arXiv subjects

Jiaxin Liu

Publications and source records attributed to Jiaxin Liu.

At least 55 records · Page 3Linked to original sources

arXiv2026

Fair Algorithms with Probing for Multi-Agent Multi-Armed Bandits

We propose a multi-agent multi-armed bandit (MA-MAB) framework aimed at ensuring fair outcomes across agents while maximizing overall system performance. A key challenge in this setting is decision-making under limited information about arm rewards. To address this, we introduce a novel probing framework that strategically gathers information about selected arms before allocation. In the offline setting, where reward distributions are known, we leverage submodular properties to design a greedy probing algorithm with a provable performance bound. For the more complex online setting, we develop an algorithm that achieves sublinear regret while maintaining fairness. Extensive experiments on synthetic and real-world datasets show that our approach outperforms baseline methods, achieving better fairness and efficiency.

cs.LG↗

arXiv2026

A multimodal vision foundation model for generalizable knee pathology

Musculoskeletal disorders represent a leading cause of global disability, creating an urgent demand for precise interpretation of medical imaging. Current artificial intelligence (AI) approaches in orthopedics predominantly rely on task-specific, supervised learning paradigms. These methods are inherently fragmented, require extensive annotated datasets, and often lack generalizability across different modalities and clinical scenarios. The development of foundation models in this field has been constrained by the scarcity of large-scale, curated, and open-source musculoskeletal datasets. To address these challenges, we introduce OrthoFoundation, a multimodal vision foundation model optimized for musculoskeletal pathology. We constructed a pre-training dataset of 1.2 million unlabeled knee X-ray and MRI images from internal and public databases. Utilizing a Dinov3 backbone, the model was trained via self-supervised contrastive learning to capture robust radiological representations. OrthoFoundation achieves state-of-the-art (SOTA) performance across 14 downstream tasks. It attained superior accuracy in X-ray osteoarthritis diagnosis and ranked first in MRI structural injury detection. The model demonstrated remarkable label efficiency, matching supervised baselines using only 50% of labeled data. Furthermore, despite being pre-trained on knee images, OrthoFoundation exhibited exceptional cross-anatomy generalization to the hip, shoulder, and ankle. OrthoFoundation represents a significant advancement toward general-purpose AI for musculoskeletal imaging. By learning fundamental, joint-agnostic radiological semantics from large-scale multimodal data, it overcomes the limitations of conventional models, which provides a robust framework for reducing annotation burdens and enhancing diagnostic accuracy in clinical practice.

cs.CV↗

arXiv2025

HeartBench: Probing Core Dimensions of Anthropomorphic Intelligence in LLMs

While Large Language Models (LLMs) have achieved remarkable success in cognitive and reasoning benchmarks, they exhibit a persistent deficit in anthropomorphic intelligence-the capacity to navigate complex social, emotional, and ethical nuances. This gap is particularly acute in the Chinese linguistic and cultural context, where a lack of specialized evaluation frameworks and high-quality socio-emotional data impedes progress. To address these limitations, we present HeartBench, a framework designed to evaluate the integrated emotional, cultural, and ethical dimensions of Chinese LLMs. Grounded in authentic psychological counseling scenarios and developed in collaboration with clinical experts, the benchmark is structured around a theory-driven taxonomy comprising five primary dimensions and 15 secondary capabilities. We implement a case-specific, rubric-based methodology that translates abstract human-like traits into granular, measurable criteria through a ``reasoning-before-scoring'' evaluation protocol. Our assessment of 13 state-of-the-art LLMs indicates a substantial performance ceiling: even leading models achieve only 60% of the expert-defined ideal score. Furthermore, analysis using a difficulty-stratified ``Hard Set'' reveals a significant performance decay in scenarios involving subtle emotional subtexts and complex ethical trade-offs. HeartBench establishes a standardized metric for anthropomorphic AI evaluation and provides a methodological blueprint for constructing high-quality, human-aligned training data.

cs.CL↗

arXiv2025

TEMP: A Memory Efficient Physical-aware Tensor Partition-Mapping Framework on Wafer-scale Chips

Large language models (LLMs) demand significant memory and computation resources. Wafer-scale chips (WSCs) provide high computation power and die-to-die (D2D) bandwidth but face a unique trade-off between on-chip memory and compute resources due to limited wafer area. Therefore, tensor parallelism strategies for wafer should leverage communication advantages while maintaining memory efficiency to maximize WSC performance. However, existing approaches fail to address these challenges. To address these challenges, we propose the tensor stream partition paradigm (TSPP), which reveals an opportunity to leverage WSCs' abundant communication bandwidth to alleviate stringent on-chip memory constraints. However, the 2D mesh topology of WSCs lacks long-distance and flexible interconnects, leading to three challenges: 1) severe tail latency, 2) prohibitive D2D traffic contention, and 3) intractable search time for optimal design. We present TEMP, a framework for LLM training on WSCs that leverages topology-aware tensor-stream partition, traffic-conscious mapping, and dual-level wafer solving to overcome hardware constraints and parallelism challenges. These integrated approaches optimize memory efficiency and throughput, unlocking TSPP's full potential on WSCs. Evaluations show TEMP achieves 1.7x average throughput improvement over state-of-the-art LLM training systems across various models.

cs.AR↗

arXiv2025

Generalized Geometry Encoding Volume for Real-time Stereo Matching

Real-time stereo matching methods primarily focus on enhancing in-domain performance but often overlook the critical importance of generalization in real-world applications. In contrast, recent stereo foundation models leverage monocular foundation models (MFMs) to improve generalization, but typically suffer from substantial inference latency. To address this trade-off, we propose Generalized Geometry Encoding Volume (GGEV), a novel real-time stereo matching network that achieves strong generalization. We first extract depth-aware features that encode domain-invariant structural priors as guidance for cost aggregation. Subsequently, we introduce a Depth-aware Dynamic Cost Aggregation (DDCA) module that adaptively incorporates these priors into each disparity hypothesis, effectively enhancing fragile matching relationships in unseen scenes. Both steps are lightweight and complementary, leading to the construction of a generalized geometry encoding volume with strong generalization capability. Experimental results demonstrate that our GGEV surpasses all existing real-time methods in zero-shot generalization capability, and achieves state-of-the-art performance on the KITTI 2012, KITTI 2015, and ETH3D benchmarks.

cs.CV↗

arXiv2025

Seeing before Observable: Potential Risk Reasoning in Autonomous Driving via Vision Language Models

Ensuring safety remains a key challenge for autonomous vehicles (AVs), especially in rare and complex scenarios. One critical but understudied aspect is the \textbf{potential risk} situations, where the risk is \textbf{not yet observable} but can be inferred from subtle precursors, such as anomalous behaviors or commonsense violations. Recognizing these precursors requires strong semantic understanding and reasoning capabilities, which are often absent in current AV systems due to the scarcity of such cases in existing driving or risk-centric datasets. Moreover, current autonomous driving accident datasets often lack annotations of the causal reasoning chains behind incidents, which are essential for identifying potential risks before they become observable. To address these gaps, we introduce PotentialRiskQA, a novel vision-language dataset designed for reasoning about potential risks prior to observation. Each sample is annotated with structured scene descriptions, semantic precursors, and inferred risk outcomes. Based on this dataset, we further propose PR-Reasoner, a vision-language-model-based framework tailored for onboard potential risk reasoning. Experimental results show that fine-tuning on PotentialRiskQA enables PR-Reasoner to significantly enhance its performance on the potential risk reasoning task compared to baseline VLMs. Together, our dataset and model provide a foundation for developing autonomous systems with improved foresight and proactive safety capabilities, moving toward more intelligent and resilient AVs.

cs.RO↗

arXiv2025

ReasonAct: Progressive Training for Fine-Grained Video Reasoning in Small Models

While recent multimodal models have shown progress in vision-language tasks, small-scale variants still struggle with the fine-grained temporal reasoning required for video understanding. We introduce ReasonAct, a method that enhances video reasoning in smaller models through a three-stage training process: first building a foundation with text-only reasoning, then fine-tuning on video, and finally refining with temporal-aware reinforcement learning. We build upon Temporal Group Relative Policy Optimization (T-GRPO) by incorporating temporal consistency modeling into policy optimization. We also propose a biomechanically-motivated sub-action decomposition mechanism that provides graduated rewards for constituent action phases. Through experiments on HMDB51, UCF-101, and Kinetics-400, our 3B-parameter model achieves 67.2%, 94.1%, and 78.9% accuracy respectively, demonstrating improvements of 17.9, 15.8, and 12.3 points over baselines. Ablation studies validate that our progressive training enables smaller models to achieve competitive video reasoning performance while maintaining computational efficiency.

cs.CV↗

arXiv2025

High-charge relativistic electrons by vacuum laser acceleration from plasma mirrors using flying focus pulses

Relativistic electron beams produced by intense lasers over short distances have important applications in high energy density physics and medical technologies. Vacuum laser acceleration with plasma mirrors injectors has garnered substantial research interest recently. However, a persistent challenge remains unresolved that electrons inevitably detach from the laser acceleration phase due to velocity mismatch. Here, we employ flying focus lasers to address this limitation. Through three-dimensional particle-in-cell simulations, we demonstrate that flying focus lasers can achieve a substantial enhancement in relativistic electron charge yield compared to conventional Gaussian lasers. This improvement stems from two key attributes: (1) The subluminal propagation velocity of the peak intensity keeps a larger electron population synchronized within the longitudinal ponderomotive acceleration region, and (2) Flying focus lasers sustain higher magnitudes of the longitudinal ponderomotive force over longer distances in comparison to Gaussian lasers. This approach offers high-charge relativistic electron sources ideal for demanding applications such as high-flux Thomson scattering and radiography.

physics.plasm-ph↗

arXiv2025

MoEntwine: Unleashing the Potential of Wafer-scale Chips for Large-scale Expert Parallel Inference

As large language models (LLMs) continue to scale up, mixture-of-experts (MoE) has become a common technology in SOTA models. MoE models rely on expert parallelism (EP) to alleviate memory bottleneck, which introduces all-to-all communication to dispatch and combine tokens across devices. However, in widely-adopted GPU clusters, high-overhead cross-node communication makes all-to-all expensive, hindering the adoption of EP. Recently, wafer-scale chips (WSCs) have emerged as a platform integrating numerous devices on a wafer-sized interposer. WSCs provide a unified high-performance network connecting all devices, presenting a promising potential for hosting MoE models. Yet, their network is restricted to a mesh topology, causing imbalanced communication pressure and performance loss. Moreover, the lack of on-wafer disk leads to high-overhead expert migration on the critical path. To fully unleash this potential, we first propose Entwined Ring Mapping (ER-Mapping), which co-designs the mapping of attention and MoE layers to balance communication pressure and achieve better performance. We find that under ER-Mapping, the distribution of cold and hot links in the attention and MoE layers is complementary. Therefore, to hide the migration overhead, we propose the Non-invasive Balancer (NI-Balancer), which splits a complete expert migration into multiple steps and alternately utilizes the cold links of both layers. Evaluation shows ER-Mapping achieves communication reduction up to 62%. NI-Balancer further delivers 54% and 22% improvements in MoE computation and communication, respectively. Compared with the SOTA NVL72 supernode, the WSC platform delivers an average 39% higher per-device MoE performance owing to its scalability to larger EP.

cs.DC↗

arXiv2025

DynCIM: Dynamic Curriculum for Imbalanced Multimodal Learning

Multimodal learning integrates complementary information from diverse modalities to enhance the decision-making process. However, the potential of multimodal collaboration remains under-exploited due to disparities in data quality and modality representation capabilities. To address this, we introduce DynCIM, a novel dynamic curriculum learning framework designed to quantify the inherent imbalances from both sample and modality perspectives. DynCIM employs a sample-level curriculum to dynamically assess each sample's difficulty according to prediction deviation, consistency, and stability, while a modality-level curriculum measures modality contributions from global and local. Furthermore, a gating-based dynamic fusion mechanism is introduced to adaptively adjust modality contributions, minimizing redundancy and optimizing fusion effectiveness. Extensive experiments on six multimodal benchmarking datasets, spanning both bimodal and trimodal scenarios, demonstrate that DynCIM consistently outperforms state-of-the-art methods. Our approach effectively mitigates modality and sample imbalances while enhancing adaptability and robustness in multimodal learning tasks. Our code is available at https://github.com/Raymond-Qiancx/DynCIM.

cs.CV↗

arXiv2025

Investigation for binary characteristics of LS I+61303 with optical polarization

We investigate the optical linear polarization caused by Thomson scattering of the stellar radiation for gamma-ray binary \lsi61, which likely contains a young pulsar. Based on the pulsar binary scenario, we model the interaction between the pulsar wind and stellar wind from the massive companion star, which creates a shock. To accurately compute the resulting polarization of the stellar wind, we develop a method for the Thomson scattering that accounts for the finite size of the companion star. By fitting the optical polarization data, we constrain the system parameters, such as eccentricity, the momentum ratio of the two winds, and mass-loss rate from the companion star. We find that (i) the predicted eccentricity $e\sim 0.1$ is smaller than the values derived from the radial velocity curve and (ii) the orbital phase of the periastron is $ν_{\rm p}=0.5-0.6$, which is consistent with the previous polarization study of Kravtsov et al. Additionally, we estimate the mass-loss rate from the companion star and the momentum ratio of two winds as $\dot{M}\sim 2\times 10^{-6}\rm M_{\odot}~{\rm year^{-1}}$ and $η>0.1$, respectively. Assuming that the pulsar wind carries the spin-down energy, the spin-down magnetic field of the putative pulsar inferred from these parameters is of the order of $B\sim 10^{14}\mathrm{G}$, which may support the highly-B pulsar or magnetar scenario for the compact object of $\rm{LS\ I} +61^{\circ}303$. We also discuss the dispersion measure under the predicted orbital geometry and provide a corresponding interpretation of the pulsed radio signal detected by FAST.

astro-ph.HE↗

arXiv2025

AdaThinkDrive: Adaptive Thinking via Reinforcement Learning for Autonomous Driving

While reasoning technology like Chain of Thought (CoT) has been widely adopted in Vision Language Action (VLA) models, it demonstrates promising capabilities in end to end autonomous driving. However, recent efforts to integrate CoT reasoning often fall short in simple scenarios, introducing unnecessary computational overhead without improving decision quality. To address this, we propose AdaThinkDrive, a novel VLA framework with a dual mode reasoning mechanism inspired by fast and slow thinking. First, our framework is pretrained on large scale autonomous driving (AD) scenarios using both question answering (QA) and trajectory datasets to acquire world knowledge and driving commonsense. During supervised fine tuning (SFT), we introduce a two mode dataset, fast answering (w/o CoT) and slow thinking (with CoT), enabling the model to distinguish between scenarios that require reasoning. Furthermore, an Adaptive Think Reward strategy is proposed in conjunction with the Group Relative Policy Optimization (GRPO), which rewards the model for selectively applying CoT by comparing trajectory quality across different reasoning modes. Extensive experiments on the Navsim benchmark show that AdaThinkDrive achieves a PDMS of 90.3, surpassing the best vision only baseline by 1.7 points. Moreover, ablations show that AdaThinkDrive surpasses both the never Think and always Think baselines, improving PDMS by 2.0 and 1.4, respectively. It also reduces inference time by 14% compared to the always Think baseline, demonstrating its ability to balance accuracy and efficiency through adaptive reasoning.

cs.CV↗

arXiv2025

Evaluating and Aligning Human Economic Risk Preferences in LLMs

Large Language Models (LLMs) are increasingly used in decision-making scenarios that involve risk assessment, yet their alignment with human economic rationality remains unclear. In this study, we investigate whether LLMs exhibit risk preferences consistent with human expectations across different personas. Specifically, we assess whether LLM-generated responses reflect appropriate levels of risk aversion or risk-seeking behavior based on individual's persona. Our results reveal that while LLMs make reasonable decisions in simplified, personalized risk contexts, their performance declines in more complex economic decision-making tasks. To address this, we propose an alignment method designed to enhance LLM adherence to persona-specific risk preferences. Our approach improves the economic rationality of LLMs in risk-related applications, offering a step toward more human-aligned AI decision-making.

econ.GN↗

arXiv2025

Reinforcement Learning with Rubric Anchors

Reinforcement Learning from Verifiable Rewards (RLVR) has emerged as a powerful paradigm for enhancing Large Language Models (LLMs), exemplified by the success of OpenAI's o-series. In RLVR, rewards are derived from verifiable signals-such as passing unit tests in code generation or matching correct answers in mathematical reasoning. While effective, this requirement largely confines RLVR to domains with automatically checkable outcomes. To overcome this, we extend the RLVR paradigm to open-ended tasks by integrating rubric-based rewards, where carefully designed rubrics serve as structured, model-interpretable criteria for automatic scoring of subjective outputs. We construct, to our knowledge, the largest rubric reward system to date, with over 10,000 rubrics from humans, LLMs, or a hybrid human-LLM collaboration. Implementing rubric-based RL is challenging; we tackle these issues with a clear framework and present an open-sourced Qwen-30B-A3B model with notable gains: 1) With only 5K+ samples, our system improves by +5.2% on open-ended benchmarks (especially humanities), outperforming a 671B DeepSeek-V3 model by +2.4%, while preserving general and reasoning abilities. 2) Our method provides fine-grained stylistic control, using rubrics as anchors to mitigate the "AI-like" tone and produce more human-like, expressive responses. We share key lessons in rubric construction, data selection, and training, and discuss limitations and future releases.

cs.AI↗

arXiv2025

HOID-R1: Reinforcement Learning for Open-World Human-Object Interaction Detection Reasoning with Multimodal Large Language Model

Understanding and recognizing human-object interaction (HOI) is a pivotal application in AR/VR and robotics. Recent open-vocabulary HOI detection approaches depend exclusively on large language models for richer textual prompts, neglecting their inherent 3D spatial understanding capabilities. To address this shortcoming, we introduce HOID-R1, the first HOI detection framework that integrates chain-of-thought (CoT) guided supervised fine-tuning (SFT) with group relative policy optimization (GRPO) within a reinforcement learning (RL) paradigm. Specifically, we initially apply SFT to imbue the model with essential reasoning capabilities, forcing the model to articulate its thought process in the output. Subsequently, we integrate GRPO to leverage multi-reward signals for policy optimization, thereby enhancing alignment across diverse modalities. To mitigate hallucinations in the CoT reasoning, we introduce an "MLLM-as-a-judge" mechanism that supervises the CoT outputs, further improving generalization. Extensive experiments show that HOID-R1 achieves state-of-the-art performance on HOI detection benchmarks and outperforms existing methods in open-world generalization to novel scenarios.

cs.CV↗

arXiv2025

MoFRR: Mixture of Diffusion Models for Face Retouching Restoration

The widespread use of face retouching on social media platforms raises concerns about the authenticity of face images. While existing methods focus on detecting face retouching, how to accurately recover the original faces from the retouched ones has yet to be answered. This paper introduces Face Retouching Restoration (FRR), a novel computer vision task aimed at restoring original faces from their retouched counterparts. FRR differs from traditional image restoration tasks by addressing the complex retouching operations with various types and degrees, which focuses more on the restoration of the low-frequency information of the faces. To tackle this challenge, we propose MoFRR, Mixture of Diffusion Models for FRR. Inspired by DeepSeek's expert isolation strategy, the MoFRR uses sparse activation of specialized experts handling distinct retouching types and the engagement of a shared expert dealing with universal retouching traces. Each specialized expert follows a dual-branch structure with a DDIM-based low-frequency branch guided by an Iterative Distortion Evaluation Module (IDEM) and a Cross-Attention-based High-Frequency branch (HFCAM) for detail refinement. Extensive experiments on a newly constructed face retouching dataset, RetouchingFFHQ++, demonstrate the effectiveness of MoFRR for FRR.

cs.CV↗

arXiv2025

BANet: Bilateral Aggregation Network for Mobile Stereo Matching

State-of-the-art stereo matching methods typically use costly 3D convolutions to aggregate a full cost volume, but their computational demands make mobile deployment challenging. Directly applying 2D convolutions for cost aggregation often results in edge blurring, detail loss, and mismatches in textureless regions. Some complex operations, like deformable convolutions and iterative warping, can partially alleviate this issue; however, they are not mobile-friendly, limiting their deployment on mobile devices. In this paper, we present a novel bilateral aggregation network (BANet) for mobile stereo matching that produces high-quality results with sharp edges and fine details using only 2D convolutions. Specifically, we first separate the full cost volume into detailed and smooth volumes using a spatial attention map, then perform detailed and smooth aggregations accordingly, ultimately fusing both to obtain the final disparity map. Experimental results demonstrate that our BANet-2D significantly outperforms other mobile-friendly methods, achieving 35.3\% higher accuracy on the KITTI 2015 leaderboard than MobileStereoNet-2D, with faster runtime on mobile devices. Code: \textcolor{magenta}{https://github.com/gangweix/BANet}.

cs.CV↗

arXiv2025

SIMBA -- A Bayesian Decision Framework for the Identification of Optimal Biomarker Subgroups for Cancer Basket Clinical Trials

We consider basket trials in which a biomarker-targeting drug may be efficacious for patients across different disease indications. Patients are enrolled if their cells exhibit some levels of biomarker expression. The threshold level is allowed to vary by indication. The proposed SIMBA method uses a decision framework to identify optimal biomarker subgroups (OBS) defined by an optimal biomarker threshold for each indication. The optimality is achieved through minimizing a posterior expected loss that balances estimation accuracy and investigator preference for broadly effective therapeutics. A Bayesian hierarchical model is proposed to adaptively borrow information across indications and enhance the accuracy in the estimation of the OBS. The operating characteristics of SIMBA are assessed via simulations and compared against a simplified version and an existing alternative method, both of which do not borrow information. SIMBA is expected to improve the identification of patient sub-populations that may benefit from a biomarker-driven therapeutics.

stat.AP↗