Searcharxiv⌕ Search

arXiv subjects

Chen Shen

Publications and source records attributed to Chen Shen.

At least 73 records · Page 4Linked to original sources

Reputation as a Solution to Cooperation Collapse in LLM-based MASs

Cooperation has long been a fundamental topic in both human society and AI systems. However, recent studies indicate that the collapse of cooperation may emerge in multi-agent systems (MASs) driven by large language models (LLMs). To address this challenge, we explore reputation systems as a remedy. We propose RepuNet, a dynamic, dual-level reputation framework that models both agent-level reputation dynamics and system-level network evolution. Specifically, driven by direct interactions and indirect gossip, agents form reputations for both themselves and their peers, and decide whether to connect or disconnect other agents for future interactions. Through three distinct scenarios, we show that RepuNet effectively avoids cooperation collapse, promoting and sustaining cooperation in LLM-based MASs. Moreover, we find that reputation systems can give rise to rich emergent behaviors in LLM-based MASs, such as the formation of cooperative clusters, the social isolation of exploitative agents, and the preference for sharing positive gossip rather than negative ones. The GitHub repository for our project can be accessed via the following link: https://github.com/RGB-0000FF/RepuNet.

cs.AI↗

Hallucination Begins Where Saliency Drops

Recent studies have examined attention dynamics in large vision-language models (LVLMs) to detect hallucinations. However, existing approaches remain limited in reliably distinguishing hallucinated from factually grounded outputs, as they rely solely on forward-pass attention patterns and neglect gradient-based signals that reveal how token influence propagates through the network. To bridge this gap, we introduce LVLMs-Saliency, a gradient-aware diagnostic framework that quantifies the visual grounding strength of each output token by fusing attention weights with their input gradients. Our analysis uncovers a decisive pattern: hallucinations frequently arise when preceding output tokens exhibit low saliency toward the prediction of the next token, signaling a breakdown in contextual memory retention. Leveraging this insight, we propose a dual-mechanism inference-time framework to mitigate hallucinations: (1) Saliency-Guided Rejection Sampling (SGRS), which dynamically filters candidate tokens during autoregressive decoding by rejecting those whose saliency falls below a context-adaptive threshold, thereby preventing coherence-breaking tokens from entering the output sequence; and (2) Local Coherence Reinforcement (LocoRE), a lightweight, plug-and-play module that strengthens attention from the current token to its most recent predecessors, actively counteracting the contextual forgetting behavior identified by LVLMs-Saliency. Extensive experiments across multiple LVLMs demonstrate that our method significantly reduces hallucination rates while preserving fluency and task performance, offering a robust and interpretable solution for enhancing model reliability. Code is available at: https://github.com/zhangbaijin/LVLMs-Saliency

cs.CV↗

A Hybrid BLE-Wi-Fi Communication Architecture for Adaptive Imaging in Wireless Capsule Endoscopy

Wireless capsule endoscopy (WCE) is fundamentally constrained by limited wireless bandwidth, resulting in low imaging resolution and frame rate, which can cause motion blur and missed lesions. Although adaptive frame-rate schemes have been explored to accommodate transient gastrointestinal (GI) motility, these approaches typically require sacrificing image resolution. The use of higher-frequency communication bands is further limited by increased tissue attenuation. To address these challenges, we propose a hybrid Bluetooth Low Energy (BLE) and WiFi communication architecture that combines the low-power operation of BLE with the high data throughput of WiFi. We systematically evaluate the performance of BLE and WiFi under tissue-mimicking conditions by measuring throughput, received signal strength indicator (RSSI), and power consumption. The results demonstrate that amplified BLE with an adaptive transmission power control strategy provides a stable frame rate at low power consumption, while 2.4 GHz WiFi operating in station mode is the most suitable high-throughput communication configuration for WCE. Compared with WiFi, BLE reduces power consumption by approximately ten times, whereas WiFi achieves up to ten times higher throughput. To reconcile these complementary trade-offs, we further introduce a hybrid system with a frame-boundary-synchronized switching mechanism to ensure lossless data transmission during BLE and WiFi transitions. Experimental results show that the switching latency from BLE to WiFi is approximately 92.66 ms, which is longer than the WiFi-to-BLE switching latency of 15.49 ms when transmitting 10 kB image payloads. Overall, the proposed hybrid BLE and WiFi system enables robust, lossless, and energy-efficient mode switching, supports adaptive imaging, and advances the development of next-generation autonomous WCE platforms.

eess.SY↗

Evolutionary Dynamics of Reputation-Based Voluntary Prisoner's Dilemma Games

Cooperation underlies many natural and artificial systems. While voluntary participation can sustain cooperation without informational assumptions, real interactions are rarely anonymous, leaving the joint effects of participation and reputation insufficiently understood. We propose a reputation-based voluntary Prisoner's Dilemma in which agents incur a monitoring cost to inspect opponents and decide whether to exit an interaction for a fixed incentive to avoid exploitation or to default to cooperation or defection. We show that reputation-conditioned exit generates multiple coexistence pathways that sustain cooperation across population structures. In well-mixed populations, cooperation persists through stable mixed coexistence, whereas in structured populations, exit-incentive-dependent regimes emerge, including local cyclic dominance and persistent oscillations. Together, these results extend voluntary participation frameworks and underscore the role of exit-incentive design in cooperative multi-agent systems.

physics.soc-ph↗

A scalable near-visible integrated photon-pair source for satellite quantum science

Quantum state distribution over vast distances is essential for global-scale quantum networks and fundamental test of quantum physics at space scale. While satellite platforms have demonstrated thousand-kilometer entanglement distribution, quantum key distribution and quantum teleportation with ground, future constellations and deep-space missions demand photon sources that are robust, compact, and power-efficient. Integrated photonics offers a scalable solution, yet a critical spectral gap persists. Although telecom-band integrated photon-pair sources are well established, near-visible photons offer distinct advantages for satellite-to-ground links by mitigating diffraction loss and maximizing the collection efficiency of optical telescopes. Scalable integrated sources in this regime have remained elusive due to the fundamental challenge of achieving anomalous dispersion in materials transparent at visible wavelengths. Here we bridge this gap by demonstrating an integrated near-visible photon-pair source based on a wide-bandgap, ultralow-loss, silicon nitride (Si$_3$N$_4$) microresonator. By engineering the dispersion of higher-order waveguide modes, we overcome the intrinsic normal dispersion limit to achieve efficient phase matching. The device exhibits a spectral brightness of 4.87$\times$10$^7$ pairs/s/mW$^2$/GHz and a narrow photon linewidth of 357 MHz. We report high-purity heralded single-photon generation with a heralding rate up to 2.3 MHz and a second-order correlation function as low as 0.0041. Furthermore, we observe energy-time entanglement with 98.4% interference visibility, violating the CHSH limit even at flux exceeding 40.6 million pairs/s. Combined with the proven radiation hardness of Si$_3$N$_4$, this source constitutes a flight-ready hardware foundation for daylight quantum communications and protocols requiring on-orbit multiphoton interference.

quant-ph↗

Bridging the Knowledge Void: Inference-time Acquisition of Unfamiliar Programming Languages for Coding Tasks

The proficiency of Large Language Models (LLMs) in coding tasks is often a reflection of their extensive pre-training corpora, which typically collapses when confronted with previously unfamiliar programming languages. Departing from data-intensive finetuning, we investigate the paradigm of Inference-time Language Acquisition (ILA), where an LLM masters an unfamiliar language through dynamic interaction with limited external resources. In this paper, we propose ILA-agent, a general ILA framework that equips LLMs with a set of behavioral primitives. By modeling essential human-like behaviors as a suite of tools, ILA-agent enables LLMs to incrementally explore, apply, and verify language knowledge through structured interactions with the official documentation and execution environment. To provide a rigorous evaluation in a low-resource setting, we construct Cangjie-bench, a multi-task benchmark based on the novel statically-typed language Cangjie. We instantiate ILA-agent for Cangjie and evaluate its performance across code generation, translation, and program repair tasks. Results using diverse LLMs demonstrate that ILA-agent significantly outperforms retrieval-augmented baselines. Further analysis of agent trajectories characterizes the emergent behavior patterns while highlighting persisting performance gaps.

cs.CL↗

Distribution-Aligned Sequence Distillation for Superior Long-CoT Reasoning

In this report, we introduce DASD-4B-Thinking, a lightweight yet highly capable, fully open-source reasoning model. It achieves SOTA performance among open-source models of comparable scale across challenging benchmarks in mathematics, scientific reasoning, and code generation -- even outperforming several larger models. We begin by critically reexamining a widely adopted distillation paradigm in the community: SFT on teacher-generated responses, also known as sequence-level distillation. Although a series of recent works following this scheme have demonstrated remarkable efficiency and strong empirical performance, they are primarily grounded in the SFT perspective. Consequently, these approaches focus predominantly on designing heuristic rules for SFT data filtering, while largely overlooking the core principle of distillation itself -- enabling the student model to learn the teacher's full output distribution so as to inherit its generalization capability. Specifically, we identify three critical limitations in current practice: i) Inadequate representation of the teacher's sequence-level distribution; ii) Misalignment between the teacher's output distribution and the student's learning capacity; and iii) Exposure bias arising from teacher-forced training versus autoregressive inference. In summary, these shortcomings reflect a systemic absence of explicit teacher-student interaction throughout the distillation process, leaving the essence of distillation underexploited. To address these issues, we propose several methodological innovations that collectively form an enhanced sequence-level distillation training pipeline. Remarkably, DASD-4B-Thinking obtains competitive results using only 448K training samples -- an order of magnitude fewer than those employed by most existing open-source efforts. To support community research, we publicly release our models and the training dataset.

cs.LG↗

Federated Learning with Partially Labeled Data: A Conditional Distillation Approach

In medical imaging, developing generalized segmentation models that can handle multiple organs and lesions is crucial. However, the scarcity of fully annotated datasets and strict privacy regulations present significant barriers to data sharing. Federated Learning (FL) allows decentralized model training, but existing FL methods often struggle with partial labeling, leading to model divergence and catastrophic forgetting. We propose ConDistFL, a novel FL framework incorporating conditional distillation to address these challenges. ConDistFL enables effective learning from partially labeled datasets, significantly improving segmentation accuracy across distributed and non-uniform datasets. In addition to its superior segmentation performance, ConDistFL maintains computational and communication efficiency, ensuring its scalability for real-world applications. Furthermore, ConDistFL demonstrates remarkable generalizability, significantly outperforming existing FL methods in out-of-federation tests, even adapting to unseen contrast phases (e.g., non-contrast CT images) in our experiments. Extensive evaluations on 3D CT and 2D chest X-ray datasets show that ConDistFL is an efficient, adaptable solution for collaborative medical image segmentation in privacy-constrained settings.

cs.CV↗

GeoCAD: Local Geometry-Controllable CAD Generation with Large Language Models

Local geometry-controllable computer-aided design (CAD) generation aims to modify local parts of CAD models automatically, enhancing design efficiency. It also ensures that the shapes of newly generated local parts follow user-specific geometric instructions (e.g., an isosceles right triangle or a rectangle with one corner cut off). However, existing methods encounter challenges in achieving this goal. Specifically, they either lack the ability to follow textual instructions or are unable to focus on the local parts. To address this limitation, we introduce GeoCAD, a user-friendly and local geometry-controllable CAD generation method. Specifically, we first propose a complementary captioning strategy to generate geometric instructions for local parts. This strategy involves vertex-based and VLLM-based captioning for systematically annotating simple and complex parts, respectively. In this way, we caption $\sim$221k different local parts in total. In the training stage, given a CAD model, we randomly mask a local part. Then, using its geometric instruction and the remaining parts as input, we prompt large language models (LLMs) to predict the masked part. During inference, users can specify any local part for modification while adhering to a variety of predefined geometric instructions. Extensive experiments demonstrate the effectiveness of GeoCAD in generation quality, validity and text-to-CAD consistency. Code will be available at https://github.com/Zhanwei-Z/GeoCAD.

cs.CV↗

AIPOM: Agent-aware Interactive Planning for Multi-Agent Systems

Large language models (LLMs) are being increasingly used for planning in orchestrated multi-agent systems. However, existing LLM-based approaches often fall short of human expectations and, critically, lack effective mechanisms for users to inspect, understand, and control their behaviors. These limitations call for enhanced transparency, controllability, and human oversight. To address this, we introduce AIPOM, a system supporting human-in-the-loop planning through conversational and graph-based interfaces. AIPOM enables users to transparently inspect, refine, and collaboratively guide LLM-generated plans, significantly enhancing user control and trust in multi-agent workflows. Our code and demo video are available at https://github.com/megagonlabs/aipom.

cs.HC↗

Extending the capillary wave model to include the effect of bending rigidity: X-ray reflectivity and diffuse scattering

The surface roughness of a thin film at a liquid interface exhibits contributions of thermally excited fluctuations. This thermal roughness depends on temperature (T), surface tension ($γ$) and elastic material properties, specifically the bending modulus ($κ$) of the film. A non-zero $κ$ suppresses the thermal roughness at small length scales compared to an interface with zero $κ$, as expressed by the power spectral density (PSD) of the thermal roughness. The description of the X-ray scattering of the standard Capillary Wave Model (CWM), that is valid for zero $κ$, is extended to include the effect of $κ$. The extended CWM (eCWM) provides a single analytical form for both the specular XRR and the diffuse scattering around the specular reflection, and recovers the expression of the CWM at its zero $κ$ limit. This new theoretical approach enables the use of single-shot grazing incidence X-ray off-specular scattering (GIXOS) measurements for characterizing the structure of thin films on a liquid surface. The eCWM analysis approach decouples the thermal roughness factor from the surface scattering signal, providing direct access to the intrinsic surface-normal structure of the film and its bending modulus. Moreover, the eCWM facilitates the calculation of reflectivity at any desired resolution (pseudo XRR approach). The transformation into pseudo XRR provides the benefit of using widely available XRR software to perform GIXOS analysis. The extended range of the vertical scattering vector (Qz) available with the GIXOS-pseudo XRR approach allows for a higher spatial resolution than with conventional XRR. Experimental results are presented for various lipid systems, showing strong agreement between conventional specular XRR and pseudo XRR methods. This agreement validates the proposed approach and highlights its utility for analyzing soft, thin films.

cond-mat.soft↗

SalaMAnder: Shapley-based Mathematical Expression Attribution and Metric for Chain-of-Thought Reasoning

Chain-of-Thought (CoT) prompting enhances the math reasoning capability of large language models (LLMs) to a large margin. However, the mechanism underlying such improvements remains unexplored. In this paper, we present \textbf{SalaMAnder} (\textbf{S}h\textbf{a}p\textbf{l}ey-b\textbf{a}sed \textbf{M}athematical Expression \textbf{A}ttribution a\textbf{nd} M\textbf{e}t\textbf{r}ic), a theoretically grounded methodology as well as a mathematically rigorous evaluation metric for quantifying component-level contributions in few-shot CoT reasoning. Concretely, we leverage the Shapley value for mathematical expression attribution and develop an efficient stratified sampling algorithm that significantly reduces the computational complexity. Besides, we develop the \textbf{CoSP} (\textbf{C}ardinality \textbf{o}f \textbf{S}hapley \textbf{P}ositives) metric through covariance analysis. Comprehensive validation across popular LLM models and diverse mathematical benchmarks demonstrates that the CoSP metric within our SalaMAnder framework exhibits a robust monotonic correlation with model performance, not only providing theoretical explanations for the empirical success of existing few-shot CoT but also establishing mathematically rigorous principles for prompt construction optimization. Furthermore, we verify the reliability of the explanation, based on which we unify the insights of previous work.

cs.AI↗

High-throughput screening of spin Hall conductivity in 2D materials

Two-dimensional (2D) materials with large spin Hall effect (SHE) have attracted significant attention due to their potential applications in next-generation spintronic devices. In this work, we perform high-throughput (HTP) calculations to obtain the spin Hall conductivity (SHC) of 4486 non-magnetic compounds in the \texttt{2Dmatpedia} database and identify six materials with SHC exceeding $500\,(\hbar/e)\,(\mathrm{S/cm})$, surpassing those of previously known materials. Detailed analysis reveals that the significant SHC can be attributed to spin-orbit coupling (SOC)-induced gap openings at Dirac-like band crossings. Additionally, the presence of mirror symmetry further enhances the SHC. Beyond the high-SHC materials, 57 topological insulators with quantized SHCs have also been identified. Our work enables rapid screening and paves the way for experimental validation, potentially accelerating the discovery of novel 2D materials optimized for spintronics applications.

cond-mat.mtrl-sci↗

Signal Lasso with Non-Convex Penalties for Efficient Network Reconstruction and Topology Inference

Inferring network structures remains an interesting question for its importance on the understanding and controlling collective dynamics of complex systems. The existing shrinking methods such as Lasso-type estimation can not suitably reveal such property. A new method recently suggested, called by {\it signal lasso} (or its updating version: adaptive signal lasso) was proposed to solve the network reconstruction problem, where the signal parameter can be shrunk to either 0 or 1 in two different directions. The signal lasso or adaptive signal lasso employed the additive penalty of signal and non-signal terms which is a convex function and easily to complementation in computation. However their methods need tuning the one or two parameters to find an optimal solution, which is time cost for large size network. In this paper we propose new signal lasso method based on two penalty functions to estimate the signal parameter and uncovering network topology in complex network with a small amount of observations. The penalty functions we introduced are non-convex function, thus coordinate descent algorithms are suggested. We find in this method the tuning parameter can be set to a large enough values such that the signal parameter can be completely shrunk either 0 or 1. The extensive simulations are conducted in linear regression models with different assumptions, the evolutionary-game-based dynamic model and Kuramoto model of synchronization problem. The advantage and disadvantage of each method are fully discussed in various conditions. Finally a real example comes from behavioral experiment is used for illustration. Our results show that signal lasso with non-convex penalties is effective and fast in estimating signal parameters in linear regression model.

math.ST↗

FEAT: A Multi-Agent Forensic AI System with Domain-Adapted Large Language Model for Automated Cause-of-Death Analysis

Forensic cause-of-death determination faces systemic challenges, including workforce shortages and diagnostic variability, particularly in high-volume systems like China's medicolegal infrastructure. We introduce FEAT (ForEnsic AgenT), a multi-agent AI framework that automates and standardizes death investigations through a domain-adapted large language model. FEAT's application-oriented architecture integrates: (i) a central Planner for task decomposition, (ii) specialized Local Solvers for evidence analysis, (iii) a Memory & Reflection module for iterative refinement, and (iv) a Global Solver for conclusion synthesis. The system employs tool-augmented reasoning, hierarchical retrieval-augmented generation, forensic-tuned LLMs, and human-in-the-loop feedback to ensure legal and medical validity. In evaluations across diverse Chinese case cohorts, FEAT outperformed state-of-the-art AI systems in both long-form autopsy analyses and concise cause-of-death conclusions. It demonstrated robust generalization across six geographic regions and achieved high expert concordance in blinded validations. Senior pathologists validated FEAT's outputs as comparable to those of human experts, with improved detection of subtle evidentiary nuances. To our knowledge, FEAT is the first LLM-based AI agent system dedicated to forensic medicine, offering scalable, consistent death certification while maintaining expert-level rigor. By integrating AI efficiency with human oversight, this work could advance equitable access to reliable medicolegal services while addressing critical capacity constraints in forensic systems.

cs.AI↗

Non-participant externalities reshape the evolution of altruistic punishment

While voluntary participation is a key mechanism that enables altruistic punishment to emerge, its explanatory power typically rests on the common assumption that non-participants have no impact on the public good. Yet, given the decentralized nature of voluntary participation, opting out does not necessarily preclude individuals from influencing the public good. Here, we revisit the role of voluntary participation by allowing non-participants to exert either positive or negative impacts on the public good. Using evolutionary analysis in a well-mixed finite population, we find that positive externalities from non-participants lower the synergy threshold required for altruistic punishment to dominate. In contrast, negative externalities raise this threshold, making altruistic punishment harder to sustain. Notably, when non-participants have positive impacts, altruistic punishment thrives only if non-participation is incentivized, whereas under negative impacts, it can persist even when non-participation is discouraged. Our findings reveal that efforts to promote altruistic punishment must account for the active role of non-participants, whose influence can make or break collective outcomes.

physics.soc-ph↗

Enhancing Chain-of-Thought Reasoning with Critical Representation Fine-tuning

Representation Fine-tuning (ReFT), a recently proposed Parameter-Efficient Fine-Tuning (PEFT) method, has attracted widespread attention for significantly improving parameter efficiency by editing representation space alone. In this work, we investigate applying ReFT to complex reasoning tasks. However, directly using the native ReFT method, which modifies fixed representations at the beginning and end of each layer, yields suboptimal performance, as these fixed-position representations have uncertain impact on the outputs. We observe that, in complex reasoning tasks, there often exist certain critical representations. These representations either integrate significant information from preceding layers or regulate subsequent layer representations. Through layer-by-layer propagation, they exert a substantial influence on the final output. Naturally, fine-tuning these critical representations has the potential to greatly enhance reasoning performance. Building upon these insights, we propose Critical Representation Fine-Tuning (CRFT), a novel method that identifies and optimizes these critical representations through information flow analysis. CRFT operates within a supervised learning framework, dynamically optimizing critical representations in a low-rank linear subspace while freezing the base model. The effectiveness and efficiency of our method are validated across eight benchmarks for arithmetic and commonsense reasoning, using LLaMA and Mistral model families. Furthermore, our method also adapts effectively to few-shot settings, boosting one-shot accuracy by 16.4%. Our work highlights the untapped potential of representation-level optimization for CoT reasoning, offering a lightweight yet powerful alternative to traditional PEFT methods.

cs.CL↗

Network reciprocity turns cheap talk into a force for cooperation

Non-binding communication is common in daily life and crucial for fostering cooperation, even though it has no direct payoff consequences. However, despite robust empirical evidence, its evolutionary basis remains poorly understood. Here, we develop a game-theoretic model in which individuals can signal an intention to cooperate before playing a Donation game. Strategies differ in how they respond to these signals, ranging from unconditional to conditional types, with the latter incurring a cognitive cost for deliberation. Through evolutionary analysis, we show that non-binding communication alone cannot sustain cooperation in well-mixed, anonymous populations, consistent with empirical observations. In contrast, structured populations support the emergence of cooperation, with conditional cooperators acting as catalysts that protect unconditional cooperators through context-dependent patterns of cyclic dominance. These findings offer an evolutionary explanation for how non-binding communication promotes cooperation and provide a modelling framework for exploring its effects in diverse social settings.

physics.soc-ph↗