avatar

Junyan Cheng

Ph.D. Candidate
Dartmouth College
jc.th@dartmouth.edu


About Me

I am a fourth-year Ph.D. candidate at Dartmouth College, advised by Prof. Peter Chin at LISP Lab, fully funded by PhD Innovation Fellowship (Thayer & Tuck collaborative, hosted by Prof. Eric R. Fossum). I interned in the AIIL Group at Microsoft Research Redmond, mentored by Dr. Jay Stokes, on autonomous software engineering, and 6 months at the Allen Institute for AI (Ai2) under Dr. Kyle Richardson on autonomous scientific discovery.

About My Research

I build complex LLM-agentic systems across forecasting, scientific discovery, software design, and sociotal dynamics. Below are live demos or webapps of my flagship systems (click to explore):

Analytica

Robust and efficient wide and deep LLM-agentic analysis and forecasting with soft logic and parallel Jupyter Notebook analysts.

Under review. Webapp coming soon!

Genesys

Distributed discovery system for novel language model architectures with autonomous design and evaluation.

NeurIPS'25 Spotlight

Apeiron

Scalable and autonomous full-lifecycle demand-optimized app synthesizer with multi-thread CUA evaluators.

Built at Microsoft Research

SocioDojo

LLM agentic system and environment for lifelong societal modeling and analysis with news feed and real-time data integration.

ICLR'24 Spotlight

My goal is to achieve autonomous systems that can accurately, reliably, and efficiently take over critical real-world domains in long-term large-scale deployments, with a holographic perception of the world dynamics. I focus on Hybrid and Neural-symbolic AI that enhances powerful NNs and LLMs with symbolic approaches, such as logic, statistics, programming, and optimization (why we still need symbols? it’s like why we need laws!).

Publications

  1. NeurIPS
    Language Modeling by Language Models
    Junyan Cheng, Peter Clark, Kyle Richardson
    The Thirty-ninth Annual Conference on Neural Information Processing Systems (NeurIPS), 2025.
    PDF Code Demo Spotlight (3.19% acceptance rate)

  2. FinAI@ICLR
    Empirical Asset Pricing with Large Language Model Agents
    Junyan Cheng, Peter Chin
    ICLR 2025 Advances in Financial AI Workshop (Long Paper track), 2025.
    PDF Code

  3. ICLR
    SocioDojo: Building Lifelong Analytical Agents with Real-world Text and Time Series
    Junyan Cheng, Peter Chin
    The Twelfth International Conference on Learning Representations (ICLR), 2024.
    PDF Code Demo Spotlight (5.02% acceptance rate)

  4. ICLR
    Bridging Neural and Symbolic Representations with Transitional Dictionary Learning
    Junyan Cheng, Peter Chin
    The Twelfth International Conference on Learning Representations (ICLR), 2024.
    PDF Code

  5. JCISE
    Physics Informed Synthetic Image Generation for Deep Learning-Based Detection of Wrinkles and Folds
    Omey M Manyar, Junyan Cheng, Reuben Levine, Vihan Krishnan, Jernej BarbiÄŤ, Satyandra K Gupta
    Journal of Computing and Information Science in Engineering
    PDF Code Best Paper Award at ASME IDETC-CIE 2022

  6. EMNLP
    Multimodal Phased Transformer for Sentiment Analysis
    Junyan Cheng*, Iordanis Fostiropoulos*, Barry Boehm, Mohammad Soleymani
    Conference on Empirical Methods in Natural Language Processing (EMNLP Main), 2021.
    PDF Code

  7. arXiv
    Graph Conditioned Sparse-Attention for Improved Source Code Understanding
    Junyan Cheng, Iordanis Fostiropoulos, Barry Boehm
    arXiv preprint, arXiv:2112.00663
    PDF Code

  8. arXiv
    GN-Transformer: Fusing Sequence and Graph Representation for Improved Code Summarization
    Junyan Cheng, Iordanis Fostiropoulos, Barry Boehm
    arXiv preprint, arXiv:2111.08874
    PDF Code

  9. RNA biology
    Predicting microRNA-disease associations using bipartite local models and hubness-aware regression
    Xing Chen*, Junyan Cheng*, Jun Yin
    RNA biology, 2018, VOL. 15, NO. 9, 1192-1205
    PDF Code


Powered by Jekyll and Minimal Light theme.