arXiv subjects

Jun Wang

Publications and source records attributed to Jun Wang.

At least 199 records · Page 11Linked to original sources

arXiv2026

Why the Brain Consolidates: Predictive Forgetting for Optimal Generalisation

Standard accounts of memory consolidation emphasise the stabilisation of stored representations, but struggle to explain representational drift, semanticisation, or the necessity of offline replay. Here we propose that high-capacity neocortical networks optimise stored representations for generalisation by reducing complexity via predictive forgetting, i.e. the selective retention of experienced information that predicts future outcomes or experience. We show that predictive forgetting formally improves information-theoretic generalisation bounds on stored representations. Under high-fidelity encoding constraints, such compression is generally unattainable in a single pass; high-capacity networks therefore benefit from temporally separated, iterative refinement of stored traces without re-accessing sensory input. We demonstrate this capacity dependence with simulations in autoencoder-based neocortical models, biologically plausible predictive coding circuits, and Transformer-based language models, and derive quantitative predictions for consolidation-dependent changes in neural representational geometry. These results identify a computational role for off-line consolidation beyond stabilisation, showing that outcome-conditioned compression optimises the retention-generalisation trade-off.

q-bio.NC↗

arXiv2026

TSEmbed: Unlocking Task Scaling in Universal Multimodal Embeddings

Despite the exceptional reasoning capabilities of Multimodal Large Language Models (MLLMs), their adaptation into universal embedding models is significantly impeded by task conflict. To address this, we propose TSEmbed, a universal multimodal embedding framework that synergizes Mixture-of-Experts (MoE) with Low-Rank Adaptation (LoRA) to explicitly disentangle conflicting task objectives. Moreover, we introduce Expert-Aware Negative Sampling (EANS), a novel strategy that leverages expert routing distributions as an intrinsic proxy for semantic similarity. By dynamically prioritizing informative hard negatives that share expert activation patterns with the query, EANS effectively sharpens the model's discriminative power and refines embedding boundaries. To ensure training stability, we further devise a two-stage learning paradigm that solidifies expert specialization before optimizing representations via EANS. TSEmbed achieves state-of-the-art performance on both the Massive Multimodal Embedding Benchmark (MMEB) and real-world industrial production datasets, laying a foundation for task-level scaling in universal multimodal embeddings.

cs.CL↗

arXiv2026

Sim2Sea: Sim-to-Real Policy Transfer for Maritime Vessel Navigation in Congested Waters

Autonomous navigation in congested maritime environments is a critical capability for a wide range of real-world applications. However, it remains an unresolved challenge due to complex vessel interactions and significant environmental uncertainties. Existing methods often fail in practical deployment due to a substantial sim-to-real gap, which stems from imprecise simulation, inadequate situational awareness, and unsafe exploration strategies. To address these, we propose \textbf{Sim2Sea}, a comprehensive framework designed to bridge simulation and real-world execution. Sim2Sea advances in three key aspects. First, we develop a GPU-accelerated parallel simulator for scalable and accurate maritime scenario simulation. Second, we design a dual-stream spatiotemporal policy that handles complex dynamics and multi-modal perception, augmented with a velocity-obstacle-guided action masking mechanism to ensure safe and efficient exploration. Finally, a targeted domain randomization scheme helps bridge the sim-to-real gap. Simulation results demonstrate that our method achieves faster convergence and safer trajectories than established baselines. In addition, our policy trained purely in simulation successfully transfers zero-shot to a 17-ton unmanned vessel operating in real-world congested waters. These results validate the effectiveness of Sim2Sea in achieving reliable sim-to-real transfer for practical autonomous maritime navigation.

cs.RO↗

arXiv2026

Swimming Under Constraints: A Safe Reinforcement Learning Framework for Quadrupedal Bio-Inspired Propulsion

Bio-inspired aquatic propulsion offers high thrust and maneuverability but is prone to destabilizing forces such as lift fluctuations, which are further amplified by six-degree-of-freedom (6-DoF) fluid coupling. We formulate quadrupedal swimming as a constrained optimization problem that maximizes forward thrust while minimizing destabilizing fluctuations. Our proposed framework, Accelerated Constrained Proximal Policy Optimization with a PID-regulated Lagrange multiplier (ACPPO-PID), enforces constraints with a PID-regulated Lagrange multiplier, accelerates learning via conditional asymmetric clipping, and stabilizes updates through cycle-wise geometric aggregation. Initialized with imitation learning and refined through on-hardware towing-tank experiments, ACPPO-PID produces control policies that transfer effectively to quadrupedal free-swimming trials. Results demonstrate improved thrust efficiency, reduced destabilizing forces, and faster convergence compared with state-of-the-art baselines, underscoring the importance of constraint-aware safe RL for robust and generalizable bio-inspired locomotion in complex fluid environments.

cs.RO↗

arXiv2026

Can a Building Work as a Reservoir: Footstep Localization with Embedded Accelerometer Networks

Using floor vibrations to accurately predict occupants' footstep locations is essential for smart building operation and privacy-preserving indoor sensing. However, existing approaches are dominated by either physics-based models that rely on simplified wave propagation assumptions and careful calibration, or data-driven methods that require large labeled datasets and often lack robustness to subject and environmental variability. This work introduces a new approach by treating an instrumented building floor as a physical reservoir computer, whose intrinsic structural dynamics can perform nonlinear spatio-temporal computation and information extraction directly. Specifically, foot strike-induced floor vibrations recorded by a distributed accelerometer network are processed using a lightweight physical reservoir computing (PRC) pipeline consisting of short waveform extraction, root-mean-square (RMS) normalization, principal component analysis (PCA), and a weighted linear readout. Results of this study, involving 2 participants and 12 accelerometers, showed that RMS normalization and PCA projection successfully extracted occupant-invariant features from floor-vibration waveform data, enabling a single linear readout to predict foot-strike location across repeated traversals and participants. Sub-meter accuracy is achieved along the hallway direction with moderate sensing coverage, while cross-participant tests achieved meter-scale accuracy without subject-specific recalibration or retraining. These findings demonstrate that building-scale structures can function as capable physical reservoir computers for intelligent monitoring.

cs.CE↗

arXiv2026

MERIT: Memory-Enhanced Retrieval for Interpretable Knowledge Tracing

Knowledge Tracing (KT) models students' evolving knowledge states to predict future performance, serving as a foundation for personalized education. While traditional deep learning models achieve high accuracy, they often lack interpretability. Large Language Models (LLMs) offer strong reasoning capabilities but struggle with limited context windows and hallucinations. Furthermore, existing LLM-based methods typically require expensive fine-tuning, limiting scalability and adaptability to new data. We propose MERIT (Memory-Enhanced Retrieval for Interpretable Knowledge Tracing), a training-free framework combining frozen LLM reasoning with structured pedagogical memory. Rather than updating parameters, MERIT transforms raw interaction logs into an interpretable memory bank. The framework uses semantic denoising to categorize students into latent cognitive schemas and constructs a paradigm bank where representative error patterns are analyzed offline to generate explicit Chain-of-Thought (CoT) rationales. During inference, a hierarchical routing mechanism retrieves relevant contexts, while a logic-augmented module applies semantic constraints to calibrate predictions. By grounding the LLM in interpretable memory, MERIT achieves state-of-the-art performance on real-world datasets without gradient updates. This approach reduces computational costs and supports dynamic knowledge updates, improving the accessibility and transparency of educational diagnosis.

cs.CL↗

arXiv2026

When Agents "Misremember" Collectively: Exploring the Mandela Effect in LLM-based Multi-Agent Systems

Recent advancements in large language models (LLMs) have significantly enhanced the capabilities of collaborative multi-agent systems, enabling them to address complex challenges. However, within these multi-agent systems, the susceptibility of agents to collective cognitive biases remains an underexplored issue. A compelling example is the Mandela effect, a phenomenon where groups collectively misremember past events as a result of false details reinforced through social influence and internalized misinformation. This vulnerability limits our understanding of memory bias in multi-agent systems and raises ethical concerns about the potential spread of misinformation. In this paper, we conduct a comprehensive study on the Mandela effect in LLM-based multi-agent systems, focusing on its existence, causing factors, and mitigation strategies. We propose MANBENCH, a novel benchmark designed to evaluate agent behaviors across four common task types that are susceptible to the Mandela effect, using five interaction protocols that vary in agent roles and memory timescales. We evaluate agents powered by several LLMs on MANBENCH to quantify the Mandela effect and analyze how different factors affect it. Moreover, we propose strategies to mitigate this effect, including prompt-level defenses (e.g., cognitive anchoring and source scrutiny) and model-level alignment-based defense, achieving an average 74.40% reduction in the Mandela effect compared to the baseline. Our findings provide valuable insights for developing more resilient and ethically aligned collaborative multi-agent systems. Code and dataset are available at https://github.com/bluedream02/Mandela-Effect.

cs.CL↗

arXiv2026

Stop Treating Collisions Equally: Qualification-Aware Semantic ID Learning for Recommendation at Industrial Scale

Semantic IDs (SIDs) are compact discrete representations derived from multimodal item features, serving as a unified abstraction for ID-based and generative recommendation. However, learning high-quality SIDs remains challenging due to two issues. (1) Collision problem: the quantized token space is prone to collisions, in which semantically distinct items are assigned identical or overly similar SID compositions, resulting in semantic entanglement. (2) Collision-signal heterogeneity: collisions are not uniformly harmful. Some reflect genuine conflicts between semantically unrelated items, while others stem from benign redundancy or systematic data effects. To address these challenges, we propose Qualification-Aware Semantic ID Learning (QuaSID), an end-to-end framework that learns collision-qualified SIDs by selectively repelling qualified conflict pairs and scaling the repulsion strength by collision severity. QuaSID consists of two mechanisms: Hamming-guided Margin Repulsion, which translates low-Hamming SID overlaps into explicit, severity-scaled geometric constraints on the encoder space; and Conflict-Aware Valid Pair Masking, which masks protocol-induced benign overlaps to denoise repulsion supervision. In addition, QuaSID incorporates a dual-tower contrastive objective to inject collaborative signals into tokenization. Experiments on public benchmarks and industrial data validate QuaSID. On public datasets, QuaSID consistently outperforms strong baselines, improving top-K ranking quality by 5.9% over the best baseline while increasing SID composition diversity. In an online A/B test on Kuaishou e-commerce with a 5% traffic split, QuaSID increases ranking GMV-S2 by 2.38% and improves completed orders on cold-start retrieval by up to 6.42%. Finally, we show that the proposed repulsion loss is plug-and-play and enhances a range of SID learning frameworks across datasets.

cs.IR↗

arXiv2026

Polarization Engineering of Second-Harmonic Generation in 3R-MoS$_2$ Waveguides

Chip-scale nonlinear optics enables strong light-matter interactions within compact devices, serving as a fundamental platform for multifunctional integrated photonics from classical optical signal processing to quantum information technologies. Transition metal dichalcogenide (TMDC) waveguides have recently emerged as a highly promising platform owing to their giant material nonlinearity and extended interaction lengths. To date, however, research has predominantly focused on conversion efficiency, leaving the mechanisms governing the polarization state of nonlinear signal largely unexplored. Here, we establish a comprehensive framework for engineering the polarization of second-harmonic generation (SHG) in 3R-MoS$_2$ waveguides. By synergizing polarization-resolved measurements with theoretical modeling, we reveal that the SHG polarization is determined by guided-mode interactions constrained by waveguide geometry and crystal symmetry, and further reshaped during propagation. We demonstrate that thickness-dependent guided-mode confinement and in-plane crystal symmetry provide robust, static control over SHG polarization, while propagation length offers a dynamic tuning knob for continuously tailoring the nonlinear output. Our findings provide a deterministic approach for on-chip polarization engineering, opening opportunities for reconfigurable nonlinear light sources and quantum photonic circuits.

physics.optics↗

arXiv2026

Coarse-to-Fine Learning of Dynamic Causal Structures

Learning the dynamic causal structure of time series is a challenging problem. Most existing approaches rely on distributional or structural invariance to uncover underlying causal dynamics, assuming stationary or partially stationary causality. However, these assumptions often conflict with the complex, time-varying causal relationships observed in real-world systems. This motivates the need for methods that address fully dynamic causality, where both instantaneous and lagged dependencies evolve over time. Such a setting poses significant challenges for the efficiency and stability of causal discovery. To address these challenges, we introduce DyCausal, a dynamic causal structure learning framework. DyCausal leverages convolutional networks to capture causal patterns within coarse-grained time windows, and then applies linear interpolation to refine causal structures at each time step, thereby recovering fine-grained and time-varying causal graphs. In addition, we propose an acyclic constraint based on matrix norm scaling, which improves efficiency while effectively constraining loops in evolving causal structures. Comprehensive evaluations on both synthetic and real-world datasets demonstrate that DyCausal achieves superior performance compared to existing methods, offering a stable and efficient approach for identifying fully dynamic causal structures from coarse to fine.

cs.LG↗

arXiv2026

MViR: Multi-View Visual-Semantic Representation for Fake News Detection

With the rise of online social networks, detecting fake news accurately is essential for a healthy online environment. While existing methods have advanced multimodal fake news detection, they often neglect the multi-view visual-semantic aspects of news, such as different text perspectives of the same image. To address this, we propose a Multi-View Visual-Semantic Representation (MViR) framework. Our approach includes a Multi-View Representation module using pyramid dilated convolution to capture multi-view visual-semantic features, a Multi-View Feature Fusion module to integrate these features with text, and multiple aggregators to extract multi-view semantic cues for detection. Experiments on benchmark datasets demonstrate the superiority of MViR. The source code of FedCoop is available at https://github.com/FlowerinZDF/FakeNews-MVIR.

cs.MM↗

arXiv2026

A Benchmark for Deep Information Synthesis

Large language model (LLM)-based agents are increasingly used to solve complex tasks involving tool use, such as web browsing, code execution, and data analysis. However, current evaluation benchmarks do not adequately assess their ability to solve real-world tasks that require synthesizing information from multiple sources and inferring insights beyond simple fact retrieval. To address this, we introduce DEEPSYNTH, a novel benchmark designed to evaluate agents on realistic, time-consuming problems that combine information gathering, synthesis, and structured reasoning to produce insights. DEEPSYNTH contains 120 tasks collected across 7 domains and data sources covering 67 countries. DEEPSYNTH is constructed using a multi-stage data collection pipeline that requires annotators to collect official data sources, create hypotheses, perform manual analysis, and design tasks with verifiable answers. When evaluated on DEEPSYNTH, 11 state-of-the-art LLMs and deep research agents achieve a maximum F1 score of 8.97 and 17.5 on the LLM-judge metric, underscoring the difficulty of the benchmark. Our analysis reveals that current agents struggle with hallucinations and reasoning over large information spaces, highlighting DEEPSYNTH as a crucial benchmark for guiding future research.

cs.AI↗

arXiv2026

ConformalNL2LTL: Translating Natural Language Instructions into Temporal Logic Formulas with Conformal Correctness Guarantees

Linear Temporal Logic (LTL) is a widely used task specification language for autonomous systems. To mitigate the significant manual effort and expertise required to define LTL-encoded tasks, several methods have been proposed for translating Natural Language (NL) instructions into LTL formulas, which, however, lack correctness guarantees. To address this, we propose a new NL-to-LTL translation method, called ConformalNL2LTL that achieves user-defined translation success rates on unseen NL commands. Our method constructs LTL formulas iteratively by solving a sequence of open-vocabulary question-answering (QA) problems using large language models (LLMs). These QA tasks are handled collaboratively by a primary and an auxiliary model. The primary model answers each QA instance while quantifying uncertainty via conformal prediction; when it is insufficiently certain according to user-defined confidence thresholds, it requests assistance from the auxiliary model and, if necessary, from the user. We demonstrate theoretically and empirically that ConformalNL2LTL achieves the desired translation accuracy while minimizing user intervention.

cs.CL↗

arXiv2026

Design and characterization of a photosensor system for the RELICS experiment

In this paper, we present the design and characterization of a photosensor system developed for the RELICS experiment. An extended dynamic range base was designed to mitigate photomultiplier tube (PMT) saturation caused by intense cosmic muon backgrounds in the surface-level RELICS detector. The system employs dual readout from the anode and the seventh dynode to extend the linear response range of the PMT. In particular, our characterization and measurements of Hamamatsu R8520-406 PMTs confirm stable operation under positive high-voltage bias, extending the linear response range by more than an order of magnitude. Furthermore, a model of PMT saturation and recovery was developed to evaluate the influence of cosmic muon signals in the RELICS detector. The results demonstrate the system capability to detect coherent elastic neutrino-nucleus scattering signals under surface-level cosmic backgrounds, and suggest the potential to extend the scientific reach of RELICS to MeV-scale interactions.

physics.ins-det↗

arXiv2026

LMSeg: Unleashing the Power of Large-Scale Models for Open-Vocabulary Semantic Segmentation

It is widely agreed that open-vocabulary-based approaches outperform classical closed-set training solutions for recognizing unseen objects in images for semantic segmentation. Existing open-vocabulary approaches leverage vision-language models, such as CLIP, to align visual features with rich semantic features acquired through pre-training on large-scale vision-language datasets. However, the text prompts employed in these methods are short phrases based on fixed templates, failing to capture comprehensive object attributes. Moreover, while the CLIP model excels at exploiting image-level features, it is less effective at pixel-level representation, which is crucial for semantic segmentation tasks. In this work, we propose to alleviate the above-mentioned issues by leveraging multiple large-scale models to enhance the alignment between fine-grained visual features and enriched linguistic features. Specifically, our method employs large language models (LLMs) to generate enriched language prompts with diverse visual attributes for each category, including color, shape/size, and texture/material. Additionally, for enhanced visual feature extraction, the SAM model is adopted as a supplement to the CLIP visual encoder through a proposed learnable weighted fusion strategy. Built upon these techniques, our method, termed LMSeg, achieves state-of-the-art performance across all major open-vocabulary segmentation benchmarks. The code will be made available soon.

cs.CV↗

arXiv2026

Plan-MCTS: Plan Exploration for Action Exploitation in Web Navigation

Large Language Models (LLMs) have empowered autonomous agents to handle complex web navigation tasks. While recent studies integrate tree search to enhance long-horizon reasoning, applying these algorithms in web navigation faces two critical challenges: sparse valid paths that lead to inefficient exploration, and a noisy context that dilutes accurate state perception. To address this, we introduce Plan-MCTS, a framework that reformulates web navigation by shifting exploration to a semantic Plan Space. By decoupling strategic planning from execution grounding, it transforms sparse action space into a Dense Plan Tree for efficient exploration, and distills noisy contexts into an Abstracted Semantic History for precise state awareness. To ensure efficiency and robustness, Plan-MCTS incorporates a Dual-Gating Reward to strictly validate both physical executability and strategic alignment and Structural Refinement for on-policy repair of failed subplans. Extensive experiments on WebArena demonstrate that Plan-MCTS achieves state-of-the-art performance, surpassing current approaches with higher task effectiveness and search efficiency.

cs.AI↗

arXiv2026

Existence and nonexistence of solutions for fractional elliptic equations arising from closed MEMS model

The objective of our paper is to investigate fractional elliptic equations of the form $(-Δ)^s u=\frac{λ}{(a-u)^2}$ within a bounded domain $Ω$, subject to zero Dirichlet boundary conditions. Here, $s\in(0,1)$, $λ>0$, and the function $a$ vanishes at the boundary while satisfying additional conditions. This problem originates from Micro-Electromechanical Systems (MEMS) devices, particularly when the elastic membrane makes contact with the ground plate at the boundary. We establish both existence and nonexistence results, illustrating how the boundary decay of the membrane influences the solutions and pull-in voltage.

math.AP↗

arXiv2026

Adaptive Milestone Reward for GUI Agents

Reinforcement Learning (RL) has emerged as a mainstream paradigm for training Mobile GUI Agents, yet it struggles with the temporal credit assignment problem inherent in long-horizon tasks. A primary challenge lies in the trade-off between reward fidelity and density: outcome reward offers high fidelity but suffers from signal sparsity, while process reward provides dense supervision but remains prone to bias and reward hacking. To resolve this conflict, we propose the Adaptive Milestone Reward (ADMIRE) mechanism. ADMIRE constructs a verifiable, adaptive reward system by anchoring trajectory to milestones, which are dynamically distilled from successful explorations. Crucially, ADMIRE integrates an asymmetric credit assignment strategy that denoises successful trajectories and scaffolds failed trajectories. Extensive experiments demonstrate that ADMIRE consistently yields over 10% absolute improvement in success rate across different base models on AndroidWorld. Moreover, the method exhibits robust generalizability, achieving strong performance across diverse RL algorithms and heterogeneous environments such as web navigation and embodied tasks.

cs.LG↗