I am a final-year Ph.D. student at the University of Exeter, expecting to graduate in 2025. My research focuses on Large Language Models (LLMs), Genomics, Software Engineering, and including biological sequence modeling, sentiment analysis, and adversarial attacks.

I am the creator of two open-source projects, OmniGenBench and PyABSA, and have published my work in top-tier conferences such as ACL, EMNLP, AAAI, and CIKM. I am passionate about advancing AI and NLP and am actively seeking new opportunities in both academia and industry.

A large-scale, in-silico benchmarking framework for genomic foundation models (GFMs). It standardizes the evaluation of GFMs and features a public leaderboard to track their performance.

A modularized framework for Aspect-Based Sentiment Analysis (ABSA). PyABSA provides pre-trained models and datasets to simplify sentiment analysis tasks for both research and production.

A text augmentation framework that enhances the quality of augmented data through hybrid instance filtering. It supports various augmentation techniques and is designed to improve model robustness.

A framework for RNA design using OmniGenome and GRPO. It integrates genomic foundation models with to facilitate RNA sequence design and structure prediction.

• PlantRNA-FM: An Interpretable RNA Foundation Model for Exploration Functional RNA Motifs in Plants Models
  Haopeng Yu, Heng Yang#, et al. | Nature Machine Intelligence (Co-first Author) | Paper | Code

• OmniGenBench: Automating Large-scale in-silico Benchmarking for Genomic Foundation Models
  Heng Yang, Jack Cole, Ke Li | arXiv 2024 | Paper | Code

• Bridging Sequence-Structure Alignment in RNA Foundation Models
  Heng Yang, Ke Li | AAAI 2025 | Paper | Code

• The Best Defense is Attack: Repairing Semantics in Textual Adversarial Examples
  Heng Yang, Ke Li | EMNLP 2024 | Paper | Code

• MP-RNA: Unleashing Multi-species RNA Foundation Model via Calibrated Secondary Structure Prediction
  Heng Yang, Ke Li | EMNLP 2024 | Paper | Code

• Modeling Aspect Sentiment Coherency via Local Sentiment Aggregation
  Heng Yang, Ke Li | EACL 2024 | Paper | Code

• PyABSA: A Modularized Framework for Reproducible Aspect-based Sentiment Analysis
  Heng Yang, Cheng Zhang, Ke Li | CIKM 2023 | Paper | Code

• InstOptima: Evolutionary Multi-objective Instruction Optimization via Large Language Model-based Instruction Operators
  Heng Yang, Ke Li | EMNLP 2023 | Paper | Code

• Boosting Text Augmentation via Hybrid Instance Filtering Framework
  Heng Yang, Ke Li | ACL 2023 | Paper | Code

• DaNuoYi: Evolutionary Multi-Task Injection Testing on Web Application Firewalls
  Ke Li, Heng Yang | IEEE Transaction on Software, 2023 | Paper | Code

• A Multi-task Learning Model for Chinese-oriented Aspect Polarity Classification and Aspect Term Extraction
  Heng Yang, Biqing Zeng, et al. | Neurocomputing, 2021 | Paper | Code

• DeBERTa-v3 Base ABSA: A model for aspect-based sentiment analysis, trained on over 30k samples.
• PlantRNA-FM: An interpretable RNA foundation model pre-trained on data from over 1,124 plant species.
• OmniGenome: A genomic model for biological sequence modeling, part of the OmniGenome project.

• Email: [email protected]
• Website: yangheng95.github.io
• DBLP: dblp.org/pid/83/415-8.html
• ACL Anthology: aclanthology.org/people/h/heng-yang