arXiv subjects

Yanan Wang

Publications and source records attributed to Yanan Wang.

At least 55 records · Page 3Linked to original sources

arXiv2025

An Empirical Study for Representations of Videos in Video Question Answering via MLLMs

Multimodal large language models have recently achieved remarkable progress in video question answering (VideoQA) by jointly processing visual, textual, and audio information. However, it remains unclear which video representations are most effective for MLLMs, and how different modalities balance task accuracy against computational efficiency. In this work, we present a comprehensive empirical study of video representation methods for VideoQA with MLLMs. We systematically evaluate single modality inputs question only, subtitles, visual frames, and audio signals as well as multimodal combinations, on two widely used benchmarks: VideoMME and LongVideoBench. Our results show that visual frames substantially enhance accuracy but impose heavy costs in GPU memory and inference latency, while subtitles provide a lightweight yet effective alternative, particularly for long videos. These findings highlight clear trade-offs between effectiveness and efficiency and provide practical insights for designing resource-aware MLLM-based VideoQA systems.

cs.IR↗

arXiv2025

A Unified Biomedical Named Entity Recognition Framework with Large Language Models

Accurate recognition of biomedical named entities is critical for medical information extraction and knowledge discovery. However, existing methods often struggle with nested entities, entity boundary ambiguity, and cross-lingual generalization. In this paper, we propose a unified Biomedical Named Entity Recognition (BioNER) framework based on Large Language Models (LLMs). We first reformulate BioNER as a text generation task and design a symbolic tagging strategy to jointly handle both flat and nested entities with explicit boundary annotation. To enhance multilingual and multi-task generalization, we perform bilingual joint fine-tuning across multiple Chinese and English datasets. Additionally, we introduce a contrastive learning-based entity selector that filters incorrect or spurious predictions by leveraging boundary-sensitive positive and negative samples. Experimental results on four benchmark datasets and two unseen corpora show that our method achieves state-of-the-art performance and robust zero-shot generalization across languages. The source codes are freely available at https://github.com/dreamer-tx/LLMNER.

cs.CL↗

arXiv2025

MARS2 2025 Challenge on Multimodal Reasoning: Datasets, Methods, Results, Discussion, and Outlook

This paper reviews the MARS2 2025 Challenge on Multimodal Reasoning. We aim to bring together different approaches in multimodal machine learning and LLMs via a large benchmark. We hope it better allows researchers to follow the state-of-the-art in this very dynamic area. Meanwhile, a growing number of testbeds have boosted the evolution of general-purpose large language models. Thus, this year's MARS2 focuses on real-world and specialized scenarios to broaden the multimodal reasoning applications of MLLMs. Our organizing team released two tailored datasets Lens and AdsQA as test sets, which support general reasoning in 12 daily scenarios and domain-specific reasoning in advertisement videos, respectively. We evaluated 40+ baselines that include both generalist MLLMs and task-specific models, and opened up three competition tracks, i.e., Visual Grounding in Real-world Scenarios (VG-RS), Visual Question Answering with Spatial Awareness (VQA-SA), and Visual Reasoning in Creative Advertisement Videos (VR-Ads). Finally, 76 teams from the renowned academic and industrial institutions have registered and 40+ valid submissions (out of 1200+) have been included in our ranking lists. Our datasets, code sets (40+ baselines and 15+ participants' methods), and rankings are publicly available on the MARS2 workshop website and our GitHub organization page https://github.com/mars2workshop/, where our updates and announcements of upcoming events will be continuously provided.

cs.CV↗

arXiv2025

Probing the Strong Gravity Region of Black Holes with eXTP

We present the novel capabilities of the enhanced X-ray Timing and Polarimetry (eXTP) mission to study the strong gravity region around stellar-mass black holes in X-ray binary systems and supermassive black holes in active galactic nuclei. eXTP can combine X-ray spectral, timing, and polarimetric techniques to study the accretion process near black holes, measure black hole masses and spins, and test Einstein's theory of General Relativity in the strong field regime. We show how eXTP can improve the current measurements of black holes of existing X-ray missions and we discuss the scientific questions that can be addressed.

astro-ph.HE↗

arXiv2025

An Emotion Recognition Framework via Cross-modal Alignment of EEG and Eye Movement Data

Emotion recognition is essential for applications in affective computing and behavioral prediction, but conventional systems relying on single-modality data often fail to capture the complexity of affective states. To address this limitation, we propose an emotion recognition framework that achieves accurate multimodal alignment of Electroencephalogram (EEG) and eye movement data through a hybrid architecture based on cross-modal attention mechanism. Experiments on the SEED-IV dataset demonstrate that our method achieve 90.62% accuracy. This work provides a promising foundation for leveraging multimodal data in emotion recognition

cs.MM↗

arXiv2025

Breather-to-soliton transitions and nonlinear wave interactions for the higher-order generalized Gerdjikov-Ivanov equation

In this paper, we systematically investigate the intricate dynamics of the breather-to-soliton transitions and nonlinear wave interactions for the higher-order generalized Gerdjikov-Ivanov equation. The transition conditions of the breather-to-soliton are established and the novel nonlinear converted waves, including the W-shaped soliton, M-shaped soliton, multi-peak soliton, anti-dark soliton and periodic wave solution are discussed. Meanwhile, the interactions among the above nonlinear converted waves are explored by choosing appropriate parameters. Furthermore, we derive the double-pole solutions exhibiting breather-to-soliton transitions and employ the asymptotic analysis method to analyze the dynamics of the asymptotic solitons for the double-pole anti-dark soliton. This work deepens the fundamental understanding of nonlinear wave metamorphosis induced by higher-order terms in integrable systems.

nlin.PS↗

arXiv2025

Delayed Launch of Ultrafast Outflows in the Tidal Disruption Event AT2020afhd

We report the detection and characterization of ultrafast outflows (UFOs) in the X-ray spectra of the tidal disruption event (TDE) AT2020afhd, based on observations from NICER, Swift, and XMM. Prominent blueshifted absorption features were detected exclusively during the intermediate phase of the event, occurring between days 172 and 212 within the first 300 days post-discovery. During this period, the UFO appeared no earlier than day 74, strengthened between days 172 and 194, and disappeared after day 215. This marks the first time that the full evolutionary sequence of X-ray outflows has been observed in a TDE. Moreover, the outflows exhibited a dramatic deceleration from ~0.19c to ~0.0097c over a span of approximately 10 days. Photoionization spectral analysis reveals an inverse correlation between outflow velocity and ionization parameter, in contradiction to the predictions from radiation pressure-driven wind. Eventually, we propose that the delayed onset of the outflows may result from an increase in the wind opening angle and/or metal enrichment, particularly iron and oxygen, during the disk formation phase.

astro-ph.HE↗

arXiv2025

CoTasks: Chain-of-Thought based Video Instruction Tuning Tasks

Despite recent progress in video large language models (VideoLLMs), a key open challenge remains: how to equip models with chain-of-thought (CoT) reasoning abilities grounded in fine-grained object-level video understanding. Existing instruction-tuned models, such as the Qwen and LLaVA series, are trained on high-level video-text pairs, often lacking structured annotations necessary for compositional, step-by-step reasoning. We propose CoTasks: Chain-of-Thought based Video Instruction Tuning Tasks, a new framework that decomposes complex video questions of existing datasets (e.g., NeXT-QA, STAR) into four entity-level foundational tasks: frame localization, entity tracking, spatial and temporal relation extraction. By embedding these intermediate CoT-style reasoning steps into the input, CoTasks enables models to explicitly perform object-centric spatiotemporal reasoning. Experiments on the NeXT-QA benchmark show that CoTasks significantly enhance inference performance: LLaVA-video-7B improves by +3.3 points in average GPT-4 evaluation score, and Qwen2.5-VL-3B gains +17.4, with large boosts in causal (+14.6), temporal (+10.9), and descriptive (+48.1) subcategories. These results demonstrate the effectiveness of CoTasks as a structured CoT-style supervision framework for improving compositional video reasoning.

cs.CV↗

arXiv2025

Observation of Cavity-Mediated Nonlinear Landau Fan and Modified Landau Level Degeneracy in Graphene Quantum Transport

Recent studies on cavity-coupled two-dimensional electron gas demonstrate that vacuum-field engineering can tailor electronic transport properties of materials. By achieving ultra-strong coupling between a terahertz resonator and mesoscopic graphene, we demonstrate that cavity vacuum fields can alter the effective degeneracies of Landau levels, resulting in a nonlinear Landau fan diagram for massless Dirac fermions while preserving quantum-Hall quantization. Specifically, by leveraging graphene's gate-tunability, we observe that quantum-Hall features, minimum longitudinal and quantized Hall conductance for a given filling factor, occur at carrier densities reduced by more than 20 percent compared to systems without cavity. Theoretical analysis attributes this effect to the virtual cavity photon mediated transitions between the non-equidistant Landau levels in graphene, significantly reducing their effective degeneracy. This study paves the way for investigating cavity quantum electrodynamics in highly tunable, atomically thin two-dimensional crystals.

cond-mat.mes-hall↗

arXiv2025

The Mini-SiTian Array: Prospects of Searching for Tidal Disruption Events

We assess the detectability of tidal disruption events (TDEs) using mock observations from the Mini-SiTian array. We select 100 host galaxy samples from a simulated galaxy catalog based on specific criteria such as redshift, BH mass, and event rate. Taking into account the site conditions and survey strategy, we simulate observations over a 440 deg$^2$ field. The results indicate that $0.53\pm 0.73$ TDEs can be detected per year when observing in both $g$ and $r$ bands with 300-second exposures every 3 days. Applying this method to the SiTian project, we expect to discover approximately 204 TDEs annually, heralding a new era in TDE science.

astro-ph.HE↗

arXiv2025

AT2021aeuk: A Repeating Partial Tidal Disruption Event Candidate in a Narrow-line Seyfert 1 galaxy

A black hole (BH) can tear apart a star that ventures within its tidal radius, producing a luminous flare as the stellar debris falls back, known as a tidal disruption event (TDE). While TDEs in quiescent galaxies are relatively well understood, identifying TDEs in active galactic nuclei (AGN) still remains a significant challenge. We present the discovery of AT2021aeuk, a transient exhibiting dual flares within around three years in a narrow-line Seyfert 1 galaxy. Multi-wavelength observations triggered during the second flare in 2023 revealed an extraordinary X-ray V-shaped light curve, strongly anti-correlated with the optical light curve and accompanied by a lag of $\sim$40 days. This behavior is inconsistent with both supernova and pure AGN origins. In addition, a new broad component emerges in the Balmer lines during the second flare, showing a clear reverberation signal to the continuum variation. We propose that the dual-flare may be linked to a repeating partial TDE (rpTDE), where the second flare results from a collision between the TDE stream and the inner accretion disk, triggering an optical flare while simultaneously partially destroying the X-ray corona. However, other mechanisms, such as a stellar-mass BH (sBH) merger within an accretion disk, could produce similar phenomena, which we cannot entirely rule out. The Vera C. Rubin Observatory will be a powerful tool for further investigating the nature of such events in the future.

astro-ph.HE↗

arXiv2025

Strawberry Robotic Operation Interface: An Open-Source Device for Collecting Dexterous Manipulation Data in Robotic Strawberry Farming

The strawberry farming is labor-intensive, particularly in tasks requiring dexterous manipulation such as picking occluded strawberries. To address this challenge, we present the Strawberry Robotic Operation Interface (SROI), an open-source device designed for collecting dexterous manipulation data in robotic strawberry farming. The SROI features a handheld unit with a modular end effector, a stereo robotic camera, enabling the easy collection of demonstration data in field environments. A data post-processing pipeline is introduced to extract spatial trajectories and gripper states from the collected data. Additionally, we release an open-source dataset of strawberry picking demonstrations to facilitate research in dexterous robotic manipulation. The SROI represents a step toward automating complex strawberry farming tasks, reducing reliance on manual labor.

cs.RO↗

arXiv2025

Reverberation evidence for Stream Collision and Delayed Disk Formation in Tidal Disruption Events

When a star passes through the tidal disruption radius of a massive black hole (BH), it can be torn apart by the tidal force of the BH, known as the Tidal Disruption Event (TDE). Since the observed UV/optical luminosity significantly exceeds the predictions of the compact disk model in classical TDE theory, two competing models, stream collision and envelope reprocessing, have been proposed to address this discrepancy. To distinguish between these models, we investigate the continuum reverberation behaviors for $\sim$30 TDEs with high-quality multi-band light curves. We found that over half of them exhibit a positive lag by a few days in UV/optical bands, indicating that their inferred sizes are significantly larger than the envelope sizes in reprocessing. Moreover, X-ray emissions not only significantly delay relative to the primary UV/optical peak but also lag behind the rebrightening bump by up to several tens of days, completely different from the X-ray illumination reprocessing. Additionally, the anti-correlated UV-optical continuum in ASASSN-15lh further disfavors the reprocessing scenario. In contrast, the model of stream collisions, combined with delayed accretion disk formation, can provide a unified explanation for the diverse TDE observations, e.g., the optical/X-ray population, the frequently observed rebrightening bump. This model describes a unification scheme wherein the UV/optical emission originates from stream collisions during the early-stage of TDE evolution and gradually transitions to being dominated by accretion disk with detectable X-ray emission in the late stage. After transitioning to a quiescent state, recurrent flares may be observed in some cases, possibly related to repeating partial TDEs.

astro-ph.HE↗

arXiv2025

Science objectives of the Einstein Probe mission

The Einstein Probe (EP) is an interdisciplinary mission of time-domain and X-ray astronomy. Equipped with a wide-field lobster-eye X-ray focusing imager, EP will discover cosmic X-ray transients and monitor the X-ray variability of known sources in 0.5-4 keV, at a combination of detecting sensitivity and cadence that is not accessible to the previous and current wide-field monitoring missions. EP can perform quick characterisation of transients or outbursts with a Wolter-I X-ray telescope onboard. In this paper, the science objectives of the Einstein Probe mission are presented. EP is expected to enlarge the sample of previously known or predicted but rare types of transients with a wide range of timescales. Among them, fast extragalactic transients will be surveyed systematically in soft X-rays, which include γ-ray bursts and their variants, supernova shock breakouts, and the predicted X-ray transients associated with binary neutron star mergers. EP will detect X-ray tidal disruption events and outbursts from active galactic nuclei, possibly at an early phase of the flares for some. EP will monitor the variability and outbursts of X-rays from white dwarfs, neutron stars and black holes in our and neighbouring galaxies at flux levels fainter than those detectable by the current instruments, and is expected to discover new objects. A large sample of stellar X-ray flares will also be detected and characterised. In the era of multi-messenger astronomy, EP has the potential of detecting the possible X-ray counterparts of gravitational wave events, neutrino sources, and ultra-high energy γ-ray and cosmic ray sources. EP is expected to help advance the studies of extreme objects/phenomena and their underlying physical processes revealed in the dynamic X-ray universe, as well as studies in other areas of X-ray astronomy.

astro-ph.HE↗

arXiv2024

TTT4Rec: A Test-Time Training Approach for Rapid Adaption in Sequential Recommendation

Sequential recommendation tasks, which aim to predict the next item a user will interact with, typically rely on models trained solely on historical data. However, in real-world scenarios, user behavior can fluctuate in the long interaction sequences, and training data may be limited to model this dynamics. To address this, Test-Time Training (TTT) offers a novel approach by using self-supervised learning during inference to dynamically update model parameters. This allows the model to adapt to new user interactions in real-time, leading to more accurate recommendations. In this paper, we propose TTT4Rec, a sequential recommendation framework that integrates TTT to better capture dynamic user behavior. By continuously updating model parameters during inference, TTT4Rec is particularly effective in scenarios where user interaction sequences are long, training data is limited, or user behavior is highly variable. We evaluate TTT4Rec on three widely-used recommendation datasets, demonstrating that it achieves performance on par with or exceeding state-of-the-art models. The codes are available at https://github.com/ZhaoqiZachYang/TTT4Rec.

cs.IR↗

arXiv2024

Analytical formula for the cross section of hadron production from $e^{+}e^{-}$ collisions around the narrow charmouinum resonances

The paper reports an analytical formula for the production cross section of $e^{+}e^{-}$ annihilation to hadrons in the vicinity of a narrow resonance, particularly in the $τ$-charm region, while considering initial state radiation. Despite some approximations in its derivation, the comparison between the analytical formula and direct integration of ISR shows good accuracy, indicating that the analytical formula meets current experimental requirements. Furthermore, the paper presents a comparison of the cross section between the analytical formula and calculations using the {\sc ConExc} Monte Carlo generator. The efficiency of the analytical formula in significantly reducing computing time makes it a favorable choice for the regression procedure to extract the parameters of narrow charmonium resonances in experiments.

hep-ph↗

arXiv2024

Multi-object event graph representation learning for Video Question Answering

Video question answering (VideoQA) is a task to predict the correct answer to questions posed about a given video. The system must comprehend spatial and temporal relationships among objects extracted from videos to perform causal and temporal reasoning. While prior works have focused on modeling individual object movements using transformer-based methods, they falter when capturing complex scenarios involving multiple objects (e.g., "a boy is throwing a ball in a hoop"). We propose a contrastive language event graph representation learning method called CLanG to address this limitation. Aiming to capture event representations associated with multiple objects, our method employs a multi-layer GNN-cluster module for adversarial graph representation learning, enabling contrastive learning between the question text and its relevant multi-object event graph. Our method outperforms a strong baseline, achieving up to 2.2% higher accuracy on two challenging VideoQA datasets, NExT-QA and TGIF-QA-R. In particular, it is 2.8% better than baselines in handling causal and temporal questions, highlighting its strength in reasoning multiple object-based events.

cs.CV↗

arXiv2024

Top-down Activity Representation Learning for Video Question Answering

Capturing complex hierarchical human activities, from atomic actions (e.g., picking up one present, moving to the sofa, unwrapping the present) to contextual events (e.g., celebrating Christmas) is crucial for achieving high-performance video question answering (VideoQA). Recent works have expanded multimodal models (e.g., CLIP, LLaVA) to process continuous video sequences, enhancing the model's temporal reasoning capabilities. However, these approaches often fail to capture contextual events that can be decomposed into multiple atomic actions non-continuously distributed over relatively long-term sequences. In this paper, to leverage the spatial visual context representation capability of the CLIP model for obtaining non-continuous visual representations in terms of contextual events in videos, we convert long-term video sequences into a spatial image domain and finetune the multimodal model LLaVA for the VideoQA task. Our approach achieves competitive performance on the STAR task, in particular, with a 78.4% accuracy score, exceeding the current state-of-the-art score by 2.8 points on the NExTQA task.

cs.CV↗