arXiv subjects

Hongsheng Chen

Publications and source records attributed to Hongsheng Chen.

At least 19 recordsLinked to original sources

arXiv2026

Non-Polynomial Wave Computation through Recurrent Resonant Scattering

Structural nonlinearity enables nonlinear input--output mappings to emerge from otherwise linear wave dynamics. Repeated interactions with an input-encoded structure can enhance such mappings, but finite-depth implementations restrict the accessible functional order. Here we show that recurrent scattering in a resonant cavity provides a distinct regime for structural nonlinearity. Repeated interactions are coherently accumulated into a non-polynomial response to structural perturbations. The resulting mapping naturally takes a Kolmogorov--Arnold form: perturbation-induced resonance shifts realize the inner univariate mappings, while the resonant spectral response provides the outer mapping. We identify two complementary physical controls of representation capacity: the resonance linewidth governs the functional richness within each branch, whereas combining multiple branches expands the accessible function space. Microwave-cavity measurements validate the two-stage mapping and demonstrate a two-branch nonlinear computation through XOR classification. Our results connect recurrent resonant scattering with controllable non-polynomial computation in linear wave systems.

physics.optics↗

arXiv2026

A Structural Proof of the Lower Bound 21 for $3\times3$ Matrix Multiplication over $\mathbb F_2$

We prove that the tensor rank of $3\times3$ matrix multiplication over $\mathbb F_2$ is at least $21$. The structural proof, independently developed by Qiushi Engine, converts occupation constraints on a single tensor factor into algebraic relations coupling all three factors. Certified quotient-rank bounds and finite geometry force any hypothetical $20$-term decomposition to have first-factor matrix-rank profile $(16,1,3)$. The ranks of the corresponding split-flattened summands therefore sum to $27$, exactly the rank of the full split flattening. Equality in rank subadditivity forces their images to form a direct sum; normalization by the inverse flattening then makes the summands pairwise annihilating idempotents. An explicit product identity for matrix multiplication implies that at most one first factor can be invertible, contradicting the three forced by the profile. The same obstruction constrains $22$-term decompositions attaining the split-rank bound. The complete proof, including the finite quotient bounds, is formalized in Lean. The accompanying research trajectory records Qiushi Engine's long-horizon autonomous research, from numerical experiments and quotient constructions to the structural proof.

cs.CC↗

arXiv2026

Data-Efficient Language Modeling: From Frontier Advancement to Principle-Guided Model Improvement

Learning from limited text requires models to use context, generalize to new inputs, and retain useful capabilities. Qiushi Engine conducted a long-horizon, end-to-end autonomous research program on BabyLM 2026 Strict-Small, within 10 million corpus words and 100 million cumulative word presentations. Three stages connected frontier advancement, principle discovery, and principle-guided model improvement. Stage I combined compact restatements, budget reinvestment, and residual incremental learning to build a frontier model. Stage II found that exact repetition and aligned restatement produce different patterns of context use, depending on target relations and prediction windows. In controlled tasks, recovering familiar performance did not ensure that unseen inputs could still use learned computations. These findings support a testable data-efficient learning principle: organize experience around the contextual dependencies needed for prediction; separately design visible information, supervision, and preservation; test learning, generalization, and retention. Stage III retained source text, masked more local clues, supervised selected targets, and preserved predictions on ordinarily masked inputs. Two continuation seeds from the same parent outperformed ordinary continuation on the complete nine-metric aggregate. Overall rose from 42.02 to 42.25 across two generations; the second achieved the highest Overall in the public Strict-Small snapshot of 8 September 2026. Further studies addressed compression, relational anchors, shared representations, and measurement. Models are available on Hugging Face; code and research records accompany the GitHub repository. Together, these stages illustrate Research RSI: recursive self-improvement of the research process. Scientific understanding and method innovations change subsequent questions and designs; new experiments test and refine them.

cs.CL↗

arXiv2026

Qiushi Engine on AstaBench E2E-Bench-Hard

This report analyzes Qiushi Engine v0.8 across all 40 test tasks in AstaBench E2E-Bench-Hard, a benchmark that requires autonomous agents to carry a research question through experimental design, code implementation, actual execution, result analysis, and report delivery. Qiushi Engine is model-configurable; this evaluation selected DeepSeek deepseek-v4pro-preview as the model backend. The official AstaBench leaderboard records a score of 0.816 and an average benchmark cost of USD 15.209 per task, while the full-precision local recomputation is $81.59 \pm 1.87$. Four tasks satisfied every rubric item, yielding a full-task completion rate of 4/40 = 10% -- 7 percentage points above, and about 3.3 times, the approximately 3% best rate reported for AstaBench's official agents. Across 507 required rubric items, 416 were satisfied (82.1%). Official scoring archives and 40 Meta-Trace records show sustained production and verification of reports, code, and experimental artifacts; the principal gaps lie in repeated runs, external dependencies, specified metrics, and ablation studies. The report explains the benchmark, system workflow, aggregate results, representative cases, and limits of interpretation.

cs.AI↗

arXiv2026

VortexChat: An agentic framework for autonomous multi-objective integrated photonic design

The advancement of modern integrated photonics is frequently bottlenecked by device design workflows that rely heavily on manual simulation and expert intuition. While inverse design offers an alternative, it remains constrained by expert supervision and a lack of end-to-end automation. To address these issues, we present VortexChat, an agentic framework for the autonomous, end-to-end inverse design of integrated photonic devices directly from natural language specifications. VortexChat couples a large language model (LLM) decision agent with topology generation, gradient-based refinement, and full-wave electromagnetic simulation. This closed-loop architecture enables the system to iteratively decompose design objectives, orchestrate computational tools, and update strategies based on feedback with minimal human intervention. Constrained by the absolute metrics of the Vortex100 Benchmark, VortexChat autonomously generates devices that strictly meet all predefined performance thresholds without any human-in-the-loop. As an experimental demonstration, we fabricated a broadband terahertz perfect vortex beam multiplexer, autonomously designed by VortexChat, with measurements confirming high-efficiency operation, high mode purity and low inter-channel crosstalk in agreement with full-wave simulations. These results demonstrate that an LLM agent can assume key aspects of expert decision-making in photonic inverse design while maintaining physical fidelity and fabrication feasibility, providing a scalable route towards autonomous design of complex integrated photonic systems.

cs.AI↗

arXiv2026

Causality Sum Rules in Conventional Scattering Matrices

Scattering matrices are the standard experimental and computational description of photonic and electromagnetic devices. Passivity is explicit in the conventional incoming-outgoing matrix, whereas causality sum rules are usually formulated only after transforming the response into auxiliary variables. Here we show that these rules can be written directly in the conventional scattering matrix by removing the time advance introduced by the reference domain. Using the earliest-arrival delay of each channel, we define a domain-delayed matrix that preserves real-frequency passivity while restoring the causal time origin. Under explicit analyticity, transparency, and regularity assumptions, this matrix becomes a Schur function, enabling a Cayley-Herglotz construction. The resulting projected and determinant bounds constrain coherent channel superpositions and aggregate multichannel loss. The framework recovers Rozanov's absorber limit and spherical-multipole sum rules, while extending causality bounds to measurable quantities including insertion loss, suppressed singular-value channels, and conditional lossless delay-bandwidth trade-offs. Our work directly connects fundamental causality theory with experimentally accessible scattering data. The initial theoretical route is autonomously explored by Qiushi Engine, an AI research system for open-ended scientific discovery, and subsequently verified, refined, and developed by the authors, demonstrating a hybrid AI-human discovery workflow.

physics.optics↗

arXiv2026

Brewster-anomaly delocalization for free-electron radiation

Localization effects are central to disordered electronics and photonics. In electronics, Anderson localization governs electron confinement in randomly perturbed lattices. Similarly, its photonic counterpart inhibits light transport via disorder -- but with a unique exception: Brewster-anomaly delocalization, where the Brewster effect prevents multiple-scattering interference and counteracts the localization. Despite extensive research in electronics and photonics separately, the intricate role of localization effects in free-electron--light interactions -- vital for lasers, accelerators, microscopy and spectroscopy, and quantum information -- remains largely unexplored. At the same time, localization effects are widely regarded as a key factor limiting the efficient coupling between free electrons and light in random media. Here we overcome this key limitation via the unconventional interplay between Brewster-anomaly delocalization and free-electron radiation. In this way, free-electron radiation can be localization-free, intense and directional even in strongly disordered, unengineered multilayers. Essentially, this delocalization-mediated free-electron radiation is remarkably invariant not only to the random-medium configuration, but also to the light frequency and the electron velocity. Our findings unlock new opportunities for particle detectors and achromatic light sources operating in easy-to-fabricate complex media at previously inaccessible frequencies.

physics.optics↗

arXiv2026

Self-healing topological streams in space-time

Topological states are renowned for their robustness against perturbations. Recent advances have further introduced momentum-gap (or time) topology in time-varying media, enabling temporal topological states. However, existing studies for momentum-gap topology are largely confined to non-propagating interface states; their potential for wave transport, especially at moving boundaries, has yet to be realized. Here, we experimentally demonstrate topological streams in space-time: space-time topological edge states that propagate along a moving boundary in a dynamically modulated time-synthetic photonic lattice. These streams arise from the coexistence of energy-gap and momentum-gap topologies and break both spatial and temporal translation symmetries, conserving neither energy nor momentum. Remarkably, they reconstruct their wave profiles after strong localized spatiotemporal obstacles, enabled by an imaginary quasienergy gap and kinematic decoupling from radiative bulk channels. Our results establish a form of self-healing spatiotemporal non-Hermitian protection, showing that the temporal dimension can transform topology from pinned states and isolated events into robust, directional and self-restoring wave transport.

physics.optics↗

arXiv2026

Fronthaul-Efficient Distributed Cooperative 3D Positioning with Quantized Latent CSI Embeddings

High-precision three-dimensional (3D) positioning in dense urban non-line-of-sight (NLOS) environments benefits significantly from cooperation among multiple distributed base stations (BSs). However, forwarding raw CSI from multiple BSs to a central unit (CU) incurs prohibitive fronthaul overhead, which limits scalable cooperative positioning in practice. This paper proposes a learning-based edge-cloud cooperative positioning framework under limited-capacity fronthaul constraints. In the proposed architecture, a neural network is deployed at each BS to compress the locally estimated CSI into a quantized representation subject to a fixed fronthaul payload. The quantized CSI is transmitted to the CU, which performs cooperative 3D positioning by jointly processing the compressed CSI received from multiple BSs. The proposed framework adopts a two-stage training strategy consisting of self-supervised local training at the BSs and end-to-end joint training for positioning at the CU. Simulation results based on a 3.5~GHz 5G NR compliant urban ray-tracing scenario with six BSs and 20~MHz bandwidth show that the proposed method achieves a mean 3D positioning error of 0.48~m and a 90th-percentile error of 0.83~m, while reducing the fronthaul payload to 6.25% of lossless CSI forwarding. The achieved performance is close to that of cooperative positioning with full CSI exchange.

eess.SP↗

arXiv2026

Observation of localization reversal and harmonic generation in nonlinear non-Hermitian skin effect

The interplay between band topology and material nonlinearity gives rise to a variety of novel phenomena, such as topological solitons and nonlinearity-induced topological phase transitions. However, most previous studies fall within the Hermitian regime, leaving the impact of nonlinearity on non-Hermitian topology seldom explored. Here, we investigate the effects of nonlinearity on the non-Hermitian skin effect, a well-known non-Hermitian phenomenon induced by the point-gap topology unique to non-Hermitian systems. Interestingly, we discover a nonlinearity-induced point-gap topological phase transition accompanied by a reversal of the skin mode localization, which is distinct from previous nonlinearity-induced line-gap topological phases. This phenomenon is experimentally demonstrated in a nonlinear microwave metamaterial, where electromagnetic waves are localized around one end of the sample under a low-intensity pump, whereas at a high-intensity pump, the waves are localized around the other end. Our results open a new route towards nonlinear topological physics in non-Hermitian systems and are promising for reconfigurable topological wave manipulation.

cond-mat.mes-hall↗

arXiv2026

Agentic metasurface design with self-correcting language-model systems

Automated metasurface design is increasingly important, and recent advances in language-model systems are opening a route toward agentic optical design. Yet modern metasurface applications, from metalenses and holography to optical computing, require long design chains spanning modeling, simulation, coding, optimization and evaluation. These chains are error-prone, whereas existing language-model-based metasurface tools remain largely limited to simple objectives, predefined pipelines or language-to-layout generation. Here we introduce MetaDesigner, a self-correcting language-model system for agentic metasurface design. From a natural-language optical objective, MetaDesigner plans the design route, retrieves domain knowledge, invokes simulation and optimization tools, generates missing tool code and identifies errors through a dedicated Verifier. We demonstrate three tasks of increasing complexity: an RGB metalens with three independent focal spots, a six-plane full-color hologram with an average structural similarity index measure (SSIM) of 0.97, and an optoelectronic hybrid neural network for image style transfer. These tasks require 74, 136 and 90 reasoning steps, respectively, and the system self-corrects errors in frequency mapping, numerical aperture estimation, network-parameter counting and loss-function description. These results establish MetaDesigner as a self-correcting route to agentic metasurface design, where language-model systems can not only execute optical design tasks but also extend, inspect and repair the design process itself.

physics.optics↗

arXiv2026

End-to-end autonomous scientific discovery on a real optical platform

Scientific research has long been human-led, driving new knowledge and transformative technologies through the continual revision of questions, methods and claims as evidence accumulates. Although large language model (LLM)-based agents are beginning to move beyond assisting predefined research workflows, none has yet demonstrated end-to-end autonomous discovery in a real physical system that produces a nontrivial result supported by experimental evidence. Here we introduce Qiushi Discovery Engine, an LLM-based agentic system for end-to-end autonomous scientific discovery on a real optical platform. Qiushi Engine combines nonlinear research phases, Meta-Trace memory and a dual-layer architecture to maintain adaptive and stable research trajectories across long-horizon investigations involving thousands of LLM-mediated reasoning, measurement and revision actions. It autonomously reproduces a published transmission-matrix experiment on a non-original platform and converts an abstract coherence-order theory into experimental observables, providing, to our knowledge, the first observation of this class of coherence-order structure. More importantly, in an open-ended study involving 145.9 million tokens, 3,242 LLM calls, 1,242 tool calls, 163 research notes and 44 scripts, Qiushi Engine proposes and experimentally validates optical bilinear interaction, a physical mechanism structurally analogous to a core operation in Transformer attention. This AI-discovered mechanism suggests a route towards high-speed, energy-efficient optical hardware for pairwise computation. To our knowledge, this is the first demonstration of an AI agentic system autonomously identifying and experimentally validating a nontrivial, previously unreported physical mechanism, marking a milestone for research-level autonomous agents.

cs.AI↗

arXiv2026

Testing holographic duality in hyperbolic lattices

The celebrated holographic duality posits a correspondence between a quantum gravity in a bulk spacetime and a quantum field theory (QFT) defined on its lower-dimensional boundary. This duality not only offers deep insights into the enigmatic nature of quantum gravity but also provides an efficient methodology for studying strongly correlated systems. However, despite its profound significance in modern physics, holographic duality remains a conjecture, and further experimental exploration is highly sought after. Here, we present the first experimental test of holographic duality between a three-dimensional bulk gravity and a two-dimensional boundary QFT using hyperbolic lattices. By experimentally measuring the classical scalar field propagator in hyperbolic circuits, we reproduce the equal-time two-point correlation function of the dual boundary conformal field theory (CFT), verifying its exponential dependence on the boundary separation and the conformal dimension-scalar mass relation. Furthermore, by leveraging the two-point correlation function, we reconstruct the entanglement entropy for a boundary CFT subsystem, confirming that it follows the Ryu-Takayanagi formula. These results constitute the first direct experimental evidence that quantum properties of the QFT can be holographically reproduced through its dual classical field in curved space. This heuristic experimental effort opens a new avenue for in-depth investigations on the holographic duality and extensive exploration of quantum-gravity-inspired phenomena in classical systems.

hep-lat↗

arXiv2026

Non-diffusion transport in decoherent non-Hermitian quasicrystals

Disorder and coherence jointly govern wave transport in complex media. In Hermitian systems, a long-established paradigm since Anderson's work holds that disorder-induced localization relies on phase-coherent interference, and that the loss of coherence inevitably suppresses localization and restores featureless diffusive transport at long times. Whether this intuition remains valid in non-Hermitian systems, where transport can be governed by dissipation rather than interference, has remained largely open. Here we theoretically and experimentally demonstrate that this paradigm fundamentally breaks down in decoherent non-Hermitian quasicrystals. Using a programmable photonic lattice with independently engineered dissipation and fully programmable dephasing, we access regimes spanning from fully coherent to fully incoherent dynamics. While decoherence washes out localization and enforces structureless diffusion in Hermitian lattices, we find that decoherent non-Hermitian quasicrystals retain nontrivial, non-diffusive transport structures even in the incoherent limit. These include dissipation-induced localization, diffusion-localization transitions, and decoherence-induced mobility edges, phenomena with no counterparts in Hermitian disordered systems. We develop a unified theoretical framework that captures the ensemble-averaged dynamics across the entire coherence landscape, continuously connecting coherent and incoherent regimes, and reveals how dissipation and decoherence cooperate to shape transport. Our results establish decoherent non-Hermitian lattices as a distinct class of transport systems, in which dissipation and incoherence generate structured, non-diffusive phases, beyond the conventional Anderson picture.

physics.optics↗

arXiv2026

Topological Valley-Reshaped Device: Bifunctional Waveguiding and Single-Beam Leaky-Wave Radiation for Terahertz Communication

Topological photonics has emerged as a powerful platform for terahertz on-chip systems due to its robust waveguiding capabilities. However, directly extracting topological valley-locked edge states into directional free-space radiation without auxiliary couplers while preserving guided-wave functionality remains a fundamental challenge. In this work, we propose and experimentally demonstrate a bifunctional topological valley-reshaped device. By introducing an angular truncation and a spatial displacement to a complete topological waveguide (TW), the resulting structure inherently retains its waveguiding capabilities. Furthermore, when operated as an isolated section, it functions as a topological leaky-wave antenna (TLWA) that exhibits directional single-lobe radiation. The TW shows low-loss guided-wave performance with an 18 GHz operating bandwidth, supporting error-free transmission up to 60 Gbps. For the TLWA, by gradually reducing the number of protective lattices that are orthogonal to the propagation direction, the valley-locked edge state becomes momentum-matched to the free-space light line, generating leaky-wave radiation. Simultaneously, reshaping of the opposite valley-locked edge state suppresses far-field side lobes and reduces reflection, yielding a clean single-beam radiation pattern with a side-lobe suppression ratio (SLSR) exceeding 15 dB. The TLWA realizes a measured peak gain of 12.5 dBi and a 19 GHz operating bandwidth. Notably, the low-dispersion property of the K-valley radiation allows the main-lobe direction to vary by only 2 degrees across the entire operating band, enabling error-free free-space reception at 24 Gbps. This bifunctional design represents a key step toward highly integrated and modular terahertz on-chip systems.

physics.optics↗

arXiv2026

Programmable Electromagnetic Space via Metasurface Clusters

The rapid evolution of next-generation communications and the Internet of Things (IoT) has catalyzed an urgent demand for governing expansive spatial environments as functional electromagnetic (EM) entities. However, deterministically programming such open EM spaces remains a formidable challenge, as current methodologies are largely confined to localized interfaces that lack the collective coordination required to orchestrate unbounded environments. Here, we introduce a general framework for the deterministic programming of EM space via cooperative metasurface clusters, achieved by mapping volumetric field interference landscapes onto a virtual nodal network. By representing excitations and meta-atoms as fully interconnected nodes, we transform intricate non-local interactions into tractable nodal states, enabling the precise quantitative synthesis of spatial scattering. This framework bridges local meta-atoms with global EM environment to program space as a functional entity, as demonstrated by a deeply coupled meta-emitter for programmable collective radiation and metasurface clusters that sculpt angle-resolved illusion spaces. By transitioning from individual components to cooperative multi-body assemblies, our work provides a scalable foundation for next-generation wireless networks, wave-based analog computing, and ambient intelligence, where space itself becomes a coherent functional and reconfigurable entity capable of holistic information management.

physics.optics↗

arXiv2026

A scalable and programmable optical neural network in a time-synthetic dimension

Programmable optical neural networks (ONNs) can offer high-throughput and energy-efficient solutions for accelerating artificial intelligence (AI) computing. However, existing ONN architectures, typically based on cascaded unitary transformations such as Mach-Zehnder interferometer meshes, face inherent scalability limitations due to spatial encoding, which causes optical components and system complexity to scale quadratically with network size. A promising solution to this challenge is the use of synthetic dimensions to enhance scalability, though experimental demonstration has remained scarce. Here, we present the first experimental demonstration of an all-optical, highly scalable, programmable ONN operating in a time-synthetic dimension. By implementing a time-cycle computation paradigm analogous to gate cycling in conventional spatial photonic circuits, our approach achieves a gate count surpassing that of state-of-the-art programmable photonic processors. Unlike conventional ONN architectures that rely on real-space wave interferences, our framework exploits time-reflection and time-refraction to perform computations, fundamentally eliminating backscattering errors through causality constraints. To bridge the gap between simulation and reality, we introduce an in-situ training framework that dynamically adapts to experimental errors, achieving performance exceeding traditional in silico learning paradigms. Our synthetic-dimension-based approach provides a compact, scalable, backscattering-free, and programmable neuromorphic computing architecture, advancing the potential for next-generation photonic AI systems.

physics.optics↗

arXiv2026

Realizing anomalous Floquet non-Abelian band topology in photonic scattering networks

The concept of multi-gap topology has recently been shown to give rise to uncharted phases beyond conventional single-gap classifications. These phases relate to band nodes with non-Abelian quaternion charges and momentum-space braiding processes characterized by new invariants such as paradigmatic Euler class, phenomena that intrinsically require at least two spatial dimensions. Extending such phases into the non-equilibrium regime is predicted to unlock even richer multi-gap topologies beyond static settings, yet their experimental realization has remained elusive due to the stringent requirements on dimensionality, symmetry, and dynamical control. Here, we theoretically demonstrate and, for the first time, experimentally realize two-dimensional (2D) Floquet non-Abelian band topology in photonic scattering networks. Within this platform, we uncover a sequence of topological phenomena unique to 2D multi-gap systems far from equilibrium, including anomalous multi-gap phases interconnected by band nodes, Floquet Euler transfer, gapped phases with anomalous Dirac string configurations, and Floquet-induced non-Abelian braiding of band nodes. In addition, we observe Floquet-periodic anomalous edge states across multiple gaps, providing experimental signatures of these sought-after 2D multi-gap Floquet topological phases. Our results establish photonic scattering networks as a practical and versatile route to non-Abelian Floquet systems, opening avenues for dynamical topological physics with braiding capability and robust photonic functionalities.

physics.optics↗