arXiv subjects

Tao Gu

Publications and source records attributed to Tao Gu.

At least 19 recordsLinked to original sources

arXiv2026

LIRA: Local Cross-Layer Information Routing for Vision-Language-Action Decoding

Vision-Language-Action (VLA) models transform representations from pretrained vision-language models (VLMs) into robot actions, yet the interface that routes intermediate VLM features into action decoders remains underexplored. Existing designs either expose only a narrow part of the representation hierarchy or rigidly match each decoder block to one VLM layer, restricting access to complementary task evidence across depths. We introduce LIRA, a local cross-layer action-conditioning mechanism that formulates VLM-to-action conditioning as depth-aware information routing. LIRA operates on task-token features and LIRA Query features derived from intermediate VLM states, then assigns each Parallel Fusion Block a depth-aligned local window centered on its corresponding VLM layer. Parallel Fusion Blocks aggregate neighboring LIRA Query features and integrate them with task-token features and proprioceptive inputs before action prediction. This routing interface leaves the backbone architecture, action decoder, and supervised training recipe unchanged. Across LIBERO, LIBERO-Plus, CALVIN ABC$\rightarrow$D, and real-world manipulation, LIRA improves the principal aggregate metrics over the VLA-Adapter baseline under the same 0.5B-parameter configuration. In zero-shot transfer to LIBERO-Plus, LIRA increases average success from 59.1% to 78.0%, an 18.9-point gain indicating improved robustness under controlled distribution shifts.

cs.RO↗

arXiv2026

Nanbeige4.2-3B: Unlocking Agentic Capabilities in a Compact Model

We present Nanbeige4.2-3B, a compact general agentic model with 3B non-embedding parameters. It delivers strong performance across code-agent, office-agent, and complex tool-use tasks while maintaining highly competitive reasoning capabilities in mathematics, coding, and science. Nanbeige4.2-3B is pretrained from scratch on 28T tokens with a Looped Transformer that reuses the layer stack to increase capacity without adding parameters. For SFT data and trajectory construction, we expand the diversity of executable environments, task assets, and agentic scaffolds through real-world deployment and large-scale synthesis. Our RL pipeline applies mixed-mode RLHF over Think and Non-Think responses to improve overall model quality and reduce failure cases, length-controlled reasoning RL to balance accuracy and reasoning efficiency, and agentic RL with outcome and process rewards to stabilize long-horizon training. Extensive evaluations show that Nanbeige4.2-3B outperforms larger models, including Qwen3.5-9B and Gemma4-12B, across diverse agentic benchmarks while remaining competitive on reasoning and alignment tasks. Performance with OpenClaw further supports its use as a compact local personal assistant.

cs.AI↗

arXiv2026

ST-Gen4D: Embedding 4D Spatiotemporal Cognition into World Model for 4D Generation

Generative models have achieved success in producing apparently coherent 2D videos, but remain challenging in the physical world due to lack of 4D spatiotemporal scale. Typically, existing 4D generative models directly embed macro scale constraints to enhance overall spatiotemporal consistency. However, these methods only ensure global appearance coherence and fail to reveal the local dynamics of the physical world. Our insight is that global appearance structure and local dynamic topology empower 4D spatiotemporal cognition, thereby enabling 4D generation with spatiotemporal regularities. In this work, we propose ST-Gen4D, a 4D generation framework with 4D spatiotemporal cognition-based world model. Our model is guided by four key designs: 1) Spatiotemporal representation. We encode various modalities into multiple representations as a feature basis. 2) Spatiotemporal cognition. We sculpture these representations into global appearance graph and local dynamic graph, and fuse them via semantic-bridged spatiotemporal fusion to obtain a 4D cognition graph. 3) Spatiotemporal reasoning. We utilize a world model to derive future state based on the 4D cognition. 4) Spatiotemporal generation. We leverage the derived cognition as condition to guide latent diffusion for 4D Gaussian generation. By deeply integrating 4D intrinsic cognition with generative priors, our model guarantees the structural rationality and topological consistency of 4D generation. Moreover, we propose ST-4D datasets by aggregating public 4D datasets and self-built subset. Extensive experiments demonstrate the superiority of our ST-Gen4D across 3D and 4D generation tasks.

cs.CV↗

arXiv2026

Continuations and Completeness in Proof-theoretic Semantics

This is a short paper about the relationship between logic and computation. More specifically, it is about a relationship between the completeness proof for intuitionistic propositional logic within the form of proof-theoretic semantics that is known as base-extension semantics and a fundamental idea from the theory of computation called continuation-passing semantics. The latter is explained herein both in terms of reduction in natural deduction and the lambda calculus and in terms of proof-search. The relationship between completeness and continuations is explored through an analysis of Sandqvist's proof of the completeness theorem as seen from the mathematical perspective of Kripke's and Heyting's semantics. Our analysis can be seen to reveal how syntactic representations of continuations embody intensional semantical intuitions about the relationship between their meaning and use. These intuitions are made precise using the tools of proof-theoretic semantics.

math.LO↗

arXiv2026

ST-$\pi$: Structured SpatioTemporal VLA for Robotic Manipulation

Vision-language-action (VLA) models have achieved great success on general robotic tasks, but still face challenges in fine-grained spatiotemporal manipulation. Typically, existing methods mainly embed spatiotemporal knowledge into visual and action representations, and directly perform a cross-modal mapping for step-level action prediction. However, such spatiotemporal reasoning remains largely implicit, making it difficult to handle multiple sequential behaviors with explicit spatiotemporal boundaries. In this work, we propose ST-$\pi$, a structured spatiotemporal VLA model for robotic manipulation. Our model is guided by two key designs: 1) Spatiotemporal VLM. We encode 4D observations and task instructions into latent spaces, and feed them into the LLM to generate a sequence of causally ordered chunk-level action prompts consisting of sub-tasks, spatial grounding and temporal grounding. 2) Spatiotemporal action expert. Conditioned on chunk-level action prompts, we design a structured dual-generator guidance to jointly model spatial dependencies and temporal causality, thus predicting step-level action parameters. Within this structured framework, the VLM explicitly plans global spatiotemporal behavior, and the action expert further refines local spatiotemporal control. In addition, we propose a real-world robotic dataset with structured spatiotemporal annotations for fine-tuning. Extensive experiments have been conducted to demonstrate the effectiveness of our model. Our code link: https://github.com/chuanhaoma/ST-pi.

cs.RO↗

arXiv2026

Cog2Gen3D: Sculpturing 3D Semantic-Geometric Cognition for 3D Generation

Generative models have achieved success in producing semantically plausible 2D images, but it remains challenging in 3D generation due to the absence of spatial geometry constraints. Typically, existing methods utilize geometric features as conditions to enhance spatial awareness. However, these methods can only model relative relationships and are prone to scale inconsistency of absolute geometry. Thus, we argue that semantic information and absolute geometry empower 3D cognition, thereby enabling controllable 3D generation for the physical world. In this work, we propose Cog2Gen3D, a 3D cognition-guided diffusion framework for 3D generation. Our model is guided by three key designs: 1) Cognitive Feature Embeddings. We encode different modalities into semantic and geometric representations and further extract logical representations. 2) 3D Latent Cognition Graph. We structure different representations into dual-stream semantic-geometric graphs and fuse them via common-based cross-attention to obtain a 3D cognition graph. 3) Cognition-Guided Latent Diffusion. We leverage the fused 3D cognition graph as the condition to guide the latent diffusion process for 3D Gaussian generation. Under this unified framework, the 3D cognition graph ensures the physical plausibility and structural rationality of 3D generation. Moreover, we construct a validation subset based on the Marble World Labs. Extensive experiments demonstrate that our Cog2Gen3D significantly outperforms existing methods in both semantic fidelity and geometric plausibility.

cs.CV↗

arXiv2026

Traceable Latent Variable Discovery Based on Multi-Agent Collaboration

Revealing the underlying causal mechanisms in the real world is crucial for scientific and technological progress. Despite notable advances in recent decades, the lack of high-quality data and the reliance of traditional causal discovery algorithms (TCDA) on the assumption of no latent confounders, as well as their tendency to overlook the precise semantics of latent variables, have long been major obstacles to the broader application of causal discovery. To address this issue, we propose a novel causal modeling framework, TLVD, which integrates the metadata-based reasoning capabilities of large language models (LLMs) with the data-driven modeling capabilities of TCDA for inferring latent variables and their semantics. Specifically, we first employ a data-driven approach to construct a causal graph that incorporates latent variables. Then, we employ multi-LLM collaboration for latent variable inference, modeling this process as a game with incomplete information and seeking its Bayesian Nash Equilibrium (BNE) to infer the possible specific latent variables. Finally, to validate the inferred latent variables across multiple real-world web-based data sources, we leverage LLMs for evidence exploration to ensure traceability. We comprehensively evaluate TLVD on three de-identified real patient datasets provided by a hospital and two benchmark datasets. Extensive experimental results confirm the effectiveness and reliability of TLVD, with average improvements of 32.67% in Acc, 62.21% in CAcc, and 26.72% in ECit across the five datasets.

cs.LG↗

arXiv2026

LQA: A Lightweight Quantized-Adaptive Framework for Vision-Language Models on the Edge

Deploying Vision-Language Models (VLMs) on edge devices is challenged by resource constraints and performance degradation under distribution shifts. While test-time adaptation (TTA) can counteract such shifts, existing methods are too resource-intensive for on-device deployment. To address this challenge, we propose LQA, a lightweight, quantized-adaptive framework for VLMs that combines a modality-aware quantization strategy with gradient-free test-time adaptation. We introduce Selective Hybrid Quantization (SHQ) and a quantized, gradient-free adaptation mechanism to enable robust and efficient VLM deployment on resource-constrained hardware. Experiments across both synthetic and real-world distribution shifts show that LQA improves overall adaptation performance by 4.5\%, uses less memory than full-precision models, and significantly outperforms gradient-based TTA methods, achieving up to 19.9$\times$ lower memory usage across seven open-source datasets. These results demonstrate that LQA offers a practical pathway for robust, privacy-preserving, and efficient VLM deployment on edge devices.

cs.AI↗

arXiv2026

HoRD: Robust Humanoid Control via History-Conditioned Reinforcement Learning and Online Distillation

Humanoid robots can suffer significant performance drops under small changes in dynamics, task specifications, or environment setup. We propose HoRD, a two-stage learning framework for robust humanoid control under domain shift. First, we train a high-performance teacher policy via history-conditioned reinforcement learning, where the policy infers latent dynamics context from recent state--action trajectories to adapt online to diverse randomized dynamics. Second, we perform online distillation to transfer the teacher's robust control capabilities into a transformer-based student policy that operates on sparse root-relative 3D joint keypoint trajectories. By combining history-conditioned adaptation with online distillation, HoRD enables a single policy to adapt zero-shot to unseen domains without per-domain retraining. Extensive experiments show HoRD outperforms strong baselines in robustness and transfer, especially under unseen domains and external perturbations. Code and project page are available at https://tonywang-0517.github.io/hord/.

cs.RO↗

arXiv2025

Energy and Memory-Efficient Federated Learning With Ordered Layer Freezing

Federated Learning (FL) has emerged as a privacy-preserving paradigm for training machine learning models across distributed edge devices in the Internet of Things (IoT). By keeping data local and coordinating model training through a central server, FL effectively addresses privacy concerns and reduces communication overhead. However, the limited computational power, memory, and bandwidth of IoT edge devices pose significant challenges to the efficiency and scalability of FL, especially when training deep neural networks. Various FL frameworks have been proposed to reduce computation and communication overheads through dropout or layer freezing. However, these approaches often sacrifice accuracy or neglect memory constraints. To this end, in this work, we introduce Federated Learning with Ordered Layer Freezing (FedOLF). FedOLF consistently freezes layers in a predefined order before training, significantly mitigating computation and memory requirements. To further reduce communication and energy costs, we incorporate Tensor Operation Approximation (TOA), a lightweight alternative to conventional quantization that better preserves model accuracy. Experimental results demonstrate that over non-iid data, FedOLF achieves at least 0.3%, 6.4%, 5.81%, 4.4%, 6.27% and 1.29% higher accuracy than existing works respectively on EMNIST (with CNN), CIFAR-10 (with AlexNet), CIFAR-100 (with ResNet20 and ResNet44), and CINIC-10 (with ResNet20 and ResNet44), along with higher energy efficiency and lower memory footprint.

cs.LG↗

arXiv2025

InstructDubber: Instruction-based Alignment for Zero-shot Movie Dubbing

Movie dubbing seeks to synthesize speech from a given script using a specific voice, while ensuring accurate lip synchronization and emotion-prosody alignment with the character's visual performance. However, existing alignment approaches based on visual features face two key limitations: (1)they rely on complex, handcrafted visual preprocessing pipelines, including facial landmark detection and feature extraction; and (2) they generalize poorly to unseen visual domains, often resulting in degraded alignment and dubbing quality. To address these issues, we propose InstructDubber, a novel instruction-based alignment dubbing method for both robust in-domain and zero-shot movie dubbing. Specifically, we first feed the video, script, and corresponding prompts into a multimodal large language model to generate natural language dubbing instructions regarding the speaking rate and emotion state depicted in the video, which is robust to visual domain variations. Second, we design an instructed duration distilling module to mine discriminative duration cues from speaking rate instructions to predict lip-aligned phoneme-level pronunciation duration. Third, for emotion-prosody alignment, we devise an instructed emotion calibrating module, which finetunes an LLM-based instruction analyzer using ground truth dubbing emotion as supervision and predicts prosody based on the calibrated emotion analysis. Finally, the predicted duration and prosody, together with the script, are fed into the audio decoder to generate video-aligned dubbing. Extensive experiments on three major benchmarks demonstrate that InstructDubber outperforms state-of-the-art approaches across both in-domain and zero-shot scenarios.

cs.SD↗

arXiv2025

Fine-Grained Privacy Extraction from Retrieval-Augmented Generation Systems via Knowledge Asymmetry Exploitation

Retrieval-augmented generation (RAG) systems enhance large language models (LLMs) by integrating external knowledge bases, but this advancement introduces significant privacy risks. Existing privacy attacks on RAG systems can trigger data leakage but often fail to accurately isolate knowledge-base-derived sentences within mixed responses. They also lack robustness when applied across multiple domains. This paper addresses these challenges by presenting a novel black-box attack framework that exploits knowledge asymmetry between RAG and standard LLMs to achieve fine-grained privacy extraction across heterogeneous knowledge landscapes. We propose a chain-of-thought reasoning strategy that creates adaptive prompts to steer RAG systems away from sensitive content. Specifically, we first decompose adversarial queries to maximize information disparity and then apply a semantic relationship scoring to resolve lexical and syntactic ambiguities. We finally train a neural network on these feature scores to precisely identify sentences containing private information. Unlike prior work, our framework generalizes to unseen domains through iterative refinement without pre-defined knowledge. Experimental results show that we achieve over 91% privacy extraction rate in single-domain and 83% in multi-domain scenarios, reducing sensitive sentence exposure by over 65% in case studies. This work bridges the gap between attack and defense in RAG systems, enabling precise extraction of private information while providing a foundation for adaptive mitigation.

cs.CR↗

arXiv2025

XTransfer: Modality-Agnostic Few-Shot Model Transfer for Human Sensing at the Edge

Deep learning for human sensing on edge systems presents significant potential for smart applications. However, its training and development are hindered by the limited availability of sensor data and resource constraints of edge systems. While transferring pre-trained models to different sensing applications is promising, existing methods often require extensive sensor data and computational resources, resulting in high costs and limited transferability. In this paper, we propose XTransfer, a first-of-its-kind method enabling modality-agnostic, few-shot model transfer with resource-efficient design. XTransfer flexibly uses pre-trained models and transfers knowledge across different modalities by (i) model repairing that safely mitigates modality shift by adapting pre-trained layers with only few sensor data, and (ii) layer recombining that efficiently searches and recombines layers of interest from source models in a layer-wise manner to restructure models. We benchmark various baselines across diverse human sensing datasets spanning different modalities. The results show that XTransfer achieves state-of-the-art performance while significantly reducing the costs of sensor data collection, model training, and edge deployment.

cs.CV↗

arXiv2025

Subspace-Based Super-Resolution Sensing for Bi-Static ISAC with Clock Asynchronism

Bi-static sensing is an attractive configuration for integrated sensing and communications (ISAC) systems; however, clock asynchronism between widely separated transmitters and receivers introduces time-varying time offsets (TO) and phase offsets (PO), posing significant challenges. This paper introduces a signal-subspace-based framework that estimates decoupled angles, delays, and complex gain sequences (CGS)-- the target-reflected signals -- for multiple dynamic target paths. The proposed framework begins with a novel TO alignment algorithm, leveraging signal subspace or covariance, to mitigate TO variations across temporal snapshots, enabling coherent delay-domain analysis. Subsequently, subspace-based methods are developed to compensate for TO residuals and to perform joint angle-delay estimation. Finally, leveraging the high resolution in the joint angle-delay domain, the framework compensates for the PO and estimates the CGS for each target. The framework can be applied to both single-antenna and multi-antenna systems. Extensive simulations and experiments using commercial Wi-Fi devices demonstrate that the proposed framework significantly surpasses existing solutions in parameter estimation accuracy and delay resolution. Notably, it uniquely achieves a super-resolution in the delay domain, with a probability-of-resolution curve tightly approaching that in synchronized systems.

eess.SP↗

arXiv2025

On the Logical Content of Knowledge Bases

Standard epistemic logics introduce a modal operator K to represent knowledge, but in doing so they presuppose the logical apparatus they aim to explain. By contrast, this paper explores how logic may be derived from the structure of knowledge itself. We begin from a pre-logical notion of a knowledge base understood as a network of inferential connections between atomic propositions. Logical constants are then defined in terms of what is supported by such a base: intrinsically, by relations that hold within it, and extrinsically, by the behaviour of those relations under extension. This yields a general semantic framework in which familiar systems (classical, intuitionistic, and various intermediate logics) arise naturally from different assumptions about the form of knowledge. This offers a reversal of the traditional explanatory order: rather than treating logic as a precondition for the articulation of knowledge, it shows how logical structure can emerge from epistemic organisation.

cs.LO↗

arXiv2024

Towards Battery-Free Wireless Sensing via Radio-Frequency Energy Harvesting

Diverse Wi-Fi-based wireless applications have been proposed, ranging from daily activity recognition to vital sign monitoring. Despite their remarkable sensing accuracy, the high energy consumption and the requirement for customized hardware modification hinder the wide deployment of the existing sensing solutions. In this paper, we propose REHSense, an energy-efficient wireless sensing solution based on Radio-Frequency (RF) energy harvesting. Instead of relying on a power-hungry Wi-Fi receiver, REHSense leverages an RF energy harvester as the sensor and utilizes the voltage signals harvested from the ambient Wi-Fi signals to enable simultaneous context sensing and energy harvesting. We design and implement REHSense using a commercial-off-the-shelf (COTS) RF energy harvester. Extensive evaluation of three fine-grained wireless sensing tasks (i.e., respiration monitoring, human activity, and hand gesture recognition) shows that REHSense can achieve comparable sensing accuracy with conventional Wi-Fi-based solutions while adapting to different sensing environments, reducing the power consumption by 98.7% and harvesting up to 4.5mW of power from RF energy.

cs.NI↗

arXiv2024

A Note on an Inferentialist Approach to Resource Semantics

A central concept within informatics is in modelling such systems for the purpose of reasoning (perhaps automated) about their behaviour and properties. To this end, one requires an interpretation of logical formulae in terms of the resources and states of the system; such an interpretation is called a 'resource semantics' of the logic. This paper shows how 'inferentialism' -- the view that meaning is given in terms of inferential behaviour -- enables a versatile and expressive framework for resource semantics. Specifically, how inferentialism seamlessly incorporates the assertion-based approach of the logic of Bunched Implications, foundational in program verification (e.g., as the basis of Separation Logic), and the renowned number-of-uses reading of Linear Logic. This integration enables reasoning about shared and separated resources in intuitive and familiar ways, as well as about the composition and interfacing of system components.

cs.LO↗

arXiv2024

Magic Clothing: Controllable Garment-Driven Image Synthesis

We propose Magic Clothing, a latent diffusion model (LDM)-based network architecture for an unexplored garment-driven image synthesis task. Aiming at generating customized characters wearing the target garments with diverse text prompts, the image controllability is the most critical issue, i.e., to preserve the garment details and maintain faithfulness to the text prompts. To this end, we introduce a garment extractor to capture the detailed garment features, and employ self-attention fusion to incorporate them into the pretrained LDMs, ensuring that the garment details remain unchanged on the target character. Then, we leverage the joint classifier-free guidance to balance the control of garment features and text prompts over the generated results. Meanwhile, the proposed garment extractor is a plug-in module applicable to various finetuned LDMs, and it can be combined with other extensions like ControlNet and IP-Adapter to enhance the diversity and controllability of the generated characters. Furthermore, we design Matched-Points-LPIPS (MP-LPIPS), a robust metric for evaluating the consistency of the target image to the source garment. Extensive experiments demonstrate that our Magic Clothing achieves state-of-the-art results under various conditional controls for garment-driven image synthesis. Our source code is available at https://github.com/ShineChen1024/MagicClothing.

cs.CV↗