arXiv subjects

Zefeng Wang

Publications and source records attributed to Zefeng Wang.

7 recordsLinked to original sources

arXiv2026

MetaSkill-Evolve: Recursive Self-Improvement of LLM Agents via Two-Timescale Meta-Skill Evolution

Recent LLM agents tackle increasingly long-horizon, open-ended tasks, and external skills, reusable procedural knowledge supplied to the agent, further extend this capability. However, a fixed, hand-authored skill is rarely optimal, and cannot adapt to the diversity of tasks an agent encounters. Self-improving agents address this by rewriting their own skill files from execution traces, yielding meaningful gains on challenging benchmarks. Yet such self-evolution remains non-recursive: it improves only the task skill (what the agent does) while the improvement procedure (how it improves) is authored once and held fixed. We introduce MetaSkill-Evolve, a two-timescale framework that makes agentic skill improvement recursive: every branch carries both a task skill $s$ and a branch-local meta-skill $m=(\psi,\sigma,\alpha,\pi,\varepsilon)$ whose five components parameterise the Analyzer, Retriever, Allocator, Proposer, and Evolver agents of the improvement pipeline. Task skills evolve on a fast loop while the meta-skill evolves on a slower one under the same pipeline applied to itself, with no additional model or objective. With all five pipeline agents sharing a single frozen backbone, MetaSkill-Evolve outperforms no-skill, static-skill, and single-level evolution baselines on three agentic benchmarks (OfficeQA, SealQA, ALFWorld), improving held-out test accuracy over the raw backbone by +23.54, +16.09, and +1.92 points respectively.

cs.AI↗

arXiv2026

DeepLook: Deeper Thinking with Lookahead

Inference-time scaling has emerged as a powerful paradigm for improving large language model reasoning, often delivering larger gains on difficult reasoning tasks than parameter scaling alone. However, existing approaches remain inefficient in how compute is allocated within a reasoning trace. Motivated by the observation that reasoning failures often exhibit an early onset of uncertainty before a wrong answer become explicit, we introduce DeepLook, a training-free monitor-and-intervene decoding framework that concentrates lookahead compute at uncertainty bottlenecks. DeepLook aggregates token-level confidence into segment-level signals, triggers when confidence drops relative to recent history, and explores candidate continuations with fixed-horizon lookahead. Branches are ranked by Average Lookahead Confidence (ALC), the average segment-level confidence over rollout continuations, then pruned and aggregated through voting. On four competition-style mathematics benchmarks across DeepSeek-R1-8B, Qwen3-32B, GPT-OSS-20B, and GPT-OSS-120B, DeepLook shifts the accuracy--token-cost Pareto frontier: it improves accuracy over DeepConf-low in 11 of 16 settings while reducing dataset-level token generation by 87.3% on average, including gains of +3.1 on AIME25 with Qwen3-32B and +8.8 on BRUMO25 with GPT-OSS-20B. These results show that selective, future-aware intervention yields substantially stronger accuracy--cost trade-offs than uniformly scaling complete reasoning trajectories. Code is available here.

cs.AI↗

arXiv2026

Meta-learning-enhanced implicit full waveform inversion

Implicit full waveform inversion (IFWI) introduces implicit neural representations to parameterize the subsurface velocity model as a continuous function of spatial coordinates, which alleviates the dependence on the initial model and improves inversion flexibility. However, IFWI still requires a large number of iterative updates for each new exploration area, leading to slow convergence, high computational cost, and a lack of mechanisms to share prior knowledge across different geological settings, thereby limiting its efficiency and generalization capability. To further accelerate convergence and enhance cross-area generalization, we propose a meta-learning-based implicit full waveform inversion method, referred to as Meta-learning-enhanced implicit full waveform inversion (Meta-IFWI). In this framework, the subsurface velocity model is represented using an implicit neural network with periodic activation functions (SIREN), while a meta-learning strategy is employed to pretrain a single network on multiple velocity inversion tasks. Through this process, the network learns shared inversion priors and rapid adaptation strategies across different geological scenarios. For a new inversion task, the pretrained Meta-IFWI model can be efficiently adapted to the observed seismic data with only a few gradient updates, significantly reducing the number of iterations required for inversion. Numerical experiments conducted on in-distribution models, including layered synthetic models and the Overthrust model, as well as out-of-distribution complex models such as Marmousi 2, demonstrate that, compared with conventional IFWI, the proposed Meta-IFWI achieves improved inversion accuracy while substantially accelerating convergence and reducing computational cost. Moreover, Meta-IFWI exhibits enhanced robustness and stronger cross-area generalization capability.

physics.geo-ph↗

arXiv2026

Two-Step Tapering-Collapse Method Enables Element-Interdiffused Cladding for Enhanced Laser Amplification in Yb:YAG Single Crystal Fibers

The development of high-power single crystal fiber (SCF) lasers is critically hindered by the lack of a reliable cladding scheme to confine the optical mode and ensure beam quality. Here, we propose and demonstrate a two-step tapering-collapse method for the first time to fabricate a high-quality cladding on Yb:YAG SCFs based on elemental interdiffusion. This in-situ formed crystalline transition layer with a graded refractive index effectively suppresses lattice mismatch and abruptly mitigates core-cladding interfacial stress. Consequently, the numerical aperture of the SCF is significantly reduced from 0.280 to 0.199. In a master oscillator power amplifier configuration, the clad SCF delivers a remarkable 46.7% enhancement in slope efficiency compared to its bare counterpart, accompanied by a substantially improved near-field beam profile. This work establishes a facile and effective route to high-performance clad SCFs, unlocking their full potential for next-generation extreme-condition lasers.

physics.optics↗

arXiv2025

All-fiber highly efficient delivery of 2 kW laser over 2.45 km hollow-core fiber

Anti-resonant hollow-core fibers (AR-HCFs) have emerged as an important medium for high-power laser delivery due to their low optical nonlinearity and high damage threshold. However, current delivery systems of high-power laser based on AR-HCFs mainly rely on free-space optical components, which limits long-term stability in dynamic environments. Here, we report an all-fiber delivery of 2 kW laser with 85.3% transmission efficiency over 2.45 km, using a self-fabricated AR-HCF with a record low transmission loss of 0.175 dB/km at 1080 nm. This represents a nearly 500-fold improvement in the power-distance product compared to reported all-fiber AR-HCF-based laser transmission systems, achieving a record transmission distance for high-power laser delivery. Notably, we observed the phenomenon of stimulated Raman scattering amplified within the silica nested tubes in AR-HCF for the first time. By effectively suppressing the Raman noise from the laser source, we achieve an all-fiber laser delivery without stimulated Raman scattering of silica glass. This work marks a significant breakthrough in multi-kilometer and multi-kilowatt power delivery that is potentially useful for industrial manufacturing, nuclear decommissioning, laser drilling of oil, particle acceleration and so on.

physics.optics↗

arXiv2025

High-brightness multimode fiber laser amplifier

Fiber lasers are widely used in various fields owing to their high efficiency, flexible transmission and excellent beam quality. In applications such as industrial manufacturing and defense systems, a higher output power is always desired. Nevertheless, the power scaling in fiber lasers is limited by nonlinear effects and transverse mode instability in conventional high-power fiber laser systems, where the laser is amplified within the fundamental fiber mode. A promising strategy to overcome these limitations is to utilize multimode fibers, which exhibit higher thresholds for both nonlinear effects and transverse mode instability, combined with wavefront shaping techniques to convert the output speckle pattern into a single concentrated spot. In this study, a high-power multimode fiber laser amplifier based on wavefront shaping is constructed and investigated, achieving a focused beam profile with a 168 W output power. The effects of objective function and the linewidth of seed laser on the system performance are also studied. Additionally, an all-fiber version of high-brightness multimode fiber laser amplifier is proposed. This work opens up new avenues for leveraging multimode fibers to achieve higher brightness in fiber lasers and may inspire other research based on wavefront shaping.

physics.optics↗

arXiv2024

Stop Reasoning! When Multimodal LLM with Chain-of-Thought Reasoning Meets Adversarial Image

Multimodal LLMs (MLLMs) with a great ability of text and image understanding have received great attention. To achieve better reasoning with MLLMs, Chain-of-Thought (CoT) reasoning has been widely explored, which further promotes MLLMs' explainability by giving intermediate reasoning steps. Despite the strong power demonstrated by MLLMs in multimodal reasoning, recent studies show that MLLMs still suffer from adversarial images. This raises the following open questions: Does CoT also enhance the adversarial robustness of MLLMs? What do the intermediate reasoning steps of CoT entail under adversarial attacks? To answer these questions, we first generalize existing attacks to CoT-based inferences by attacking the two main components, i.e., rationale and answer. We find that CoT indeed improves MLLMs' adversarial robustness against the existing attack methods by leveraging the multi-step reasoning process, but not substantially. Based on our findings, we further propose a novel attack method, termed as stop-reasoning attack, that attacks the model while bypassing the CoT reasoning process. Experiments on three MLLMs and two visual reasoning datasets verify the effectiveness of our proposed method. We show that stop-reasoning attack can result in misled predictions and outperform baseline attacks by a significant margin.

cs.CV↗