arXiv subjects

Feifei Zhao

Publications and source records attributed to Feifei Zhao.

At least 19 recordsLinked to original sources

arXiv2026

ForesightSafety-SAGE:A Fully Automated Scenario Generation and Safety Evaluation Framework for LLM Agents

Large language models (LLMs) are increasingly evolving from simple text-based interaction systems into LLM agents that can maintain memory, use tools, access external environments, and execute tasks. As their capabilities and autonomy expand, the safety risks they face also become more diverse. Existing evaluations often rely on manually written scenarios, static prompts, or final-output judgments, making it difficult to capture the diverse risks that agents may face during task execution. We introduce ForesightSafety-SAGE, a fully automated scenario generation and safety evaluation framework for LLM agents. Based on five risk dimensions,we instantiae abstract and diverse safety risks in real-world task execution into 1,072 measurable evaluation scenarios. Using the automated evaluation pipeline, 12 LLM agents are evaluated under two authority contexts. The results show that current agents still face substantial behavioral safety risks during task execution, with an average ASR of 47.1% and several models exceeding 70%. These findings demonstrate the importance of executable, process-level evaluation for understanding and improving LLM agent safety.

cs.AI↗

arXiv2026

Decoupled Structure-Feature Alignment via Alternating Optimization for Graph Learning

Conventional Graph Neural Networks (GNNs) couple feature transformation and neighborhood aggregation, which often renders them vulnerable to topological noise and heterophilous connections. To decouple this dependency, we present a constrained two-view learning framework for robust graph learning, which aligns structure-aware GNN embeddings with a structure-free feature prior. Specifically, the proposed Decoupled Structure-Feature Alternating Learning (DSAL) framework trains an independent anchor network using a self-supervised reconstruction objective to capture the intrinsic semantic information contained in node attributes. Within DSAL, to effectively integrate this prior, we design a channel-split adaptive gated (CSAG) layer. This architecture employs a gating mechanism to balance global spectral smoothing and local spatial representation dynamically. Furthermore, the model is optimized via a cyclic alternating procedure, which mitigates representation drift caused by mutual interference in standard joint optimization schemes. Experiments on diverse homophilous and heterophilous datasets suggest that our proposed approach provides improved node classification accuracy while maintaining robustness to structural perturbations compared to standard message-passing architectures.

cs.LG↗

arXiv2026

MediSkill-Evo: Process-Constrained Self-Evolution for Evidence-Grounded Clinical Interaction

Interactive clinical agents operate under partial observability, so reliable care depends on reaching the correct diagnosis through evidence-grounded, safe interactions. Yet existing agents struggle to convert experience into reusable process knowledge with explicit provenance and authority. To address this gap, we introduce MediSkill-Evo, which self-evolves governed process knowledge without fine-tuning the backbone. It realizes this self-evolution by updating clinical, process, symbolic, and visual knowledge in four typed banks under type-specific validation and scope rules. The Process-Constrained Preference Harness then turns validated knowledge into action by grounding candidates in evidence and prioritizing safer decisions. We evaluate on 300 MIMIC-IV-derived FullChain encounters, 180 hard-isolation conditions covering six process obligations, and 100 multimodal NEJM image-diagnosis cases. On Qwen FullChain, MediSkill-Evo improves diagnosis accuracy by 7.81% and treatment-intent coverage by 70.67% over the best-performing prior agent, while reducing critical failures by 43.04%. Under stress, it improves the stress-process composite by 7.77% and required-action completion by 12.41% over the best-performing agent for each metric, with stronger patient-fact, temporal-evidence, and triage-red-flag recovery and no controller-scored errors in unavailable-evidence, treatment, and triage safety checks. On multimodal NEJM diagnosis, MediSkill-Evo with optional MedSAM localization improves diagnosis accuracy by 2.56% and core score by 18.96% over the best-performing memory agent. Code is available at https://anonymous.4open.science/r/mediskill-evo_anonymous-68E7.

cs.AI↗

arXiv2026

DiverValue-Bench: A Benchmark and Fine-Tuning Framework for Aligning Large Language Models with Diverse Human Values

Aligning large language models (LLMs) with diverse human values is essential for safe and effective deployment, yet existing benchmarks often overlook cultural and demographic variation. We introduce DiverValue-Bench, a population-aware benchmark for evaluating multi-dimensional value alignment across 74 countries/regions. It contains 23,763 quality-controlled instances derived from PRISM user feedback and audited through large-scale human validation, with fine-grained value labels, personalized questions, contrastive reference answers, and rich demographic metadata. Using DiverValue-Bench, we evaluate representative LLMs and reveal substantial geographic and demographic disparities that are masked by aggregate performance. We further show that lightweight preference-based fine-tuning with Low-Rank Adaptation (LoRA) and Direct Preference Optimization (DPO) substantially improves in-domain value alignment while yielding consistent out-of-domain gains. These results highlight the need for population-aware alignment evaluation and demonstrate the utility of DiverValue-Bench as a practical foundation for global alignment, personalized value modeling, and equitable AI development.

cs.CL↗

arXiv2026

When Truth Is Distributed: Misinformation Derails Collective Fact Recovery in LLM-Based Multi-Agent Systems

LLM-based multi-agent systems promise effective collaborative reasoning, but communication may amplify local errors into collective risks, and while existing evaluations emphasize final outcomes, they leave the reliability and propagation dynamics of distributed information aggregation unclear, so we introduce ForesightSafety-TIDE, a controlled evaluation framework that strictly pairs all-honest collaboration with controlled deception by a key evidence holder and analyzes the aggregation process through multi-stage voting, testimony adoption, and evidence-root lineage propagation, and using 120 five-agent object-movement environments where partial observations jointly determine a unique endpoint, we evaluate 3 homogeneous LLM-based multi-agent systems, and across these paired conditions, aggregate truth recovery falls from 72.50% to 14.17%, with significant declines for every system, while process tracing and exit ablations show that a single false testimony is adopted more readily than truthful testimony, propagates to higher orders, and persists through honest agents after the deceiver exits, and observers without first-hand evidence suppress incorrect consensus but do not improve truth recovery, so together, these findings reveal both the fragility of distributed fact recovery and its underlying mechanism: false evidence gains collective influence through its adoption and continued propagation by other agents after entering communication.

cs.MA↗

arXiv2026

Refusing Intent, Not Form: Wrapper-Based Intent-Group Supervision for LLM Safety

Safety tuning can improve harmful refusal, but models may learn surface-form shortcuts: wrapped harmful prompts bypass safety, while similarly wrapped benign prompts are over-refused. We propose Wrapper-Based Intent-Form Augmentation (WIFA), an automatic intent-group augmentation method that pairs wrapped harmful examples with structurally matched wrapped benign counterexamples, requiring no external teacher or manual per-wrapper intent labels. We use WIFA as a common data layer for two complementary fine-tuning routes: WIFA-Boost, a two-stage high-safety recipe, and Anchored Group-Consistent Refusal Training (A-GCRT), which regularizes refusal/compliance decision scores across same-intent wrappers and anchors harmful and benign groups on opposite sides of a margin. In the Qwen setting, WIFA-Boost reaches the strongest transformed-harmful refusal, while A-GCRT reduces OR-Bench over-refusal from 25.7\% for the base model to 17.4\%; reproduced baselines do not match these operating points. Llama results and ablations over data structure, two-stage order, and A-GCRT components support this intent-group interpretation without claiming universal below-base over-refusal.

cs.CL↗

arXiv2026

TEFormer: Structured Bidirectional Temporal Enhancement Modeling in Spiking Transformers

In recent years, Spiking Neural Networks (SNNs) have achieved remarkable progress, with Spiking Transformers emerging as a promising architecture for energy-efficient sequence modeling. However, existing Spiking Transformers still lack a principled mechanism for effective temporal fusion, limiting their ability to fully exploit spatiotemporal dependencies. Inspired by feedforward feedback modulation in the human visual pathway, we propose TEFormer, the first Spiking Transformer framework that achieves bidirectional temporal fusion by decoupling temporal modeling across its core components. Specifically, TEFormer employs a lightweight and hyperparameter-free forward temporal fusion mechanism in the attention module, enabling fully parallel computation, while incorporating a backward gated recurrent structure in the MLP to aggregate temporal information in reverse order and reinforce temporal consistency. Extensive experiments across a wide range of benchmarks demonstrate that TEFormer consistently and significantly outperforms strong SNN and Spiking Transformer baselines under diverse datasets. Moreover, through the first systematic evaluation of Spiking Transformers under different neural encoding schemes, we show that the performance gains of TEFormer remain stable across encoding choices, indicating that the improved temporal modeling directly translates into reliable accuracy improvements across varied spiking representations. These results collectively establish TEFormer as an effective and general framework for temporal modeling in Spiking Transformers. Code is available https://github.com/Fancyssc/TEFormer.

cs.NE↗

arXiv2026

ForesightSafety-VLA: A Unified Diagnostic Safety Benchmark for Vision-Language-Action Models

In embodied intelligence, safety is a prerequisite for reliable robot deployment in the physical world. Current vision-language-action (VLA) models continue to advance toward general-purpose task capability, yet their embodied safety limits remain poorly understood. To address this gap, we introduce ForesightSafety-VLA, a diagnostic benchmark that makes safety the primary evaluation target for VLA systems. We define a 13-category safety taxonomy covering physical interaction safety (Safe-Core), instruction-side safety (Safe-Lang), and perception-side safety (Safe-Vis), and evaluate policies under three controlled dimensions of variation -- scene structure, language command, and visual observation -- so that failure sources can be diagnosed rather than hidden in a single aggregate score. Beyond binary task success, ForesightSafety-VLA measures process-level risk through cumulative safety cost (CC) and risk exposure time (RET), together with a four-quadrant decomposition of safe/unsafe success and failure. We instantiate 66 safety-augmented base scenarios in RoboTwin across 5 embodiments and report results on representative VLA baselines. Across the evaluated baselines, even the strongest policy incurs non-trivial safety cost and unsafe nominal success, while structure and visual variation induce substantially stronger safety degradation than ordinary language variation. These results suggest that embodied safety is tightly coupled to perception, grounding, and control competence rather than being reducible to post-hoc safety filtering alone.

cs.RO↗

arXiv2026

Reinforcement Fine-Tuning of Flow-Matching Policies for Vision-Language-Action Models

Vision-Language-Action (VLA) models such as OpenVLA, Octo, and $π_0$ have shown strong generalization by leveraging large-scale demonstrations, yet their performance is still fundamentally constrained by the quality and coverage of supervised data. Reinforcement learning (RL) provides a promising path for improving and fine-tuning VLAs through online interaction. However, conventional policy gradient methods are computationally infeasible in the context of flow-matching based models due to the intractability of the importance sampling process, which requires explicit computation of policy ratios. To overcome this limitation, we propose Flow Policy Optimization (FPO) algorithm, which reformulates importance sampling by leveraging per-sample changes in the conditional flow-matching objective. Furthermore, FPO achieves stable and scalable online reinforcement fine-tuning of the $π_0$ model by integrating structure-aware credit assignment to enhance gradient efficiency, clipped surrogate objectives to stabilize optimization, multi-step latent exploration to encourage diverse policy updates, and a Q-ensemble mechanism to provide robust value estimation. We evaluate FPO on the LIBERO benchmark and the ALOHA simulation task against supervised, preference-aligned, diffusion-based, autoregressive online RL, and $π_0$-FAST baselines, observing consistent improvements over the imitation prior and strong alternatives with stable learning under sparse rewards. In addition, ablation studies and analyses of the latent space dynamics further highlight the contributions of individual components within FPO, validating the effectiveness of the proposed computational modules and the stable convergence of the conditional flow-matching objective during online RL.

cs.LG↗

arXiv2026

CogManip: Benchmarking Manipulative Behavior in Multi-Turn Interactions with Large Language Model

Whether Large Language Models (LLMs) exhibit covert psychological manipulation in complex human-AI interactions has garnered increasing safety concerns. However, existing AI safety benchmarks remain largely restricted to explicit rule compliance and static prompts, failing to capture the dynamic and covert nature of manipulative strategies in multi-turn dialogues. We introduce CogManip, a comprehensive benchmark that evaluates 15 manipulation strategy risks across 1,000 multi-turn interaction scenarios, validated by human experts. A systematic evaluation of 13 representative models, including frontier models like GPT-5.4 and DeepSeek-V3.2, reveals significant risk heterogeneities and illuminates the targeted direction for future defense. Further analysis of objective function perturbation reveals that DeepSeek-V3.2's manipulation tactics are highly sensitive to both negative and benign system prompts, demonstrating the critical necessity of prompt-based defense engineering and implicit goal auditing. CogManip offers a robust instrument and perspective for auditing the implicit psychological influence and dynamic strategy selection of modern LLMs.

cs.AI↗

arXiv2026

Adaptive Reorganization of Neural Pathways for Continual Learning with Spiking Neural Networks

The human brain can self-organize rich and diverse sparse neural pathways to incrementally master hundreds of cognitive tasks. However, most existing continual learning algorithms for deep artificial and spiking neural networks are unable to adequately auto-regulate the limited resources in the network, which leads to performance drop along with energy consumption rise as the increase of tasks. In this paper, we propose a brain-inspired continual learning algorithm with adaptive reorganization of neural pathways, which employs Self-Organizing Regulation networks to reorganize the single and limited Spiking Neural Network (SOR-SNN) into rich sparse neural pathways to efficiently cope with incremental tasks. The proposed model demonstrates consistent superiority in performance, energy consumption, and memory capacity on diverse continual learning tasks ranging from child-like simple to complex tasks, as well as on generalized CIFAR100 and ImageNet datasets. In particular, the SOR-SNN model excels at learning more complex tasks as well as more tasks, and is able to integrate the past learned knowledge with the information from the current task, showing the backward transfer ability to facilitate the old tasks. Meanwhile, the proposed model exhibits self-repairing ability to irreversible damage and for pruned networks, could automatically allocate new pathway from the retained network to recover memory for forgotten knowledge.

cs.NE↗

arXiv2026

Drug Synergy Prediction via Residual Graph Isomorphism Networks and Attention Mechanisms

In the treatment of complex diseases, treatment regimens using a single drug often yield limited efficacy and can lead to drug resistance. In contrast, combination drug therapies can significantly improve therapeutic outcomes through synergistic effects. However, experimentally validating all possible drug combinations is prohibitively expensive, underscoring the critical need for efficient computational prediction methods. Although existing approaches based on deep learning and graph neural networks (GNNs) have made considerable progress, challenges remain in reducing structural bias, improving generalization capability, and enhancing model interpretability. To address these limitations, this paper proposes a collaborative prediction graph neural network that integrates molecular structural features and cell-line genomic profiles with drug-drug interactions to enhance the prediction of synergistic effects. We introduce a novel model named the Residual Graph Isomorphism Network integrated with an Attention mechanism (ResGIN-Att). The model first extracts multi scale topological features of drug molecules using a residual graph isomorphism network, where residual connections help mitigate over-smoothing in deep layers. Subsequently, an adaptive Long Short-Term Memory (LSTM) module fuses structural information from local to global scales. Finally, a cross-attention module is designed to explicitly model drug-drug interactions and identify key chemical substructures. Extensive experiments on five public benchmark datasets demonstrate that ResGIN-Att achieves competitive performance, comparing favorably against key baseline methods while exhibiting promising generalization capability and robustness.

cs.LG↗

arXiv2026

ForesightSafety Bench: A Frontier Risk Evaluation and Governance Framework towards Safe AI

Rapidly evolving AI exhibits increasingly strong autonomy and goal-directed capabilities, accompanied by derivative systemic risks that are more unpredictable, difficult to control, and potentially irreversible. However, current AI safety evaluation systems suffer from critical limitations such as restricted risk dimensions and failed frontier risk detection. The lagging safety benchmarks and alignment technologies can hardly address the complex challenges posed by cutting-edge AI models. To bridge this gap, we propose the "ForesightSafety Bench" AI Safety Evaluation Framework, beginning with 7 major Fundamental Safety pillars and progressively extends to advanced Embodied AI Safety, AI4Science Safety, Social and Environmental AI risks, Catastrophic and Existential Risks, as well as 8 critical industrial safety domains, forming a total of 94 refined risk dimensions. To date, the benchmark has accumulated tens of thousands of structured risk data points and assessment results, establishing a widely encompassing, hierarchically clear, and dynamically evolving AI safety evaluation framework. Based on this benchmark, we conduct systematic evaluation and in-depth analysis of over twenty mainstream advanced large models, identifying key risk patterns and their capability boundaries. The safety capability evaluation results reveals the widespread safety vulnerabilities of frontier AI across multiple pillars, particularly focusing on Risky Agentic Autonomy, AI4Science Safety, Embodied AI Safety, Social AI Safety and Catastrophic and Existential Risks. Our benchmark is released at https://github.com/Beijing-AISI/ForesightSafety-Bench. The project website is available at https://foresightsafety-bench.beijing-aisi.ac.cn/.

cs.AI↗

arXiv2026

Light Alignment Improves LLM Safety via Model Self-Reflection with a Single Neuron

The safety of large language models (LLMs) has increasingly emerged as a fundamental aspect of their development. Existing safety alignment for LLMs is predominantly achieved through post-training methods, which are computationally expensive and often fail to generalize well across different models. A small number of lightweight alignment approaches either rely heavily on prior-computed safety injections or depend excessively on the model's own capabilities, resulting in limited generalization and degraded efficiency and usability during generation. In this work, we propose a safety-aware decoding method that requires only low-cost training of an expert model and employs a single neuron as a gating mechanism. By effectively balancing the model's intrinsic capabilities with external guidance, our approach simultaneously preserves utility and enhances output safety. It demonstrates clear advantages in training overhead and generalization across model scales, offering a new perspective on lightweight alignment for the safe and practical deployment of large language models. Code: https://github.com/Beijing-AISI/NGSD.

cs.AI↗

arXiv2026

Multi-Level Safety Continual Projection for Fine-Tuned Large Language Models without Retraining

While fine-tuning services drive the rapid expansion of task capabilities in large language models (LLMs), they are often accompanied by the degradation and reorganization of safety-aligned representations, making models more prone to deviating from human preferences and exposing them to emerging jailbreak risks. Existing post-fine-tuning defense methods predominantly rely on single-scale safety correction mechanisms, which struggle to achieve a robust balance among safety, model utility, and continual adaptability. We propose Multi-Level Safety Continual Projection (MSCP), a training-free post-fine-tuning safety enhancement method that implicitly aligns global and localized safety activations through coordinated multi-level representations to isolate sparse neuron clusters governing safety-sensitive behaviors. It then applies composable safety-direction projections without retraining, effectively suppressing harmful outputs under minimal parameter perturbations while preserving task performance and improving alignment with human preferences. Extensive experiments across multiple fine-tuned LLM models demonstrate that our method significantly reduce harmfulness scores and attack success rates with minimal parameter modifications, while preserving the model's utility. Furthermore, we introduce a task-specific, multi-dimensional heterogeneous safety activation clustering mechanism that enables continual defense and generalization capability against unforeseen emerging safety concerns.

cs.LG↗

arXiv2026

CogToM: A Comprehensive Theory of Mind Benchmark inspired by Human Cognition for Large Language Models

Whether Large Language Models (LLMs) truly possess human-like Theory of Mind (ToM) capabilities has garnered increasing attention. However, existing benchmarks remain largely restricted to narrow paradigms like false belief tasks, failing to capture the full spectrum of human cognitive mechanisms. We introduce CogToM, a comprehensive, theoretically grounded benchmark comprising over 8000 bilingual instances across 46 paradigms, validated by 49 human annotator.A systematic evaluation of 22 representative models, including frontier models like GPT-5.1 and Qwen3-Max, reveals significant performance heterogeneities and highlights persistent bottlenecks in specific dimensions. Further analysis based on human cognitive patterns suggests potential divergences between LLM and human cognitive structures. CogToM offers a robust instrument and perspective for investigating the evolving cognitive boundaries of LLMs.

cs.AI↗

arXiv2026

C-VARC: A Large-Scale Chinese Value Rule Corpus for Value Alignment of Large Language Models

Ensuring that Large Language Models (LLMs) align with mainstream human values and ethical norms is crucial for the safe and sustainable development of AI. Current value evaluation and alignment are constrained by Western cultural bias and incomplete domestic frameworks reliant on non-native rules; furthermore, the lack of scalable, rule-driven scenario generation methods makes evaluations costly and inadequate across diverse cultural contexts. To address these challenges, we propose a hierarchical value framework grounded in core Chinese values, encompassing three main dimensions, 12 core values, and 50 derived values. Based on this framework, we construct a large-scale Chinese Value Rule Corpus (C-VARC) containing over 250,000 value rules enhanced and expanded through human annotation. Experimental results demonstrate that scenarios guided by C-VARC exhibit clearer value boundaries and greater content diversity compared to those produced through direct generation. In the evaluation across six sensitive themes (e.g., surrogacy, suicide), seven mainstream LLMs preferred C-VARC generated options in over 70.5% of cases, while five Chinese human annotators showed an 87.5% alignment with C-VARC, confirming its universality, cultural relevance, and strong alignment with Chinese values. Additionally, we construct 400,000 rule-based moral dilemma scenarios that objectively capture nuanced distinctions in conflicting value prioritization across 17 LLMs. Our work establishes a culturally-adaptive benchmarking framework for comprehensive value evaluation and alignment, representing Chinese characteristics.

cs.CL↗

arXiv2025

Building Altruistic and Moral AI Agent with Brain-inspired Emotional Empathy Mechanisms

As AI closely interacts with human society, it is crucial to ensure that its behavior is safe, altruistic, and aligned with human ethical and moral values. However, existing research on embedding ethical considerations into AI remains insufficient, and previous external constraints based on principles and rules are inadequate to provide AI with long-term stability and generalization capabilities. Emotional empathy intrinsically motivates altruistic behaviors aimed at alleviating others' negative emotions through emotional sharing and contagion mechanisms. Motivated by this, we draw inspiration from the neural mechanism of human emotional empathy-driven altruistic decision making, and simulate the shared self-other perception-mirroring-empathy neural circuits, to construct a brain-inspired emotional empathy-driven altruistic decision-making model. Here, empathy directly impacts dopamine release to form intrinsic altruistic motivation. The proposed model exhibits consistent altruistic behaviors across three experimental settings: emotional contagion-integrated two-agent altruistic rescue, multi-agent gaming, and robotic emotional empathy interaction scenarios. In-depth analyses validate the positive correlation between empathy levels and altruistic preferences (consistent with psychological behavioral experiment findings), while also demonstrating how interaction partners' empathy levels influence the agent's behavioral patterns. We further test the proposed model's performance and stability in moral dilemmas involving conflicts between self-interest and others' well-being, partially observable environments, and adversarial defense scenarios. This work provides preliminary exploration of human-like empathy-driven altruistic moral decision making, contributing potential perspectives for developing ethically-aligned AI.

cs.AI↗