arXiv subjects

Qiang Ji

Publications and source records attributed to Qiang Ji.

At least 19 recordsLinked to original sources

arXiv2026

Skills Know Their Neighbors: Cluster-Contrastive Capability Pages for Skill Retrieval

As skill libraries grow, large language model agents must retrieve reusable skills from candidates that often share the same topic and vocabulary but implement different capabilities. Retrieval is limited not only by the scorer but also by the text being scored: a document may describe what a skill does without stating which similar requests should be routed elsewhere. We formalize a skill's capability as its \emph{executable region}, the set of queries it can solve, and view its document as a lossy observation of that region. This view exposes a document-imposed component of retrieval error that cannot be removed by improving the retriever alone. We therefore propose \emph{Capability Pages}, cluster-contrastive skill representations containing a positive trigger $\Tpos$, a negative boundary $\Tneg$, and a discriminative body $B$. An offline compiler compares neighboring skills to write these fields. At inference time, the index uses $\Tpos$ and $B$ for candidate recall, while the router uses $\Tneg$ to reject confusable alternatives. On SRA-Bench, which contains 26{,}262 skills and 5{,}400 questions from six datasets, Capability Pages improve Recall@10 for all five tested retrievers, with a mean gain of $2.94$ points. Adding $\Tneg$ to candidate cards improves end-to-end task success by $3.62$ points on average across four executors and six datasets. A transfer evaluation on Chinese SSL-SkillDiscovery reaches $73.07\%$ MRR@50 using the same encoder across conditions. Capability Pages require no modification to the online models; they improve routing by rewriting the offline skill library.

cs.IR↗

arXiv2026

Can Vision-Language Models Reason about AI Edits in Images?

Detection and localization of AI-tampered images are critical for trustworthy AI, yet modern generative models have made such manipulations increasingly difficult to identify. While traditional binary classifiers can detect image tampering, they lack interpretability and generalization. Vision-Language Models (VLMs) offer a promising alternative due to their strong visual understanding and reasoning capabilities; however, existing approaches typically rely on supervised finetuning with curated explanations rather than exploiting their inherent reasoning capabilities. In this work, we investigate whether VLMs can be trained to reason about AI-generated image edits using reinforcement learning (RL) rather than explicit reasoning supervision. Motivated by the success in Group Relative Policy Optimization (GRPO), an RL technique that incentivizes the model to reason by generating thinking traces prior to giving the final answer, we propose a GRPO-based training framework that utilizes simple accuracy and format rewards. Given an input image, the model produces a structured reasoning trace and predicts whether the image has been tampered with. A lightweight segmentation model is then guided by the reasoning output to generate pixel-level localization masks. Experiments across multiple image manipulation datasets demonstrate that our approach achieves competitive detection and localization performance compared to state-of-the-art image forgery detectors, despite requiring substantially weaker supervision. We introduce effective intersection over union (eff-IoU), a unified metric to jointly evaluate detection and localization. These results suggest that reinforcement learning provides an effective and scalable mechanism for teaching VLMs to reason about AI-generated content.

cs.CV↗

arXiv2026

Optimal sizing of a hydrogen-based direct reduced iron-electric arc furnace system integrated with methanol synthesis toward zero-carbon steel production

A hydrogen-based direct reduced iron-electric arc furnace system integrated with methanol synthesis (H2-DRI-EAF-MeOH) provides a feasible pathway toward zero-carbon steel production. However, the volatility and intermittency of renewable energy sources (RES) may lead to capacity oversizing, while conventional annualized-cost and levelized-cost metrics are insufficient to evaluate investment return. To address these issues, this paper develops a fractional programming-based optimal sizing model for the H2-DRI-EAF-MeOH system. Hourly production rate limits are decoupled from annual production capacities and formulated as independent decision variables. Theoretical analysis demonstrates that increasing hourly production rate limits expands the multi-period operating feasible region, increases the theoretical upper bound on RES matching, and provides an investment-efficiency condition for increasing production rate limits. The annualized net return on investment (ANROI) is then adopted to represent investment return, and the resulting mixed-integer linear fractional programming model is solved using the Dinkelbach method. Case studies show that compared with the annualized net return (ANR) benchmark, the ANROI-based sizing yields higher internal rates of return (IRR), reaching 20.67% and 17.08% under the on-grid and off-grid scenarios, respectively. As the equivalent annual operating hours decrease from 8000 to 4000 h, the IRR increases from 19.68% to 20.67% under the on-grid scenario and from 12.64% to 17.08% under the off-grid scenario, while the corresponding battery storage capacities decrease by 113.56 and 796.31 MWh. The off-grid zero-carbon constraint enables CO2-to-MeOH conversion but reduces the system IRR to 14.99%, indicating that the resulting revenues are insufficient to offset the additional costs.

eess.SY↗

arXiv2026

Skill Is Not Document: Query-Conditioned Compatibility for LLM Agent Skill Routing

Large language model agents increasingly rely on reusable skills, making skill retrieval a critical front-end component of agent systems. Skill retrieval, however, is not ordinary document retrieval: a useful top-$K$ result must contain individually relevant skills that also form an executable set for the current query. Existing benchmarks and training pipelines largely supervise pairwise relevance and discard the rejection decisions produced when a language model judges a sampled skill combination to be implausible. We introduce R3-Skill, a Chinese--English benchmark that retains these rejections as query-conditioned compatibility supervision. R3-Skill contains 10,246 deduplicated skills, 41,592 accepted queries, and 32,828 rejected annotations across four language directions; all multi-skill test labels were independently reviewed by multiple experts, and 15,962 parseable rejections are organized into an eight-class taxonomy. We further propose a two-stage system composed of R3-Embedding, a multi-positive bi-encoder for large-pool recall, and R3-Reranker, a cross-encoder trained with graded ListNet supervision. Our analysis shows that this signal is stage-dependent, helping cross-encoder reranking while providing no benefit for the tested bi-encoder objective. On R3-Skill, the complete pipeline achieves $75.39\%$ Hit@1, $81.97\%$ NDCG@10, and $33.27\%$ Set-Compat, a $36.6\%$ relative gain over the strongest reranking baseline. It also obtains $83.87\%$ NDCG@10 on SkillRet, demonstrating transfer beyond R3-Skill.

cs.IR↗

arXiv2026

Inter-Stance: A Dyadic Multimodal Corpus for Conversational Stance Analysis

Social interactions dominate our perceptions of the world and shape our daily behavior by attaching social meaning to acts as simple and spontaneous as gestures, facial expressions, voice, and speech. People mimic and otherwise respond to each other's postures, facial expressions, mannerisms, and other verbal and nonverbal behavior, and form appraisals or evaluations in the process. Yet, no publicly-available dataset includes multimodal recordings and self-report measures of multiple persons in social interaction. Dyadic recordings and annotation are lacking. We present a new data corpus of multimodal dyadic interaction (45 dyads, 90 persons) that includes synchronized multi-modality behavior (2D face video, 3D face geometry, thermal spectrum dynamics, voice and speech behavior, physiology (PPG, EDA, heart-rate, blood pressure, and respiration), and self-reported affect of all participants in a communicative interaction scenario. Two types of dyads are included: persons with shared past history and strangers. Annotations include social signals, agreement, disagreement, and neutral stance. With a potent emotion induction, these multimodal data will enable novel modeling of multimodal interpersonal behavior. We present extensive experiments to evaluate multimodal dyadic communication of dyads with and without interpersonal history, and their affect. This new database will make multimodal modeling of social interaction never possible before. The dataset includes 20TB of multimodal data to share with the research community.

cs.CV↗

arXiv2026

ControlFoley: Unified and Controllable Video-to-Audio Generation with Cross-Modal Conflict Handling

Recent advances in video-to-audio (V2A) generation enable high-quality audio synthesis from visual content, yet achieving robust and fine-grained controllability remains challenging. Existing methods suffer from weak textual controllability under visual-text conflict and imprecise stylistic control due to entangled temporal and timbre information in reference audio. Moreover, the lack of standardized benchmarks limits systematic evaluation. We propose ControlFoley, a unified multimodal V2A framework that enables precise control over video, text, and reference audio. We introduce a joint visual encoding paradigm that integrates CLIP with a spatio-temporal audio-visual encoder to improve alignment and textual controllability. We further propose temporal-timbre decoupling to suppress redundant temporal cues while preserving discriminative timbre features. In addition, we design a modality-robust training scheme with unified multimodal representation alignment (REPA) and random modality dropout. We also present VGGSound-TVC, a benchmark for evaluating textual controllability under varying degrees of visual-text conflict. Extensive experiments demonstrate state-of-the-art performance across multiple V2A tasks, including text-guided, text-controlled, and audio-controlled generation. ControlFoley achieves superior controllability under cross-modal conflict while maintaining strong synchronization and audio quality, and shows competitive or better performance compared to an industrial V2A system. Code, models, datasets, and demos are available at: https://github.com/xiaomi-research/controlfoley.

cs.MM↗

arXiv2026

A Process-Aware Demand Response Evaluation Framework for Hydrogen-Integrated Zero-Carbon Steel Plants Coupled with Methanol Production

High penetration of renewables (RES) and the retirement of thermal units aggravate flexibility scarcity in power systems. Hydrogen-based low-carbon steel production systems possess substantial demand response (DR) potential. This paper proposes a process-aware DR evaluation framework for hydrogen-integrated zero-carbon steel plants coupled with methanol production (H2-DRI-EAF-MeOH). First, a novel H2-DRI-EAF-MeOH architecture is introduced to eliminate residual emissions via methanol synthesis. Integrated energy-material flows are formulated to reflect coupling interactions governing DR potential. Second, to capture electric arc furnace (EAF) operational constraints while preserving tractability, an operating feasible region model is developed and validated using field data from a pure hydrogen direct reduced iron and EAF plant, yielding a 4.1% average relative error. Third, a process-aware DR potential evaluation model is formulated, incorporating a nonlinear asymmetric penalty and an adaptive rolling mechanism to reflect operators' aversion to process deviations and avoid myopic scheduling. Finally, dual-side evaluation metrics are established to quantify grid-side delivered DR capacity and ramping risks, making load-side unit-level regulation behaviors observable. Case studies show the proposed framework achieves an average effective delivered DR capacity of 178.3 MW, improves RES-load matching from 0.257 to 0.587, and reduces costs by 15.68% compared to the baseline. Furthermore, the exponential asymmetric penalty mitigates extreme tail risks of process deviations. Ultimately, this work provides a theoretical foundation for leveraging RES-steel-chemical synergies to mitigate flexibility scarcity.

eess.SY↗

arXiv2026

Demand response potential evaluation of a zero carbon hydrogen metallurgy system considering shaft furnace's flexibility

The increasing penetration of intermittent renewable energy sources and the retirement of thermal units have widened the power system flexibility gap. Industrial demand response (DR) driven by real-time pricing is widely regarded as a viable solution. In this paper, we propose a framework to quantify the DR potential of a zero-carbon hydrogen metallurgy system (ZCHMS) considering shaft furnace's flexibility. First, we model the shaft furnace as a constrained flexible load and validate the model via simulation, achieving a root mean square error of 4.48\% of the rated load. Second, we formulate a DR potential evaluation method that determines baseline and DR-based production scheduling schemes by minimizing operating cost subject to production orders. Finally, the numerical results show that compared with the baseline, DR-based ZCHMS reduces operating cost by 6.6\%, incentivizing demand-side management in ironmaking and strengthening power-ironmaking synergies.

eess.SY↗

arXiv2026

Superbunched random fiber laser

Photon superbunching, distinguished by second-order coherence values far exceeding the Gaussian thermal limit, represents a highly desirable resource for quantum optics and correlation-based imaging technologies. However, existing approaches typically rely on fragile experimental platforms, inefficient nonlinear conversion processes, or mechanically complex optical architectures. Here, we demonstrate a fully fiber-integrated superbunched random fiber laser (SRFL) in which intrinsic Rayleigh scattering cooperatively interacts with cascaded stimulated Brillouin scattering and quasi-phase-matched four-wave mixing to tailor extreme photon statistics. The SRFL generates a multi-wavelength comb, in which individual spectral components exhibit widely tunable photon bunching, with the second-order coherence g(2)(0) continuously controlled from ~1 to ~26 by tuning the pump power, spectral order and diffusion length. Moreover, we establish a direct correlation between photonic phase transitions (quantified by the Parisi overlap order parameter) and the emergence of superbunching, thereby bridging macroscopic disorder physics and microscopic photon statistics. Finally, we employ the superbunched emission for temporal ghost imaging, realizing high-fidelity temporal object reconstruction with a substantial reduction in required ensemble averaging. These findings validate random fiber lasers as a robust, scalable, and integrated platform for generating extreme photon statistics and unlock new avenues for correlation-enhanced photonic sensing and quantum optics investigations in complex photonic systems.

physics.optics↗

arXiv2025

Let It Flow: Agentic Crafting on Rock and Roll, Building the ROME Model within an Open Agentic Learning Ecosystem

Agentic crafting requires LLMs to operate in real-world environments over multiple turns by taking actions, observing outcomes, and iteratively refining artifacts. Despite its importance, the open-source community lacks a principled, end-to-end ecosystem to streamline agent development. We introduce the Agentic Learning Ecosystem (ALE), a foundational infrastructure that optimizes the production pipeline for agentic model. ALE consists of three components: ROLL, a post-training framework for weight optimization; ROCK, a sandbox environment manager for trajectory generation; and iFlow CLI, an agent framework for efficient context engineering. We release ROME, an open-source agent grounded by ALE and trained on over one million trajectories. Our approach includes data composition protocols for synthesizing complex behaviors and a novel policy optimization algorithm, Interaction-Perceptive Agentic Policy Optimization (IPA), which assigns credit over semantic interaction chunks rather than individual tokens to improve long-horizon training stability. Empirically, we evaluate ROME within a structured setting and introduce Terminal Bench Pro, a benchmark with improved scale and contamination control. ROME demonstrates strong performance across benchmarks like SWE-bench Verified and Terminal Bench, proving the effectiveness of ALE.

cs.AI↗

arXiv2025

Diffusion-based 3D Hand Motion Recovery with Intuitive Physics

While 3D hand reconstruction from monocular images has made significant progress, generating accurate and temporally coherent motion estimates from videos remains challenging, particularly during hand-object interactions. In this paper, we present a novel 3D hand motion recovery framework that enhances image-based reconstructions through a diffusion-based and physics-augmented motion refinement model. Our model captures the distribution of refined motion estimates conditioned on initial ones, generating improved sequences through an iterative denoising process. Instead of relying on scarce annotated video data, we train our model only using motion capture data without images. We identify valuable intuitive physics knowledge during hand-object interactions, including key motion states and their associated motion constraints. We effectively integrate these physical insights into our diffusion model to improve its performance. Extensive experiments demonstrate that our approach significantly improves various frame-wise reconstruction methods, achieving state-of-the-art (SOTA) performance on existing benchmarks.

cs.CV↗

arXiv2025

International Tourism and Global Biodiversity Risks

The impact of international tourism on biodiversity risks has received considerable attention, yet quantitative research in this field remains relatively limited. This study constructs a biodiversity risk index for 155 countries and regions spanning the years 2001 to 2019, analysing how international tourism influences biodiversity risks in destination countries. The results indicate that the growth of international tourism significantly elevates biodiversity risks, with these effects displaying both lagging and cumulative characteristics. Furthermore, spatial analysis shows that international tourism also intensifies biodiversity risks in neighbouring countries. The extent of its impact varies according to the tourism model and destination. In addition, government regulations and international financial assistance play a crucial role in mitigating the biodiversity risks associated with international tourism.

econ.GN↗

arXiv2025

Energy-carbon comprehensive efficiency evaluation of hydrogen metallurgy system considering low-temperature waste heat recovery

To address the lack of energy-carbon efficiency evaluation and the underutilization of low-temperature waste heat in traditional direct reduction iron (DRI) production, this paper proposes a novel zero-carbon hydrogen metallurgy system that integrates the recovery and utilization of low-temperature and high-temperature waste heat, internal energy, and cold energy during hydrogen production, storage, reaction and circulation. Firstly, the detailed mathematical models are developed to describe energy and exergy characteristics of the operational components in the proposed zero-carbon hydrogen metallurgy system. Additionally, energy efficiency, exergy efficiency, and energy-carbon efficiency indices are introduced from a full life-cycle perspective of energy flow, avoiding the overlaps in energy inputs and outputs. Subsequently, the efficiency metrics of the proposed zero-carbon hydrogen metallurgy system are then compared with those of traditional DRI production systems with H$_2$/CO ratios of 6:4 and 8:2. The comparative results demonstrate the superiority and advancement of the proposed zero-carbon hydrogen metallurgy system. Finally, sensitivity analysis reveals that the overall electricity energy generated by incorporating the ORC and expander equipments exceeds the heat energy recovered from the furnace top gas, highlighting the energy potential of waste energy utilization.

eess.SY↗

arXiv2024

Weakly-Supervised 3D Hand Reconstruction with Knowledge Prior and Uncertainty Guidance

Fully-supervised monocular 3D hand reconstruction is often difficult because capturing the requisite 3D data entails deploying specialized equipment in a controlled environment. We introduce a weakly-supervised method that avoids such requirements by leveraging fundamental principles well-established in the understanding of the human hand's unique structure and functionality. Specifically, we systematically study hand knowledge from different sources, including biomechanics, functional anatomy, and physics. We effectively incorporate these valuable foundational insights into 3D hand reconstruction models through an appropriate set of differentiable training losses. This enables training solely with readily-obtainable 2D hand landmark annotations and eliminates the need for expensive 3D supervision. Moreover, we explicitly model the uncertainty that is inherent in image observations. We enhance the training process by exploiting a simple yet effective Negative Log Likelihood (NLL) loss that incorporates uncertainty into the loss function. Through extensive experiments, we demonstrate that our method significantly outperforms state-of-the-art weakly-supervised methods. For example, our method achieves nearly a 21\% performance improvement on the widely adopted FreiHAND dataset.

cs.CV↗

arXiv2024

Parallel fast random bit generation based on spectrotemporally uncorrelated Brillouin random fiber lasing oscillation

Correlations existing between spectral components in multi-wavelength lasers have been the key challenge that hinders these laser sources from being developed to chaotic comb entropy sources for parallel random bit generation. Herein, spectrotemporally uncorrelated multi-order Stokes/anti-Stokes emissions are achieved by cooperatively exploiting nonlinear optical processes including cascaded stimulated Brillouin scattering and quasi-phase-matched four-wave mixing in a Brillouin random fiber laser. Chaotic instabilities induced by random mode resonance are enhanced and disorderly redistributed among different lasing lines through complex nonlinear optical interactions, which comprehensively releases the inherent correlation among multiple Stokes/anti-Stokes emission lines, realizing a chaotic frequency comb with multiple spectrotemporally uncorrelated channels. Parallel fast random bit generation is fulfilled with 31 channels, single-channel bit rate of 35-Gbps and total bit rate of 1.085-Tbps. National Institute of Standards and Technology statistic tests verify the randomness of generated bit streams. This work, in a simple and efficient way, breaks the correlation barrier for utilizing multi-wavelength laser to achieve high-quality spectrotemporally uncorrelated chaotic laser source, opening new avenues for achieving greatly accelerated random bit generation through parallelization and potentially revolutionizing the current architecture of secure communication and high-performance computation.

physics.optics↗

arXiv2024

Epistemic Uncertainty Quantification For Pre-trained Neural Network

Epistemic uncertainty quantification (UQ) identifies where models lack knowledge. Traditional UQ methods, often based on Bayesian neural networks, are not suitable for pre-trained non-Bayesian models. Our study addresses quantifying epistemic uncertainty for any pre-trained model, which does not need the original training data or model modifications and can ensure broad applicability regardless of network architectures or training techniques. Specifically, we propose a gradient-based approach to assess epistemic uncertainty, analyzing the gradients of outputs relative to model parameters, and thereby indicating necessary model adjustments to accurately represent the inputs. We first explore theoretical guarantees of gradient-based methods for epistemic UQ, questioning the view that this uncertainty is only calculable through differences between multiple models. We further improve gradient-driven UQ by using class-specific weights for integrating gradients and emphasizing distinct contributions from neural network layers. Additionally, we enhance UQ accuracy by combining gradient and perturbation methods to refine the gradients. We evaluate our approach on out-of-distribution detection, uncertainty calibration, and active learning, demonstrating its superiority over current state-of-the-art UQ methods for pre-trained models.

cs.LG↗

arXiv2024

PhysPT: Physics-aware Pretrained Transformer for Estimating Human Dynamics from Monocular Videos

While current methods have shown promising progress on estimating 3D human motion from monocular videos, their motion estimates are often physically unrealistic because they mainly consider kinematics. In this paper, we introduce Physics-aware Pretrained Transformer (PhysPT), which improves kinematics-based motion estimates and infers motion forces. PhysPT exploits a Transformer encoder-decoder backbone to effectively learn human dynamics in a self-supervised manner. Moreover, it incorporates physics principles governing human motion. Specifically, we build a physics-based body representation and contact force model. We leverage them to impose novel physics-inspired training losses (i.e., force loss, contact loss, and Euler-Lagrange loss), enabling PhysPT to capture physical properties of the human body and the forces it experiences. Experiments demonstrate that, once trained, PhysPT can be directly applied to kinematics-based estimates to significantly enhance their physical plausibility and generate favourable motion forces. Furthermore, we show that these physically meaningful quantities translate into improved accuracy of an important downstream task: human action recognition.

cs.CV↗

arXiv2023

Effective Causal Discovery under Identifiable Heteroscedastic Noise Model

Capturing the underlying structural causal relations represented by Directed Acyclic Graphs (DAGs) has been a fundamental task in various AI disciplines. Causal DAG learning via the continuous optimization framework has recently achieved promising performance in terms of both accuracy and efficiency. However, most methods make strong assumptions of homoscedastic noise, i.e., exogenous noises have equal variances across variables, observations, or even both. The noises in real data usually violate both assumptions due to the biases introduced by different data collection processes. To address the issue of heteroscedastic noise, we introduce relaxed and implementable sufficient conditions, proving the identifiability of a general class of SEM subject to these conditions. Based on the identifiable general SEM, we propose a novel formulation for DAG learning that accounts for the variation in noise variance across variables and observations. We then propose an effective two-phase iterative DAG learning algorithm to address the increasing optimization difficulties and to learn a causal DAG from data with heteroscedastic variable noise under varying variance. We show significant empirical gains of the proposed approaches over state-of-the-art methods on both synthetic data and real data.

cs.LG↗