arXiv subjects

Yifei Yang

Publications and source records attributed to Yifei Yang.

At least 37 records · Page 2Linked to original sources

arXiv2026

PROSPECT: Unified Streaming Vision-Language Navigation via Semantic--Spatial Fusion and Latent Predictive Representation

Multimodal large language models (MLLMs) have advanced zero-shot end-to-end Vision-Language Navigation (VLN), yet robust navigation requires not only semantic understanding but also predictive modeling of environment dynamics and spatial structure. We propose PROSPECT, a unified streaming navigation agent that couples a streaming Vision-Language-Action (VLA) policy with latent predictive representation learning. PROSPECT uses CUT3R as a streaming 3D foundation spatial encoder to produce long-context, absolute-scale spatial features, and fuses them with SigLIP semantic features via cross-attention. During training, we introduce learnable stream query tokens that query the streaming context and predict next-step 2D and 3D latent features (rather than pixels or explicit modalities), supervised in the latent spaces of frozen SigLIP and CUT3R teachers. The predictive branch shapes internal representations without inference overhead. Experiments on VLN-CE benchmarks and real-robot deployment demonstrate state-of-the-art performance and improved long-horizon robustness under diverse lighting. We will release code for the community soon.

cs.CV↗

arXiv2025

Rethinking Analytical Processing in the GPU Era

The era of GPU-powered data analytics has arrived. In this paper, we argue that recent advances in hardware (e.g., larger GPU memory, faster interconnect and IO, and declining cost) and software (e.g., composable data systems and mature libraries) have removed the key barriers that have limited the wider adoption of GPU data analytics. We present Sirius, a prototype open-source GPU-native SQL engine that offers drop-in acceleration for diverse data systems. Sirius treats GPU as the primary engine and leverages libraries like libcudf for high-performance relational operators. It provides drop-in acceleration for existing databases by leveraging the standard Substrait query representation, replacing the CPU engine without changing the user-facing interface. Sirius achieves 8.3x and 7.4x better cost efficiency on TPC-H and ClickBench, respectively, when integrated with single-node DuckDB, and delivers up to 12.5x speedup when integrated with Apache Doris distributed engine.

cs.DB↗

arXiv2025

Seeing to Act, Prompting to Specify: A Bayesian Factorization of Vision Language Action Policy

The pursuit of out-of-distribution generalization in Vision-Language-Action (VLA) models is often hindered by catastrophic forgetting of the Vision-Language Model (VLM) backbone during fine-tuning. While co-training with external reasoning data helps, it requires experienced tuning and data-related overhead. Beyond such external dependencies, we identify an intrinsic cause within VLA datasets: modality imbalance, where language diversity is much lower than visual and action diversity. This imbalance biases the model toward visual shortcuts and language forgetting. To address this, we introduce BayesVLA, a Bayesian factorization that decomposes the policy into a visual-action prior, supporting seeing-to-act, and a language-conditioned likelihood, enabling prompt-to-specify. This inherently preserves generalization and promotes instruction following. We further incorporate pre- and post-contact phases to better leverage pre-trained foundation models. Information-theoretic analysis formally validates our effectiveness in mitigating shortcut learning. Extensive experiments show superior generalization to unseen instructions, objects, and environments compared to existing methods. Project page is available at: https://xukechun.github.io/papers/BayesVLA.

cs.RO↗

arXiv2025

Enhancing Channel Estimation for OTFS systems using Sparse Bayesian Learning with Adaptive Threshold

Orthogonal time frequency space (OTFS) modulation is a two-dimensional modulation scheme designed in the delay-Doppler (DD) domain, exhibiting superior performance over orthogonal frequency division multiplexing (OFDM) modulation in environments with high Doppler frequency shifts. We investigated the channel estimation in the DD domain of OTFS systems, modeling it as a sparse signal recovery problem. Subsequently, within the existing sparse Bayesian learning framework, we proposed an adaptive Bayesian threshold-based active denoising mechanism. Combined with inverse-free sparse Bayesian learning, this effectively addresses the pseudo-peak issue in low signal-to-noise ratio (SNR) scenarios while maintaining low complexity. The simulation results demonstrate that this algorithm outperforms existing channel estimation algorithms in terms of anti-noise performance and complexity.

cs.IT↗

arXiv2025

Predicate Transfer: Efficient Pre-Filtering on Multi-Join Queries

This paper presents predicate transfer, a novel method that optimizes join performance by pre-filtering tables to reduce the join input sizes. Predicate transfer generalizes Bloom join, which conducts pre-filtering within a single join operation, to multi-table joins such that the filtering benefits can be significantly increased. Predicate transfer is inspired by the seminal theoretical results by Yannakakis, which uses semi-joins to pre-filter acyclic queries. Predicate transfer generalizes the theoretical results to any join graphs and use Bloom filters to replace semi-joins leading to significant speedup. Evaluation shows predicate transfer can outperform Bloom join by 3.1x on average on TPC-H benchmark.

cs.DB↗

arXiv2025

AutoHall: Automated Factuality Hallucination Dataset Generation for Large Language Models

Large language models (LLMs) have gained broad applications across various domains but still struggle with hallucinations. Currently, hallucinations occur frequently in the generation of factual content and pose a great challenge to trustworthy LLMs. However, hallucination detection is hindered by the laborious and expensive manual annotation of hallucinatory content. Meanwhile, as different LLMs exhibit distinct types and rates of hallucination, the collection of hallucination datasets is inherently model-specific, which also increases the cost. To address this issue, this paper proposes a method called $\textbf{AutoHall}$ for $\underline{Auto}$matically constructing model-specific $\underline{Hall}$ucination datasets based on existing fact-checking datasets. The empirical results reveal variations in hallucination proportions and types among different models. Moreover, we introduce a zero-resource and black-box hallucination detection method based on self-contradiction to recognize the hallucination in our constructed dataset, achieving superior detection performance compared to baselines. Further analysis on our dataset provides insight into factors that may contribute to LLM hallucinations. Our codes and datasets are publicly available at https://github.com/zouyingcao/AutoHall.

cs.CL↗

arXiv2025

One Join Order Does Not Fit All: Reducing Intermediate Results with Per-Split Query Plans

Minimizing intermediate results is critical for efficient multi-join query processing. Although the seminal Yannakakis algorithm offers strong guarantees for acyclic queries, cyclic queries remain an open challenge. In this paper, we propose SplitJoin, a framework that introduces split as a first-class query operator. By partitioning input tables into heavy and light parts, SplitJoin allows different data partitions to use distinct query plans, with the goal of reducing intermediate sizes using existing binary join engines. We systematically explore the design space for split-based optimizations, including threshold selection, split strategies, and join ordering after splits. Implemented as a front-end to DuckDB and Umbra, SplitJoin achieves substantial improvements: on DuckDB, SplitJoin completes 43 social network queries (vs. 29 natively), achieving 2.1x faster runtime and 7.9x smaller intermediates on average (up to 13.6x and 74x, respectively); on Umbra, it completes 45 queries (vs. 35), achieving 1.3x speedups and 1.2x smaller intermediates on average (up to 6.1x and 2.1x, respectively).

cs.DB↗

arXiv2025

LDMD with Temporally Adaptive Segmentation

Dynamic mode decomposition (DMD) is a widely used data-driven algorithm for predicting the future states of dynamical systems. However, its standard formulation often struggles with poor long-term predictive accuracy. To address this limitation, we propose a localized DMD (LDMD) framework that improves prediction performance by integrating DMD's strong linear forecasting capabilities with time-domain segmentation techniques. In this framework, the temporal domain is segmented into multiple subintervals, within which snapshot matrices are constructed and localized predictions are performed. We first present the localized DMD method with predefined segmentation, and then explore an adaptive segmentation strategy to further enhance computational efficiency and prediction robustness. Furthermore, we conduct an error analysis that provides the upper bound of the local and global truncation error for the proposed framework. The effectiveness of LDMD is demonstrated on four benchmark problems-Burgers', Allen-Cahn, nonlinear Schrodinger, and Maxwell's equations. Numerical results show that LDMD significantly enhances long-term predictive accuracy while preserving high computational efficiency.

math.NA↗

arXiv2025

Localized Dynamic Mode Decomposition with Temporally Adaptive Segmentation

math.NA↗

arXiv2025

Lagrangian Motion Fields for Long-term Motion Generation

Long-term motion generation is a challenging task that requires producing coherent and realistic sequences over extended durations. Current methods primarily rely on framewise motion representations, which capture only static spatial details and overlook temporal dynamics. This approach leads to significant redundancy across the temporal dimension, complicating the generation of effective long-term motion. To overcome these limitations, we introduce the novel concept of Lagrangian Motion Fields, specifically designed for long-term motion generation. By treating each joint as a Lagrangian particle with uniform velocity over short intervals, our approach condenses motion representations into a series of "supermotions" (analogous to superpixels). This method seamlessly integrates static spatial information with interpretable temporal dynamics, transcending the limitations of existing network architectures and motion sequence content types. Our solution is versatile and lightweight, eliminating the need for neural network preprocessing. Our approach excels in tasks such as long-term music-to-dance generation and text-to-motion generation, offering enhanced efficiency, superior generation quality, and greater diversity compared to existing methods. Additionally, the adaptability of Lagrangian Motion Fields extends to applications like infinite motion looping and fine-grained controlled motion generation, highlighting its broad utility. Video demonstrations are available at https://plyfager.github.io/LaMoG.

cs.CV↗

arXiv2025

Toward Embodiment Equivariant Vision-Language-Action Policy

Vision-language-action policies learn manipulation skills across tasks, environments and embodiments through large-scale pre-training. However, their ability to generalize to novel robot configurations remains limited. Most approaches emphasize model size, dataset scale and diversity while paying less attention to the design of action spaces. This leads to the configuration generalization problem, which requires costly adaptation. We address this challenge by formulating cross-embodiment pre-training as designing policies equivariant to embodiment configuration transformations. Building on this principle, we propose a framework that (i) establishes a embodiment equivariance theory for action space and policy design, (ii) introduces an action decoder that enforces configuration equivariance, and (iii) incorporates a geometry-aware network architecture to enhance embodiment-agnostic spatial reasoning. Extensive experiments in both simulation and real-world settings demonstrate that our approach improves pre-training effectiveness and enables efficient fine-tuning on novel robot embodiments. Our code is available at https://github.com/hhcaz/e2vla

cs.RO↗

arXiv2025

Efficient Alignment of Unconditioned Action Prior for Language-conditioned Pick and Place in Clutter

We study the task of language-conditioned pick and place in clutter, where a robot should grasp a target object in open clutter and move it to a specified place. Some approaches learn end-to-end policies with features from vision foundation models, requiring large datasets. Others combine foundation models in a zero-shot setting, suffering from cascading errors. In addition, they primarily leverage vision and language foundation models, focusing less on action priors. In this paper, we aim to develop an effective policy by integrating foundation priors from vision, language, and action. We propose A$^2$, an action prior alignment method that aligns unconditioned action priors with 3D vision-language priors by learning one attention layer. The alignment formulation enables our policy to train with less data and preserve zero-shot generalization capabilities. We show that a shared policy for both pick and place actions enhances the performance for each task, and introduce a policy adaptation scheme to accommodate the multi-modal nature of actions. Extensive experiments in simulation and the real-world show that our policy achieves higher task success rates with fewer steps for both pick and place tasks in clutter, effectively generalizing to unseen objects and language instructions. Videos and codes are available at https://xukechun.github.io/papers/A2.

cs.RO↗

arXiv2025

TOP: Time Optimization Policy for Stable and Accurate Standing Manipulation with Humanoid Robots

Humanoid robots have the potential capability to perform a diverse range of manipulation tasks, but this is based on a robust and precise standing controller. Existing methods are either ill-suited to precisely control high-dimensional upper-body joints, or difficult to ensure both robustness and accuracy, especially when upper-body motions are fast. This paper proposes a novel time optimization policy (TOP), to train a standing manipulation control model that ensures balance, precision, and time efficiency simultaneously, with the idea of adjusting the time trajectory of upper-body motions but not only strengthening the disturbance resistance of the lower-body. Our approach consists of three parts. Firstly, we utilize motion prior to represent upper-body motions to enhance the coordination ability between the upper and lower-body by training a variational autoencoder (VAE). Then we decouple the whole-body control into an upper-body PD controller for precision and a lower-body RL controller to enhance robust stability. Finally, we train TOP method in conjunction with the decoupled controller and VAE to reduce the balance burden resulting from fast upper-body motions that would destabilize the robot and exceed the capabilities of the lower-body RL policy. The effectiveness of the proposed approach is evaluated via both simulation and real world experiments, which demonstrate the superiority on standing manipulation tasks stably and accurately. The project page can be found at https://anonymous.4open.science/w/top-258F/.

cs.RO↗

arXiv2025

Demonstration of Electron-Mediated Voltage-Controlled Exchange Coupling in Perpendicular Magnetic Tunnel Junctions

Electron-mediated voltage control of exchange coupling (EM-VCEC) has been proposed as a mechanism for magnetization switching via modulation of spin-dependent electron reflection. However, its experimental verification has been challenging due to the coexistence of slower, voltage-induced ionic effects. Here, we fabricate magnetic tunnel junction (MTJ) devices that enable nanosecond timescale voltage application. Our results reveal rapid exchange coupling modulation on the nanosecond timescale, consistent with an electronic origin. The observed enhancement and saturation at low temperatures further rule out ionic migration, conclusively confirming the electronic nature of the mechanism. These results establish EM-VCEC as a viable mechanism for fast and energy-efficient voltage-driven magnetic switching.

physics.app-ph↗

arXiv2025

PGPO: Enhancing Agent Reasoning via Pseudocode-style Planning Guided Preference Optimization

Large Language Model (LLM) agents have demonstrated impressive capabilities in handling complex interactive problems. Existing LLM agents mainly generate natural language plans to guide reasoning, which is verbose and inefficient. NL plans are also tailored to specific tasks and restrict agents' ability to generalize across similar tasks. To this end, we explore pseudocode-style plans (P-code Plan) to capture the structural logic of reasoning. We find that P-code Plan empowers LLM agents with stronger generalization ability and more efficiency. Inspired by this finding, we propose a pseudocode-style Planning Guided Preference Optimization method called PGPO for effective agent learning. With two planning-oriented rewards, PGPO further enhances LLM agents' ability to generate high-quality P-code Plans and subsequent reasoning. Experiments show that PGPO achieves superior performance on representative agent benchmarks and outperforms the current leading baselines. Analyses reveal the advantage of PGPO in reducing action errors and omissions during reasoning.

cs.AI↗

arXiv2025

Coexistence of unconventional spin Hall effect and antisymmetric planar Hall effect in IrO$_2$

Crystal symmetry plays an important role in the Hall effects. Unconventional spin Hall effect (USHE), characterized by Dresselhaus and out-of-plane spins, has been observed in materials with low crystal symmetry. Recently, antisymmetric planar Hall effect (APHE) was discovered in rutile RuO2 and IrO2 (101) thin films, which also exhibit low crystal symmetry. In this study, we report the observation of both USHE and APHE in IrO2 (111) films, using spin-torque ferromagnetic resonance (ST-FMR) and harmonic Hall measurements, respectively. Notably, the unconventional spin torque efficiency from Dresselhaus spin was more than double that of a previous report. Additionally, the temperature dependence of APHE suggests that it arises from the Lorentz force, constrained by crystal symmetry. Symmetry analysis supports the coexistence of USHE and APHE and demonstrates that both originate from the crystal symmetry of IrO2 (111), paving the way for a deeper understanding of Hall effects and related physical phenomena.

cond-mat.mes-hall↗

arXiv2025

Debunking the Myth of Join Ordering: Toward Robust SQL Analytics

Join order optimization is critical in achieving good query performance. Despite decades of research and practice, modern query optimizers could still generate inferior join plans that are orders of magnitude slower than optimal. Existing research on robust query processing often lacks theoretical guarantees on join-order robustness while sacrificing query performance. In this paper, we rediscover the recent Predicate Transfer technique from a robustness point of view. We introduce two new algorithms, LargestRoot and SafeSubjoin, and then propose Robust Predicate Transfer (RPT) that is provably robust against arbitrary join orders of an acyclic query. We integrated Robust Predicate Transfer with DuckDB, a state-of-the-art analytical database, and evaluated against all the queries in TPC-H, JOB, and TPC-DS benchmarks. Our experimental results show that RPT improves join-order robustness by orders of magnitude compared to the baseline. With RPT, the largest ratio between the maximum and minimum execution time out of random join orders for a single acyclic query is only 1.6x (the ratio is close to 1 for most evaluated queries). Meanwhile, applying RPT also improves the end-to-end query performance by 1.5x (per-query geometric mean). We hope that this work sheds light on solving the practical join ordering problem.

cs.DB↗

arXiv2025

How Deep is Love in LLMs' Hearts? Exploring Semantic Size in Human-like Cognition

How human cognitive abilities are formed has long captivated researchers. However, a significant challenge lies in developing meaningful methods to measure these complex processes. With the advent of large language models (LLMs), which now rival human capabilities in various domains, we are presented with a unique testbed to investigate human cognition through a new lens. Among the many facets of cognition, one particularly crucial aspect is the concept of semantic size, the perceived magnitude of both abstract and concrete words or concepts. This study seeks to investigate whether LLMs exhibit similar tendencies in understanding semantic size, thereby providing insights into the underlying mechanisms of human cognition. We begin by exploring metaphorical reasoning, comparing how LLMs and humans associate abstract words with concrete objects of varying sizes. Next, we examine LLMs' internal representations to evaluate their alignment with human cognitive processes. Our findings reveal that multi-modal training is crucial for LLMs to achieve more human-like understanding, suggesting that real-world, multi-modal experiences are similarly vital for human cognitive development. Lastly, we examine whether LLMs are influenced by attention-grabbing headlines with larger semantic sizes in a real-world web shopping scenario. The results show that multi-modal LLMs are more emotionally engaged in decision-making, but this also introduces potential biases, such as the risk of manipulation through clickbait headlines. Ultimately, this study offers a novel perspective on how LLMs interpret and internalize language, from the smallest concrete objects to the most profound abstract concepts like love. The insights gained not only improve our understanding of LLMs but also provide new avenues for exploring the cognitive abilities that define human intelligence.

cs.CL↗