process Reward Informed Tree Rollout for Effective Multi-Turn RL
Xintong Li, Sha Li, Yuwei Zhang +8
Reinforcement learning (RL) has become a key approach for training LLM agents, yet popular methods such as GRPO/RLOO rely on multiple independently sampled complete trajectories fo…
Schema-Guided Event Graph Completion
Hongwei Wang, Zixuan Zhang, Sha Li +5
We tackle a new task, event graph completion, which aims to predict missing event nodes for event graphs. Existing link prediction or graph completion methods have difficulty deali…
Open Relation and Event Type Discovery with Type Abstraction
Sha Li, Heng Ji, Jiawei Han
Conventional closed-world information extraction (IE) approaches rely on human ontologies to define the scope for extraction. As a result, such approaches fall short when applied t…
A Review on Serious Games for ADHD
Yuanyuan Zheng, Rongyang Li, Sha Li +3
Attention deficit and hyperactivity disorder (ADHD) have two main characteristics: inattention and impulsivity.It has many obstacles to the normal development of children and is ve…
The Future is not One-dimensional: Complex Event Schema Induction by Graph Modeling for Event Prediction
Manling Li, Sha Li, Zhenhailong Wang +5
Event schemas encode knowledge of stereotypical structures of events and their connections. As events unfold, schemas are crucial to act as a scaffolding. Previous work on event sc…
Knowledge Overshadowing Causes Amalgamated Hallucination in Large Language Models
Yuji Zhang, Sha Li, Jiateng Liu +5
Hallucination is often regarded as a major impediment for using large language models (LLMs), especially for knowledge-intensive tasks. Even when the training corpus consists solel…
TEXT2DB: Integration-Aware Information Extraction with Large Language Model Agents
Yizhu Jiao, Sha Li, Sizhe Zhou +2
The task of information extraction (IE) is to extract structured knowledge from text. However, it is often not straightforward to utilize IE output due to the mismatch between the…
Enhanced Knowledge Selection for Grounded Dialogues via Document Semantic Graphs
Sha Li, Mahdi Namazifar, Di Jin +4
Providing conversation models with background knowledge has been shown to make open-domain dialogues more informative and engaging. Existing models treat knowledge selection as a s…
WebCoach: Self-Evolving Web Agents with Cross-Session Memory Guidance
Genglin Liu, Shijie Geng, Sha Li +4
Multimodal LLM-powered agents have recently demonstrated impressive capabilities in web navigation, enabling agents to complete complex browsing tasks across diverse domains. Howev…
Schema-Guided Culture-Aware Complex Event Simulation with Multi-Agent Role-Play
Sha Li, Revanth Gangi Reddy, Khanh Duy Nguyen +7
Complex news events, such as natural disasters and socio-political conflicts, require swift responses from the government and society. Relying on historical events to project the f…
A Review on Serious Games for Phobia
Sha Li, Peichen Yang, Rongyang Li +4
Phobia is a widespread mental illness, and severe phobias can seriously impact patients daily lives. One-session Exposure Treatment (OST) has been used to treat phobias in the earl…
MACAROON: Training Vision-Language Models To Be Your Engaged Partners
Shujin Wu, Yi R. Fung, Sha Li +3
Large vision-language models (LVLMs), while proficient in following instructions and responding to diverse questions, invariably generate detailed responses even when questions are…
Open-Vocabulary Argument Role Prediction for Event Extraction
Yizhu Jiao, Sha Li, Yiqing Xie +3
The argument role in event extraction refers to the relation between an event and an argument participating in it. Despite the great progress in event extraction, existing studies…
From Pixels to Policies: Reinforcing Spatial Reasoning in Language Models for Content-Aware Layout Design
Sha Li, Stefano Petrangeli, Yu Shen +1
We introduce LaySPA, a reinforcement learning framework that equips large language models (LLMs) with explicit and interpretable spatial reasoning for content-aware graphic layout…
Dynamic Global Memory for Document-level Argument Extraction
Xinya Du, Sha Li, Heng Ji
Extracting informative arguments of events from news articles is a challenging problem in information extraction, which requires a global contextual understanding of each document.…
Nonlinear effects in modulated quantum optomechanics
Tai-Shuang Yin, Xin-You Lü, Li-Li Zheng +3
The nonlinear quantum regime is crucial for implementing interesting quantum effects, which have wide applications in modern quantum science. Here we propose an effective method to…
RESIN-EDITOR: A Schema-guided Hierarchical Event Graph Visualizer and Editor
Khanh Duy Nguyen, Zixuan Zhang, Reece Suchocki +5
In this paper, we present RESIN-EDITOR, an interactive event graph visualizer and editor designed for analyzing complex events. Our RESIN-EDITOR system allows users to render and f…
Evidence for local spots of viscous electron flow in graphene at moderate mobility
Sayanti Samaddar, Jeff Strasdas, Kevin JanÃen +8
Dominating electron-electron scattering enables viscous electron flow exhibiting hydrodynamic current density patterns such as Poiseuille profiles or vortices. The viscous regime h…
Document-Level Event Argument Extraction by Conditional Generation
Sha Li, Heng Ji, Jiawei Han
Event extraction has long been treated as a sentence-level task in the IE community. We argue that this setting does not match human information-seeking behavior and leads to incom…
Relation Learning on Social Networks with Multi-Modal Graph Edge Variational Autoencoders
Carl Yang, Jieyu Zhang, Haonan Wang +5
While node semantics have been extensively explored in social networks, little research attention has been paid to profile edge semantics, i.e., social relations. Ideal edge semant…
GLEN: General-Purpose Event Detection for Thousands of Types
Qiusi Zhan, Sha Li, Kathryn Conger +3
The progress of event extraction research has been hindered by the absence of wide-coverage, large-scale datasets. To make event extraction systems more accessible, we build a gene…
HiGitClass: Keyword-Driven Hierarchical Classification of GitHub Repositories
Yu Zhang, Frank F. Xu, Sha Li +4
GitHub has become an important platform for code sharing and scientific exchange. With the massive number of repositories available, there is a pressing need for topic-based search…
Switchable dynamics in the deep-strong-coupling regime
Li-Li Zheng, Qian Bin, Zhi-Ming Zhan +3
We investigate theoretically the dynamics of the system that consists of a cascade three-level emitter interacting with a single-mode resonator in the deep-strong-coupling regime.…
: Structure-Originated Reasoning Data Improves Long-Context Reasoning Ability of Large Language Models
Quyet V. Do, Thinh Pham, Nguyen Nguyen +3
We study a pipeline that curates reasoning data from initial structured data for improving long-context reasoning in large language models (LLMs). Our approach, , constructs…
How Do Large Language Models Learn Concepts During Continual Pre-Training?
Barry Menglong Yao, Sha Li, Yunzhi Yao +4
Human beings primarily understand the world through concepts (e.g., dog), abstract mental representations that structure perception, reasoning, and learning. However, how large lan…
Oreo: A Plug-in Context Reconstructor to Enhance Retrieval-Augmented Generation
Sha Li, Naren Ramakrishnan
Retrieval-Augmented Generation (RAG) aims to augment the capabilities of Large Language Models (LLMs) by retrieving and incorporate external documents or chunks prior to generation…
Defining a New NLP Playground
Sha Li, Chi Han, Pengfei Yu +8
The recent explosion of performance of large language models (LLMs) has changed the field of Natural Language Processing (NLP) more abruptly and seismically than any other shift in…
P4E: Few-Shot Event Detection as Prompt-Guided Identification and Localization
Sha Li, Liyuan Liu, Yiqing Xie +2
We propose P4E, an identify-and-localize event detection framework that integrates the best of few-shot prompting and structured prediction. Our framework decomposes event detectio…
Establishing Knowledge Preference in Language Models
Sizhe Zhou, Sha Li, Yu Meng +3
Language models are known to encode a great amount of factual knowledge through pretraining. However, such knowledge might be insufficient to cater to user requests, requiring the…
Instruct and Extract: Instruction Tuning for On-Demand Information Extraction
Yizhu Jiao, Ming Zhong, Sha Li +4
Large language models with instruction-following capabilities open the door to a wider group of users. However, when it comes to information extraction - a classic task in natural…
Agentic Conversational Search with Contextualized Reasoning via Reinforcement Learning
Fengran Mo, Yifan Gao, Sha Li +7
Large Language Models (LLMs) have become a popular interface for human-AI interaction, supporting information seeking and task assistance through natural, multi-turn dialogue. To r…
EVEDIT: Event-based Knowledge Editing with Deductive Editing Boundaries
Jiateng Liu, Pengfei Yu, Yuji Zhang +3
The dynamic nature of real-world information necessitates efficient knowledge editing (KE) in large language models (LLMs) for knowledge updating. However, current KE approaches, w…
TopoFE: topology-aware LLM-guided Automated Feature Engineering
Sha Li, Naren Ramakrishnan
Automatic feature engineering (AutoFE) for tabular learning can be naturally formulated as a program synthesis problem, where the objective is to discover predictive feature transf…
Ferroelectric control of antiferromagnetism via coordination swapping in A2Mo3O8 (A= Mn, Fe, Co)
Yaxin Gao, Sha Li, Menghao Wu
Transition metal molybdenum oxides A2Mo3O8 (A= Mn, Fe, Co) are known to be polar magnets where A ions are located in either octahedrally or tetrahedrally coordinated sites. In this…
Non-Sequential Graph Script Induction via Multimedia Grounding
Yu Zhou, Sha Li, Manling Li +4
Online resources such as WikiHow compile a wide range of scripts for performing everyday tasks, which can assist models in learning to reason about procedures. However, the scripts…
LLMs as Layout Designers: Enhanced Spatial Reasoning for Content-Aware Layout Generation
Sha Li, Stefano Petrangeli, Yu Shen +2
While Large Language Models (LLMs) have demonstrated impressive reasoning and planning abilities in textual domains and can effectively follow instructions for complex tasks, their…
Human-in-the-Loop Schema Induction
Tianyi Zhang, Isaac Tham, Zhaoyi Hou +12
Schema induction builds a graph representation explaining how events unfold in a scenario. Existing approaches have been based on information retrieval (IR) and information extract…
Disentangling High Harmonic Generation from Surface and Bulk States of a Topological Insulator
Sha Li, Wenyi Zhou, Kazi A. Imroz +9
The discovery of topological phases has introduced a new dimension to materials science. Three-dimensional (3D) topological insulators (TIs) are a remarkable class of matter that i…
Graphene-Quantum Dot Hybrid Photodetectors from 200 mm Wafer Scale Processing
Sha Li, Zhenxing Wang, Bianca Robertz +14
A 200 mm processing platform for the large-scale production of graphene field-effect transistor-quantum dot (GFET-QD) hybrid photodetectors is demonstrated. Comprehensive statistic…
Stepwise Penalization for Length-Efficient Chain-of-Thought Reasoning
Xintong Li, Sha Li, Rongmei Lin +10
Large reasoning models improve with more test-time computation, but often overthink, producing unnecessarily long chains-of-thought that raise cost without improving accuracy. Prio…
If LLM Is the Wizard, Then Code Is the Wand: A Survey on How Code Empowers Large Language Models to Serve as Intelligent Agents
Ke Yang, Jiateng Liu, John Wu +9
The prominent large language models (LLMs) of today differ from past language models not only in size, but also in the fact that they are trained on a combination of natural langua…
Paxion: Patching Action Knowledge in Video-Language Foundation Models
Zhenhailong Wang, Ansel Blume, Sha Li +5
Action knowledge involves the understanding of textual, visual, and temporal aspects of actions. We introduce the Action Dynamics Benchmark (ActionBench) containing two carefully d…
FanChuan: A Multilingual and Graph-Structured Benchmark For Parody Detection and Analysis
Yilun Zheng, Sha Li, Fangkun Wu +9
Parody is an emerging phenomenon on social media, where individuals imitate a role or position opposite to their own, often for humor, provocation, or controversy. Detecting and an…
Eider: Empowering Document-level Relation Extraction with Efficient Evidence Extraction and Inference-stage Fusion
Yiqing Xie, Jiaming Shen, Sha Li +2
Document-level relation extraction (DocRE) aims to extract semantic relations among entity pairs in a document. Typical DocRE methods blindly take the full document as input, while…
OpenPI-C: A Better Benchmark and Stronger Baseline for Open-Vocabulary State Tracking
Xueqing Wu, Sha Li, Heng Ji
Open-vocabulary state tracking is a more practical version of state tracking that aims to track state changes of entities throughout a process without restricting the state space a…