Qingyang Zhang

Qingyang Zhang, Yiming Yang, Jingqing Ruan et al.

Goal-Conditioned Hierarchical Reinforcement Learning (GCHRL) is a promising paradigm to address the exploration-exploitation dilemma in reinforcement learning. It decomposes the source task into subgoal conditional subtasks and conducts exploration and exploitation in the subgoal space. The effectiveness of GCHRL heavily relies on subgoal representation functions and subgoal selection strategy. However, existing works often overlook the temporal coherence in GCHRL when learning latent subgoal representations and lack an efficient subgoal selection strategy that balances exploration and exploitation. This paper proposes HIerarchical reinforcement learning via dynamically building Latent Landmark graphs (HILL) to overcome these limitations. HILL learns latent subgoal representations that satisfy temporal coherence using a contrastive representation learning objective. Based on these representations, HILL dynamically builds latent landmark graphs and employs a novelty measure on nodes and a utility measure on edges. Finally, HILL develops a subgoal selection strategy that balances exploration and exploitation by jointly considering both measures. Experimental results demonstrate that HILL outperforms state-of-the-art baselines on continuous control tasks with sparse rewards in sample efficiency and asymptotic performance. Our code is available at https://github.com/papercode2022/HILL.

Qingyang Zhang, Haitao Wu, Changqing Zhang et al.

The inherent challenge of multimodal fusion is to precisely capture the cross-modal correlation and flexibly conduct cross-modal interaction. To fully release the value of each modality and mitigate the influence of low-quality multimodal data, dynamic multimodal fusion emerges as a promising learning paradigm. Despite its widespread use, theoretical justifications in this field are still notably lacking. Can we design a provably robust multimodal fusion method? This paper provides theoretical understandings to answer this question under a most popular multimodal fusion framework from the generalization perspective. We proceed to reveal that several uncertainty estimation solutions are naturally available to achieve robust multimodal fusion. Then a novel multimodal fusion framework termed Quality-aware Multimodal Fusion (QMF) is proposed, which can improve the performance in terms of classification accuracy and model robustness. Extensive experimental results on multiple benchmarks can support our findings.

Rose E. Wang, Qingyang Zhang, Carly Robinson et al.

Scaling high-quality tutoring remains a major challenge in education. Due to growing demand, many platforms employ novice tutors who, unlike experienced educators, struggle to address student mistakes and thus fail to seize prime learning opportunities. Our work explores the potential of large language models (LLMs) to close the novice-expert knowledge gap in remediating math mistakes. We contribute Bridge, a method that uses cognitive task analysis to translate an expert's latent thought process into a decision-making model for remediation. This involves an expert identifying (A) the student's error, (B) a remediation strategy, and (C) their intention before generating a response. We construct a dataset of 700 real tutoring conversations, annotated by experts with their decisions. We evaluate state-of-the-art LLMs on our dataset and find that the expert's decision-making model is critical for LLMs to close the gap: responses from GPT4 with expert decisions (e.g., "simplify the problem") are +76% more preferred than without. Additionally, context-sensitive decisions are critical to closing pedagogical gaps: random decisions decrease GPT4's response quality by -97% than expert decisions. Our work shows the potential of embedding expert thought processes in LLM generations to enhance their capability to bridge novice-expert knowledge gaps. Our dataset and code can be found at: \url{https://github.com/rosewang2008/bridge}.

Yun Luo, Futing Wang, Qianjia Cheng et al.

The transition from symbolic manipulation to science-grade reasoning represents a pivotal frontier for Large Language Models (LLMs), with physics serving as the critical test anchor for binding abstract logic to physical reality. Physics demands that a model maintain physical consistency with the laws governing the universe, a task that fundamentally requires multimodal perception to ground abstract logic in reality. At the Olympiad level, diagrams are often constitutive rather than illustrative, containing essential constraints, such as boundary conditions and spatial symmetries, that are absent from the text. To bridge this visual-logical gap, we introduce P1-VL, a family of open-source vision-language models engineered for advanced scientific reasoning. Our method harmonizes Curriculum Reinforcement Learning, which employs progressive difficulty expansion to stabilize post-training, with Agentic Augmentation, enabling iterative self-verification at inference. Evaluated on HiPhO, a rigorous benchmark of 13 exams from 2024-2025, our flagship P1-VL-235B-A22B becomes the first open-source Vision-Language Model (VLM) to secure 12 gold medals and achieves the state-of-the-art performance in the open-source models. Our agent-augmented system achieves the No.2 overall rank globally, trailing only Gemini-3-Pro. Beyond physics, P1-VL demonstrates remarkable scientific reasoning capacity and generalizability, establishing significant leads over base models in STEM benchmarks. By open-sourcing P1-VL, we provide a foundational step toward general-purpose physical intelligence to better align visual perceptions with abstract physical laws for machine scientific discovery.

Yuxuan Chen, Haoyan Yang, Hengkai Pan et al.

Breast ultrasound is essential for detecting and diagnosing abnormalities, with radiology reports summarizing key findings like lesion characteristics and malignancy assessments. Extracting this critical information is challenging due to the unstructured nature of these reports, with varied linguistic styles and inconsistent formatting. While proprietary LLMs like GPT-4 are effective, they are costly and raise privacy concerns when handling protected health information. This study presents a pipeline for developing an in-house LLM to extract clinical information from radiology reports. We first use GPT-4 to create a small labeled dataset, then fine-tune a Llama3-8B model on it. Evaluated on clinician-annotated reports, our model achieves an average F1 score of 84.6%, which is on par with GPT-4. Our findings demonstrate the feasibility of developing an in-house LLM that not only matches GPT-4's performance but also offers cost reductions and enhanced data privacy.

Qingyang Zhang, Haitao Wu, Changqing Zhang et al.

Existing methods to enhance the reasoning capability of large language models predominantly rely on supervised fine-tuning (SFT) followed by reinforcement learning (RL) on reasoning-specific data. These approaches critically depend on external supervisions--such as labeled reasoning traces, verified golden answers, or pre-trained reward models. In this work, we propose Entropy Minimized Policy Optimization (\ours), which makes an early attempt at fully unsupervised LLM reasoning incentivization. By continuously minimizing the predictive entropy of LLMs on unlabeled questions in a latent semantic space, \ours achieves competitive performance compared to supervised counterparts on both mathematical and free-form natural reasoning tasks. Specifically, without any supervised signals, \ours boosts the accuracy of Qwen2.5-Math-7B Base from 30.7\% to 48.1\% on mathematical benchmarks and improves the accuracy of Qwen2.5-7B Base from 32.1\% to 50.1\% on MMLU-Pro. Primary experiments and analysis are also provided to interpret the effectiveness of \ours. Code is available at https://github.com/QingyangZhang/EMPO.

Jiacheng Chen, Qianjia Cheng, Fangchen Yu et al. · tsinghua

Recent progress in large language models (LLMs) has moved the frontier from puzzle-solving to science-grade reasoning-the kind needed to tackle problems whose answers must stand against nature, not merely fit a rubric. Physics is the sharpest test of this shift, which binds symbols to reality in a fundamental way, serving as the cornerstone of most modern technologies. In this work, we manage to advance physics research by developing large language models with exceptional physics reasoning capabilities, especially excel at solving Olympiad-level physics problems. We introduce P1, a family of open-source physics reasoning models trained entirely through reinforcement learning (RL). Among them, P1-235B-A22B is the first open-source model with Gold-medal performance at the latest International Physics Olympiad (IPhO 2025), and wins 12 gold medals out of 13 international/regional physics competitions in 2024/2025. P1-30B-A3B also surpasses almost all other open-source models on IPhO 2025, getting a silver medal. Further equipped with an agentic framework PhysicsMinions, P1-235B-A22B+PhysicsMinions achieves overall No.1 on IPhO 2025, and obtains the highest average score over the 13 physics competitions. Besides physics, P1 models also present great performance on other reasoning tasks like math and coding, showing the great generalibility of P1 series.

Qingyang Zhang, Qiuxuan Feng, Joey Tianyi Zhou et al.

Out-of-distribution (OOD) detection is essential for model trustworthiness which aims to sensitively identify semantic OOD samples and robustly generalize for covariate-shifted OOD samples. However, we discover that the superior OOD detection performance of state-of-the-art methods is achieved by secretly sacrificing the OOD generalization ability. Specifically, the classification accuracy of these models could deteriorate dramatically when they encounter even minor noise. This phenomenon contradicts the goal of model trustworthiness and severely restricts their applicability in real-world scenarios. What is the hidden reason behind such a limitation? In this work, we theoretically demystify the ``\textit{sensitive-robust}'' dilemma that lies in many existing OOD detection methods. Consequently, a theory-inspired algorithm is induced to overcome such a dilemma. By decoupling the uncertainty learning objective from a Bayesian perspective, the conflict between OOD detection and OOD generalization is naturally harmonized and a dual-optimal performance could be expected. Empirical studies show that our method achieves superior performance on standard benchmarks. To our best knowledge, this work is the first principled OOD detection method that achieves state-of-the-art OOD detection performance without compromising OOD generalization ability. Our code is available at \href{https://github.com/QingyangZhang/DUL}{https://github.com/QingyangZhang/DUL}.

Qingyang Zhang, Yake Wei, Zongbo Han et al.

Multimodal fusion focuses on integrating information from multiple modalities with the goal of more accurate prediction, which has achieved remarkable progress in a wide range of scenarios, including autonomous driving and medical diagnosis. However, the reliability of multimodal fusion remains largely unexplored especially under low-quality data settings. This paper surveys the common challenges and recent advances of multimodal fusion in the wild and presents them in a comprehensive taxonomy. From a data-centric view, we identify four main challenges that are faced by multimodal fusion on low-quality data, namely (1) noisy multimodal data that are contaminated with heterogeneous noises, (2) incomplete multimodal data that some modalities are missing, (3) imbalanced multimodal data that the qualities or properties of different modalities are significantly different and (4) quality-varying multimodal data that the quality of each modality dynamically changes with respect to different samples. This new taxonomy will enable researchers to understand the state of the field and identify several potential directions. We also provide discussion for the open problems in this field together with interesting future research directions.

Qingyang Zhang, Yatao Bian, Xinke Kong et al.

Machine learning models must continuously self-adjust themselves for novel data distribution in the open world. As the predominant principle, entropy minimization (EM) has been proven to be a simple yet effective cornerstone in existing test-time adaption (TTA) methods. While unfortunately its fatal limitation (i.e., overconfidence) tends to result in model collapse. For this issue, we propose to Conservatively Minimize the Entropy (COME), which is a simple drop-in replacement of traditional EM to elegantly address the limitation. In essence, COME explicitly models the uncertainty by characterizing a Dirichlet prior distribution over model predictions during TTA. By doing so, COME naturally regularizes the model to favor conservative confidence on unreliable samples. Theoretically, we provide a preliminary analysis to reveal the ability of COME in enhancing the optimization stability by introducing a data-adaptive lower bound on the entropy. Empirically, our method achieves state-of-the-art performance on commonly used benchmarks, showing significant improvements in terms of classification accuracy and uncertainty estimation under various settings including standard, life-long and open-world TTA, i.e., up to $34.5\%$ improvement on accuracy and $15.1\%$ on false positive rate.

Qingyang Zhang, Xinke Kong, Haitao Wu et al.

Test-time training (TTT) adapts model parameters on unlabeled test instances during inference time, which continuously extends capabilities beyond the reach of offline training. Despite initial gains, existing TTT methods for LRMs plateau quickly and do not benefit from additional test-time compute. Without external calibration, the self-generated reward signal increasingly drifts as the policy model evolves, leading to both performance plateaus and diversity collapse. We propose TEMPO, a TTT framework that interleaves policy refinement on unlabeled questions with periodic critic recalibration on a labeled dataset. By formalizing this alternating procedure through the Expectation-Maximization (EM) algorithm, we reveal that prior methods can be interpreted as incomplete variants that omit the crucial recalibration step. Reintroducing this step tightens the evidence lower bound (ELBO) and enables sustained improvement. Across diverse model families (Qwen3 and OLMO3) and reasoning tasks, TEMPO improves OLMO3-7B on AIME 2024 from 33.0% to 51.1% and Qwen3-14B from 42.3% to 65.8%, while maintaining high diversity.

Sijia Cui, Aiyao He, Shuai Xu et al.

Function calling enables large language models (LLMs) to interact with external systems by leveraging tools and APIs. When faced with multi-step tool usage, LLMs still struggle with tool selection, parameter generation, and tool-chain planning. Existing methods typically rely on manually designing task-specific demonstrations, or retrieving from a curated library. These approaches demand substantial expert effort and prompt engineering becomes increasingly complex and inefficient as tool diversity and task difficulty scale. To address these challenges, we propose a self-guided method, Stepwise Experience Recall (SEER), which performs fine-grained, stepwise retrieval from a continually updated experience pool. Instead of relying on static or manually curated library, SEER incrementally augments the experience pool with past successful trajectories, enabling continuous expansion of the pool and improved model performance over time. Evaluated on the ToolQA benchmark, SEER achieves an average improvement of 6.1% on easy and 4.7% on hard questions. We further test SEER on $τ$-bench, which includes two real-world domains. Powered by Qwen2.5-7B and Qwen2.5-72B models, SEER demonstrates substantial accuracy gains of 7.44% and 23.38%, respectively.

Wei Yang, Jinwei Xiao, Hongming Zhang et al.

Recent advancements in Large Language Models (LLMs) have driven growing interest in LLM-based agents for complex planning tasks. To avoid costly agent training, many studies adopted memory mechanism that enhances LLM with offline experiences or online trajectory analysis. However, existing works focus on single-granularity memory derived from dynamic environmental interactions, which are inherently constrained by the quality of the collected experiences. This limitation, in turn, constrain the diversity of knowledge and the flexibility of planning. We propose Coarse-to-Fine Grounded Memory (\Ours{}), a novel framework that grounds coarse-to-fine memories with LLM, thereby fully leverage them for flexible adaptation to diverse scenarios. \Ours{} grounds environmental information into coarse-grained focus points to guide experience collection in training tasks, followed by grounding of actionable hybrid-grained tips from each experience. At inference, \Ours{} retrieves task-relevant experiences and tips to support planning. When facing environmental anomalies, the LLM grounds the current situation into fine-grained key information, enabling flexible self-QA reflection and plan correction.

Zhichao Wang, Xinhai Chen, Qinglin Wang et al.

Partial differential equations (PDEs) form the mathematical foundation for modeling physical systems in science and engineering, where numerical solutions demand rigorous accuracy-efficiency tradeoffs. Mesh movement techniques address this challenge by dynamically relocating mesh nodes to rapidly-varying regions, enhancing both simulation accuracy and computational efficiency. However, traditional approaches suffer from high computational complexity and geometric inflexibility, limiting their applicability, and existing supervised learning-based approaches face challenges in zero-shot generalization across diverse PDEs and mesh topologies.In this paper, we present an Unsupervised and Generalizable Mesh Movement Network (UGM2N). We first introduce unsupervised mesh adaptation through localized geometric feature learning, eliminating the dependency on pre-adapted meshes. We then develop a physics-constrained loss function, M-Uniform loss, that enforces mesh equidistribution at the nodal level.Experimental results demonstrate that the proposed network exhibits equation-agnostic generalization and geometric independence in efficient mesh adaptation. It demonstrates consistent superiority over existing methods, including robust performance across diverse PDEs and mesh geometries, scalability to multi-scale resolutions and guaranteed error reduction without mesh tangling.

Qingyang Zhang, Rose E. Wang, Ana T. Ribeiro et al.

Educator attention is critical for student success, yet how educators distribute their attention across students remains poorly understood due to data and methodological constraints. This study presents the first large-scale computational analysis of educator attention patterns, leveraging over 1 million educator utterances from virtual group tutoring sessions linked to detailed student demographic and academic achievement data. Using natural language processing techniques, we systematically examine the recipient and nature of educator attention. Our findings reveal that educators often provide more attention to lower-achieving students. However, disparities emerge across demographic lines, particularly by gender. Girls tend to receive less attention when paired with boys, even when they are the lower achieving student in the group. Lower-achieving female students in mixed-gender pairs receive significantly less attention than their higher-achieving male peers, while lower-achieving male students receive significantly and substantially more attention than their higher-achieving female peers. We also find some differences by race and English learner (EL) status, with low-achieving Black students receiving additional attention only when paired with another Black student but not when paired with a non-Black peer. In contrast, higher-achieving EL students receive disproportionately more attention than their lower-achieving EL peers. This work highlights how large-scale interaction data and computational methods can uncover subtle but meaningful disparities in teaching practices, providing empirical insights to inform more equitable and effective educational strategies.

Jingqing Ruan, Kaishen Wang, Qingyang Zhang et al.

Many complicated real-world tasks can be broken down into smaller, more manageable parts, and planning with prior knowledge extracted from these simplified pieces is crucial for humans to make accurate decisions. However, replicating this process remains a challenge for AI agents and naturally raises two questions: How to extract discriminative knowledge representation from priors? How to develop a rational plan to decompose complex problems? Most existing representation learning methods employing a single encoder structure are fragile and sensitive to complex and diverse dynamics. To address this issue, we introduce a multiple-encoder and individual-predictor regime to learn task-essential representations from sufficient data for simple subtasks. Multiple encoders can extract adequate task-relevant dynamics without confusion, and the shared predictor can discriminate the task characteristics. We also use the attention mechanism to generate a top-k subtask planning tree, which customizes subtask execution plans in guiding complex decisions on unseen tasks. This process enables forward-looking and globality by flexibly adjusting the depth and width of the planning tree. Empirical results on a challenging platform composed of some basic simple tasks and combinatorially rich synthetic tasks consistently outperform some competitive baselines and demonstrate the benefits of our design.

Shouyong Jiang, Yong Wang, Yaru Hu et al.

Dynamic multi-objective optimisation (DMO) handles optimisation problems with multiple (often conflicting) objectives in varying environments. Such problems pose various challenges to evolutionary algorithms, which have popularly been used to solve complex optimisation problems, due to their dynamic nature and resource restrictions in changing environments. This paper proposes vector autoregressive evolution (VARE) consisting of vector autoregression (VAR) and environment-aware hypermutation to address environmental changes in DMO. VARE builds a VAR model that considers mutual relationship between decision variables to effectively predict the moving solutions in dynamic environments. Additionally, VARE introduces EAH to address the blindness of existing hypermutation strategies in increasing population diversity in dynamic scenarios where predictive approaches are unsuitable. A seamless integration of VAR and EAH in an environment-adaptive manner makes VARE effective to handle a wide range of dynamic environments and competitive with several popular DMO algorithms, as demonstrated in extensive experimental studies. Specially, the proposed algorithm is computationally 50 times faster than two widely-used algorithms (i.e., TrDMOEA and MOEA/D-SVR) while producing significantly better results.

Feilong Chen, Minglun Han, Haozhi Zhao et al.

Large language models (LLMs) have demonstrated remarkable language abilities. GPT-4, based on advanced LLMs, exhibits extraordinary multimodal capabilities beyond previous visual language models. We attribute this to the use of more advanced LLMs compared with previous multimodal models. Unfortunately, the model architecture and training strategies of GPT-4 are unknown. To endow LLMs with multimodal capabilities, we propose X-LLM, which converts Multi-modalities (images, speech, videos) into foreign languages using X2L interfaces and inputs them into a large Language model (ChatGLM). Specifically, X-LLM aligns multiple frozen single-modal encoders and a frozen LLM using X2L interfaces, where ``X'' denotes multi-modalities such as image, speech, and videos, and ``L'' denotes languages. X-LLM's training consists of three stages: (1) Converting Multimodal Information: The first stage trains each X2L interface to align with its respective single-modal encoder separately to convert multimodal information into languages. (2) Aligning X2L representations with the LLM: single-modal encoders are aligned with the LLM through X2L interfaces independently. (3) Integrating multiple modalities: all single-modal encoders are aligned with the LLM through X2L interfaces to integrate multimodal capabilities into the LLM. Our experiments show that X-LLM demonstrates impressive multimodel chat abilities, sometimes exhibiting the behaviors of multimodal GPT-4 on unseen images/instructions, and yields a 84.5\% relative score compared with GPT-4 on a synthetic multimodal instruction-following dataset. And we also conduct quantitative tests on using LLM for ASR and multimodal ASR, hoping to promote the era of LLM-based speech recognition.

Shuwei Zhang, Maiqi Tang, Qingyang Zhang et al.

In our project, we focus on NLP-based hybrid recommendation systems. Our data is from Yelp Data. For our hybrid recommendation system, we have two major components: the first part is to embed the reviews with the Bert model and word2vec model; the second part is the implementation of an item-based collaborative filtering algorithm to compute the similarity of each review under different categories of restaurants. In the end, with the help of similarity scores, we are able to recommend users the most matched restaurant based on their recorded reviews. The coding work is split into several parts: selecting samples and data cleaning, processing, embedding, computing similarity, and computing prediction and error. Due to the size of the data, each part will generate one or more JSON files as the milestone to reduce the pressure on memory and the communication between each part.

Chenggong Zhang, Juan Song, Qingyang Zhang et al.

This paper describes an approach to the facial action unit (AU) detection. In this work, we present our submission to the Field Affective Behavior Analysis (ABAW) 2021 competition. The proposed method uses the pre-trained JAA model as the feature extractor, and extracts global features, face alignment features and AU local features on the basis of multi-scale features. We take the AU local features as the input of the graph convolution to further consider the correlation between AU, and finally use the fused features to classify AU. The detected accuracy was evaluated by 0.5*accuracy + 0.5*F1. Our model achieves 0.674 on the challenging Aff-Wild2 database.

Liangkai Liu, Sidi Lu, Ren Zhong et al.

The recent proliferation of computing technologies (e.g., sensors, computer vision, machine learning, and hardware acceleration), and the broad deployment of communication mechanisms (e.g., DSRC, C-V2X, 5G) have pushed the horizon of autonomous driving, which automates the decision and control of vehicles by leveraging the perception results based on multiple sensors. The key to the success of these autonomous systems is making a reliable decision in real-time fashion. However, accidents and fatalities caused by early deployed autonomous vehicles arise from time to time. The real traffic environment is too complicated for current autonomous driving computing systems to understand and handle. In this paper, we present state-of-the-art computing systems for autonomous driving, including seven performance metrics and nine key technologies, followed by twelve challenges to realize autonomous driving. We hope this paper will gain attention from both the computing and automotive communities and inspire more research in this direction.