Searcharxiv⌕ Search

arXiv subjects

Yuchen Hou

Publications and source records attributed to Yuchen Hou.

10 recordsLinked to original sources
arXiv2026

FedQHD: Closed-Form Function-Space Federated Reinforcement Learning

Federated reinforcement learning enables decentralized agents to collaboratively improve policies or value estimates without exchanging raw trajectories. However, FedAvg-style parameter averaging is not function-space consistent: when clients use heterogeneous encoders or even identical nonlinear networks, averaged parameters need not correspond to the weighted average of client value functions in any common function space. We propose FedQHD, a federated Q-learning method using hyperdimensional (random-feature) state encoders with a linear readout, so that Q-functions are nonlinear in state yet linear in trainable parameters. This linear structure enables closed-form aggregation. With a shared encoder, the function-space consensus update coincides exactly with weighted averaging of local readout matrices. With heterogeneous encoders, the server constructs a global teacher by averaging client Q-values on a shared anchor-state set, and each client compiles this teacher into its local representation via a single ridge projection. We formalize the federation gap -- the error incurred when compiling a federated teacher into a heterogeneous client representation -- relative to a client-specific oracle projection. We show that this gap decomposes into subspace misalignment, anchor-set conditioning, and regularization bias. We further identify the anchor-to-dimension ratio $m \geq D_i$ as the well-conditioned regime in which the gap reduces to a multiple of the encoder heterogeneity floor. On four continuous-state, discrete-action control benchmarks, FedQHD matches or outperforms FedAvg-style baselines and distillation-based alternatives while requiring substantially less computation, and the empirical dependence of the federation gap on encoder dimension matches our theoretical analysis.

cs.LG↗
arXiv2026

Beyond Neural Activity Prediction: Probing Latent Representations in Mouse V1 Digital Twins

Digital twins of sensory cortex serve as powerful response oracles. Although prediction accuracy is the central metric by which these models are evaluated, it provides limited insight into the latent representations that support those predictions. This becomes increasingly important as digital twins are used as in silico experimental systems for stimulus design and hypothesis generation: models with similar prediction accuracy may rely on different latent representations. We address this gap by systematically probing a family of digital twins of mouse V1 trained to predict neural activity from naturalistic videos recorded in freely moving mice. The models share the same training data and neural-prediction objective, but differ in visual-encoder architecture. For each frozen model, we characterize latent representations along three levels: (i) linear decodability from controlled visual probes of orientation, contrast, and motion; (ii) latent-unit tuning to canonical visual features including orientation selectivity, contrast response, spatial-frequency tuning; and (iii) population geometry of hidden-layer activity. Across architectures, better neural-response prediction correlates with stronger probe accuracy. Additionally, highly predictive models exhibit flatter hidden-population eigenspectra, indicating higher-dimensional representations closer to population-geometry signatures reported in mouse V1. Although these representational properties covary with prediction accuracy across architectures, digital twins with comparable prediction scores can still differ substantially in probe performance and latent-unit tuning. These results establish multi-level representational probing as a complement to standard neural-prediction evaluation, providing a framework for understanding digital twins not only as predictors, but also as substrates for studying visual computations.

q-bio.NC↗
arXiv2026

LangGap: Diagnosing and Closing the Language Gap in Vision-Language-Action Models

Vision-Language-Action (VLA) models achieve over 95% success on standard benchmarks. However, through systematic experiments, we find that current state-of-the-art VLA models largely ignore language instructions. Prior work lacks: (1) systematic semantic perturbation diagnostics, (2) a benchmark that forces language understanding by design, and (3) linguistically diverse training data. This paper constructs the LangGap benchmark, based on a four-dimensional semantic perturbation method -- varying instruction semantics while keeping the tabletop layout fixed -- revealing language understanding deficits in {\pi}0.5. Existing benchmarks like LIBERO assign only one task per layout, underutilizing available objects and target locations; LangGap fully diversifies pick-and-place tasks under identical layouts, forcing models to truly understand language. Experiments show that targeted data augmentation can partially close the language gap -- success rate improves from 0% to 90% with single-task training, and 0% to 28% with multi-task training. However, as semantic diversity of extended tasks increases, model learning capacity proves severely insufficient; even trained tasks perform poorly. This reveals a fundamental challenge for VLA models in understanding diverse language instructions -- precisely the long-term value of LangGap.

cs.RO↗
arXiv2026

AgentIF-OneDay: A Task-level Instruction-Following Benchmark for General AI Agents in Daily Scenarios

The capacity of AI agents to effectively handle tasks of increasing duration and complexity continues to grow, demonstrating exceptional performance in coding, deep research, and complex problem-solving evaluations. However, in daily scenarios, the perception of these advanced AI capabilities among general users remains limited. We argue that current evaluations prioritize increasing task difficulty without sufficiently addressing the diversity of agentic tasks necessary to cover the daily work, life, and learning activities of a broad demographic. To address this, we propose AgentIF-OneDay, aimed at determining whether general users can utilize natural language instructions and AI agents to complete a diverse array of daily tasks. These tasks require not only solving problems through dialogue but also understanding various attachment types and delivering tangible file-based results. The benchmark is structured around three user-centric categories: Open Workflow Execution, which assesses adherence to explicit and complex workflows; Latent Instruction, which requires agents to infer implicit instructions from attachments; and Iterative Refinement, which involves modifying or expanding upon ongoing work. We employ instance-level rubrics and a refined evaluation pipeline that aligns LLM-based verification with human judgment, achieving an 80.1% agreement rate using Gemini-3-Pro. AgentIF-OneDay comprises 104 tasks covering 767 scoring points. We benchmarked four leading general AI agents and found that agent products built based on APIs and ChatGPT agents based on agent RL remain in the first tier simultaneously. Leading LLM APIs and open-source models have internalized agentic capabilities, enabling AI application teams to develop cutting-edge Agent products.

cs.CL↗
arXiv2026

AION: Aerial Indoor Object-Goal Navigation Using Dual-Policy Reinforcement Learning

Object-Goal Navigation (ObjectNav) requires an agent to autonomously explore an unknown environment and navigate toward target objects specified by a semantic label. While prior work has primarily studied zero-shot ObjectNav under 2D locomotion, extending it to aerial platforms with 3D locomotion capability remains underexplored. Aerial robots offer superior maneuverability and search efficiency, but also introduce new challenges in spatial perception, dynamic control, and safety assurance. In this paper, we propose AION for vision-based aerial ObjectNav without relying on external localization or global maps. AION is an end-to-end dual-policy reinforcement learning (RL) framework that decouples exploration and goal-reaching behaviors into two specialized policies. We evaluate AION on the AI2-THOR benchmark and further assess its real-time performance in IsaacSim using high-fidelity drone models. Experimental results show that AION achieves superior performance across comprehensive evaluation metrics in exploration, navigation efficiency, and safety. The project is available at https://github.com/Zichen-Yan/AION.

cs.RO↗
arXiv2025

Mouse vs. AI: A Neuroethological Benchmark for Visual Robustness and Neural Alignment

Visual robustness under real-world conditions remains a critical bottleneck for modern reinforcement learning agents. In contrast, biological systems such as mice show remarkable resilience to environmental changes, maintaining stable performance even under degraded visual input with minimal exposure. Inspired by this gap, we propose the Mouse vs. AI: Robust Foraging Competition, a novel bioinspired visual robustness benchmark to test generalization in reinforcement learning (RL) agents trained to navigate a virtual environment toward a visually cued target. Participants train agents to perform a visually guided foraging task in a naturalistic 3D Unity environment and are evaluated on their ability to generalize to unseen, ecologically realistic visual perturbations. What sets this challenge apart is its biological grounding: real mice performed the same task, and participants receive both behavioral performance data and large-scale neural recordings (over 19,000 neurons across visual cortex) for benchmarking. The competition features two tracks: (1) Visual Robustness, assessing generalization across held-out visual perturbations; and (2) Neural Alignment, evaluating how well agents' internal representations predict mouse visual cortical activity via a linear readout. We provide the full Unity environment, a fog-perturbed training condition for validation, baseline proximal policy optimization (PPO) agents, and a rich multimodal dataset. By bridging reinforcement learning, computer vision, and neuroscience through a shared, behaviorally grounded task, this challenge advances the development of robust, generalizable, and biologically inspired AI.

q-bio.NC↗
arXiv2025

GNNs as Predictors of Agentic Workflow Performances

Agentic workflows invoked by Large Language Models (LLMs) have achieved remarkable success in handling complex tasks. However, optimizing such workflows is costly and inefficient in real-world applications due to extensive invocations of LLMs. To fill this gap, this position paper formulates agentic workflows as computational graphs and advocates Graph Neural Networks (GNNs) as efficient predictors of agentic workflow performances, avoiding repeated LLM invocations for evaluation. To empirically ground this position, we construct FLORA-Bench, a unified platform for benchmarking GNNs for predicting agentic workflow performances. With extensive experiments, we arrive at the following conclusion: GNNs are simple yet effective predictors. This conclusion supports new applications of GNNs and a novel direction towards automating agentic workflow optimization. All codes, models, and data are available at https://github.com/youngsoul0731/Flora-Bench.

cs.CL↗
arXiv2024

Self-Evolving Multi-Agent Collaboration Networks for Software Development

LLM-driven multi-agent collaboration (MAC) systems have demonstrated impressive capabilities in automatic software development at the function level. However, their heavy reliance on human design limits their adaptability to the diverse demands of real-world software development. To address this limitation, we introduce EvoMAC, a novel self-evolving paradigm for MAC networks. Inspired by traditional neural network training, EvoMAC obtains text-based environmental feedback by verifying the MAC network's output against a target proxy and leverages a novel textual backpropagation to update the network. To extend coding capabilities beyond function-level tasks to more challenging software-level development, we further propose rSDE-Bench, a requirement-oriented software development benchmark, which features complex and diverse software requirements along with automatic evaluation of requirement correctness. Our experiments show that: i) The automatic requirement-aware evaluation in rSDE-Bench closely aligns with human evaluations, validating its reliability as a software-level coding benchmark. ii) EvoMAC outperforms previous SOTA methods on both the software-level rSDE-Bench and the function-level HumanEval benchmarks, reflecting its superior coding capabilities. The benchmark can be downloaded at https://yuzhu-cai.github.io/rSDE-Bench/.

cs.SE↗
arXiv2024

Predicting the Temporal Dynamics of Prosthetic Vision

Retinal implants are a promising treatment option for degenerative retinal disease. While numerous models have been developed to simulate the appearance of elicited visual percepts ("phosphenes"), these models often either focus solely on spatial characteristics or inadequately capture the complex temporal dynamics observed in clinical trials, which vary heavily across implant technologies, subjects, and stimulus conditions. Here we introduce two computational models designed to accurately predict phosphene fading and persistence under varying stimulus conditions, cross-validated on behavioral data reported by nine users of the Argus II Retinal Prosthesis System. Both models segment the time course of phosphene perception into discrete intervals, decomposing phosphene fading and persistence into either sinusoidal or exponential components. Our spectral model demonstrates state-of-the-art predictions of phosphene intensity over time (r = 0.7 across all participants). Overall, this study lays the groundwork for enhancing prosthetic vision by improving our understanding of phosphene temporal dynamics.

cs.CE↗