Haojia Sun

Weiwei Sun, Xuhui Zhou, Jiarui Liu et al.

Humans learn social norms and behaviors from verbal feedback (e.g., a parent saying "that was rude" or a friend explaining "here's why that hurt"). Yet, learning from feedback for LLMs has largely focused on domains like code and math, where RL rewards are directly verifiable and condensed into scalar values. As LLMs are increasingly used to simulate human behavior, e.g., standing in for users, patients, students, and other personas, there is a pressing need to make them more human-like, which requires embracing a fundamentally different kind of signal: feedback that is verbal, subjective, and multi-faceted. We present DITTO, a model trained by treating verbal feedback as a first-class signal in reinforcement learning. After each rollout, DITTO receives verbal feedback and generates a feedback-conditioned improved rollout; both outputs are jointly optimized with GRPO, distilling verbal guidance into the base policy without requiring feedback at test time. We also introduce SOUL (Simulation gym Of hUman-Like behavior), a unified benchmark and training data suite spanning 10 tasks across six categories: Theory of Mind, character role play, social skill, learner simulation, user simulation, and persona simulation. DITTO achieves an average 36% improvement over the base model and exceeds GPT-5.4 on 6 of 10 SOUL benchmarks, demonstrating that RL with verbal feedback is a promising direction for training LLMs to simulate human behavior.

Yuxin Zuo, Haojia Sun, Yongyi Hu et al.

Influence maximization aims to find a subset of seeds that maximize the influence spread under a given budget. In this paper, we mainly address the data-driven version of this problem, where the diffusion model is not given but needs to be inferred from the history cascades. Several previous works have addressed this topic in a statistical way and provided efficient algorithms with theoretical guarantee. However, in their settings, though the diffusion parameters are inferred, they still need users to preset the diffusion model, which can be an intractable problem in real-world practices. In this paper, we reformulate the problem on the attributed network and leverage the node attributes to estimate the closeness between the connected nodes. Specifically, we propose a machine learning-based framework, named DSCom, to address this problem in a heuristic way. Under this framework, we first infer the users' relationship from the diffusion dataset through attention mechanism and then leverage spectral clustering to overcome the influence overlap problem in the lack of exact diffusion formula. Compared to the previous theoretical works, we carefully designed empirical experiments with parameterized diffusion models based on real-world social networks, which prove the efficiency and effectiveness of our algorithm.

Xiaoyan Bai, Alexander Baumgartner, Haojia Sun et al.

Reproducibility crises across sciences highlight the limitations of the paper-centric review system in assessing the rigor and reproducibility of research. AI agents that autonomously design and generate large volumes of research outputs exacerbate these challenges. In this work, we address the growing challenges of scalability and rigor by flipping the dynamic and developing AI agents as research evaluators. We propose the first execution-grounded evaluation framework that verifies research beyond narrative review by examining code and data alongside the paper. We use mechanistic interpretability research as a testbed, build standardized research output, and develop MechEvalAgent, an automated evaluation framework that assesses the coherence of the experimental process, the reproducibility of results, and the generalizability of findings. We show that our framework achieves above 80% agreement with human judges, identifies substantial methodological problems, and surfaces 51 additional issues that human reviewers miss. Our work demonstrates the potential of AI agents to transform research evaluation and pave the way for rigorous scientific practices.

Haojia Sun, Yaqi Wang, Shuting Zhang

We designed a Retrieval-Augmented Generation (RAG) system to provide large language models with relevant documents for answering domain-specific questions about Pittsburgh and Carnegie Mellon University (CMU). We extracted over 1,800 subpages using a greedy scraping strategy and employed a hybrid annotation process, combining manual and Mistral-generated question-answer pairs, achieving an inter-annotator agreement (IAA) score of 0.7625. Our RAG framework integrates BM25 and FAISS retrievers, enhanced with a reranker for improved document retrieval accuracy. Experimental results show that the RAG system significantly outperforms a non-RAG baseline, particularly in time-sensitive and complex queries, with an F1 score improvement from 5.45% to 42.21% and recall of 56.18%. This study demonstrates the potential of RAG systems in enhancing answer precision and relevance, while identifying areas for further optimization in document retrieval and model training.