arXiv subjects

Tong Li

Publications and source records attributed to Tong Li.

At least 55 records · Page 3Linked to original sources

arXiv2026

Constraints on light dark matter from primordial black hole evaporation at dark matter direct detection experiments

Primordial black holes (PBHs) are able to produce light dark matter (DM) particles via Hawking radiation, and yield a flux of boosted DM that can be probed at underground DM direct detection experiments. We analyze both galactic and extragalactic contributions to the differential flux of light DM from PBH evaporation, and then compute the expected event rate from PBH boosted DM scattering off electrons or nuclei after taking into account the attenuation effect. Using recent data from DM direct detection experiments XENONnT, PandaX-4T and LZ, we set constraints on both DM-electron and DM-nucleus scattering cross sections, as well as the fraction of DM composed of PBHs $f_{\rm PBH}$ for $9\times10^{14}-1\times10^{16}\,\mathrm{g}$ PBHs that are not fully evaporated today. We also investigate the spectral evolution induced by Hawking evaporation throughout the evaporation and post-evaporation regimes. The constraints on the PBH mass are then extended into the $1\times10^{13}-6\times10^{14}\,\mathrm{g}$ window for fully evaporated PBHs.

hep-ph↗

arXiv2026

Online Statistical Inference of Constant Sample-averaged Q-Learning

Reinforcement learning algorithms have been widely used for decision-making tasks in various domains. However, the performance of these algorithms can be impacted by high variance and instability, particularly in environments with noise or sparse rewards. In this paper, we propose a framework to perform statistical online inference for a sample-averaged Q-learning approach. We adapt the functional central limit theorem (FCLT) for the modified algorithm under some general conditions and then construct confidence intervals for the Q-values via random scaling. We conduct experiments to perform inference on both the modified approach and its traditional counterpart, Q-learning using random scaling and report their coverage rates and confidence interval widths on two problems: a grid world problem as a simple toy example and a dynamic resource-matching problem as a real-world example for comparison between the two solution approaches.

stat.ML↗

arXiv2026

LSA: A Long-Short-term Aspect Interest Transformer for Aspect-Based Recommendation

Aspect-based recommendation methods extract aspect terms from reviews, such as price, to model fine-grained user preferences on items, making them a critical approach in personalized recommender systems. Existing methods utilize graphs to represent the relationships among users, items, and aspect terms, modeling user preferences based on graph neural networks. However, they overlook the dynamic nature of user interests - users may temporarily focus on aspects they previously paid little attention to - making it difficult to assign accurate weights to aspect terms for each user-item interaction. In this paper, we propose a long-short-term aspect interest Transformer (LSA) for aspect-based recommendation, which effectively captures the dynamic nature of user preferences by integrating both long-term and short-term aspect interests. Specifically, the short-term interests model the temporal changes in the importance of recently interacted aspect terms, while the long-term interests consider global behavioral patterns, including aspects that users have not interacted with recently. Finally, LSA combines long- and short-term interests to evaluate the importance of aspects within the union of user and item aspect neighbors, therefore accurately assigns aspect weights for each user-item interaction. Experiments conducted on four real-world datasets demonstrate that LSA improves MSE by 2.55% on average over the best baseline.

cs.IR↗

arXiv2026

A Statistically Reliable Optimization Framework for Bandit Experiments in Scientific Discovery

Scientific experimentation is largely driven by statistical hypothesis testing to determine significant differences in interventions. Traditionally, experimenters allocate samples uniformly between each intervention. However, such an approach may lead to suboptimal outcomes - multi-armed bandits (MABs) addresses this problem by allocating samples adaptively to maximize outcomes. Yet, two challenges have hindered the use of MABs in scientific domains. First, common hypothesis tests (e.g., $t$-tests) become invalid under adaptive sampling without correction, leading to inflated type~I and type~II errors. This is an understudied problem, and prior solutions suffer from issues such as low statistical power which prevent adoption in many practical settings. Second, practitioners must explicitly balance cumulative reward with statistical efficiency, yet no general methodology exists to quantify this trade-off across algorithms. In this paper, we study assumption modification and critical region correction approaches for hypothesis testing that enable common tests to be applied to adaptively collected data. We provide heuristic justification for its power efficiency and show in simulation that it achieves higher power than existing approaches. Further, we derive a theoretically and practically motivated objective function for adaptive experiment evaluation, which we integrate into a unified experimental framework. Our framework asks experimenters to specify an experiment extension cost for their problem, and based on that enables our proposed optimization procedure to select the bandit algorithm that best balances reward and power in their setting. We show that our approach enables practitioners to improve outcomes with only slightly more steps than uniform randomization, while retaining statistical validity.

stat.AP↗

arXiv2026

Hiding in Plain Sight: A Steganographic Approach to Stealthy LLM Jailbreaks

Jailbreak attacks pose a serious threat to Large Language Models (LLMs) by bypassing their safety mechanisms. A truly advanced jailbreak is defined not only by its effectiveness but, more critically, by its stealthiness. However, existing methods face a fundamental trade-off between semantic stealth (hiding malicious intent) and linguistic stealth (appearing natural), leaving them vulnerable to detection. To resolve this trade-off, we propose StegoAttack, a framework that leverages steganography. The core insight is to embed a harmful query within a benign, semantically coherent paragraph. This design provides semantic stealth by concealing the existence of malicious content and ensures linguistic stealth by maintaining the natural fluency of the cover paragraph. We evaluate StegoAttack on four state-of-the-art, safety-aligned LLMs, including GPT-5 and Gemini-3, and benchmark it against eight leading jailbreak methods. Our results show that StegoAttack achieves an average attack success rate (ASR) of 95.50%, outperforming existing baselines across all four models. Critically, its ASR drops by less than 27.00% under external detectors, while maintaining natural language distribution. This demonstrates that steganography effectively decouples linguistic and semantic stealth, thereby posing a fully concealed yet highly effective security threat. The code is available at https://github.com/GenggengSvan/StegoAttack

cs.CR↗

arXiv2026

Evaluating Long-Horizon Memory for Multi-Party Collaborative Dialogues

Long-term conversational memory in practical LLM applications is inherently collaborative: information is produced by multiple participants, scattered across groups and channels, revised over time, and implicitly grounded in roles and social context. Yet there is currently no established benchmark that evaluates memory under interaction patterns resembling real-world deployment, as existing benchmarks largely focus on dyadic or single-topic dialogues. In this paper, we introduce EverMemBench, the first benchmark designed for long-horizon collaborative memory, built from multi-party, multi-group conversations spanning over one million tokens with dense cross-topic interleaving, temporally evolving decisions, and role-conditioned personas. EverMemBench evaluates memory systems using 2400 QA pairs across three dimensions essential for real applications: fine-grained recall, memory awareness, and user profile understanding. Our evaluation reveals fundamental limitations of current systems: multi-hop reasoning collapses under multi-party attribution even with oracle evidence (26% accuracy), temporal reasoning fails without explicit version semantics beyond timestamps, and memory awareness is bottlenecked by retrieval, as similarity-based methods miss implicitly relevant information. EverMemBench thus represents a concrete step toward realistic evaluation of LLM memory and a cornerstone benchmark for developing next-generation LLMs that reason over time, roles, and collaborative interaction structure. Our benchmark and code are publicly available at https://github.com/EverMind-AI/EverMemBench.

cs.CL↗

arXiv2026

All Cities are Equal: A Unified Human Mobility Generation Model Enabled by LLMs

Synthetic human mobility generation is gaining traction as an ethical and practical approach to supporting the data needs of intelligent urban systems. Existing methods perform well primarily in data-rich cities, while their effectiveness declines significantly in cities with limited data resources. However, the ability to generate reliable human mobility data should not depend on a city's size or available resources, all cities deserve equal consideration. To address this open issue, we propose UniMob, a unified human mobility generation model across cities. UniMob is composed of three main components: an LLM-powered travel planner that derives high-level, temporally-aware, and semantically meaningful travel plans; a unified spatial embedding module that projects the spatial regions of various cities into a shared representation space; and a diffusion-based mobility generator that captures the joint spatiotemporal characteristics of human movement, guided by the derived travel plans. We evaluate UniMob extensively using two real-world datasets covering five cities. Comprehensive experiments show that UniMob significantly outperforms state-of-the-art baselines, achieving improvements of over 30\% across multiple evaluation metrics. Further analysis demonstrates UniMob's robustness in both zero- and few-shot scenarios, underlines the importance of LLM guidance, verifies its privacy-preserving nature, and showcases its applicability for downstream tasks.

cs.ET↗

arXiv2026

VLM-DEWM: Dynamic External World Model for Verifiable and Resilient Vision-Language Planning in Manufacturing

Vision-language model (VLM) shows promise for high-level planning in smart manufacturing, yet their deployment in dynamic workcells faces two critical challenges: (1) stateless operation, they cannot persistently track out-of-view states, causing world-state drift; and (2) opaque reasoning, failures are difficult to diagnose, leading to costly blind retries. This paper presents VLM-DEWM, a cognitive architecture that decouples VLM reasoning from world-state management through a persistent, queryable Dynamic External World Model (DEWM). Each VLM decision is structured into an Externalizable Reasoning Trace (ERT), comprising action proposal, world belief, and causal assumption, which is validated against DEWM before execution. When failures occur, discrepancy analysis between predicted and observed states enables targeted recovery instead of global replanning. We evaluate VLM-DEWM on multi-station assembly, large-scale facility exploration, and real-robot recovery under induced failures. Compared to baseline memory-augmented VLM systems, VLM DEWM improves state-tracking accuracy from 56% to 93%, increases recovery success rate from below 5% to 95%, and significantly reduces computational overhead through structured memory. These results establish VLM-DEWM as a verifiable and resilient solution for long-horizon robotic operations in dynamic manufacturing environments.

cs.RO↗

arXiv2026

PankRAG: Enhancing Graph Retrieval via Globally Aware Query Resolution and Dependency-Aware Reranking Mechanism

Recent graph-based RAG approaches leverage knowledge graphs by extracting entities from a query to fetch their associated relationships and metadata. However, relying solely on entity extraction often results in the misinterpretation or omission of latent critical information and relationships. This can lead to the retrieval of irrelevant or contradictory content, as well as the exclusion of essential information, thereby increasing hallucination risks and undermining the quality of generated responses. In this paper, we propose PankRAG, a framework designed to capture and resolve the latent relationships within complex queries that prior methods overlook. It achieves this through a synergistic combination of a globally-aware hierarchical resolution pathway and a dependency-aware reranking mechanism. PankRAG first generates a globally aware resolution pathway that captures parallel and progress relationships, guiding LLMs to resolve queries through a hierarchical reasoning path. Additionally, its dependency-aware reranking mechanism utilizes resolved sub-question dependencies to augment and validate the retrieved content of the current unresolved sub-question. Experimental results demonstrate that PankRAG consistently outperforms existing state-of-the-art methods, underscoring its generalizability.

cs.CL↗

arXiv2026

Production of Light Dark Particles from Nonlinear Compton Scattering Between Intense Laser and Muon or Proton Beam

The laser of an intense electromagnetic field promotes the studies of strong-field particle physics in high-intensity frontier. Particle accelerator facilities in the world produce high-quality muon and proton beams. In this work, we propose the nonlinear Compton scattering to light dark particles through the collision between intense laser pulse and muon or proton beam. We take light dark photon and axion-like particle as illustrative dark particles. The cross sections of relevant nonlinear Compton scattering to dark photon or axion-like particle are calculated. We also analyze the background processes with missing neutrinos. The prospective sensitivity shows that the laser-induced process provides a complementary and competitive search of new invisible particles lighter than about 1 MeV.

hep-ph↗

arXiv2026

Can Large Language Models Identify Implicit Suicidal Ideation? An Empirical Evaluation

We present a comprehensive evaluation framework for assessing Large Language Models' (LLMs) capabilities in suicide prevention, focusing on two critical aspects: the Identification of Implicit Suicidal ideation (IIS) and the Provision of Appropriate Supportive responses (PAS). We introduce \ourdata, a novel dataset of 1,308 test cases built upon psychological frameworks including D/S-IAT and Negative Automatic Thinking, alongside real-world scenarios. Through extensive experiments with 8 widely used LLMs under different contextual settings, we find that current models struggle significantly with detecting implicit suicidal ideation and providing appropriate support, highlighting crucial limitations in applying LLMs to mental health contexts. Our findings underscore the need for more sophisticated approaches in developing and evaluating LLMs for sensitive psychological applications.

cs.CL↗

arXiv2026

EverMemOS: A Self-Organizing Memory Operating System for Structured Long-Horizon Reasoning

Large Language Models (LLMs) are increasingly deployed as long-term interactive agents, yet their limited context windows make it difficult to sustain coherent behavior over extended interactions. Existing memory systems often store isolated records and retrieve fragments, limiting their ability to consolidate evolving user states and resolve conflicts. We introduce EverMemOS, a self-organizing memory operating system that implements an engram-inspired lifecycle for computational memory. Episodic Trace Formation converts dialogue streams into MemCells that capture episodic traces, atomic facts, and time-bounded Foresight signals. Semantic Consolidation organizes MemCells into thematic MemScenes, distilling stable semantic structures and updating user profiles. Reconstructive Recollection performs MemScene-guided agentic retrieval to compose the necessary and sufficient context for downstream reasoning. Experiments on LoCoMo and LongMemEval show that EverMemOS achieves state-of-the-art performance on memory-augmented reasoning tasks. We further report a profile study on PersonaMem v2 and qualitative case studies illustrating chat-oriented capabilities such as user profiling and Foresight. Code is available at https://github.com/EverMind-AI/EverMemOS.

cs.AI↗

arXiv2026

Practical Poisoning Attacks against Retrieval-Augmented Generation

Large language models (LLMs) have demonstrated impressive natural language processing abilities but face challenges such as hallucination and outdated knowledge. Retrieval-Augmented Generation (RAG) has emerged as a state-of-the-art approach to mitigate these issues. While RAG enhances LLM outputs, it remains vulnerable to poisoning attacks. Recent studies show that injecting poisoned text into the knowledge database can compromise RAG systems, but most existing attacks assume that the attacker can insert a sufficient number of poisoned texts per query to outnumber correct-answer texts in retrieval, an assumption that is often unrealistic. To address this limitation, we propose CorruptRAG, a practical poisoning attack against RAG systems in which the attacker injects only a single poisoned text, enhancing both feasibility and stealth. Extensive experiments conducted on multiple large-scale datasets demonstrate that CorruptRAG achieves higher attack success rates than existing baselines.

cs.CR↗

arXiv2026

Solutions to axion electrodynamics with electric-magnetic duality in supersymmetric Seiberg-Witten theory

Axion and magnetic monopole are among the most fascinating candidates for physics beyond the Standard Model. The potential connection between axion and magnetic monopole stems from the Witten effect and is revealed by non-standard axion electrodynamics. Non-standard axion electrodynamics under electric-magnetic duality modifies conventional axion Maxwell equations and motivates intriguing axion-photon phenomenology. A calculable ultraviolet model of Peccei-Quinn axion coupled to magnetic monopoles and electric charges was proposed based on $\mathcal{N}=2$ supersymmetric Seiberg-Witten (SW) theory with manifest electric-magnetic duality. In this work, we aim to investigate the solutions to the non-linear axion electrodynamics from SW axion model and propose relevant detection strategies for non-trivial axion-photon couplings. Based on the infrared Lagrangian of SW axion, we derive the electromagetic (EM) equations of motion. We also analyze the moduli space coordinate in SW theory and find out the reliabe parameter space. We then solve the resultant axion Maxwell equations with an external EM field. The observable axion-induced EM fields are obtained analytically and then numerically computed. Finally, we propose the detection strategy with an LC circuit and show the prospective sensitivity to SW axion-photon couplings.

hep-ph↗

arXiv2025

ESG Beliefs of Large Language Models: Evidence and Impact

We examine whether large language models (LLMs) hold systematic beliefs about environmental, social, and governance (ESG) issues and how these beliefs compare with-and potentially influence-those of human market participants. Based on established surveys originally administered to professional and retail investors, we show that major LLMs exhibit a strong pro-ESG orientation. Compared with human investors, LLMs assign greater financial relevance for ESG performance, expect larger return premia for high-ESG firms, and display a stronger willingness to sacrifice financial returns for ESG improvements. These preferences are highly uniform and values-driven, in contrast to heterogeneous human views. Using a large dataset of analyst reports, we further show that sell-side analysts become significantly more optimistic about high-ESG firms after adopting LLMs for research. Our findings reveal that LLMs embed distinct, coherent ESG beliefs and that these beliefs can shape human judgments, highlighting a new channel through which AI adoption may influence financial markets.

physics.soc-ph↗

arXiv2025

AutoRec: Accelerating Loss Recovery for Live Streaming in a Multi-Supplier Market

Due to the limited permissions for upgrading dualside (i.e., server-side and client-side) loss tolerance schemes from the perspective of CDN vendors in a multi-supplier market, modern large-scale live streaming services are still using the automatic-repeat-request (ARQ) based paradigm for loss recovery, which only requires server-side modifications. In this paper, we first conduct a large-scale measurement study with up to 50 million live streams. We find that loss shows dynamics and live streaming contains frequent on-off mode switching in the wild. We further find that the recovery latency, enlarged by the ubiquitous retransmission loss, is a critical factor affecting live streaming's client-side QoE (e.g., video freezing). We then propose an enhanced recovery mechanism called AutoRec, which can transform the disadvantages of on-off mode switching into an advantage for reducing loss recovery latency without any modifications on the client side. AutoRec allows users to customize overhead tolerance and recovery latency tolerance and adaptively adjusts strategies as the network environment changes to ensure that recovery latency meets user demands whenever possible while keeping overhead under control. We implement AutoRec upon QUIC and evaluate it via testbed and real-world commercial services deployments. The experimental results demonstrate the practicability and profitability of AutoRec.

cs.NI↗

arXiv2025

Practical Framework for Privacy-Preserving and Byzantine-robust Federated Learning

Federated Learning (FL) allows multiple clients to collaboratively train a model without sharing their private data. However, FL is vulnerable to Byzantine attacks, where adversaries manipulate client models to compromise the federated model, and privacy inference attacks, where adversaries exploit client models to infer private data. Existing defenses against both backdoor and privacy inference attacks introduce significant computational and communication overhead, creating a gap between theory and practice. To address this, we propose ABBR, a practical framework for Byzantine-robust and privacy-preserving FL. We are the first to utilize dimensionality reduction to speed up the private computation of complex filtering rules in privacy-preserving FL. Additionally, we analyze the accuracy loss of vector-wise filtering in low-dimensional space and introduce an adaptive tuning strategy to minimize the impact of malicious models that bypass filtering on the global model. We implement ABBR with state-of-the-art Byzantine-robust aggregation rules and evaluate it on public datasets, showing that it runs significantly faster, has minimal communication overhead, and maintains nearly the same Byzantine-resilience as the baselines.

cs.CR↗

arXiv2025

Performant Synchronization in Geo-Distributed Databases

The deployment of databases across geographically distributed regions has become increasingly critical for ensuring data reliability and scalability. Recent studies indicate that distributed databases exhibit significantly higher latency than single-node databases, primarily due to consensus protocols maintaining data consistency across multiple nodes. We argue that synchronization cost constitutes the primary bottleneck for distributed databases, which is particularly pronounced in wide-area networks (WAN). Fortunately, we identify opportunities to optimize synchronization costs in real production environments: (1) network clustering phenomena, (2) triangle inequality violations in transmission, and (3) redundant data transfers. Based on these observations, we propose GeoCoCo, a synchronization acceleration framework for cross-region distributed databases. First, GeoCoCo presents a group rescheduling strategy that adapts to real-time network conditions to maximize WAN transmission efficiency. Second, GeoCoCo introduces a task-preserving data filtering method that reduces data volume transmitted over the WAN. Finally, GeoCoCo develops a consistency-guaranteed transmission framework integrating grouping and pruning. Extensive evaluations in both trace-driven simulations and real-world deployments demonstrate that GeoCoCo reduces synchronization cost-primarily by lowering WAN bandwidth usage-by up to 40.3%, and increases system throughput by up to 14.1% in GeoGauss.

cs.DB↗