arXiv subjects

Junqi Wang

Publications and source records attributed to Junqi Wang.

At least 19 recordsLinked to original sources

arXiv2026

Scaling and Renormalization in Neural Networks: An Effective Renormalization-Group Description of Grokking

Grokking is a delayed transition from memorization to generalization in deep neural networks (NNs). We examine whether scaling and renormalization group (RG) theories can describe this dynamics, establishing a quantitative physical framework for neural generalization transitions. For modular addition on a two-layer NN, an ensemble of trained models defines an order parameter (the squared weight norm) and fluctuation proxies for the heat capacity and the susceptibility, with training epoch and weight decay serving as thermal and field control coordinates. To bridge neural architectures to real-space RG, we construct an exact coupling matrix via a second-order expansion of the loss with respect to per-neuron norms and use its Fiedler vector to define an effective geometry. This spectral ordering yields a monotonic spatial decay of coupling strength that is absent under random permutations. We then execute real-space RG by block-spin aggregating adjacent neurons and parameterizing the coupling matrix into off-diagonal band strengths. Linearizing this local RG map yields relevant scaling exponents and five derived critical exponents, with direct empirical estimates matching RG predictions within 1-10%. Overall, our framework provides a consistent effective description of grokking, suggesting the scaling and renormalization group formalism as a tool to investigate NN dynamics.

cs.LG↗

arXiv2026

Broadband soliton microcombs in the biological window enabled by a resonant coupler

Broadband soliton microcombs near 1 $μ$m offer coherent, evenly spaced optical lines for measurements in the biological window. However, combining broad spectral coverage with closely spaced comb lines increases the pump-power requirement and makes thermally stable soliton access more difficult. Here, we demonstrate broadband silicon nitride (Si3N4) soliton microcombs enabled by a resonant coupler that enhances pump delivery to the nonlinear resonator. The device generates a spectrum spanning 930-1380 nm with an approximately 98-GHz comb-line spacing at on-chip pump powers of tens of milliwatts, while providing reliable thermally stable access to the single-soliton state. These results provide a route to compact, power-efficient microcomb sources that combine broad spectral coverage with closely spaced comb lines for spectroscopy and imaging in the biological window.

physics.optics↗

arXiv2026

Power-Bandwidth Scaling of Resonantly Coupled Soliton Microcombs

A soliton microcomb requires increasing pump power as its optical bandwidth is broadened. Resonant pumping through an auxiliary microresonator can reduce the power required to sustain a soliton, but simultaneously increases the power required for soliton formation. We show that this competition leads to optimal inter-resonator coupling, and the predicted minimum input pump power scales as the two-thirds power of the comb bandwidth, in contrast to the quadratic scaling under direct pumping. Experiments support the opposing power trends, providing a design rule for power-efficient, ultra-broadband soliton microcombs.

physics.optics↗

arXiv2026

Representation Learning with Quantum Signal Processing

Representation learning begins when training changes the features that define similarity between data. A frozen-kernel model only reweights a fixed geometry. We establish quantum signal processing (QSP) as a solvable quantum model of the representation-learning regime. At arbitrary depth, we compute the exact mean and variance of its quantum neural tangent kernel, revealing an input-dependent angular geometry whose diagonal remains non-self-averaging even when the underlying unitary approaches Haar randomness. We also prove a sparse-data guarantee for the full nonlinear gradient flow without freezing or ensemble-averaging the kernel: the realized dynamics converges to an integrable scalar flow with a time-dependent kernel closure and explicit convergence times. A finite-depth speed limit holds for every data set and trajectory. At higher data density, numerical results show coupled evolution beyond both the scalar and frozen-kernel descriptions. These results give a controlled theory of learned quantum data geometry with provable training dynamics beyond the frozen limit.

quant-ph↗

arXiv2026

Contamination Means Overestimation? A Fine-Grained Empirical Study in Code Intelligence

In recent years, code intelligence has gained increasing importance in the field of automated software engineering. Meanwhile, the widespread adoption of Pretrained Language Models (PLMs) and Large Language Models (LLMs) has raised concerns regarding data contamination and its potential impact on model performance evaluation. Previous studies mainly focused on sample-level contamination, ignoring partial contamination scenarios that are pervasive in code intelligence. This paper fills this gap and presents a systematic empirical study to investigate the fine-grained data contamination on mainstream code tasks. Our study involves diverse representative PLMs: RoBERTa and GPT-2, and LLMs: LLaMA and StarCoder, covering three major tasks: code translation, code generation, and code summarization, across two Programming Languages (PLs): Java and Python. We categorize contamination scenarios into four types according to the code intelligence practice, namely input-only, output-only, unpaired, and paired contamination settings, and construct corresponding experimental and control groups for exploration. Experimental results show that, under the pre-training, fine-tuning, and inference paradigm adopted by PLMs, even deliberately injecting paired contamination does not lead to significant performance overestimation. But direct inference or small-scale fine-tuning uncovers the contamination effects. In contrast, LLMs with pre-training and inference paradigm are significantly affected by the paired contamination. Apart from the above, other contamination scenarios have no impact on both PLMs and LLMs. Our findings challenge the conventional belief that contamination inevitably leads to performance overestimation, providing new insights into the evaluation and deployment of code intelligence models.

cs.SE↗

arXiv2026

When AI meets quantum information: A comprehensive review

Artificial intelligence (AI) and quantum information (QI) are rapidly co-evolving. AI is becoming a practical tool for learning, designing, controlling, and verifying quantum systems, while QI offers new computational models, representational structures, and learning-theoretic questions for AI. This survey reviews the interface from both directions. In the AI for QI direction, we organize recent progress around the central tasks of extracting information from limited measurements, training and discovering quantum algorithms, stabilizing noisy hardware, automating experimental and programming workflows, and extending learning-based methods to sensing and networking. In the QI for AI direction, we examine how quantum computation and quantum-inspired structures affect learning through algorithmic speedups, expressivity, trainability, generalization, neural-network design, and tensor-network representations. We close by identifying cross-cutting challenges in reproducibility, scalability, hardware realism, and co-design, arguing that progress will depend on tighter integration of theory, experiment, and hybrid quantum--classical systems.

quant-ph↗

arXiv2026

Program gain and loss for broadband soliton microcombs

Soliton microcombs provide compact, broadband, coherent light sources for precision metrology, spectroscopy, communications, and microwave photonics. Extending their spectral span while retaining useful output power remains challenging and often requires impractically high pump power. Existing approaches mainly tailor the dispersion and pumping conditions, but they do not exploit the coupling spectrum as a programmable aspect of soliton operation. Here we introduce a meta-coupler whose lithographically programmed coupling spectrum concentrates strong pump access near the pumped resonance while leaving most comb lines close to the intrinsic loss rate. Si$_3$N$_4$ microresonators incorporating a meta-coupler exhibit broader circulating soliton spectra, nearly twofold larger 3 dB soliton bandwidths, up to about 12 dB higher central comb-line power, and up to about fivefold greater emitted comb power, without an additional pump-power penalty. Our work unlocks gain and loss as simultaneous programmable knobs for realizing high-performance soliton microcombs.

physics.optics↗

arXiv2026

BACH-V: Bridging Abstract and Concrete Human-Values in Large Language Models

Do large language models (LLMs) genuinely understand abstract concepts, or merely manipulate them as statistical patterns? We introduce an abstraction-grounding framework that decomposes conceptual understanding into three capacities: interpretation of abstract concepts (Abstract-Abstract, A-A), grounding of abstractions in concrete events (Abstract-Concrete, A-C), and application of abstract principles to regulate concrete decisions (Concrete-Concrete, C-C). Using human values as a testbed - given their semantic richness and centrality to alignment - we employ probing (detecting value traces in internal activations) and steering (modifying representations to shift behavior). Across six open-source LLMs and ten value dimensions, probing shows that diagnostic probes trained solely on abstract value descriptions reliably detect the same values in concrete event narratives and decision reasoning, demonstrating cross-level transfer. Steering reveals an asymmetry: intervening on value representations causally shifts concrete judgments and decisions (A-C, C-C), yet leaves abstract interpretations unchanged (A-A), suggesting that encoded abstract values function as stable anchors rather than malleable activations. These findings indicate LLMs maintain structured value representations that bridge abstraction and action, providing a mechanistic and operational foundation for building value-driven autonomous AI systems with more transparent, generalizable alignment and control.

cs.CL↗

arXiv2026

Simple Role Assignment is Extraordinarily Effective for Safety Alignment

Principle-based alignment often lacks context sensitivity and completeness. Grounded in Theory of Mind, we propose role conditioning as a compact alternative: social roles (e.g., mother, judge) implicitly encode both values and the cognitive schemas required to apply them. We introduce a training-free pipeline featuring a role-conditioned generator and iterative role-based critics for refinement. Across five model families, our approach consistently outperforms principle-based, Chain-of-Thought (CoT) and other baselines across benchmarks. Notably, it reduces unsafe outputs on the WildJailbreak benchmark from 81.4\% to 3.6\% with DeepSeek-V3. Not only for common safety benchmarks, it consistently applies for agentic safety tasks. These results establish role assignment as a powerful, interpretable paradigm for AI alignment and LLM-as-a-Judge construction.

cs.CY↗

arXiv2025

ValuePilot: A Two-Phase Framework for Value-Driven Decision-Making

Personalized decision-making is essential for human-AI interaction, enabling AI agents to act in alignment with individual users' value preferences. As AI systems expand into real-world applications, adapting to personalized values beyond task completion or collective alignment has become a critical challenge. We address this by proposing a value-driven approach to personalized decision-making. Human values serve as stable, transferable signals that support consistent and generalizable behavior across contexts. Compared to task-oriented paradigms driven by external rewards and incentives, value-driven decision-making enhances interpretability and enables agents to act appropriately even in novel scenarios. We introduce ValuePilot, a two-phase framework consisting of a dataset generation toolkit (DGT) and a decision-making module (DMM). DGT constructs diverse, value-annotated scenarios from a human-LLM collaborative pipeline. DMM learns to evaluate actions based on personal value preferences, enabling context-sensitive, individualized decisions. When evaluated on previously unseen scenarios, DMM outperforms strong LLM baselines, including GPT-5, Claude-Sonnet-4, Gemini-2-flash, and Llama-3.1-70b, in aligning with human action choices. Our results demonstrate that value-driven decision-making is an effective and extensible engineering pathway toward building interpretable, personalized AI agents.

cs.AI↗

arXiv2025

Communication-ready high-power soliton microcombs in highly-dispersive Fabry-Perot-microresonators

Microcombs generated in optical microresonators are widely regarded as promising light sources for next-generation communication systems, but the optical power available per comb line has so far fallen short of practical requirements. Here we introduce an integrated Fabry-Pérot microresonator platform that overcomes fundamental dispersion-engineering constraints and enables bright soliton microcombs with unprecedented power per line. The resonator is defined by chirped Bragg gratings that provide exceptionally large anomalous group-velocity dispersion, allowing more than ten comb lines to reach the milliwatt level. These combs can be used directly in coherent communication systems without additional amplification, achieving an aggregate data rate of 2 Tb/s. Once integrated, our high-power soliton microcombs could be instantly ready for communications as well as a broad range of practical comb-based applications.

physics.optics↗

arXiv2025

Integrated soliton microcombs beyond the turnkey limit

Soliton microcombs generated in optical microresonators are accelerating the transition of optical frequency combs from laboratory instruments to industrial platforms. Self injection locking (SIL) enables direct driving of soliton microcombs by integrated lasers, providing turnkey initiation and improved coherence, but it also pins the pump close to resonance, limiting both spectral span and tuning flexibility. Here we theoretically and experimentally demonstrate that introducing a thermally tunable auxiliary microresonator extends the bandwidth of SIL soliton microcombs. By engineering hybridization of the pumped resonance, we achieve deterministic access to single soliton states and then push operation into a far detuned regime inaccessible to direct initiation. The resulting combs reach a near 200 nm span at a 25 GHz repetition rate, while preserving the SIL-enabled noise suppression throughout. Moreover, the added degree of freedom afforded by the coupled resonator architecture enables orthogonal control of the comb's repetition rate and center frequency. These advances expand the spectral reach and controllability of integrated soliton microcombs for information processing and precision metrology.

physics.optics↗

arXiv2025

Raman-induced dynamics of ultrafast microresonator solitons

Soliton microcombs are evolving towards octave-spanning for $f$-$2f$ self-referencing and expanding applications in spectroscopy and timekeeping. As spectra broaden and pulses shorten, the Raman-induced soliton self-frequency shift (SSFS) becomes a principal limitation: it reduces pump-to-comb conversion efficiency, constrains achievable span, and can, in extremes, preclude stationary operation. We develop a complementary theory of SSFS in microresonators that remains valid when the soliton duration $τ_s$ is shorter than the Raman response timescale. The theory predicts a reduced dependence of the SSFS on $τ_s$ which also expands the soliton existence range. Such predictions are validated by numerical simulations and by experiments on Si$_3$N$_4$ microresonators. Our results provide practical guidelines for engineering efficient and broadband soliton microcombs.

physics.optics↗

arXiv2025

Power-efficient ultra-broadband soliton microcombs in resonantly-coupled microresonators

The drive to miniaturize optical frequency combs for practical deployment has spotlighted microresonator solitons as a promising chip-scale candidate. However, these soliton microcombs could be very power-hungry when their span increases, especially with fine comb spacings. As a result, realizing an octave-spanning comb at microwave repetition rates for direct optical-microwave linkage is considered not possible for photonic integration due to the high power requirements. Here, we introduce the concept of resonant-coupling to soliton microcombs to reduce pump consumption significantly. Compared to conventional waveguide-coupled designs, we demonstrate (i) a threefold increase in spectral span for high-power combs and (ii) up to a tenfold reduction in repetition frequency for octave-spanning operation. This configuration is compatible with laser integration and yields reliable, turnkey soliton generation. By eliminating the long-standing pump-power bottleneck, microcombs will soon become readily available for portable optical clocks, massively parallel data links, and field-deployable spectrometers.

physics.optics↗

arXiv2025

Are the Values of LLMs Structurally Aligned with Humans? A Causal Perspective

As large language models (LLMs) become increasingly integrated into critical applications, aligning their behavior with human values presents significant challenges. Current methods, such as Reinforcement Learning from Human Feedback (RLHF), typically focus on a limited set of coarse-grained values and are resource-intensive. Moreover, the correlations between these values remain implicit, leading to unclear explanations for value-steering outcomes. Our work argues that a latent causal value graph underlies the value dimensions of LLMs and that, despite alignment training, this structure remains significantly different from human value systems. We leverage these causal value graphs to guide two lightweight value-steering methods: role-based prompting and sparse autoencoder (SAE) steering, effectively mitigating unexpected side effects. Furthermore, SAE provides a more fine-grained approach to value steering. Experiments on Gemma-2B-IT and Llama3-8B-IT demonstrate the effectiveness and controllability of our methods.

cs.CL↗

arXiv2025

Compact Turnkey Soliton Microcombs at Microwave Rates via Wafer-Scale Fabrication

Soliton microcombs generated in nonlinear microresonators facilitate the photonic integration of timing, frequency synthesis, and astronomical calibration functionalities. For these applications, low-repetition-rate soliton microcombs are essential as they establish a coherent link between optical and microwave signals. However, the required pump power typically scales with the inverse of the repetition rate, and the device footprint scales with the inverse of square of the repetition rate, rendering low-repetition-rate soliton microcombs challenging to integrate within photonic circuits. This study designs and fabricates silicon nitride microresonators on 4-inch wafers with highly compact form factors. The resonator geometries are engineered from ring to finger and spiral shapes to enhance integration density while attaining quality factors over 10^7. Driven directly by an integrated laser, soliton microcombs with repetition rates below 10 GHz are demonstrated via turnkey initiation. The phase noise performance of the synthesized microwave signals reaches -130 dBc/Hz at 100 kHz offset frequency for 10 GHz carrier frequencies. This work enables the high-density integration of soliton microcombs for chip-based microwave photonics and spectroscopy applications.

physics.optics↗

arXiv2024

Cosmological Bell Tests with Decoherence Effects

The inflationary universe creates particle pairs, which are entangled in their momenta due to momentum conservation. Operators involving the momenta of the fluctuations can be rewritten into pseudo-spin operators, such as the Gour-Khanna-Mann-Revzen (GKMR) pseudo-spin. Making use of these pseudo-spin operators, cosmological Bell inequalities can be formulated. The violation of these Bell inequalities indicates the quantum nature of primordial fluctuations. In this work, we focus on primordial curvature perturbations. Since curvature perturbations arise from gravity, their action includes the Gibbons-Hawking-York boundary term. We clarify the role of the boundary term in selecting suitable initial conditions for linear perturbations. After that, we proceed to the interactions of cosmological perturbations, including the bulk and boundary interaction terms, which introduce decoherence effects. These decoherence effects change the expectation value of the Bell operator, and gradually restore the Bell inequality. We describe this process by a ``Bell test curve'', which offers a window around 5 e-folds for testing the quantum origin of cosmological perturbations. We also explore the possibility of extracting the information of the decoherence rate and the structure of primordial interactions from the Bell test curve.

hep-th↗

arXiv2024

Score Neural Operator: A Generative Model for Learning and Generalizing Across Multiple Probability Distributions

Most existing generative models are limited to learning a single probability distribution from the training data and cannot generalize to novel distributions for unseen data. An architecture that can generate samples from both trained datasets and unseen probability distributions would mark a significant breakthrough. Recently, score-based generative models have gained considerable attention for their comprehensive mode coverage and high-quality image synthesis, as they effectively learn an operator that maps a probability distribution to its corresponding score function. In this work, we introduce the $\emph{Score Neural Operator}$, which learns the mapping from multiple probability distributions to their score functions within a unified framework. We employ latent space techniques to facilitate the training of score matching, which tends to over-fit in the original image pixel space, thereby enhancing sample generation quality. Our trained Score Neural Operator demonstrates the ability to predict score functions of probability measures beyond the training space and exhibits strong generalization performance in both 2-dimensional Gaussian Mixture Models and 1024-dimensional MNIST double-digit datasets. Importantly, our approach offers significant potential for few-shot learning applications, where a single image from a new distribution can be leveraged to generate multiple distinct images from that distribution.

cs.LG↗