Suyuchen Wang, Jinlin Wang, Xinyu Wang, Shiqi Li, Xiangru Tang, Sirui Hong, Xiao-Wen Chang, Chenglin Wu, Bang Liu. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Jaehoon Yun, Jiwoong Sohn, Jungwoo Park, Hyunjae Kim, Xiangru Tang, Daniel Shao, Yong Hoe Koo, Ko Minhyeok, Qingyu Chen, Mark Gerstein, Michael Moor, Jaewoo Kang. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.

Software is one of the most powerful tools that we humans have at our disposal; it allows a skilled programmer to interact with the world in complex and profound ways. At the same time, thanks to improvements in large language models (LLMs), there has also been a rapid development in AI agents that interact with and effect change in their surrounding environments. In this paper, we introduce OpenHands, a platform for the development of powerful and flexible AI agents that interact with the world in similar ways to a human developer: by writing code, interacting with a command line, and browsing the web. We describe how the platform allows for the implementation of new agents, utilization of various LLMs, safe interaction with sandboxed environments for code execution, and incorporation of evaluation benchmarks. Based on our currently incorporated benchmarks, we perform an evaluation of agents over 13 challenging tasks, including software engineering (e.g., SWE-Bench) and web browsing (e.g., WebArena), amongst others. Released under the permissive MIT license, OpenHands is a community project spanning academia and industry with more than 2K contributions from over 186 contributors in less than six months of development, and will improve going forward.

Chemical reasoning usually involves complex, multi-step processes that demand precise calculations, where even minor errors can lead to cascading failures. Furthermore, large language models (LLMs) encounter difficulties handling domain-specific formulas, executing reasoning steps accurately, and integrating code ef- effectively when tackling chemical reasoning tasks. To address these challenges, we present ChemAgent, a novel framework designed to improve the performance of LLMs through a dynamic, self-updating library. This library is developed by decomposing chemical tasks into sub-tasks and compiling these sub-tasks into a structured collection that can be referenced for future queries. Then, when presented with a new problem, ChemAgent retrieves and refines pertinent information from the library, which we call memory, facilitating effective task decomposition and the generation of solutions. Our method designs three types of memory and a library-enhanced reasoning component, enabling LLMs to improve over time through experience. Experimental results on four chemical reasoning datasets from SciBench demonstrate that ChemAgent achieves performance gains of up to 46% (GPT-4), significantly outperforming existing methods. Our findings suggest substantial potential for future applications, including tasks such as drug discovery and materials science. Our code can be found at https://github.com/gersteinlab/ChemAgent.

Yilun Zhao, Yitao Long, Hongjun Liu, Ryo Kamoi, Linyong Nan, Lyuhao Chen, Yixin Liu, Xiangru Tang, Rui Zhang, Arman Cohan. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

Yilun Zhao, Yitao Long, Tintin Jiang, Chengye Wang, Weiyuan Chen, Hongjun Liu, Xiangru Tang, Yiming Zhang, Chen Zhao, Arman Cohan. Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing. 2024.

Finetuning large language models (LLMs) on instructions leads to vast performance improvements on natural language tasks. We apply instruction tuning using code, leveraging the natural structure of Git commits, which pair code changes with human instructions. We compile CommitPack: 4 terabytes of Git commits across 350 programming languages. We benchmark CommitPack against other natural and synthetic code instructions (xP3x, Self-Instruct, OASST) on the 16B parameter StarCoder model, and achieve state-of-the-art performance among models not trained on OpenAI outputs, on the HumanEval Python benchmark (46.2% pass@1). We further introduce HumanEvalPack, expanding the HumanEval benchmark to a total of 3 coding tasks (Code Repair, Code Explanation, Code Synthesis) across 6 languages (Python, JavaScript, Java, Go, C++, Rust). Our models, OctoCoder and OctoGeeX, achieve the best performance across HumanEvalPack among all permissive models, demonstrating CommitPack's benefits in generalizing to a wider set of languages and natural coding tasks. Code, models and data are freely available at https://github.com/bigcode-project/octopack.

Despite the advancements of open-source large language models (LLMs), e.g., LLaMA, they remain significantly limited in tool-use capabilities, i.e., using external tools (APIs) to fulfill human instructions. The reason is that current instruction tuning largely focuses on basic language tasks but ignores the tool-use domain. This is in contrast to the excellent tool-use capabilities of state-of-the-art (SOTA) closed-source LLMs, e.g., ChatGPT. To bridge this gap, we introduce ToolLLM, a general tool-use framework encompassing data construction, model training, and evaluation. We first present ToolBench, an instruction-tuning dataset for tool use, which is constructed automatically using ChatGPT. Specifically, the construction can be divided into three stages: (i) API collection: we collect 16,464 real-world RESTful APIs spanning 49 categories from RapidAPI Hub; (ii) instruction generation: we prompt ChatGPT to generate diverse instructions involving these APIs, covering both single-tool and multi-tool scenarios; (iii) solution path annotation: we use ChatGPT to search for a valid solution path (chain of API calls) for each instruction. To enhance the reasoning capabilities of LLMs, we develop a novel depth-first search-based decision tree algorithm. It enables LLMs to evaluate multiple reasoning traces and expand the search space. Moreover, to evaluate the tool-use capabilities of LLMs, we develop an automatic evaluator: ToolEval. Based on ToolBench, we fine-tune LLaMA to obtain an LLM ToolLLaMA, and equip it with a neural API retriever to recommend appropriate APIs for each instruction. Experiments show that ToolLLaMA demonstrates a remarkable ability to execute complex instructions and generalize to unseen APIs, and exhibits comparable performance to ChatGPT. Our ToolLLaMA also demonstrates strong zero-shot generalization ability in an out-of-distribution tool-use dataset: APIBench.

Chunyuan Deng, Yilun Zhao, Xiangru Tang, Mark Gerstein, Arman Cohan. Proceedings of the 2024 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2024.

Artificial intelligence (AI)-driven methods can vastly improve the historically costly drug design process, with various generative models already in widespread use. Generative models for de novo drug design, in particular, focus on the creation of novel biological compounds entirely from scratch, representing a promising future direction. Rapid development in the field, combined with the inherent complexity of the drug design process, creates a difficult landscape for new researchers to enter. In this survey, we organize de novo drug design into two overarching themes: small molecule and protein generation. Within each theme, we identify a variety of subtasks and applications, highlighting important datasets, benchmarks, and model architectures and comparing the performance of top models. We take a broad approach to AI-driven drug design, allowing for both micro-level comparisons of various methods within each subtask and macro-level observations across different fields. We discuss parallel challenges and approaches between the two applications and highlight future directions for AI-driven de novo drug design as a whole. An organized repository of all covered sources is available at https://github.com/gersteinlab/GenAI4Drug.

SUMMARY: Pretrained large language models (LLMs) have significantly improved code generation. As these models scale up, there is an increasing need for the output to handle more intricate tasks and to be appropriately specialized to particular domains. Here, we target bioinformatics due to the amount of domain knowledge, algorithms, and data operations this discipline requires. We present BioCoder, a benchmark developed to evaluate LLMs in generating bioinformatics-specific code. BioCoder spans much of the field, covering cross-file dependencies, class declarations, and global variables. It incorporates 1026 Python functions and 1243 Java methods extracted from GitHub, along with 253 examples from the Rosalind Project, all pertaining to bioinformatics. Using topic modeling, we show that the overall coverage of the included code is representative of the full spectrum of bioinformatics calculations. BioCoder incorporates a fuzz-testing framework for evaluation. We have applied it to evaluate various models including InCoder, CodeGen, CodeGen2, SantaCoder, StarCoder, StarCoder+, InstructCodeT5+, GPT-3.5, and GPT-4. Furthermore, we fine-tuned one model (StarCoder), demonstrating that our training dataset can enhance the performance on our testing benchmark (by >15% in terms of Pass@K under certain prompt configurations and always >3%). The results highlight two key aspects of successful models: (i) Successful models accommodate a long prompt (>2600 tokens) with full context, including functional dependencies. (ii) They contain domain-specific knowledge of bioinformatics, beyond just general coding capability. This is evident from the performance gain of GPT-3.5/4 compared to the smaller models on our benchmark (50% versus up to 25%). AVAILABILITY AND IMPLEMENTATION: All datasets, benchmark, Docker images, and scripts required for testing are available at: https://github.com/gersteinlab/biocoder and https://biocoder-benchmark.github.io/.

MOTIVATION: The current paradigm of deep learning models for the joint representation of molecules and text primarily relies on 1D or 2D molecular formats, neglecting significant 3D structural information that offers valuable physical insight. This narrow focus inhibits the models' versatility and adaptability across a wide range of modalities. Conversely, the limited research focusing on explicit 3D representation tends to overlook textual data within the biomedical domain. RESULTS: We present a unified pre-trained language model, MolLM, that concurrently captures 2D and 3D molecular information alongside biomedical text. MolLM consists of a text Transformer encoder and a molecular Transformer encoder, designed to encode both 2D and 3D molecular structures. To support MolLM's self-supervised pre-training, we constructed 160K molecule-text pairings. Employing contrastive learning as a supervisory signal for learning, MolLM demonstrates robust molecular representation capabilities across four downstream tasks, including cross-modal molecule and text matching, property prediction, captioning, and text-prompted molecular editing. Through ablation, we demonstrate that the inclusion of explicit 3D representations improves performance in these downstream tasks. AVAILABILITY AND IMPLEMENTATION: Our code, data, pre-trained model weights, and examples of using our model are all available at https://github.com/gersteinlab/MolLM. In particular, we provide Jupyter Notebooks offering step-by-step guidance on how to use MolLM to extract embeddings for both molecules and text.

Niklas Muennighoff, Thomas Wang, Lintang Sutawika, Adam Roberts, Stella Biderman, Teven Le Scao, M Saiful Bari, Sheng Shen, Zheng Xin Yong, Hailey Schoelkopf, Xiangru Tang, Dragomir Radev, Alham Fikri Aji, Khalid Almubarak, Samuel Albanie, Zaid Alyafeai, Albert Webson, Edward Raff, Colin Raffel. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

Yilun Zhao, Chen Zhao, Linyong Nan, Zhenting Qi, Wenlin Zhang, Xiangru Tang, Boyu Mi, Dragomir Radev. Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2023.

Yilun Zhao, Zhenting Qi, Linyong Nan, Boyu Mi, Yixin Liu, Weijin Zou, Simeng Han, Ruizhe Chen, Xiangru Tang, Yumo Xu, Dragomir Radev, Arman Cohan. Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing. 2023.

Fast-developing fields such as Artificial Intelligence (AI) often outpace the efforts of encyclopedic sources such as Wikipedia, which either do not completely cover recently-introduced topics or lack such content entirely. As a result, methods for automatically producing content are valuable tools to address this information overload. We show that recent advances in pretrained language modeling can be combined for a two-stage extractive and abstractive approach for Wikipedia lead paragraph generation. We extend this approach to generate longer Wikipedia-style summaries with sections and examine how such methods struggle in this application through detailed studies with 100 reference human-collected surveys. This is the first study on utilizing web resources for long Wikipedia-style summaries to the best of our knowledge.

Xiangru Tang, Alexander Fabbri, Haoran Li, Ziming Mao, Griffin Adams, Borui Wang, Asli Celikyilmaz, Yashar Mehdad, Dragomir Radev. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.

Xiangru Tang, Arjun Nair, Borui Wang, Bingyao Wang, Jai Desai, Aaron Wade, Haoran Li, Asli Celikyilmaz, Yashar Mehdad, Dragomir Radev. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.

Abstract Existing table question answering datasets contain abundant factual questions that primarily evaluate a QA system’s comprehension of query and tabular data. However, restricted by their short-form answers, these datasets fail to include question–answer interactions that represent more advanced and naturally occurring information needs: questions that ask for reasoning and integration of information pieces retrieved from a structured knowledge source. To complement the existing datasets and to reveal the challenging nature of the table-based question answering task, we introduce FeTaQA, a new dataset with 10K Wikipedia-based {table, question, free-form answer, supporting table cells} pairs. FeTaQA is collected from noteworthy descriptions of Wikipedia tables that contain information people tend to seek; generation of these descriptions requires advanced processing that humans perform on a daily basis: Understand the question and table, retrieve, integrate, infer, and conduct text planning and surface realization to generate an answer. We provide two benchmark methods for the proposed task: a pipeline method based on semantic parsing-based QA systems and an end-to-end method based on large pretrained text generation models, and show that FeTaQA poses a challenge for both methods.

Linyong Nan, Dragomir Radev, Rui Zhang, Amrit Rau, Abhinand Sivaprasad, Chiachun Hsieh, Xiangru Tang, Aadit Vyas, Neha Verma, Pranav Krishna, Yangxiaokang Liu, Nadia Irwanto, Jessica Pan, Faiaz Rahman, Ahmad Zaidi, Mutethia Mutuma, Yasin Tarabar, Ankit Gupta, Tao Yu, Yi Chern Tan, Xi Victoria Lin, Caiming Xiong, Richard Socher, Nazneen Fatema Rajani. Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2021.

Code generation is a model-driven engineering approach that enables developers to generate source code automatically and achieves extremely high development productivity.Specifically, generating code from a descriptive text reduces the time and expense of software development significantly.However, the performance of existing methods is not satisfying, since they are either of low accuracy (lack of specifics of the generated code) or too complicated (lack of efficiency in training).In this work, we proposed three novel methods by combining neural architectures and syntax rules, aiming at explicitly capturing the syntactical characteristics of target code.First, we proposed three models based on the Combination of Deep learning and Syntax rules (CDS models).Then, we evaluated CDS models with BLEU metric by comparing our models with existing methods.The results show that our models outperform existing methods for the challenging code generation task.Finally, we conducted a comparative study between the three CDS models.With further analysis we provided advice on the choice of neural architectures by considering both task accuracy and efficiency.Experimental results show that (1) there is a trade-off between speed and accuracy of the model, and (2) one of our CDS models (i.e., the CDS-POOLING model) outperforms other existing methods for the challenging code generation task.