arXiv subjects

Bin Chen

Publications and source records attributed to Bin Chen.

At least 37 records · Page 2Linked to original sources

arXiv2026

Adaptive Probabilistic Constellation Shaping based on Enumerative Sphere Shaping for FSO Channel with Turbulence and Pointing Errors

Free-space optical (FSO) transmission enables fast, secure, and efficient next-generation communications with abundant spectrum resources. However, atmospheric turbulence, pointing errors, path loss, and atmospheric loss induce random attenuation, challenging link reliability. Adaptive coded modulation technology enhances spectrum utilization and reliability. We propose an adaptive probabilistic constellation shaping (A-PCS) coherent system utilizing enumerative sphere shaping (ESS) for distribution matcher (DM). With PCS-64QAM, the system achieves continuous rate control from conventional QPSK-equivalent to 64QAM spectral efficiency, providing quasi-continuous control with granularities of approximately $0.05$~bits/4D for spectral efficiency and $0.1$~dB for the post-FEC SNR threshold, and a maximum control depth of $12.5$~dB. Leveraging ESS for efficient sequence utilization, it offers higher spectral efficiency and finer control granularity than constant composition distribution matcher (CCDM)-based A-PCS systems. We further model and analyze the FSO channel, presenting calculations and comparisons of outage probability and ergodic capacity under varying turbulence intensities and pointing errors. Results demonstrate 99.999~\% reliability at maximum $σ_\mathrm{R}^2 = 1.02$ and $σ_\mathrm{s} = 0.51~\mathrm{m}$, meeting requirements under severe turbulence and large pointing errors. {Furthermore, under non-ideal delayed channel state information (CSI) feedback conditions, the system adapts to varying turbulence coherence times and feedback delays, with results showing that finer modulation granularity (provided by A-PCS-ESS) enhances immunity to feedback delay, maintaining a performance advantage over conventional adaptive schemes across a range of channel environments and delay values.

eess.SP↗

arXiv2026

Dense Is Not Enough: Hierarchical Supervision Allocation for Long-Horizon On-Policy Distillation

On-policy distillation (OPD) transfers the capabilities of a large language model to a smaller student by providing teacher supervision on the student's own rollouts. In long-horizon agentic tasks, however, uniform token-level matching can allocate supervision poorly: a large local discrepancy need not improve future behavior, while consequential guidance may be beyond the current student's reach or fail to persist without privileged input. We formulate long-horizon OPD as hierarchical supervision allocation and argue that productive guidance lies at the intersection of future utility and current learnability. Crucially, this intersection evolves as the student learns. Based on this principle, we propose LENS-OPD, a coarse-to-fine framework that organizes supervision through Locate, Validate, and Refine. Locate adapts trajectory exposure to the student's evolving competence and proposes a candidate decision for intervention. Validate tests whether teacher guidance at that decision improves the same student's subsequent behavior. Refine internalizes the beneficial guided behavior into the deployable policy and concentrates token-level supervision on decisive teacher-student conflicts within the validated turn. These stages are nested: each finer allocation is conditioned on the coarser decision, rather than being optimized as an independent importance score. Experiments across multiple long-horizon agent benchmarks and student-teacher configurations show that LENS-OPD consistently improves task performance over vanilla OPD and strong curriculum- and selection-based baselines. Our results suggest that effective long-horizon distillation requires teaching at the right depth, the right decision, and the right token.

cs.CL↗

arXiv2026

MergeHEIR: Mitigating Multimodal Hallucinations as the Tax of Model Merging

Model merging consolidates task-specialized experts into a single deployable model. However, we show that such capability consolidation incurs a merging tax of increased hallucination: across 8 model-merging methods, every merged checkpoint exhibits a higher hallucination rate than the average of its constituent experts. An intuitive approach is to adapt existing hallucination-mitigation methods to the post-merge model, yet this unconstrained adaptation disrupts inherited capabilities, creating a tension between hallucination mitigation and expertise retention. To tackle this challenge, we introduce MergeHEIR, a post-merge adaptation framework designed to reduce this merging tax while preserving expertise inherited from initial experts. Using small expert-task calibration sets, MergeHEIR constructs layer-wise null-space projectors via SVD from task-specific activations collected from the merged checkpoint, and periodically projects the accumulated post-merge displacement onto the resulting null spaces to preserve inherited expertise. Theoretically, we establish minimum-distortion and maximum-dimensionality guarantees, characterize the threshold-controlled adaptation-retention trade-off, and extend perturbation guarantees beyond finite calibration data. Across 24 paired comparisons spanning three MLLM configurations and 8 model-merging methods, MergeHEIR consistently mitigates hallucination while largely preserving inherited expertise, demonstrating a more favorable hallucination-retention trade-off.

cs.AI↗

arXiv2026

C2P-VAR: Continual and Compositional Personalization in Visual Autoregressive Models

Visual autoregressive (VAR) models have recently emerged as an efficient paradigm for text-to-image generation, yet their personalization capabilities remain largely limited to static, single-concept settings. In practice, users may continuously introduce new concepts and wish to compose multiple personalized concepts within a single image. Such scenarios pose two fundamental challenges: catastrophic forgetting during sequential personalization and feature interference during multi-concept composition. In this work, we study continual and compositional personalization in VAR models and propose C2P-VAR, a unified framework addressing both challenges. For continual personalization, we introduce C2PVAR-S, which identifies concept-relevant parameters from gradient magnitudes and dynamically updates their selection during training. To preserve previously learned concepts, C2PVAR-S applies regularization only to parameters shared by the current and historical concepts, thereby reducing unnecessary interference without introducing additional model components. For multi-concept personalization, we further propose C2PVAR-M, which employs parallel global and concept-specific branches with spatially localized feature fusion and logit aggregation to achieve controllable concept placement and reduce feature entanglement. Extensive experiments on continual and multi-concept personalization demonstrate that C2P-VAR consistently outperforms existing baselines in subject fidelity, while maintaining competitive text alignment and introducing negligible storage and inference overhead. Our results establish a unified framework for scalable and controllable personalization of visual autoregressive models.

cs.CV↗

arXiv2026

IndustryLLM: Failure-Driven LLM Training for Industrial Procurement

Industrial procurement requires language models to bridge informal buyer jargon, sparse marketplace attributes, and authoritative engineering standards under strict safety tolerances. We present IndustryLLM, an open-weight industrial language model trained from Qwen3.5-35B-A3B-Base (35B total parameters with ~3B activated per token, with the vision encoder frozen). Rather than relying on generic text scaling, we introduce a failure-driven adaptation recipe spanning continued pre-training (CPT) and supervised fine-tuning (SFT). CPT leverages a curated ~100B-token corpus integrating 5B tokens of national standards (e.g., GB/T) and technical archives, 10B tokens of de-identified real-world industrial transaction and inquiry records, and 60B tokens of general replay. To overcome register mismatch and factual brittleness, we systematically reconstruct an estimated 20B-token domain subset via multi-register rewriting across 10 genres and 8 writing styles, confidence-routed minimal factual editing, and error-targeted QA synthesis (resolving colloquial typos like '42-luo-mu' -> 42CrMo, expanding ambiguous codes like '16674' -> GB/T 16674, and clarifying conflicting dimensional specs). For downstream deployment, we formalize an evidence-gated constraint-evaluation interface enforcing three-valued logic where unverified product evidence remains unknown rather than satisfied. Offline evaluations demonstrate consistent gains on procurement-query structuring (+2.97 percentage points in exact match, 95% CI [2.11, 3.86] in No-Think mode), while randomized online A/B experiments in production yield substantial improvements (+4.25% GMV, +8.3% satisfied inquiries) alongside a latency reduction from 6-7 s to 1.5 s. Model weights and configs are released at https://huggingface.co/alibaba-multimodal-industrial-ai/IndustryLLM.

cs.AI↗

arXiv2026

PANGU I: A Parthenon-Based Framework for Fixed-Spacetime GRMHD

Existing general-relativistic magnetohydrodynamics (GRMHD) codes offer complementary capabilities, but these capabilities remain distributed across different frameworks. To unify them, we introduce PANGU--Parthenon-Based Astrophysics for Numerical Relativity and GRMHD in a Unified Framework--a modular computational framework that integrates a shared solver across physical regimes, flexible spacetime geometry, and extensible computational infrastructure within a single architecture. In the first paper of this series, we present and validate the fixed-spacetime GRMHD component of PANGU. The shared solver combines finite-volume Godunov methods with constrained transport on uniform and block-structured, statically refined meshes, and covers hydrodynamics and magnetohydrodynamics in Newtonian, special-relativistic, and general-relativistic regimes. PANGU provides a flexible geometry interface supporting coordinate selection and either on-the-fly or precomputed evaluation of prescribed stationary spacetimes, demonstrated here in Kerr spacetime with Cartesian Kerr-Schild and modified Kerr-Schild coordinates; additional components can be incorporated without restructuring the core solver. Verification covers all three regimes on uniform and statically refined meshes: smooth Newtonian and special-relativistic problems approach second order against their analytic solutions, with deliberately matched double-precision trajectories agreeing with AthenaK to roundoff; in fixed-spacetime GRMHD, the stationary analytic solutions are preserved to second order, and the turbulent accretion reproduces the same flow and flux scales as BHAC. Together, these results establish the fixed-spacetime matter-sector foundation for the broader PANGU framework. Numerical-relativity and coupled dynamical-spacetime GRMHD capabilities will be presented in subsequent work.

astro-ph.HE↗

arXiv2026

Holographic description of RN-like black hole in bumblebee gravity

We study the holographic description of the RN-like black hole in bumblebee gravity, which is a vector-tensor theory with spontaneous Lorentz violation. By combining the first laws of the outer and inner horizons, the thermodynamics method yields the right- and left-moving entropies, temperatures, and chemical potentials of the dual two-dimensional conformal field theory (CFT), determines the central charges from the mass-independent entropy product, and reproduces the outer and inner horizon entropies through the Cardy formula. An independent dynamical test based on the hidden conformal symmetry of a charged scalar probe confirms these results. A central outcome is the equality of the right- and left-moving central charges, $c_R=c_L$, which we trace to the spontaneous character of Lorentz breaking. Because the action of bumblebee gravity preserves diffeomorphism and local Lorentz invariance, and the symmetry is broken only by a vector vacuum expectation value, no chirally asymmetric structure arises in the dual theory, in sharp contrast to topologically massive gravity in $(2+1)$ dimensions. All CFT data depend smoothly on the Lorentz-violating parameter and reduce to the known RN results when Lorentz violation is switched off.

hep-th↗

arXiv2026

KPolaris: GPU-accelerated Polarized Radiative Transfer in General Relativity

Polarized black-hole images carry information about both the magnetized plasma and the curved spacetime through which the radiation propagates. We present KPolaris, an open-source GPU-accelerated general-relativistic polarized radiative-transfer code built on the Kokkos framework. It supports analytic plasma models and GRMHD simulation data, with both fast-light and slow-light radiative transfer. The code also provides diagnostics that connect image structure to the emitting plasma and to propagation effects along the rays. These include emissivity-weighted plasma maps, emission contributions from selected regions, selected-ray histories, and sensitivities of the Stokes parameters to plasma parameters. Numerical convergence tests and comparisons with an independent radiative-transfer code assess the accuracy of the calculations. GPU acceleration reduces the computational cost of image generation, facilitating multi-frequency and time-dependent studies of polarized black-hole images.

astro-ph.HE↗

arXiv2026

Dense Dump Codec: Error-Controlled, Random-Access Compression of GRMHD Time Series for Slow-Light Radiative Transfer

Black-hole movies provide a unique venue to study the time evolution of accreting plasma. Modeling this evolution with slow-light radiative transfer requires closely spaced simulation outputs, because the plasma changes as light propagates through it. Saving these outputs in full is costly, whereas sparse sampling introduces time-interpolation errors. We present the Dense Dump Codec (DDC), a compression method for dense GRMHD time series. DDC stores exact anchor states and compactly represents the intermediate evolution, while allowing direct access to selected times and variables. Applied to eight production sequences spanning SANE and MAD simulations at several black-hole spins with 0.1M output, DDC archives are about 10 times smaller than the original dense data and even about 50% smaller than the original 0.5M output in aggregate. We further integrate a native DDC reader into polarized slow-light radiative transfer. Tests at 86, 230, and 345 GHz show that DDC may reduce aggregate Stokes-image errors by 54-79% relative to slow-light calculations using original 0.5M input. These results indicate that DDC enables high-cadence slow-light calculations without the prohibitive cost of saving every dense GRMHD state in its original form.

astro-ph.HE↗

arXiv2026

From Bits to Beliefs: Recoverable Semantic Fingerprints for Black-Box Verification of Large Language Models

Open-weight large language models (LLMs) can be copied, modified, and redeployed behind black-box APIs, making post-release ownership verification difficult. Existing black-box fingerprints often rely on secret query-key pairs that reproduce predefined responses, and can therefore be easily disrupted by fine-tuning, pruning, quantization, model merging, and serving-time prompt changes. We propose SimPrint, a recoverable semantic fingerprinting framework for black-box LLM ownership verification. Rather than relying on isolated exact matches, SimPrint encodes a private owner signature into a coded semantic fingerprint domain, distributing ownership evidence across natural binary question-answering probes. It implants only base-deviating probes through a low-interference batch update that preserves the original model behavior, and later recovers the signature by parsing suspect-model responses into reliable bits or erasures with an error-correcting recovery mechanism. Because verification only uses input-output queries, SimPrint remains applicable when model weights or activations are inaccessible. Experiments on three open-weight LLMs show that SimPrint reliably recovers the owner signature in both clean and modified settings, remains robust under fine-tuning, pruning, quantization, model merging, and serving-time perturbations, and maintains comparable downstream utility.

cs.CR↗

arXiv2026

The solvability of the $L_p$-dual mixed Euclidean-Gauss Minkowski problem

This paper investigates the $L_p$-dual mixed Euclidean-Gauss Minkowski problem associated with the $L_p$-dual mixed Euclidean-Gaussian curvature measure. The solvability of this problem is equivalent to that of a class of Monge-Ampère type equations on the unit sphere. Under smooth assumptions, We establish the existence of solutions for $p<1$, $q<0$, and $q<p$ and uniqueness of solutions for $p\geq q$ to the aforementioned Monge-Ampère type equations, respectively. Furthermore, we obtain a complete solution concerning the existence part of this problem when $p<1$ and $q<0$.

math.AP↗

arXiv2026

Privacy Leakage on DNNs: A Survey of Model Inversion Attacks and Defenses

Deep Neural Networks (DNNs) have revolutionized various domains with their exceptional performance across numerous applications. However, Model Inversion (MI) attacks, which disclose private information about the training dataset by abusing access to the trained models, have emerged as a formidable privacy threat. Given a trained network, these attacks enable adversaries to reconstruct high-fidelity data that closely aligns with the private training samples, posing significant privacy concerns. Despite the rapid advances in the field, we lack a comprehensive and systematic overview of existing MI attacks and defenses. To fill this gap, this paper thoroughly investigates this realm and presents a holistic survey. Firstly, our work briefly reviews early MI studies on traditional machine learning scenarios. We then elaborately analyze and compare numerous recent attacks and defenses on Deep Neural Networks (DNNs) across multiple modalities and learning tasks. By meticulously analyzing their distinctive features, we summarize and classify these methods into different categories and provide a novel taxonomy. Finally, this paper discusses promising research directions and presents potential solutions to open issues. To facilitate further study on MI attacks and defenses, we have implemented an open-source model inversion toolbox on GitHub (https://github.com/ffhibnese/Model-Inversion-Attack-ToolBox).

cs.CV↗

arXiv2026

Distorting Kerr Images with Parity-Odd Scalar Hair

We investigate thin-disk imaging of Kerr black holes with synchronized scalar hair, focusing on backreacted parity-odd excited states of a complex scalar field minimally coupled to Einstein gravity. The spacetime displays a core-double-torus lensing structure, with a central black hole surrounded by two scalar clouds. We study the dependence of the images on hair strength and viewing angle, identifying a weak-hair regime close to Kerr. With increasing hair, the photon ring and shadow region shrink and become more distorted. In the strong-hair regime, gravitational lensing produces new features, including multiple disconnected shadow components, crescent-shaped structures, and signatures of chaotic lensing. For nearly edge-on viewing angles, repeated equatorial crossings generate nested ring-like patterns. These results highlight possible geometric signatures of black holes with excited scalar hair.

gr-qc↗

arXiv2026

Behavior2Value: Benchmarking and Empowering LLMs for Consumer Value Measurement from E-commerce Behaviors

Human values are deep motivational orientations that shape human behaviors. In e-commerce, they reveal the stable drivers behind users' purchase decisions. Compared with short-term interests, consumer values better explain how users evaluate products before purchase. However, consumer values are often implicit in complex and fragmented behavioral trajectories, leaving value measurement from e-commerce behaviors largely underexplored. To this end, we propose the Behavior-to-Value (B2V) task, which aims to identify consumer values from e-commerce behavioral trajectories. Centered on this task, we first construct the E-commerce Consumption Value Taxonomy (ECVT) and introduce B2V-Bench, the first B2V dataset and benchmark, based on anonymized Taobao behavioral logs. B2V-Bench consists of real-world purchase decision episodes, covering 25 types of purchase behaviors, along with corresponding consumer value orientations manifested in each episode. To improve consumer value measurement accuracy, we further present B2V-Verifier, a behavior-to-value measurement model based on Value Verification Tuning, which learns to assess whether behaviors provide sufficient evidence for each value inference. Experiments show that B2V-Verifier outperforms strong LLM baselines, improving multi-label classification by 34\%. The dataset and code will be publicly released upon acceptance.

cs.CL↗

arXiv2026

Heisenberg-like critical behavior in the quasi-two-dimensional metallic ferromagnet LaCrSb3

LaCrSb3 is a material exhibiting both quasi-two-dimensional spin fluctuations and three-dimensional magnetic interaction characteristics. By measuring the isothermal magnetization of single-crystals and conducting a systematic critical behavior analysis, we clarify the critical properties of its ferromagnetic phase transition and the intrinsic magnetic interaction mechanism. Based on high-precision isothermal magnetization data measured in the vicinity of the critical point, the Curie temperature for the ferromagnetic-paramagnetic phase transition is determined to be TC = 126 K, with the critical exponents obtained as $β$ = 0.376, $γ$ = 1.417 and $δ$ = 4.76 via the self-consistent iterative method based on the Arrott-Noakes equation. The reliability of these critical exponents is verified by the Widom scaling law, the magnetic state scaling equation and other analyses. A comparison with theoretical models demonstrates that the critical behavior of the magnetic phase transition in this system basically belongs to the universality class of the three-dimensional Heisenberg model. This conclusion is further confirmed by the distance-dependent decay behavior of the exchange interaction J(r), revealing the dominant role of isotropic direct exchange interactions in this system. Finally, drawing on research findings of other quasi-two-dimensional magnetic materials, this work proposes that LaCrSb3 may exhibit finite-temperature magnetic order in the two-dimensional limit, thereby possessing important theoretical research significance and promising practical application prospects.

cond-mat.str-el↗

arXiv2026

Ground States of Nonlinear Fermionic Systems: From Power-Law Interactions to Logarithmic Sobolev Inequality

We consider ground states of a two-component logarithmic fermionic system in $\mathbb{R}^d$, where $d\ge 1$ is arbitrary. We prove that up to translations and scalings, ground states of the logarithmic system are the $L^\infty $-limits of ground states for a two-component $2p-1$ power-law fermionic system as $p\searrow 1$. As a byproduct, we also establish a sharp logarithmic Sobolev inequality for orthonormal functions in $\mathbb{R}^d$, whose optimizers are, up to scalings, the minimizers of a constraint variational problem associated with the logarithmic system.

math.AP↗

arXiv2026

A new product entropy

We propose a new product entropy, defined as the Rényi (or von Neumann) entropy of a normalized product operator constructed from two density matrices. We establish a duality showing that the SVD entanglement entropy of a subsystem for two pure states is exactly equivalent to the product entropy of the complementary subsystem. This connection provides both a transparent physical interpretation in terms of the spectral diversity of the subsystem state product and a computationally efficient route that bypasses the reduced transition matrix. For low-lying eigenstates, we derive analytical expressions for the subsystem product entropy between the ground state and primary excitations in two-dimensional conformal field theories, explicitly verified against the critical Ising chain. In quantum quench dynamics, the quasiparticle picture yields time evolution in the scaling limit: following a global quench, the subsystem product entropy exhibits distinct sequences of thermalization and revivals, whereas under a local operator quench, it develops characteristic plateaus whose constant values are determined by the inserted operator. Extensive numerical calculations on the critical Ising chain confirm the analytical predictions with excellent accuracy.

hep-th↗

arXiv2026

Free field realization of the BMS Ising model

In this work, we study the inhomogeneous BMS free fermion theory, and show that it gives a free field realization of the BMS Ising model. We find that besides the BMS symmetry there exists an anisotropic scaling symmetry in BMS free fermion theory. As a result, the symmetry of the theory gets enhanced to an infinite dimensional symmetry generated by a new type of BMS-Kac-Moody algebra, different from the one found in the BMS free scalar model. Besides the different coupling of the $u(1)$ Kac-Moody current to the BMS algebra, the Kac-Moody level is nonvanishing now such that the corresponding modules are further enlarged to BMS-Kac-Moody staggered modules. We show that there exists an underlying $W(2,2,1)$ structure in the operator product expansion of the currents, and the BMS-Kac-Moody staggered modules can be viewed as highest-weight modules of this $W$-algebra. Moreover we obtain the BMS Ising model by a fermion-boson duality. This BMS Ising model is not a minimal model with respect to BMS$_3$, since the minimal model construction based on BMS Kac determinant always leads to chiral Virasoro minimal models. Instead, the underlying algebra of the BMS Ising model is the $W(2,2,1)$-algebra, which can be understood as a quantum conformal BMS$_3$ algebra.

hep-th↗