Ziru Chen

Lingbo Mo, Shijie Chen, Ziru Chen et al. · microsoft-research

We introduce TacoBot, a user-centered task-oriented digital assistant designed to guide users through complex real-world tasks with multiple steps. Covering a wide range of cooking and how-to tasks, we aim to deliver a collaborative and engaging dialogue experience. Equipped with language understanding, dialogue management, and response generation components supported by a robust search engine, TacoBot ensures efficient task assistance. To enhance the dialogue experience, we explore a series of data augmentation strategies using LLMs to train advanced neural models continuously. TacoBot builds upon our successful participation in the inaugural Alexa Prize TaskBot Challenge, where our team secured third place among ten competing teams. We offer TacoBot as an open-source framework that serves as a practical example for deploying task-oriented dialogue systems.

Shijie Chen, Ziru Chen, Xiang Deng et al. · microsoft-research

We present TacoBot, a task-oriented dialogue system built for the inaugural Alexa Prize TaskBot Challenge, which assists users in completing multi-step cooking and home improvement tasks. TacoBot is designed with a user-centered principle and aspires to deliver a collaborative and accessible dialogue experience. Towards that end, it is equipped with accurate language understanding, flexible dialogue management, and engaging response generation. Furthermore, TacoBot is backed by a strong search engine and an automated end-to-end test suite. In bootstrapping the development of TacoBot, we explore a series of data augmentation strategies to train advanced neural language processing models and continuously improve the dialogue experience with collected real conversations. At the end of the semifinals, TacoBot achieved an average rating of 3.55/5.0.

Ziru Chen, Michael White, Raymond Mooney et al. · microsoft-research

In this paper, we examine how large language models (LLMs) solve multi-step problems under a language agent framework with three components: a generator, a discriminator, and a planning method. We investigate the practical utility of two advanced planning methods, iterative correction and tree search. We present a comprehensive analysis of how discrimination accuracy affects the overall performance of agents when using these two methods or a simpler method, re-ranking. Experiments on two tasks, text-to-SQL parsing and mathematical reasoning, show that: (1) advanced planning methods demand discriminators with at least 90% accuracy to achieve significant improvements over re-ranking; (2) current LLMs' discrimination abilities have not met the needs of advanced planning methods to achieve such improvements; (3) with LLM-based discriminators, advanced planning methods may not adequately balance accuracy and efficiency. For example, compared to the other two methods, tree search is at least 10--20 times slower but leads to negligible performance gains, which hinders its real-world applications. Code and data are available at https://github.com/OSU-NLP-Group/llm-planning-eval.

Bo Peng, Xinyi Ling, Ziru Chen et al.

With tremendous efforts on developing effective e-commerce models, conventional e-commerce models show limited success in generalist e-commerce modeling, and suffer from unsatisfactory performance on new users and new products - a typical out-of-domain generalization challenge. Meanwhile, large language models (LLMs) demonstrate outstanding performance in generalist modeling and out-of-domain generalizability in many fields. Toward fully unleashing their power for e-commerce, in this paper, we construct ECInstruct, the first open-sourced, large-scale, and high-quality benchmark instruction dataset for e-commerce. Leveraging ECInstruct, we develop eCeLLM, a series of e-commerce LLMs, by instruction-tuning general-purpose LLMs. Our comprehensive experiments and evaluation demonstrate that eCeLLM models substantially outperform baseline models, including the most advanced GPT-4, and the state-of-the-art task-specific models in in-domain evaluation. Moreover, eCeLLM exhibits excellent generalizability to out-of-domain settings, including unseen products and unseen instructions, highlighting its superiority as a generalist e-commerce model. Both the ECInstruct dataset and the eCeLLM models show great potential in empowering versatile and effective LLMs for e-commerce. ECInstruct and eCeLLM models are publicly accessible through https://ninglab.github.io/eCeLLM.

Ziru Chen, Dongdong Chen, Ruinan Jin et al.

Recently, there have been significant research interests in training large language models (LLMs) with reinforcement learning (RL) on real-world tasks, such as multi-turn code generation. While online RL tends to perform better than offline RL, its higher training cost and instability hinders wide adoption. In this paper, we build on the observation that multi-turn code generation can be formulated as a one-step recoverable Markov decision process and propose contextual bandit learning with offline trajectories (Cobalt), a new method that combines the benefits of online and offline RL. Cobalt first collects code generation trajectories using a reference LLM and divides them into partial trajectories as contextual prompts. Then, during online bandit learning, the LLM is trained to complete each partial trajectory prompt through single-step code generation. Cobalt outperforms two multi-turn online RL baselines based on GRPO and VeRPO, and substantially improves R1-Distill 8B and Qwen3 8B by up to 9.0 and 6.2 absolute Pass@1 scores on LiveCodeBench. Also, we analyze LLMs' in-context reward hacking behaviors and augment Cobalt training with perturbed trajectories to mitigate this issue. Overall, our results demonstrate Cobalt as a promising solution for iterative decision-making tasks like multi-turn code generation. Our code and data are available at https://github.com/OSU-NLP-Group/cobalt.

Hanane Nour Moussa, Yifei Li, Zhuoyang Li et al.

Despite recent progress in language models and agents for scientific data-driven discovery, further advancing their capabilities is held back by the absence of verifiable environments representing real-world scientific tasks.To fill this gap, we introduce D3-Gym, the first automatically constructed dataset with verifiable environments for scientific Data-Driven Discovery. D3-Gym comprises (1) 565 tasks sourced from 239 real scientific repositories across four disciplines where (2) each task is equipped with a natural language instruction, an executable environment with pre-installed dependencies, input dataset and artifact previews, a reference code solution, and an automatically synthesized evaluation script. Rigorous evaluation of the quality of the verification signal in D3-Gym confirms that our evaluation scripts achieve 87.5% agreement with human-annotated gold standards and strong alignment in domain-specific evaluation logic, showing their scientific soundness. Further, training on trajectories sampled from D3-Gym yields consistent and substantial gains across Qwen3 models of varying sizes on ScienceAgentBench, boosting Qwen3-32B by 7.8 absolute points and substantially shrinking the gap with strong proprietary models. All D3-Gym artifacts (environments, creation workflow, trajectories, and models) can be found at https://github.com/OSU-NLP-Group/D3-Gym.

Ziru Chen, Salvatore Talarico, Qing Xia et al.

In next generation of Wi-Fi networks Multiple Access Point Coordination (MAPC) is poised to significantly enhance the network performance by enabling a set of Access Points (APs) to coordinate with each other through advanced coordinating schemes so that to reduce inter-AP contention and congestion. This paper focuses on defining a framework to facilitate the coordination across multi-APs when these employ Coordinated Spatial Reuse (C-SR). In this case, the coordinating APs may need to reciprocally adjust their scheduling strategy, power control and link adaptation to meet specific Quality of Service (QoS) requirements, which by using classical approaches leads to high overhead due to negotiations needed across APs, and requires complex solutions in order to properly optimize the network across all the parameters in play. In this matter, a two layer Multi-Armed Bandit (MAB) algorithm has been proposed to optimize such a network while preserving the fair use of resources across all nodes. The validity of this holistic approach is confirmed by system level simulations, which show that the proposed algorithm not only improves the network in terms of sum-throughput, but also allows to enhance fairness, making this a robust solution for next-generation of Wi-Fi networks.

Qianheng Zhang, Song Gao, Chen Wei et al.

Recent advances in large language models (LLMs) have fueled growing interest in automating geospatial analysis and GIS workflows, yet their actual capabilities remain uncertain. In this work, we call for rigorous evaluation of LLMs on well-defined geoprocessing tasks before making claims about full GIS automation. To this end, we present GeoAnalystBench, a benchmark of 50 Python-based tasks derived from real-world geospatial problems and carefully validated by GIS experts. Each task is paired with a minimum deliverable product, and evaluation covers workflow validity, structural alignment, semantic similarity, and code quality (CodeBLEU). Using this benchmark, we assess both proprietary and open source models. Results reveal a clear gap: proprietary models such as ChatGPT-4o-mini achieve high validity 95% and stronger code alignment (CodeBLEU 0.39), while smaller open source models like DeepSeek-R1-7B often generate incomplete or inconsistent workflows (48.5% validity, 0.272 CodeBLEU). Tasks requiring deeper spatial reasoning, such as spatial relationship detection or optimal site selection, remain the most challenging across all models. These findings demonstrate both the promise and limitations of current LLMs in GIS automation and provide a reproducible framework to advance GeoAI research with human-in-the-loop support.

Shijie Chen, Ziru Chen, Huan Sun et al.

Despite remarkable progress in text-to-SQL semantic parsing in recent years, the performance of existing parsers is still far from perfect. Specifically, modern text-to-SQL parsers based on deep learning are often over-confident, thus casting doubt on their trustworthiness when deployed for real use. In this paper, we propose a parser-independent error detection model for text-to-SQL semantic parsing. Using a language model of code as its bedrock, we enhance our error detection model with graph neural networks that learn structural features of both natural language questions and SQL queries. We train our model on realistic parsing errors collected from a cross-domain setting, which leads to stronger generalization ability. Experiments with three strong text-to-SQL parsers featuring different decoding mechanisms show that our approach outperforms parser-dependent uncertainty metrics. Our model could also effectively improve the performance and usability of text-to-SQL semantic parsers regardless of their architectures. (Our implementation is available at https://github.com/OSU-NLP-Group/Text2SQL-Error-Detection)

Ziru Chen, Shijie Chen, Michael White et al.

Despite recent progress in text-to-SQL parsing, current semantic parsers are still not accurate enough for practical use. In this paper, we investigate how to build automatic text-to-SQL error correction models. Noticing that token-level edits are out of context and sometimes ambiguous, we propose building clause-level edit models instead. Besides, while most language models of code are not specifically pre-trained for SQL, they know common data structures and their operations in programming languages such as Python. Thus, we propose a novel representation for SQL queries and their edits that adheres more closely to the pre-training corpora of language models of code. Our error correction model improves the exact set match accuracy of different parsers by 2.4-6.5 and obtains up to 4.3 point absolute improvement over two strong baselines. Our code and data are available at https://github.com/OSU-NLP-Group/Auto-SQL-Correction.

Botao Yu, Frazier N. Baker, Ziru Chen et al.

To enhance large language models (LLMs) for chemistry problem solving, several LLM-based agents augmented with tools have been proposed, such as ChemCrow and Coscientist. However, their evaluations are narrow in scope, leaving a large gap in understanding the benefits of tools across diverse chemistry tasks. To bridge this gap, we develop ChemToolAgent, an enhanced chemistry agent over ChemCrow, and conduct a comprehensive evaluation of its performance on both specialized chemistry tasks and general chemistry questions. Surprisingly, ChemToolAgent does not consistently outperform its base LLMs without tools. Our error analysis with a chemistry expert suggests that: For specialized chemistry tasks, such as synthesis prediction, we should augment agents with specialized tools; however, for general chemistry questions like those in exams, agents' ability to reason correctly with chemistry knowledge matters more, and tool augmentation does not always help.

Sayash Kapoor, Benedikt Stroebl, Peter Kirgis et al. · microsoft-research, princeton

AI agents have been developed for complex real-world tasks from coding to customer service. But AI agent evaluations suffer from many challenges that undermine our understanding of how well agents really work. We introduce the Holistic Agent Leaderboard (HAL) to address these challenges. We make three main contributions. First, we provide a standardized evaluation harness that orchestrates parallel evaluations across hundreds of VMs, reducing evaluation time from weeks to hours while eliminating common implementation bugs. Second, we conduct three-dimensional analysis spanning models, scaffolds, and benchmarks. We validate the harness by conducting 21,730 agent rollouts across 9 models and 9 benchmarks in coding, web navigation, science, and customer service with a total cost of about $40,000. Our analysis reveals surprising insights, such as higher reasoning effort reducing accuracy in the majority of runs. Third, we use LLM-aided log inspection to uncover previously unreported behaviors, such as searching for the benchmark on HuggingFace instead of solving a task, or misusing credit cards in flight booking tasks. We share all agent logs, comprising 2.5B tokens of language model calls, to incentivize further research into agent behavior. By standardizing how the field evaluates agents and addressing common pitfalls in agent evaluation, we hope to shift the focus from agents that ace benchmarks to agents that work reliably in the real world.

Yifei Li, Hanane Nour Moussa, Ziru Chen et al.

Despite long-standing efforts in accelerating scientific discovery with AI, building AI co-scientists remains challenging due to limited high-quality data for training and evaluation. To tackle this data scarcity issue, we present AutoSDT, an automatic pipeline that collects high-quality coding tasks in real-world data-driven discovery workflows. AutoSDT leverages the coding capabilities and parametric knowledge of LLMs to search for diverse sources, select ecologically valid tasks, and synthesize accurate task instructions and code solutions. Using our pipeline, we construct AutoSDT-5K, a dataset of 5,404 coding tasks for data-driven discovery that covers four scientific disciplines and 756 unique Python packages. To the best of our knowledge, AutoSDT-5K is the only automatically collected and the largest open dataset for data-driven scientific discovery. Expert feedback on a subset of 256 tasks shows the effectiveness of AutoSDT: 93% of the collected tasks are ecologically valid, and 92.2% of the synthesized programs are functionally correct. Trained on AutoSDT-5K, the Qwen2.5-Coder-Instruct LLM series, dubbed AutoSDT-Coder, show substantial improvement on two challenging data-driven discovery benchmarks, ScienceAgentBench and DiscoveryBench. Most notably, AutoSDT-Coder-32B reaches the same level of performance as GPT-4o on ScienceAgentBench with a success rate of 7.8%, doubling the performance of its base model. On DiscoveryBench, it lifts the hypothesis matching score to 8.1, bringing a 17.4% relative improvement and closing the gap between open-weight models and GPT-4o.

Lucas Carrit Delgado Pinheiro, Ziru Chen, Bruno Caixeta Piazza et al.

While task-specific demonstrations show early success in applying large language models (LLMs) to automate some astronomical research tasks, they only provide incomplete views of all necessary capabilities in solving astronomy problems, calling for more thorough understanding of LLMs' strengths and limitations. So far, existing benchmarks and evaluations focus on simple question-answering that primarily tests astronomical knowledge and fails to evaluate the complex reasoning required for real-world research in the discipline. Here, we address this gap by systematically benchmarking five state-of-the-art LLMs on the International Olympiad on Astronomy and Astrophysics (IOAA) exams, which are designed to examine deep conceptual understanding, multi-step derivations, and multimodal analysis. With average scores of 85.6% and 84.2%, Gemini 2.5 Pro and GPT-5 (the two top-performing models) not only achieve gold medal level performance but also rank in the top two among ~200-300 participants in all four IOAA theory exams evaluated (2022-2025). In comparison, results on the data analysis exams show more divergence. GPT-5 still excels in the exams with an 88.5% average score, ranking top 10 among the participants in the four most recent IOAAs, while other models' performances drop to 48-76%. Furthermore, our in-depth error analysis underscores conceptual reasoning, geometric reasoning, and spatial visualization (52-79% accuracy) as consistent weaknesses among all LLMs. Hence, although LLMs approach peak human performance in theory exams, critical gaps must be addressed before they can serve as autonomous research agents in astronomy.

Kai Wang, Siyi Chen, Weicong Pang et al.

Land-cover underpins ecosystem services, hydrologic regulation, disaster-risk reduction, and evidence-based land planning; timely, accurate land-cover maps are therefore critical for environmental stewardship. Remote sensing-based land-cover classification offers a scalable route to such maps but is hindered by scarce and imbalanced annotations and by geometric distortions in high-resolution scenes. We propose LC4-DViT (Land-cover Creation for Land-cover Classification with Deformable Vision Transformer), a framework that combines generative data creation with a deformation-aware Vision Transformer. A text-guided diffusion pipeline uses GPT-4o-generated scene descriptions and super-resolved exemplars to synthesize class-balanced, high-fidelity training images, while DViT couples a DCNv4 deformable convolutional backbone with a Vision Transformer encoder to jointly capture fine-scale geometry and global context. On eight classes from the Aerial Image Dataset (AID)-Beach, Bridge, Desert, Forest, Mountain, Pond, Port, and River-DViT achieves 0.9572 overall accuracy, 0.9576 macro F1-score, and 0.9510 Cohen' s Kappa, improving over a vanilla ViT baseline (0.9274 OA, 0.9300 macro F1, 0.9169 Kappa) and outperforming ResNet50, MobileNetV2, and FlashInternImage. Cross-dataset experiments on a three-class SIRI-WHU subset (Harbor, Pond, River) yield 0.9333 overall accuracy, 0.9316 macro F1, and 0.8989 Kappa, indicating good transferability. An LLM-based judge using GPT-4o to score Grad-CAM heatmaps further shows that DViT' s attention aligns best with hydrologically meaningful structures. These results suggest that description-driven generative augmentation combined with deformation-aware transformers is a promising approach for high-resolution land-cover mapping.

Yueqi Song, Ketan Ramaneti, Zaid Sheikh et al.

Public research results on large-scale supervised finetuning of AI agents remain relatively rare, since the collection of agent training data presents unique challenges. In this work, we argue that the bottleneck is not a lack of underlying data sources, but that a large variety of data is fragmented across heterogeneous formats, tools, and interfaces. To this end, we introduce the agent data protocol (ADP), a light-weight representation language that serves as an "interlingua" between agent datasets in diverse formats and unified agent training pipelines downstream. The design of ADP is expressive enough to capture a large variety of tasks, including API/tool use, browsing, coding, software engineering, and general agentic workflows, while remaining simple to parse and train on without engineering at a per-dataset level. In experiments, we unified a broad collection of 13 existing agent training datasets into ADP format, and converted the standardized ADP data into training-ready formats for multiple agent frameworks. We performed SFT on these data, and demonstrated an average performance gain of ~20% over corresponding base models, and delivers state-of-the-art or near-SOTA performance on standard coding, browsing, tool use, and research benchmarks, without domain-specific tuning. All code and data are released publicly, in the hope that ADP could help lower the barrier to standardized, scalable, and reproducible agent training.

Siyi Chen, Kai Wang, Weicong Pang et al.

Open-set land-cover analysis in remote sensing requires the ability to achieve fine-grained spatial localization and semantically open categorization. This involves not only detecting and segmenting novel objects without categorical supervision but also assigning them interpretable semantic labels through multimodal reasoning. In this study, we introduce OSDA, an integrated three-stage framework for annotation-free open-set land-cover discovery, segmentation, and description. The pipeline consists of: (1) precise discovery and mask extraction with a promptable fine-tuned segmentation model (SAM), (2) semantic attribution and contextual description via a two-phase fine-tuned multimodal large language model (MLLM), and (3) LLM-as-judge and manual scoring of the MLLMs evaluation. By combining pixel-level accuracy with high-level semantic understanding, OSDA addresses key challenges in open-world remote sensing interpretation. Designed to be architecture-agnostic and label-free, the framework supports robust evaluation across diverse satellite imagery without requiring manual annotation. Our work provides a scalable and interpretable solution for dynamic land-cover monitoring, showing strong potential for automated cartographic updating and large-scale earth observation analysis.

Chang-You Tai, Ziru Chen, Tianshu Zhang et al.

In-context learning with large language models (LLMs) has recently caught increasing attention due to its superior few-shot performance on various tasks. However, its performance on text-to-SQL parsing still has much room for improvement. In this paper, we hypothesize that a crucial aspect of LLMs to improve for text-to-SQL parsing is their multi-step reasoning ability. Thus, we systematically study how to enhance LLMs' reasoning ability through chain of thought (CoT) style prompting, including the original chain-of-thought prompting (Wei et al., 2022b) and least-to-most prompting (Zhou et al., 2023). Our experiments demonstrate that iterative prompting as in Zhou et al. (2023) may be unnecessary for text-to-SQL parsing, and using detailed reasoning steps tends to have more error propagation issues. Based on these findings, we propose a new CoT-style prompting method for text-to-SQL parsing. It brings 5.2 and 6.5 point absolute gains on the Spider development set and the Spider Realistic set, respectively, compared to the standard prompting method without reasoning steps; 2.4 and 1.5 point absolute gains, compared to the least-to-most prompting method.

Xiang Yue, Boshi Wang, Ziru Chen et al.

A recent focus of large language model (LLM) development, as exemplified by generative search engines, is to incorporate external references to generate and support its claims. However, evaluating the attribution, i.e., verifying whether the generated statement is fully supported by the cited reference, remains an open problem. Although human evaluation is common practice, it is costly and time-consuming. In this paper, we investigate the automatic evaluation of attribution given by LLMs. We begin by defining different types of attribution errors, and then explore two approaches for automatic evaluation: prompting LLMs and fine-tuning smaller LMs. The fine-tuning data is repurposed from related tasks such as question answering, fact-checking, natural language inference, and summarization. We manually curate a set of test examples covering 12 domains from a generative search engine, New Bing. Our results on this curated test set and simulated examples from existing benchmarks highlight both promising signals and challenges. We hope our problem formulation, testbeds, and findings will help lay the foundation for future studies on this important problem.