arXiv subjects

Yang Yu

Publications and source records attributed to Yang Yu.

At least 19 recordsLinked to original sources

arXiv2026

Gauge-invariant thermodynamic description of rotating black strings in 5D minimal supergravity via Kaluza-Klein reduction

We develop a gauge-invariant formulation of the thermodynamics of rotating black strings in five-dimensional minimal supergravity through Kaluza-Klein reduction. At the two-derivative level, we consistently reduce the bulk action and the Gibbons-Hawking-York boundary term, carefully accounting for the surface contributions induced by the Chern-Simons coupling. The resulting four-dimensional theory is $\mathcal{N}=2$ supergravity coupled to a single vector multiplet, with prepotential $P=\frac{1}{3\sqrt{3}}\frac{X_1^3}{X_0}$. Employing the Hamiltonian formalism, we establish the quantum statistical relation between the Euclidean on-shell action and the Gibbs free energy. We further compute the leading four-derivative corrections to the thermodynamics of a rotating magnetic black string using the Reall-Santos method. The resulting corrections are in precise agreement with previous results obtained directly for the Kerr-Newman black hole in four-dimensional $\mathcal{N}=2$ supergravity.

hep-th↗

arXiv2026

From Association to Causation: Improving Retrieval Precision of Retrieval-Augmented Generation via Causal Relations and an Attention Mechanism

Retrieval-Augmented Generation (RAG) grounds LLM generation on retrieved documents, but the standard terminal retrieval stage--dense-vector similarity, optionally followed by reranking--often returns documents that share keywords with the query without containing the needed information, a failure mode that grows with the knowledge base. We trace it to a conceptual gap: similarity captures only associational relations, whereas the documents that matter are linked to the query causally. We model the terminal retrieval stage with a causal graph grounded in Reichenbach's common cause principle: the keywords shared by the query and a retrieved document form a latent common cause A, and the document's residual keywords form a latent set B linking the document to the ideal output. Since a retrieved document is a collider (A -> d <- B), retrieval itself opens an associational path between the query and B, which licenses a training-free, attention-style re-scoring rule: the cosine similarity between the query embedding and the weighted centroid embedding of B. Unlike causality-enhanced RAG variants that model causal relations inside the knowledge content, our graph models the causal structure of the retrieval process itself. On a real 471-document enterprise knowledge base, the method promotes a relevant guideline from rank 6 to the top 3; on a controlled diagnostic corpus reproducing the keyword-stuffing regime, it improves the mean target rank from 2.88 to 1.25, while a trained cross-encoder reranker barely helps (2.63). Conversely, on three BEIR benchmarks the score underperforms the similarity baseline, delineating the applicability boundary: the method guards the keyword-stuffing regime of growing proprietary knowledge bases and complements neural rerankers; a corpus-level calibration gate selects the correct regime with >= 95% reliability. A fully local testbed demonstrates deployability.

cs.AI↗

arXiv2026

Force Sensing in a Recoverable Bosonic Quantum Memory

In continuous-variable quantum information, sensing and quantum error correction in the same mode face an intrinsic trade-off: sensors must accumulate weak external displacements, whereas bosonic codes typically suppress such shifts as errors. Here, we establish compatibility by constructing finite-energy bosonic codes that enable a single oscillator to sense an external force while actively protecting an unknown logical qubit. By encoding logical amplitudes across discrete wave-packet sites and capturing the force signal in a shared internal coordinate, we achieve an endpoint probe readout that is independent of the logical state under continuous photon loss $γ$. Crucially, binomial site recovery cancels packet-dependent probe distortions alongside logical errors: for a $(2r+2)$-packet code, memory infidelity is suppressed to $O(γ^{r+1})$ while residual memory-probe correlations scale as $O(γ^{r+2})$. Finally, superconducting quantum circuit benchmarks under competing photon-number dephasing and Kerr nonlinearity identify preferred encoding regimes for hardware implementation.

quant-ph↗

arXiv2026

Scalar-induced gravitational waves in spatially covariant gravity with two tensorial degrees of freedom

We study scalar-induced gravitational waves (SIGWs) in a spatially covariant gravity model with two tensorial degrees of freedom, minimally coupled to a $k$-essence scalar field. For the parameter choice considered during radiation domination, the quadratic tensor action coincides with that of general relativity (GR). We derive the scalar transfer functions at finite $σ\equiv C_2\bar N$ and in the large-$|σ|$ limit. We then construct the source function of SIGWs. We further examine a cubic extrinsic-curvature interaction that modifies the source function of SIGWs without affecting the background or quadratic actions. We calculate the fractional energy density of SIGWs for monochromatic and log-normal primordial curvature power spectra. We find that, for the positive values of $σ$ studied, finite-$σ$ effects suppress the spectrum over a broad range of scales relative to the result in the $σ\rightarrow \infty$ limit. In this limit, the cubic interaction suppresses the spectrum at small wavenumbers and enhances it at larger wavenumbers. These results show that modified scalar dynamics and non-linear interactions can alter the fractional energy density of SIGWs even when the quadratic tensor action is identical to that of GR.

gr-qc↗

arXiv2026

VisGeo-PipeTrack: A Visual Geometric Servo Control Framework with MPC-RCAPID Control for UAV-Based Dynamic Pipes Tracking

While unmanned aerial vehicles (UAVs) are increasingly deployed for pipeline inspection, prevailing research predominantly assumes static pipes. In real-world industrial environments, pipes often exhibit dynamic movements, such as vibrations. This imposes more stringent requirements on UAVs' adaptive control and computational efficiency than in static scenarios. Therefore, this paper proposes a servo control framework named VisGeo-PipeTrack, which utilizes geometric constraints to enable UAV tracking of dynamic pipes. To achieve efficient UAV positioning, YOLOv5 is employed to detect labels on the pipe. The positions are derived from visual geometric constraints based on these detected labels to enable precise localization. To address tracking control challenges, a model predictive control based real-time corrected adaptive PID (MPC-RCAPID) controller is proposed. This controller employs the Recursive Least Squares method for real-time model estimation and integrates rolling horizon optimization to dynamically adjust PID parameters. Experiments conducted on both the AirSim simulator and the Tello UAV platform validate the effectiveness of the proposed framework. Simulation comparisons with standard PID, fractional-order PID, adaptive PID, and fuzzy PID demonstrate that MPC-RCAPID achieves the shortest settling time (3.26 s) and the lowest average tracking error (3.52%). In physical experiments on the Tello platform, compared with standard PID control, MPC-RCAPID reduces the average tracking error by 55.52%, 26.72%, and 26.52% across the three test dimensions. These results demonstrate that VisGeo-PipeTrack facilitates efficient autonomous tracking in dynamic pipe environments.

cs.RO↗

arXiv2026

Discrete Approximation to Time-changed Brownian Motions

We develop a general discrete approximation scheme for time-changed Brownian motions on $\mathbb{R}^d$. Our approximation scheme works for any smooth measure with full quasi-support on $\mathbb{R}^d$ with suitable initial distributions. Under some mild conditions on the smooth measure, the discrete approximation scheme works for every starting point. Our results in particular give a discrete approximation scheme for Liouville Brownian motions.

math.PR↗

arXiv2026

Adaptive Aerial Loco-Manipulation Policy Learning via LLM-Assisted Proactive Task Generation

Loco-manipulation extends the workspace of conventional manipulation by combining mobility with physical interaction, while aerial loco-manipulation further expands this capability into three-dimensional space, enabling robots to reach environments inaccessible to ground-based systems. Moving to an aerial platform, however, gives rise to a demanding control problem, since the vehicle must track trajectories accurately in three dimensions while the motion of the arm and its interaction with the environment continuously shift the center of mass, alter the inertia, and impose force and moment disturbances on the body. Conventional controllers address these effects through explicit models and estimation, whereas learning-based controllers infer dynamics from experience, yet control rewards provide no direct target for the response information retained from history, and fixed task sampling can leave persistent tracking difficulties insufficiently explored. We present ALMA, an aerial loco-manipulation framework for adaptive flight control under external arm commands. ALMA employs a contextual flight policy that adapts from recent measurements and actions, using action-conditioned prediction of future measurement representations to train its context to retain response information relevant to flight control. To expand the experience available for this adaptation, an evolutionary process uses a large language model (LLM) to analyze observed control difficulties and propose new structures of arm motion for subsequent training. In simulation, ALMA reduces position root-mean-square error (RMSE) by up to 24.7% relative to the strongest baseline, while the same policy transfers zero-shot to the aerial manipulator and reduces hover position RMSE by 79.6% during periodic arm motion relative to the real-world baseline.

cs.RO↗

arXiv2026

Toward Collective Model Intelligence: Characterizing and Coordinating Models

Independently developed AI models accumulate diverse and complementary capabilities, but a growing model collection does not automatically realize Collective AI. Addressing this gap requires understanding model capabilities with limited evaluation, organizing complementary models to collaborate on diverse tasks, and executing model calls efficiently under resource constraints. We present a systematic approach to pursuing Collective Model Intelligence through three connected components. First, Model Eval constructs evidence-based capability profiles by combining query-level performance estimation with cross-domain prediction, reducing the need for exhaustive evaluation. Second, Mio (Models in One) organizes complementary model candidates and supports task-conditioned, cost-aware selection and collaboration as task requirements evolve. Third, Mos (Model Serving OS) separates model selection from execution and provides endpoint scheduling and shared model deployment and distribution. Experiments show lower capability-estimation errors under limited evaluation. Across nine public benchmarks, Mio 1.0 Lite achieves the highest average score and lowest average cost among compared methods, with per-request costs 73.7-97.5% lower than those of the best-scoring alternatives. A prewarmed shared cache accelerates model-file retrieval by up to 23.3 times. This work lays a foundation for Collective Model Intelligence by characterizing and coordinating the diverse capabilities of an expanding model universe.

cs.AI↗

arXiv2026

Decouple to Forget: Semantic Mediation for Sensitive Unlearning in Large Language Models

Machine unlearning for large language models seeks to remove specified knowledge while preserving general capability, yet current methods often treat an entire response as the unit of forgetting. This coarse granularity causes unnecessary utility loss when the forget target is only a small set of sensitive facts embedded within otherwise benign explanations. In this paper, we introduce Decouple to Forget (D2F), a targeted unlearning method inspired by semantic mediation that separates sensitive factual content from non-sensitive meaning and surface expression. For each example to be forgotten, D2F extracts a fact-level "mediator" representation from the answer, identifies which parts are sensitive, and removes only those parts while leaving the rest of the structure intact. We then generate counterfactual answers grounded in this mediator that preserve non-sensitive content but exclude sensitive facts, employing preference-based contrastive optimization to reduce the likelihood of information leakage while maintaining helpfulness and fluency. Experiments across various benchmark datasets show that D2F achieves a better balance between forgetting and retention than strong baseline methods. Our code is available at https://github.com/JingyueCong/D2F.

cs.CL↗

arXiv2026

TRIAGE: Direction-Aware Mismatch Stabilization of Native NVFP4 Reinforcement Learning

Low-precision execution can substantially accelerate reinforcement learning (RL) for large language models, but discrepancies between learner and sampler execution can destabilize policy optimization. In this paper, we characterize the interaction between mismatch and the policy-gradient direction, distinguishing locally amplifying from contracting update contributions that mismatch magnitude alone cannot identify. In native NVFP4 runs, we observe an early imbalance between the two amplifying regions, favoring negative-advantage, negative-gap updates. Their tail tokens become concentrated in a small fraction of response segments before mismatch spreads globally. Motivated by these findings, we introduce TRIAGE, a direction-aware stabilization method that uses segment-level diagnosis to selectively rebalance policy-gradient updates and applies bounded repair to residual severe mismatch. TRIAGE modifies the optimization objective while retaining native NVFP4 weight-and activation 4-bit (W4A4) forward execution on both the sampler and learner. Experiments on Qwen3-4B and Qwen3-30B-A3B show stable optimization throughout the evaluated training horizon and achieve full precision level performance across five mathematical reasoning benchmarks, while native NVFP4 with TRIAGE provides up to 2.3x higher rollout throughput than BF16.

cs.LG↗

arXiv2026

Quantum Simulation of a Hyperbolically Driven Quantum System with Superconducting Circuits

Hyperbolic geometry hosts a plethora of exotic quantum phenomena ranging from holographic duality to novel topological phases. However, investigating these effects experimentally remains challenging due to the Euclidean constraints of natural crystalline materials. While recent synthetic lattices have achieved static realizations, they often lack the dynamical tunability required to explore the full geometric landscape. Here, we overcome these limitations by implementing a quantum simulation of hyperbolic geometry within the parameter space of a superconducting quantum circuit. We construct a Hyperbolically Driven Quantum System (HDQS) by mapping the hyperbolic properties onto the dynamically controllable Hamiltonian parameters. By probing the systems' response under geometric quantum driving, we elucidate how the intrinsic geometry of the parameter space dictates observable physical quantities. Furthermore, we extend the theoretical framework to realize ensemble simulation. This generalization enables us to harness the fast coherent control of superconducting qubits to capture essential geometric characteristics, establishing a versatile platform for exploring quantum dynamics on hyperbolic manifolds.

quant-ph↗

arXiv2026

Beyond Captions: Context-Grounded Reconstruction for Biomedical Multimodal Continued Pretraining

Biomedical figures are explained not by captions alone but by body-text passages that discuss them. Yet current multimodal corpora typically reduce figures to isolated image-caption pairs, discarding this crucial context. Existing pipelines either omit this context or append it without enforcing the figure references that support each attachment, which can create unsupported image-text attachments and incoherent discourse. We introduce context-grounded reconstruction, a source-grounded framework that converts PubMed Central Open Access (PMC-OA) records into referentially coherent interleaved sequences. It recovers captions and source text, attaches context only through article-native figure references, repairs non-contiguous context, and prunes unsupported images. Starting from these reconstructed sequences, PMC-InterCPT first filters records for text quality and medical relevance, then applies evidence-aware allocation to form a 9.63B-token corpus for continued pretraining (CPT) of generative medical MLLMs. With fixed supervised fine-tuning (SFT), PMC-InterCPT improves Qwen3.5-4B-Base by 1.46 medical-average points and 3.11 general/scientific-average points over a token-matched raw source control, and surpasses a 42% larger raw-data run. Gains transfer to Qwen3.5-2B-Base and LLaVA-OneVision-1.5-4B-Base. Controlled ablations show that context-grounded reconstruction, rather than simply appending article context, is central to useful biomedical multimodal CPT.

cs.CL↗

arXiv2026

InfiMed2: A Generalist Medical Multimodal Foundation Model from Contextual Evidence and Stability-Aware Supervision

Recent medical multimodal models have benefited from larger corpora, broader modality coverage, and stronger reasoning-oriented training, yet effective data design across continued pretraining (CPT) and post-training remains challenging. Medical sources vary substantially in structure, granularity, and information density, and their utility shifts as training progresses from broad knowledge acquisition to late-stage consolidation. Meanwhile, post-training is often dominated by short-form visual question answering, providing limited supervision for informative and answer-consistent explanations. We introduce InfiMed2, a family of 4B and 27B generalist medical multimodal foundation models built around stage-aware data design. We curate a 55.68B-token corpus that combines broad clinical knowledge with context-rich biomedical visual evidence through source-specific processing. Our CPT pipeline first adapts the vision encoder, then builds broad medical knowledge, and finally transitions to an evidence-focused data mixture during learning-rate decay. For supervised fine-tuning (SFT), we regenerate visual question-answering responses using answer stability, answer-masked reconstruction, and correctness-constrained selection to produce more informative and answer-consistent supervision. The 4B model is further optimized with reinforcement learning with verifiable rewards (RLVR). Across five medical multimodal benchmarks, InfiMed2-4B achieves 66.73% mean accuracy after RLVR, surpassing the larger Qwen3.5-9B, while InfiMed2-27B reaches 73.72%, the highest among the evaluated open-weight models.

cs.CL↗

arXiv2026

miniTE: Diagnosing and Bridging the Execution Gap in Native NVFP4 Reinforcement Learning

Native NVFP4 promises to accelerate reinforcement learning (RL) for large language models (LLMs), but achieving stable learning across separate training and sampling engines remains challenging. We show that matching model weights and numerical formats is insufficient to ensure consistent policy evaluation, highlighting the importance of cross-engine numerical execution. We introduce miniTE, a lightweight framework that combines execution diagnostics, cross-engine numerical alignment, and efficient native W4A4 (4-bit weights and activations) GEMM computation for LLM RL. It improves numerical execution consistency without modifying the RL objective, while reducing implementation overhead to realize the benefits of native low-bit computation. Long-horizon RL experiments on Qwen3-30B-A3B demonstrate stable training and benchmark performance close to BF16, outperforming naive Transformer Engine (TE) NVFP4 in evaluation quality. On single NVIDIA B300 GPU node, the optimized learner reaches 33,540 tokens/s, yielding 1.39x and 1.47x the throughput of BF16 and TE NVFP4, respectively, for complete learner updates. We released the framework, kernels, diagnostics, and training configurations to make native low-bit RL easier.

cs.LG↗

arXiv2026

MedHorizon: Towards Long-context Medical Video Understanding in the Wild

Medical multimodal large language models (MLLMs) have advanced image understanding and short-video analysis, but real clinical review often requires full-procedure video understanding. Unlike general long videos, medical procedures contain highly redundant anatomical views, while decisive evidence is temporally sparse, spatially subtle, and context dependent. Existing benchmarks often assume this evidence has already been localized through images, short clips, or pre-segmented videos, leaving the retrieval-before-reasoning problem under-tested. We introduce MedHorizon, an in-the-wild benchmark for long-context medical video understanding. MedHorizon preserves 759 hours of full-length clinical procedures and provides 1,253 evidence-grounded multiple-choice questionsthat jointly evaluate sparse evidence understanding and multi-hop clinical reasoning. Its evidence is extremely sparse, with only 0.166% evidence frames on average, requiring models to search noisy procedural streams before interpreting and aggregating findings. We evaluate representative general-domain, medical-domain, and long-video MLLMs. The best model reaches only 41.1% accuracy, showing that current systems remain far from robust full-procedure understanding. Further analysis yields four key findings: performance does not scale reliably with more frames, evidence retrieval and clinical interpretation remain primary bottlenecks; these bottlenecks are rooted in weak procedural reasoning and attention drift under redundancy, and generic sampling methods only partially balances local detail with global coverage. MedHorizon provides a rigorous testbed for MLLMs that retrieve sparse evidence and reason over complete clinical workflows.

cs.CV↗

arXiv2026

Development of a continuous-wave photocathode very-high-frequency electron gun for $\rm S^3FEL$

A very-high-frequency (VHF) electron gun operating at a resonant frequency of 216.667 MHz has been developed at Tsinghua University for the Shenzhen Superconducting Soft X-ray Free-Electron Laser ($\rm S^3FEL$) facility. Building upon the SHINE gun design, the cavity profile was optimized to achieve a higher cathode electric-field gradient and a higher accelerating voltage at comparable input power. The revised cavity geometry also yields improved multipacting performance. In addition, thermal analysis was carried out to guide the water-cooling design, and enhanced cooling was implemented in the vicinity of the cathode. During high-power conditioning, 82 kW of continuous-wave radio-frequency power was successfully coupled into the gun, corresponding to a cathode gradient of 29.9 MV/m and an accelerating voltage of 856 kV. The gun voltage surpasses the previous world record for room-temperature VHF guns. The maximum dark current measured by the Faraday cup at the gun exit was only 6.8 nA.

physics.acc-ph↗

arXiv2026

Constraining Particle Stiffness in Asteroid Regolith from Early-Time High-Speed Penetrator Dynamics under Local Granular Variability

Mechanical properties of rubble-pile asteroid regolith remain poorly constrained because contact responses depend on both material properties and local particle configuration. This study examines whether early-time high-speed penetrator dynamics can constrain particle stiffness, represented by the particle Young's modulus E, within a controlled discrete-element model. A bidirectionally coupled EDEM-Adams discrete-element-multibody model is used to screen six parameters: Young's modulus, Poisson's ratio, coefficient of restitution, static friction coefficient, rolling friction coefficient, and adhesion strength. Five Young's-modulus levels are then tested at the same 17 spatial locations in one settled granular bed, giving 85 simulations with location-matched comparisons across modulus levels. Within the 75-90 m s^-1 speed range, varying Young's modulus yields the clearest and most systematic differences in probe deceleration, whereas the other parameters have weaker effects over the tested ranges. Validation by withholding all modulus cases at one location in turn indicates that the early response is more informative for distinguishing broad modulus ranges than adjacent modulus levels. Mesoscale analysis links the response differences to concentrated interface load sharing and spatially extended three-dimensional high-capacity contact paths as Young's modulus increases. The results provide a numerical basis for narrowing the particle Young's-modulus range from high-speed penetration responses despite local granular variability. Quantitative application to real asteroid regolith requires further experimental validation.

astro-ph.EP↗

arXiv2026

FusionEval: Benchmarking Model Fusion for Large Language Models

Model fusion aims to integrate the capabilities of multi-expert models into a single model. Existing evaluations offer limited coverage of reasoning with long chains of thought and tool use, while strong average performance can mask substantial losses in individual expert capabilities. We introduce FusionEval, a benchmark that provides controlled expert pools, integrates fusion methods with and without additional training, and offers a unified evaluation suite. The pools contain domain experts independently trained from Qwen3 and Llama base models across math, code, science, instruction following, and tool use. The suite evaluates capability retention and computational cost on eight public benchmarks under comparable data access rules, reporting normalized average and worst performance, runtime, and peak GPU memory usage. Extensive experiments show that training-free methods preserve reasoning performance but substantially degrade tool use at both model scales. In contrast, MOPD preserves tool use and achieves a more balanced capability profile, but requires additional data preparation, inference, and training. Jointly retaining reasoning, instruction following, and tool use at low computational cost remains a challenge. We hope FusionEval supports future research on efficient model fusion that jointly preserves expert capabilities. Our code is available at https://github.com/Baicaihaochi/FusionEval_Benchmark.

cs.CL↗