Searcharxiv⌕ Search

arXiv subjects

Zhuo Liu

Publications and source records attributed to Zhuo Liu.

At least 19 recordsLinked to original sources
arXiv2026

Photonic magnetism and altermagnetism without magnetic materials

Magnetism, a fundamental property of solids arising from the collective alignment of electronic spins of magnetic atoms, has recently been expanded by the discovery of altermagnetism, a symmetry-driven phase distinct from both ferromagnetism and antiferromagnetism. Initial efforts to emulate altermagnetism in photonic systems are also being pursued, primarily through the magneto-optic response of materials; however, their intrinsic frequency limitations pose challenges for extending photonic altermagnetism into the optical regime, thereby restricting their material choices and optical applications. Here, we establish a microscopic framework of photonic magnetism through group theory symmetry analysis, defining photonic counterparts of electron spin and magnetic atoms. This approach enables the realization of photonic ferromagnetic, antiferromagnetic, and altermagnetic phases within structured optical lattices constructed by purely dielectric, nonmagnetic materials, which not only broadens the photonic magnetism concept, but also lifts the material and frequency limitations. Analytical models, numerical simulations, and experimental measurements reveal spin-momentum locking directly corresponding to their electronic analogs, showing d-, g-, and i-wave circular dichroism (CD) splitting bands at telecom wavelengths (around 1550 nm). These results demonstrate that altermagnetic order, and magnetism more broadly, can be reconstructed in photonic systems through symmetry and mode coupling alone, opening a new route towards spin-dependent light control and bosonic analogs of quantum magnetic phenomena in dielectric photonic crystals.

physics.optics↗
arXiv2026

RISE-RL: Rubric-Informed Selective Exploration for Open-Ended Reinforcement Learning

Aligning Large Language Models (LLMs) for open-ended tasks is challenging because responses must satisfy multidimensional criteria without following a single correct generation trajectory. Existing rubric-based reinforcement learning (RL) methods compress fine-grained criterion-level feedback into scalar rewards, making persistent capability gaps difficult to target under limited on-policy exploration. We propose $\textbf{RISE-RL}$ (Rubric-Informed Selective Exploration), which uses repeatedly missed rubric criteria to elicit privileged trajectories that are difficult to discover through unguided exploration alone. RISE-RL retains only trajectories whose complete-rubric reward exceeds the mean reward of natural rollouts, and then re-evaluates them under the original prompt to emphasize behaviors that remain weakly supported by the natural policy. The resulting guidance signal is optimized through a separate auxiliary objective and removed once its additional benefit diminishes. Experiments with 4B and 14B models across writing, chat, health, and science show that RISE-RL achieves the highest mean score on every evaluated benchmark under guidance-free evaluation. Compared with standard Rubric-RL, it improves the average score by 1.3 points at the 4B scale and $\textbf{3.3 points at the 14B scale}$, including a $\textbf{6.0-point}$ gain on CreativeWriting-V3. It also improves creative-writing diversity and yields gains on objectively scored medical and scientific benchmarks. These results indicate that selective internalization through reward filtering and policy support shaping is effective for open-ended reinforcement learning.

cs.AI↗
arXiv2026

Fast Numbers, Slow Language: Bridging Quantitative and Qualitative Earnings Signals

Earnings announcements release two types of information sequentially: quantitative surprise (numeric earnings-per-share (EPS)/revenue versus analyst estimate) arrives first in press releases and financial news, processed by algorithmic traders within minutes; qualitative language (management tone, guidance, question-and-answer (Q&A) credibility) arrives 30-90 min later in the earnings conference call transcript (ECT), requiring human interpretation overnight. Financial economists have studied quantitative surprise for 50 years; natural language processing (NLP) researchers have studied qualitative ECT signals for a decade. Despite studying the same event, the two communities used incompatible frameworks: different targets (return vs. volatility), trading setups (long top-decile and short bottom-decile vs. trade-all), and metrics (return spread between top and bottom 20% (Q5-Q1) vs. mean squared error (MSE)), making direct comparison and connection challenging. We bridge these communities with EarningsInOne, the first corpus aligning earnings news, ECTs, and intraday and next-day prices across SP 1500 (broad U.S. equity universe, 2022-2025). Applying unified trading and evaluation tools to both signal types, we confirm a clean speed separation, fast numbers, slow language: quantitative surprise peaks at announcement and is largely eliminated by the next market open; qualitative ECT sentiment peaks on the next trading day, real and tradeable, but hidden under prior transcript-based evaluation that optimised sign-agnostic volatility with pointwise MSE.

cs.CL↗
arXiv2026

Scaling the Horizon, Not the Parameters: Reaching Trillion-Parameter Performance with a 35B Agent

We introduce Agents-A1, a 35B Mixture-of-Experts Agentic Model that reaches trillion-parameter-level performance by scaling the agent horizon. We investigate agent-horizon scaling from two perspectives: scaling long-horizon trajectories and scaling heterogeneous agent abilities. To support this goal, we build a long-horizon knowledge-action infrastructure that connects external knowledge, actions, observations, and verifier outcomes, producing agentic trajectories with an average length of 45K tokens. Based on this, we train Agents-A1 with a three-stage recipe. First, we perform full-domain supervised fine-tuning to align the base model with broad agentic behaviors. Second, we train domain-level teacher models to capture specialized expertise in each domain. Third, we propose a multi-teacher domain-routed on-policy distillation with salient vocabulary alignment to improve knowledge transfer efficiency across different domains, unifying six heterogeneous domains into one deployable student model. Agents-A1 achieves strong and broad performance for long-horizon agent benchmarks. Compared with 1T-parameter model such as Kimi-K2.6 and DeepSeek-V4-pro, Agents-A1 achieves leading results on SEAL-0 (56.4), IFBench (80.6), HiPhO (46.4), FrontierScience-Olympiad (79.0), and MolBench-Bind (56.8), and remains highly competitive on SciCode (44.3), HLE (47.6) and BrowseComp (75.5). We hope this work provides the community with a practical path for scaling the horizon using a 35B agent that can reach or match the performance of 1T models on long-horizon tasks.

cs.CL↗
arXiv2026

Agents-K1: Towards Agent-native Knowledge Orchestration

Current LLM-based research agents have advanced through agent orchestration, yet largely overlook scientific knowledge orchestration. Existing works often reduce papers to abstracts, surface mentions, and flat \texttt{cites} edges, omitting key entities, claims, evidence, mechanisms, and method lineages essential for scientific reasoning. To this end, we introduce \textbf{Agents-K1}, an end-to-end knowledge orchestration pipeline that converts raw documents into agent-native scientific knowledge graphs. Agents-K1 integrates three components under a unifying theoretical foundation: a multimodal parser whose five-module schema captures entities, multimodal evidence, citations, and typed inter-entity relations across the full paper rather than abstracts alone; a 4B information-extraction backbone trained with GRPO under a rule-based reward; and a graphanything CLI, a tri-source agent interface that unifies web search, multimodal graph retrieval, and cross-document traversal. On top of this, we process 2.46 million scientific papers across six subjects to produce \textbf{Scholar-KG}, of which we release a one-million-paper subset, and the full Scholar-KG is accessible via the SCP link below. The same pipeline can be extended to general-domain corpora and to schema-conformant data synthesis. Extensive experiments demonstrate that Agents-K1 achieves superior performance in scientific information extraction, knowledge graph construction, and multi-hop scientific reasoning.

cs.AI↗
arXiv2026

Style or Content? Evaluating Style Classifiers with Controlled Content Overlap

Style classifiers can use content cues that correlate with style labels in naturally collected data, yet we lack a systematic way to measure this reliance. We study this problem with a controlled content overlap setup built on parallel Bible translations. Specifically, we define the overlap parameter $\alpha$ as the normalized residual of mutual information between content identity and style label, so that it measures how much content is shared across style classes: from no shared content ($\alpha=0$) to fully shared content ($\alpha=1$). Cross-overlap evaluation of RoBERTa-based classifiers shows that low-overlap models degrade when content cues are removed, while high-overlap models transfer more robustly. A cross-style content retrieval probe further shows that content becomes less recoverable as $\alpha$ increases, with training dynamics showing this removal occurs gradually. Together, these results suggest that controlled overlap provides a simple diagnostic for separating style learning from content shortcuts.

cs.CL↗
arXiv2026

Compiler-Guided Adaptive Proof Search with Cross-Model Synergy on Context-Dependent Theorem Proving

Theorem proving in real-world Lean 4 projects is challenging because proofs often depend on project-specific context. While iterative refinement can use compiler errors to repair failed proofs, reusing failed attempts requires careful search control: some proofs provide better starting points than others, and later revisions may degrade a partially correct proof. We propose a compiler-guided proof search framework that balances exploration and exploitation. It explores diverse starting points through dual-model generation and stagnation-triggered resampling, while exploiting promising proof states through current-best refinement guided by compiler-grounded pairwise comparison. Experiments on seven real-world Lean 4 projects from miniCTX-v2 show that our method achieves a better effectiveness--efficiency tradeoff than pass@k baselines. Within the pass@32 budget, our method improves average pass rate by 12.8 percentage points while reducing LLM calls by 21.9%.

cs.CL↗
arXiv2026

RABot: Reinforcement-Guided Graph Augmentation for Imbalanced and Noisy Social Bot Detection

Social bot detection is pivotal for safeguarding the integrity of online information ecosystems. Although recent graph neural network (GNN) solutions achieve strong results, they remain hindered by two practical challenges: (i) severe class imbalance arising from the high cost of generating bots, and (ii) topological noise introduced by bots that skillfully mimic human behavior and forge deceptive links. We propose the Reinforcement-guided graph Augmentation social Bot detector (RABot), a multi-granularity graph-augmentation framework that addresses both issues in a unified manner. RABot employs a neighborhood-aware oversampling strategy that linearly interpolates minority-class embeddings within local subgraphs, thereby stabilizing the decision boundary under low-resource regimes. Concurrently, a reinforcement-learning-driven edge-filtering module combines similarity-based edge features with adaptive threshold optimization to excise spurious interactions during message passing, yielding a cleaner topology. Extensive experiments on three real-world benchmarks and four GNN backbones demonstrate that RABot consistently surpasses state-of-the-art baselines. In addition, since its augmentation and filtering modules are orthogonal to the underlying architecture, RABot can be seamlessly integrated into existing GNN pipelines to boost performance with minimal overhead.

cs.SI↗
arXiv2026

EpicCBR: Item-Relation-Enhanced Dual-Scenario Contrastive Learning for Cold-Start Bundle Recommendation

Bundle recommendation aims to recommend a set of items to users for overall consumption. Existing bundle recommendation models primarily depend on observed user-bundle interactions, limiting exploration of newly-emerged bundles that are constantly created. It pose a critical representation challenge for current bundle methods, as they usually treat each bundle as an independent instance, while neglecting to fully leverage the user-item (UI) and bundle-item (BI) relations over popular items. To alleviate it, in this paper we propose a multi-view contrastive learning framework for cold-start bundle recommendation, named EpicCBR. Specifically, it precisely mine and utilize the item relations to construct user profiles, identifying users likely to engage with bundles. Additionally, a popularity-based method that characterizes the features of new bundles through historical bundle information and user preferences is proposed. To build a framework that demonstrates robustness in both cold-start and warm-start scenarios, a multi-view graph contrastive learning framework capable of integrating these diverse scenarios is introduced to ensure the model's generalization capability. Extensive experiments conducted on three popular benchmarks showed that EpicCBR outperforms state-of-the-art by a large margin (up to 387%), sufficiently demonstrating the superiority of the proposed method in cold-start scenario. The code and dataset can be found in the GitHub repository: https://github.com/alexlovecoding/EpicCBR.

cs.IR↗
arXiv2025

TROJail: Trajectory-Level Optimization for Multi-Turn Large Language Model Jailbreaks with Process Rewards

Large language models have seen widespread adoption, yet they remain vulnerable to multi-turn jailbreak attacks, threatening their safe deployment. This has led to the task of training automated multi-turn attackers to probe model safety vulnerabilities. However, existing approaches typically rely on turn-level optimization, which is insufficient for learning long-term attack strategies. To bridge this gap, we formulate this task as a multi-turn reinforcement learning problem, directly optimizing the harmfulness of the final-turn response as the outcome reward. To address the sparse supervision of the outcome reward, we introduce TROJail, which employs two process rewards to evaluate the utility of intermediate prompts and integrate them into advantage estimation. These rewards (1) penalize overly harmful prompts that trigger the model's refusal mechanism, and (2) encourage steering the semantic relevance of responses toward the targeted harmful content. Experimental results show improved attack success rates across multiple models and benchmarks, highlighting the effectiveness of our approach. The code is available at https://github.com/xxiqiao/TROJail. Warning: This paper contains examples of harmful content.

cs.AI↗
arXiv2025

Suppression of inverse magnetic energy transfer in collisionless marginally magnetized plasmas

We investigate the inverse cascade of magnetic energy in decaying, collisionless plasmas with moderate to high-$\beta$ values via first-principles numerical simulations and analytical theory. We find that pressure-anisotropy-driven instabilities, in particular the firehose instability, suppress reconnection-driven coalescence of magnetic structures (i.e., inverse transfer) by nullifying magnetic tension. This suppression leaves such structures elongated and confined to scales comparable to the Larmor radius of the particles. The presence of a magnetic guide field of sufficient strength, or a greater scale separation between the initial size of the magnetic structures and the Larmor radius, restores the system's ability to inverse transfer magnetic energy. These results reveal that inverse energy transfer in collisionless plasmas is not guaranteed, but instead sensitively depends on magnetization. In the astrophysical context, this identifies a kinetic mechanism by which Weibel-generated seed fields may fail to merge consistently, potentially limiting their role in cosmic magnetogenesis.

physics.plasm-ph↗
arXiv2025

AdvEvo-MARL: Shaping Internalized Safety through Adversarial Co-Evolution in Multi-Agent Reinforcement Learning

LLM-based multi-agent systems excel at planning, tool use, and role coordination, but their openness and interaction complexity also expose them to jailbreak, prompt-injection, and adversarial collaboration. Existing defenses fall into two lines: (i) self-verification that asks each agent to pre-filter unsafe instructions before execution, and (ii) external guard modules that police behaviors. The former often underperforms because a standalone agent lacks sufficient capacity to detect cross-agent unsafe chains and delegation-induced risks; the latter increases system overhead and creates a single-point-of-failure-once compromised, system-wide safety collapses, and adding more guards worsens cost and complexity. To solve these challenges, we propose AdvEvo-MARL, a co-evolutionary multi-agent reinforcement learning framework that internalizes safety into task agents. Rather than relying on external guards, AdvEvo-MARL jointly optimizes attackers (which synthesize evolving jailbreak prompts) and defenders (task agents trained to both accomplish their duties and resist attacks) in adversarial learning environments. To stabilize learning and foster cooperation, we introduce a public baseline for advantage estimation: agents within the same functional group share a group-level mean-return baseline, enabling lower-variance updates and stronger intra-group coordination. Across representative attack scenarios, AdvEvo-MARL consistently keeps attack-success rate (ASR) below 20%, whereas baselines reach up to 38.33%, while preserving-and sometimes improving-task accuracy (up to +3.67% on reasoning tasks). These results show that safety and utility can be jointly improved without relying on extra guard agents or added system overhead.

cs.AI↗
arXiv2025

Toward Practical Equilibrium Propagation: Brain-inspired Recurrent Neural Network with Feedback Regulation and Residual Connections

Brain-like intelligent systems need brain-like learning methods. Equilibrium Propagation (EP) is a biologically plausible learning framework with strong potential for brain-inspired computing hardware. However, existing im-plementations of EP suffer from instability and prohibi-tively high computational costs. Inspired by the structure and dynamics of the brain, we propose a biologically plau-sible Feedback-regulated REsidual recurrent neural network (FRE-RNN) and study its learning performance in EP framework. Feedback regulation enables rapid convergence by reducing the spectral radius. The improvement in con-vergence property reduces the computational cost and train-ing time of EP by orders of magnitude, delivering perfor-mance on par with backpropagation (BP) in benchmark tasks. Meanwhile, residual connections with brain-inspired topologies help alleviate the vanishing gradient problem that arises when feedback pathways are weak in deep RNNs. Our approach substantially enhances the applicabil-ity and practicality of EP in large-scale networks that un-derpin artificial intelligence. The techniques developed here also offer guidance to implementing in-situ learning in physical neural networks.

cs.NE↗
arXiv2025

An Ohsawa-Takegoshi-type $L^2$ extension for upper semi-continuous $L^2$-optimal functions

In this article, we obtain an Ohsawa-Takegoshi-type $L^2$-extension for upper semi-continuous $L^2$-optimal functions via a Lebesgue-type differentiation theorem. As applications, we give a characterization of plurisubharmonic functions via the multiple coarse $L^2$-estimate property for (strongly) upper semi-continuous functions and show that (strongly) upper semi-continuous $L^2$-optimal functions satisfy Skoda's integrability theorem and the strong openness property.

math.CV↗
arXiv2025

Multi-Robot Assembly of Deformable Linear Objects Using Multi-Modal Perception

Industrial assembly of deformable linear objects (DLOs) such as cables offers great potential for many industries. However, DLOs pose several challenges for robot-based automation due to the inherent complexity of deformation and, consequentially, the difficulties in anticipating the behavior of DLOs in dynamic situations. Although existing studies have addressed isolated subproblems like shape tracking, grasping, and shape control, there has been limited exploration of integrated workflows that combine these individual processes. To address this gap, we propose an object-centric perception and planning framework to achieve a comprehensive DLO assembly process throughout the industrial value chain. The framework utilizes visual and tactile information to track the DLO's shape as well as contact state across different stages, which facilitates effective planning of robot actions. Our approach encompasses robot-based bin picking of DLOs from cluttered environments, followed by a coordinated handover to two additional robots that mount the DLOs onto designated fixtures. Real-world experiments employing a setup with multiple robots demonstrate the effectiveness of the approach and its relevance to industrial scenarios.

cs.RO↗
arXiv2025

TreeRare: Syntax Tree-Guided Retrieval and Reasoning for Knowledge-Intensive Question Answering

In real practice, questions are typically complex and knowledge-intensive, requiring Large Language Models (LLMs) to recognize the multifaceted nature of the question and reason across multiple information sources. Iterative and adaptive retrieval, where LLMs decide when and what to retrieve based on their reasoning, has been shown to be a promising approach to resolve complex, knowledge-intensive questions. However, the performance of such retrieval frameworks is limited by the accumulation of reasoning errors and misaligned retrieval results. To overcome these limitations, we propose TreeRare (Syntax Tree-Guided Retrieval and Reasoning), a framework that utilizes syntax trees to guide information retrieval and reasoning for question answering. Following the principle of compositionality, TreeRare traverses the syntax tree in a bottom-up fashion, and in each node, it generates subcomponent-based queries and retrieves relevant passages to resolve localized uncertainty. A subcomponent question answering module then synthesizes these passages into concise, context-aware evidence. Finally, TreeRare aggregates the evidence across the tree to form a final answer. Experiments across five question answering datasets involving ambiguous or multi-hop reasoning demonstrate that TreeRare achieves substantial improvements over existing state-of-the-art methods.

cs.CL↗
arXiv2025

Assistant-Guided Mitigation of Teacher Preference Bias in LLM-as-a-Judge

LLM-as-a-Judge employs large language models (LLMs), such as GPT-4, to evaluate the quality of LLM-generated responses, gaining popularity for its cost-effectiveness and strong alignment with human evaluations. However, training proxy judge models using evaluation data generated by powerful teacher models introduces a critical yet previously overlooked issue: teacher preference bias, where the proxy judge model learns a biased preference for responses from the teacher model. To tackle this problem, we propose a novel setting that incorporates an additional assistant model, which is not biased toward the teacher model's responses, to complement the training data. Building on this setup, we introduce AGDe-Judge, a three-stage framework designed to debias from both the labels and feedbacks in the training data. Extensive experiments demonstrate that AGDe-Judge effectively reduces teacher preference bias while maintaining strong performance across six evaluation benchmarks. Code is available at https://github.com/Liuz233/AGDe-Judge.

cs.CL↗