arXiv subjects

Jihao Huang

Publications and source records attributed to Jihao Huang.

17 recordsLinked to original sources

arXiv2026

CompanionBench: A Theory-Anchored, Real-World-Grounded Benchmark for AI Emotional Companionship

LLM companions are deployed at scale in personally consequential settings, yet poorly evaluated. Existing benchmarks use hand-authored scenarios and prompted simulators, aggregate empathy into one score, and overlook judge biases such as same-family favoritism and scale drift. We introduce CompanionBench, an interactive bilingual benchmark. To our knowledge, it is the first companion benchmark to ground both its scenarios and a trained user simulator in de-identified real-world data. A hidden disclosure gate branches each persona's trajectory on the agent's own behavior, controlling the interaction state space without scripting dialogue. We operationalize ten capabilities derived from 25 theories across psychology and counseling, four of them not graded explicitly by prior work: holding ambiguity, selfobject responsiveness, positive resonance and calibrated challenge. Agents are assessed on two complementary axes: a subjective ten-capability rubric and a deterministic measure of whether deeper disclosure was earned. A cross-family panel dilutes same-family favoritism; an Item Response Theory model separates agent quality from judge severity. Theory fixes what to measure and how personas are structured; real data supply events, history, and profiles -- coverage from theory, authenticity from data. Rankings are reproducible in both languages (rho = 0.996 ZH / 0.953 EN). Evaluating 28 agents reveals capability-level differences obscured by aggregate scores. Emotion regulation and calibrated challenge remain common weaknesses; holding ambiguity discriminates most. Role-play agents rank near the bottom: immersion does not imply relational competence. Across agents, the dominant failure mode is substituting surface warmth for substantive relational support. We will release 500 Chinese-English parallel pairs and the evaluation code.

cs.CL↗

arXiv2026

GDEPO: Group Dual-dynamic and Equal-right Advantage Policy Optimization with Enhanced Training Data Utilization for Sample-Constrained Reinforcement Learning

Automated Theorem Proving (ATP) represents a fundamental challenge in Artificial Intelligence (AI), requiring the construction of machine-verifiable proofs in formal languages such as Lean to evaluate AI reasoning capabilities. Reinforcement learning (RL), particularly the high-performance Group Relative Policy Optimization (GRPO) algorithm, has emerged as a mainstream approach for this task. However, in ATP scenarios, GRPO faces two critical issues: when composite rewards are used, its relative advantage estimation may conflict with the binary feedback from the formal verifier; meanwhile, its static sampling strategy may discard entire batches of data if no valid proof is found, resulting in zero contribution to model updates and significant data waste. To address these limitations, we propose Group Dual-dynamic and Equal-right-advantage Policy Optimization (GDEPO), a method incorporating three core mechanisms: 1) dynamic additional sampling, which resamples invalid batches until a valid proof is discovered; 2) equal-right advantage, decoupling the sign of the advantage function (based on correctness) from its magnitude (modulated by auxiliary rewards) to ensure stable and correct policy updates; and 3) dynamic additional iterations, applying extra gradient steps to initially failed but eventually successful samples to accelerate learning on challenging cases. Experiments conducted on three datasets of varying difficulty (MinF2F-test, MathOlympiadBench, PutnamBench) confirm the effectiveness of GDEPO, while ablation studies validate the necessity of its synergistic components. The proposed method enhances data utilization and optimization efficiency, offering a novel training paradigm for ATP.

cs.AI↗

arXiv2025

DataFlow: An LLM-Driven Framework for Unified Data Preparation and Workflow Automation in the Era of Data-Centric AI

The rapidly growing demand for high-quality data in Large Language Models (LLMs) has intensified the need for scalable, reliable, and semantically rich data preparation pipelines. However, current practices remain dominated by ad-hoc scripts and loosely specified workflows, which lack principled abstractions, hinder reproducibility, and offer limited support for model-in-the-loop data generation. To address these challenges, we present DataFlow, a unified and extensible LLM-driven data preparation framework. DataFlow is designed with system-level abstractions that enable modular, reusable, and composable data transformations, and provides a PyTorch-style pipeline construction API for building debuggable and optimizable dataflows. The framework consists of nearly 200 reusable operators and six domain-general pipelines spanning text, mathematical reasoning, code, Text-to-SQL, agentic RAG, and large-scale knowledge extraction. To further improve usability, we introduce DataFlow-Agent, which automatically translates natural-language specifications into executable pipelines via operator synthesis, pipeline planning, and iterative verification. Across six representative use cases, DataFlow consistently improves downstream LLM performance. Our math, code, and text pipelines outperform curated human datasets and specialized synthetic baselines, achieving up to +3\% execution accuracy in Text-to-SQL over SynSQL, +7\% average improvements on code benchmarks, and 1--3 point gains on MATH, GSM8K, and AIME. Moreover, a unified 10K-sample dataset produced by DataFlow enables base models to surpass counterparts trained on 1M Infinity-Instruct data. These results demonstrate that DataFlow provides a practical and high-performance substrate for reliable, reproducible, and scalable LLM data preparation, and establishes a system-level foundation for future data-centric AI development.

cs.LG↗

arXiv2025

LADY: Linear Attention for Autonomous Driving Efficiency without Transformers

End-to-end autonomous driving has emerged as a promising paradigm. However, state-of-the-art methods rely heavily on Transformer architectures. The inherent quadratic complexity of Transformers restricts their ability to model long-range spatial and temporal dependencies, particularly on resource-constrained edge platforms. Given the inherent demand for efficient temporal modeling in autonomous driving, this computational bottleneck severely constrains real-time deployment. While linear attention mechanisms offer a computationally efficient alternative, existing architectures are predominantly limited to self-attention, lacking the cross-modal capabilities essential for autonomous driving. In this work, we propose LADY, the first fully linear attention-based generative model for end-to-end autonomous driving. LADY incorporates a novel, lightweight linear cross-attention (LICA) mechanism to enable effective cross-modal interaction while preserving linearity. A key advantage of our framework is its ability to fuse long-range temporal contexts during inference with constant computational and memory costs ($O(1)$), regardless of the historical sequence length. Experiments on the NAVSIM and Bench2Drive benchmarks demonstrate that LADY achieves performance comparable to state-of-the-art methods, delivering competitive planning accuracy with significantly reduced latency. Furthermore, efficiency benchmarking on edge devices validates the model's feasibility for resource-limited scenarios.

cs.AI↗

arXiv2025

FlipVQA: Scaling Multi-modal Instruction Tuning via Textbook-to-Knowledge Synthesis

Textbooks are among the richest repositories of human-verified reasoning knowledge, yet their complex layouts contain multi-column typesetting, cross-page question answer separation, and interleaved figures, make automated extraction of structured QA and VQA pairs extremely challenging. Existing alternatives either synthesize data from scratch, which lacks authentic problem contexts, or rely on costly expert annotation that cannot scale. We propose $\textbf{FlipVQA-Miner}$, an automated pipeline that resolves long-range logical dependencies and cross-page discontinuities in OCR-parsed documents, recovering coherent question--answer--figure associations even when answers reside in separate companion volumes. A subsequent multi-stage curation pipeline transforms these raw extractions into AI-ready supervision signals. Using FlipVQA-Miner, we construct $\textbf{FlipVQA-83K}$, comprising 83K QA and VQA pairs spanning 11 academic disciplines, at a $\textbf{50$\times$}$ cost saving compared to manual annotation while maintaining high structural fidelity ($F_1 > 0.96$). Models fine-tuned on FlipVQA-83K demonstrate significantly improved reasoning ability and cross-domain generalization, establishing a scalable paradigm for human-knowledge-grounded data curation. Our dataset and the complete data generating and curating methods can be found in https://github.com/OpenDCAI/DataFlow-VQA .

cs.AI↗

arXiv2025

FSDP: Fast and Safe Data-Driven Overtaking Trajectory Planning for Head-to-Head Autonomous Racing Competitions

Generating overtaking trajectories in autonomous racing is a challenging task, as the trajectory must satisfy the vehicle's dynamics and ensure safety and real-time performance running on resource-constrained hardware. This work proposes the Fast and Safe Data-Driven Planner to address this challenge. Sparse Gaussian predictions are introduced to improve both the computational efficiency and accuracy of opponent predictions. Furthermore, the proposed approach employs a bi-level quadratic programming framework to generate an overtaking trajectory leveraging the opponent predictions. The first level uses polynomial fitting to generate a rough trajectory, from which reference states and control inputs are derived for the second level. The second level formulates a model predictive control optimization problem in the Frenet frame, generating a trajectory that satisfies both kinematic feasibility and safety. Experimental results on the F1TENTH platform show that our method outperforms the State-of-the-Art, achieving an 8.93% higher overtaking success rate, allowing the maximum opponent speed, ensuring a smoother ego trajectory, and reducing 74.04% computational time compared to the Predictive Spliner method. The code is available at: https://github.com/ZJU-DDRX/FSDP.

cs.RO↗

arXiv2025

Efficient and Real-Time Motion Planning for Robotics Using Projection-Based Optimization

Generating motions for robots interacting with objects of various shapes is a complex challenge, further complicated by the robot geometry and multiple desired behaviors. While current robot programming tools (such as inverse kinematics, collision avoidance, and manipulation planning) often treat these problems as constrained optimization, many existing solvers focus on specific problem domains or do not exploit geometric constraints effectively. We propose an efficient first-order method, Augmented Lagrangian Spectral Projected Gradient Descent (ALSPG), which leverages geometric projections via Euclidean projections, Minkowski sums, and basis functions. We show that by using geometric constraints rather than full constraints and gradients, ALSPG significantly improves real-time performance. Compared to second-order methods like iLQR, ALSPG remains competitive in the unconstrained case. We validate our method through toy examples and extensive simulations, and demonstrate its effectiveness on a 7-axis Franka robot, a 6-axis P-Rob robot and a 1:10 scale car in real-world experiments. Source codes, experimental data and videos are available on the project webpage: https://sites.google.com/view/alspg-oc

cs.RO↗

arXiv2025

Dynamic Collision Avoidance Using Velocity Obstacle-Based Control Barrier Functions

Designing safety-critical controllers for acceleration-controlled unicycle robots is challenging, as control inputs may not appear in the constraints of control Lyapunov functions(CLFs) and control barrier functions (CBFs), leading to invalid controllers. Existing methods often rely on state-feedback-based CLFs and high-order CBFs (HOCBFs), which are computationally expensive to construct and fail to maintain effectiveness in dynamic environments with fast-moving, nearby obstacles. To address these challenges, we propose constructing velocity obstacle-based CBFs (VOCBFs) in the velocity space to enhance dynamic collision avoidance capabilities, instead of relying on distance-based CBFs that require the introduction of HOCBFs. Additionally, by extending VOCBFs using variants of VO, we enable reactive collision avoidance between robots. We formulate a safety-critical controller for acceleration-controlled unicycle robots as a mixed-integer quadratic programming (MIQP), integrating state-feedback-based CLFs for navigation and VOCBFs for collision avoidance. To enhance the efficiency of solving the MIQP, we split the MIQP into multiple sub-optimization problems and employ a decision network to reduce computational costs. Numerical simulations demonstrate that our approach effectively guides the robot to its target while avoiding collisions. Compared to HOCBFs, VOCBFs exhibit significantly improved dynamic obstacle avoidance performance, especially when obstacles are fast-moving and close to the robot. Furthermore, we extend our method to distributed multi-robot systems.

cs.RO↗

arXiv2024

Safe Dynamic Motion Generation in Configuration Space Using Differentiable Distance Fields

Generating collision-free motions in dynamic environments is a challenging problem for high-dimensional robotics, particularly under real-time constraints. Control Barrier Functions (CBFs), widely utilized in safety-critical control, have shown significant potential for motion generation. However, for high-dimensional robot manipulators, existing QP formulations and CBF-based methods rely on positional information, overlooking higher-order derivatives such as velocities. This limitation may lead to reduced success rates, decreased performance, and inadequate safety constraints. To address this, we construct time-varying CBFs (TVCBFs) that consider velocity conditions for obstacles. Our approach leverages recent developments on distance fields for articulated manipulators, a differentiable representation that enables the mapping of objects' position and velocity into the robot's joint space, offering a comprehensive understanding of the system's interactions. This allows the manipulator to be treated as a point-mass system thus simplifying motion generation tasks. Additionally, we introduce a time-varying control Lyapunov function (TVCLF) to enable whole-body contact motions. Our approach integrates the TVCBF, TVCLF, and manipulator physical constraints within a unified QP framework. We validate our method through simulations and comparisons with state-of-the-art approaches, demonstrating its effectiveness on a 7-axis Franka robot in real-world experiments.

cs.RO↗

arXiv2024

Velocity Obstacle for Polytopic Collision Avoidance for Distributed Multi-robot Systems

Obstacle avoidance for multi-robot navigation with polytopic shapes is challenging. Existing works simplify the system dynamics or consider it as a convex or non-convex optimization problem with positive distance constraints between robots, which limits real-time performance and scalability. Additionally, generating collision-free behavior for polytopic-shaped robots is harder due to implicit and non-differentiable distance functions between polytopes. In this paper, we extend the concept of velocity obstacle (VO) principle for polytopic-shaped robots and propose a novel approach to construct the VO in the function of vertex coordinates and other robot's states. Compared with existing work about obstacle avoidance between polytopic-shaped robots, our approach is much more computationally efficient as the proposed approach for construction of VO between polytopes is optimization-free. Based on VO representation for polytopic shapes, we later propose a navigation approach for distributed multi-robot systems. We validate our proposed VO representation and navigation approach in multiple challenging scenarios including large-scale randomized tests, and our approach outperforms the state of art in many evaluation metrics, including completion rate, deadlock rate, and the average travel distance.

cs.RO↗

arXiv2024

GeoPro-VO: Dynamic Obstacle Avoidance with Geometric Projector Based on Velocity Obstacle

Optimization-based approaches are widely employed to generate optimal robot motions while considering various constraints, such as robot dynamics, collision avoidance, and physical limitations. It is crucial to efficiently solve the optimization problems in practice, yet achieving rapid computations remains a great challenge for optimization-based approaches with nonlinear constraints. In this paper, we propose a geometric projector for dynamic obstacle avoidance based on velocity obstacle (GeoPro-VO) by leveraging the projection feature of the velocity cone set represented by VO. Furthermore, with the proposed GeoPro-VO and the augmented Lagrangian spectral projected gradient descent (ALSPG) algorithm, we transform an initial mixed integer nonlinear programming problem (MINLP) in the form of constrained model predictive control (MPC) into a sub-optimization problem and solve it efficiently. Numerical simulations are conducted to validate the fast computing speed of our approach and its capability for reliable dynamic obstacle avoidance.

cs.RO↗

arXiv2024

When ChatGPT is gone: Creativity reverts and homogeneity persists

ChatGPT has been evidenced to enhance human performance in creative tasks. Yet, it is still unclear if this boosting effect sustains with and without ChatGPT. In a pre-registered seven-day lab experiment and a follow-up survey after 30 days of experiment completion, we examined the impacts of ChatGPT presence and absence on sustained creativity using a text dataset of 3302 creative ideas and 427 creative solutions from 61 college students. Participants in the treatment group used ChatGPT in creative tasks, while those in the control group completed the tasks by themselves. The findings show that although the boosting effect of ChatGPT was consistently observed over a five-day creative journey, human creative performance reverted to baseline when ChatGPT was down on the 7th and the 30th day. More critically, the use of ChatGPT in creative tasks resulted in increasingly homogenized contents, and this homogenization effect persisted even when ChatGPT was absence. These findings pose a challenge to the prevailing argument that ChatGPT can enhance human creativity. In fact, generative AI like ChatGPT lends to human with a temporary rise in creative performance but boxes human creative capability in the long run, highlighting the imperative for cautious generative AI integration in creative endeavors.

cs.CL↗

arXiv2023

Whole-body Dynamic Collision Avoidance with Time-varying Control Barrier Functions

Recently, there has been increasing attention in robot research towards the whole-body collision avoidance. In this paper, we propose a safety-critical controller that utilizes time-varying control barrier functions (time varying CBFs) constructed by Robo-centric Euclidean Signed Distance Field (RC-ESDF) to achieve dynamic collision avoidance. The RC-ESDF is constructed in the robot body frame and solely relies on the robot's shape, eliminating the need for real-time updates to save computational resources. Additionally, we design two control Lyapunov functions (CLFs) to ensure that the robot can reach its destination. To enable real-time application, our safety-critical controller which incorporates CLFs and CBFs as constraints is formulated as a quadratic program (QP) optimization problem. We conducted numerical simulations on two different dynamics of an L-shaped robot to verify the effectiveness of our proposed approach.

cs.RO↗

arXiv2023

Optimization-based Motion Planning for Autonomous Parking Considering Dynamic Obstacle: A Hierarchical Framework

This paper introduces a hierarchical framework that integrates graph search algorithms and model predictive control to facilitate efficient parking maneuvers for Autonomous Vehicles (AVs) in constrained environments. In the high-level planning phase, the framework incorporates scenario-based hybrid A* (SHA*), an optimized variant of traditional Hybrid A*, to generate an initial path while considering static obstacles. This global path serves as an initial guess for the low-level NLP problem. In the low-level optimizing phase, a nonlinear model predictive control (NMPC)-based framework is deployed to circumvent dynamic obstacles. The performance of SHA* is empirically validated through 148 simulation scenarios, and the efficacy of the proposed hierarchical framework is demonstrated via a real-time parallel parking simulation.

cs.RO↗

arXiv2023

Fast Path Planning for Autonomous Vehicle Parking with Safety-Guarantee using Hamilton-Jacobi Reachability

We present a fast planning architecture called Hamilton-Jacobi-based bidirectional A* (HJBA*) to solve general tight parking scenarios. The algorithm is a two-layer composed of a high-level HJ-based reachability analysis and a lower-level bidirectional A* search algorithm. In high-level reachability analysis, a backward reachable tube (BRT) concerning vehicle dynamics is computed by the HJ analysis and it intersects with a safe set to get a safe reachable set. The safe set is defined by constraints of positive signed distances for obstacles in the environment and computed by solving QP optimization problems offline. For states inside the intersection set, i.e., the safe reachable set, the computed backward reachable tube ensures they are reachable subjected to system dynamics and input bounds, and the safe set guarantees they satisfy parking safety with respect to obstacles in different shapes. For online computation, randomized states are sampled from the safe reachable set, and used as heuristic guide points to be considered in the bidirectional A* search. The bidirectional A* search is paralleled for each randomized state from the safe reachable set. We show that the proposed two-level planning algorithm is able to solve different parking scenarios effectively and computationally fast for typical parking requests. We validate our algorithm through simulations in large-scale randomized parking scenarios and demonstrate it to be able to outperform other state-of-the-art parking planning algorithms.

cs.RO↗

arXiv2023

Obstacle Avoidance for Unicycle-Modelled Mobile Robots with Time-varying Control Barrier Functions

In this paper, we propose a safety-critical controller based on time-varying control barrier functions (CBFs) for a robot with an unicycle model in the continuous-time domain to achieve navigation and dynamic collision avoidance. Unlike previous works, our proposed approach can control both linear and angular velocity to avoid collision with obstacles, overcoming the limitation of confined control performance due to the lack of control variable. To ensure that the robot reaches its destination, we also design a control Lyapunov function (CLF). Our safety-critical controller is formulated as a quadratic program (QP) optimization problem that incorporates CLF and CBFs as constraints, enabling real-time application for navigation and dynamic collision avoidance. Numerical simulations are conducted to verify the effectiveness of our proposed approach.

cs.RO↗

arXiv2022

Search-Based Path Planning Algorithm for Autonomous Parking:Multi-Heuristic Hybrid A*

This paper proposed a novel method for autonomous parking. Autonomous parking has received a lot of attention because of its convenience, but due to the complex environment and the non-holonomic constraints of vehicle, it is difficult to get a collision-free and feasible path in a short time. To solve this problem, this paper introduced a novel algorithm called Multi-Heuristic Hybrid A* (MHHA*) which incorporates the characteristic of Multi-Heuristic A* and Hybrid A*. So it could provide the guarantee for completeness, the avoidance of local minimum and sub-optimality, and generate a feasible path in a short time. And this paper also proposed a new collision check method based on coordinate transformation which could improve the computational efficiency. The performance of the proposed method was compared with Hybrid A* in simulation experiments and its superiority has been proved.

cs.RO↗