arXiv subjects

Chenghao Wang

Publications and source records attributed to Chenghao Wang.

At least 19 recordsLinked to original sources

arXiv2026

MOSS-VL Technical Report

We present MOSS-VL, an open vision-language model family that treats real-time interaction -- perceiving while it speaks -- as a first-class capability. It is co-designed across the stack: the language decoder attends to vision only through gated cross-attention, so the model can naturally see incoming frames while generating; a synthesized interaction corpus supervises when to speak, when to stay silent, and when to revise; and a staged curriculum concentrates all real-time-specific training in one light final stage over a strong offline foundation. Offline, MOSS-VL-Instruct is competitive at comparable scale and leads temporal-reasoning video sets. Across four streaming benchmarks, MOSS-VL-Realtime posts the best average on three (second on the fourth) among open-source streaming models, sweeping the three subsets that squarely test proactive behavior -- 66.0 vs. 37.5 for the best baseline on OmniMMI Proactive Alerting. With 11.3B parameters but visual tokens outside the decoded sequence, MOSS-VL widens its time-to-first-token advantage over same-backbone Qwen3-VL-8B from 2.8x to 5.1x as visual context grows. We release all five checkpoints, the training curriculum, and the real-time inference code at https://github.com/OpenMOSS/MOSS-VL.

cs.CV↗

arXiv2026

ICBCBench: An Industry Consortium Benchmark for Financial Deep Research

With the rapid advancement of Deep Research Agents in knowledge-intensive domains such as finance, establishing reliable and domain-aligned evaluation standards remains a critical challenge. Existing benchmarks focus on either closed-ended question answering or open-ended report evaluation, failing to jointly capture retrieval-reasoning accuracy and end-to-end research quality required in real-world workflows. We introduce ICBCBench, a consortium-driven benchmark for financial deep research, developed in collaboration with domain experts from a broad range of financial institutions and academia, involving over 50 experts across more than 40 organizations. It adopts a dual-track paradigm integrating objective tasks with verifiable answers and subjective long-form report evaluation, enabling complementary assessment of retrieval-reasoning accuracy and end-to-end report quality in terms of expert alignment, citation consistency, and source quality. Experiments on state-of-the-art DRAs and large language models reveal substantial gaps in complex reasoning, factual grounding, and report quality, highlighting the challenges of achieving industry-level performance. Our dataset and evaluation framework are available at https://github.com/DeepFin-Intelligence/ICBCBench.

cs.CE↗

arXiv2026

MOSS-Video-Preview: Toward Real-Time Video Understanding via Cross-Attention

Video understanding is shifting from the offline paradigm -- taking a fully recorded video as input and producing a single answer after it ends -- toward real-time interaction, in which the model perceives new frames while still replying, revises its answer as new evidence appears, and remains silent when there is nothing to say. We present MOSS-Video-Preview to validate this paradigm. Our central claim is that perception must not be blocked by generation; its natural realization is a two-channel architecture. We argue that a cross-attention backbone is better suited to real-time vision-language fusion than the prevailing decoder-only design: visual features enter through a side channel rather than joining the autoregressive sequence, so perception and generation run on separate, non-blocking pathways -- reducing the frequency of visual processing and exposing a clean channel-wise interface for independent compression. We complement this with a data synthesis pipeline that converts dense captions into real-time understanding QA whose answers are revised to match what the model has perceived so far, and we specialize an offline model on these data to elicit real-time behavior. Our model trails the strong Qwen2.5-VL-7B baseline overall -- a gap we attribute primarily to data and scale rather than the architecture -- yet attains competitive offline video and multimodal understanding, remains robust on the spatial and fine-grained temporal reasoning central to real-time use, and acquires behaviors that offline models lack: continuous perception, answer revision, and timely silence. On a single H200 with 256 frames per video, it achieves about a 5x speedup in time to first token and 2.7x higher decoding throughput, with negligible degradation in offline ability. Our study of paradigm, architecture, and data outlines a viable path toward real-time video understanding.

cs.CV↗

arXiv2026

A Unified and Reproducible Experimentation Framework for Speech Understanding

Speech foundation models and Speech LLMs have advanced speech understanding, yet deployment-oriented model selection is hindered by non-comparable evaluations caused by mismatched post-processing, and by training results that are hard to reproduce across data scales and pipelines. We present SURE, a unified experimentation framework that standardizes prediction formats, normalization, and scoring. SURE evaluates strong systems across paradigms, from conventional pipelines to Speech LLMs, on representative tasks under realistic acoustic and linguistic stressors. Beyond evaluation, SURE introduces an agent-assisted training conversion flow that maps paper and code into versioned, runnable training pipelines under a unified protocol on matched open-data subsets. Overall, SURE improves comparability and reproducibility for deployment-oriented evaluation.

eess.AS↗

arXiv2026

Audio-Mind: An Auditable Agentic Framework for Audio Understanding

Audio agents extend large audio-language models (LALMs) by decomposing audio questions into tool calls, intermediate evidence, and iterative reasoning steps. However, as LALMs become stronger, the key challenge shifts from enabling tool use to determining when agentic evidence acquisition genuinely benefits audio understanding. We propose Audio-Mind, an auditable and pluggable framework for conditional evidence acquisition in audio understanding. Audio-Mind dynamically combines a strong frontend with planner-guided tool use, preserving frontend judgment when initial evidence is sufficient while acquiring bounded external evidence for questions with unresolved evidence gaps. Experiments on MMAR and MSU-Bench show that Audio-Mind outperforms prior audio-agent baselines, reaching 80.4% accuracy on MMAR and 82.8% accuracy on MSU-Bench. A matched-backbone comparison highlights why this design matters: under strong audio frontends, agentic decomposition can become an orchestration bottleneck when the workflow does not preserve the frontend's holistic audio-grounded judgment. Beyond accuracy, Audio-Mind produces higher-quality, auditable reasoning traces that expose uncertainty, tool evidence, and answer rationales, offering a potential basis for more reliable audio-QA annotation and error analysis.

eess.AS↗

arXiv2026

Can AI Reduce Acculturative Stress? Exploring the Role of AI-Mediated Speaking Practice in Chinese International Students' Perceived Language Insufficiency, Social Isolation, and Academic Pressure

This study examined whether AI-mediated speaking practice can reduce acculturative stress among Chinese international students in UK universities. Using a sequential explanatory mixed-methods design, 126 participants were randomly assigned to an experimental group, which completed a four-week intervention using EAP Talk, an AI-assisted English for Academic Purposes speaking platform offering role play, scenario-based practice, free talk, and automated feedback, or a control group, which continued usual academic and English-learning activities. Pre- and post-test questionnaires measured perceived language insufficiency, social isolation, and academic pressure, while semi-structured interviews with 20 experimental-group participants contextualised the quantitative findings. Linear mixed-effects models showed that the experimental group experienced significantly greater reductions than the control group across all three outcomes, with the strongest effect on perceived language insufficiency. Interview findings suggested that EAP Talk supported low-stakes rehearsal, communicative confidence, academic speaking preparation, and greater willingness to initiate social interaction. However, participants also noted that AI-mediated practice could not fully reproduce authentic human interaction, disciplinary feedback, or broader institutional support. The findings suggest that AI-mediated speaking practice can function as a supplementary scaffold for reducing communication-related dimensions of acculturative stress, but should be integrated with peer interaction, teacher feedback, and wider support services.

cs.HC↗

arXiv2026

A sequential explanatory mixed-methods study on the acceptance of a social robot for EFL speaking practice among Chinese primary school students: Insights from the Computers Are Social Actors (CASA) paradigm

This study investigates Chinese primary school students' acceptance of a social robot for English-as-a-foreign-language (EFL) speaking practice through a sequential explanatory mixed-methods design. Integrating the Technology Acceptance Model (TAM) and the Computers Are Social Actors (CASA) paradigm, the research explores both functional and social factors influencing learners' behavioural intention to use the robot. Quantitative data from 436 students were analysed using structural equation modelling, followed by qualitative interviews with twelve students to interpret the findings. Results show that perceived enjoyment and ease of use are the strongest predictors of acceptance, while social attributes such as warmth, anthropomorphism, and social presence significantly enhance enjoyment. Perceived intelligence affects usefulness but not ease of use. The findings suggest that emotional and social engagement are central to young learners' acceptance of educational robots, highlighting the importance of designing socially intelligent technologies that promote motivation and speaking confidence in EFL learning contexts.

cs.HC↗

arXiv2026

TASU2: Controllable CTC Simulation for Alignment and Low-Resource Adaptation of Speech LLMs

Speech LLM post-training increasingly relies on efficient cross-modal alignment and robust low-resource adaptation, yet collecting large-scale audio-text pairs remains costly. Text-only alignment methods such as TASU reduce this burden by simulating CTC posteriors from transcripts, but they provide limited control over uncertainty and error rate, making curriculum design largely heuristic. We propose \textbf{TASU2}, a controllable CTC simulation framework that simulates CTC posterior distributions under a specified WER range, producing text-derived supervision that better matches the acoustic decoding interface. This enables principled post-training curricula that smoothly vary supervision difficulty without TTS. Across multiple source-to-target adaptation settings, TASU2 improves in-domain and out-of-domain recognition over TASU, and consistently outperforms strong baselines including text-only fine-tuning and TTS-based augmentation, while mitigating source-domain performance degradation.

eess.AS↗

arXiv2025

AIMNET: An IoT-Empowered Digital Twin for Continuous Gas Emission Monitoring and Early Hazard Detection

A Digital Twin (DT) framework to enhance carbon-based gas plume monitoring is critical for supporting timely and effective mitigation responses to environmental hazards such as industrial gas leaks, or wildfire outbreaks carrying large carbon emissions. We present AIMNET, a one-of-a-kind DT framework that integrates a built-in-house Internet of Things (IoT)-based continuous sensing network with a physics-based multi-scale weather-gas transport model, that enables high-resolution and real-time simulation and detection of carbon gas emissions. AIMNET features a three-layer system architecture: (i) physical world: custom-built devices for continuous monitoring; (ii) bidirectional information feedback links: intelligent data transmission and reverse control; and (iii) digital twin world: AI-driven analytics for prediction, anomaly detection, and dynamic weather-gas coupled molecule transport modeling. Designed for scalable, energy-efficient deployment in remote environments, AIMNET architecture is realized through a small-scale distributed sensing network over an oil and gas production basin. To demonstrate the high-resolution, fast-responding concept, an equivalent mobile-based emission monitoring network was deployed around a wastewater treatment plant that constantly emits methane plumes. Our preliminary results through which, have successfully captured the methane emission events whose dynamics have been further resolved by the tiered model simulations. This work supports our position that AIMNET provides a promising DT framework for reliable, real-time monitoring and predictive risk assessment. In the end, we also discuss key implementation challenges and outline future directions for advancing such a new DT framework for translation deployment.

cs.NI↗

arXiv2025

Guiding Energy-Efficient Locomotion through Impact Mitigation Rewards

Animals achieve energy-efficient locomotion by their implicit passive dynamics, a marvel that has captivated roboticists for decades.Recently, methods incorporated Adversarial Motion Prior (AMP) and Reinforcement learning (RL) shows promising progress to replicate Animals' naturalistic motion. However, such imitation learning approaches predominantly capture explicit kinematic patterns, so-called gaits, while overlooking the implicit passive dynamics. This work bridges this gap by incorporating a reward term guided by Impact Mitigation Factor (IMF), a physics-informed metric that quantifies a robot's ability to passively mitigate impacts. By integrating IMF with AMP, our approach enables RL policies to learn both explicit motion trajectories from animal reference motion and the implicit passive dynamic. We demonstrate energy efficiency improvements of up to 32%, as measured by the Cost of Transport (CoT), across both AMP and handcrafted reward structure.

cs.RO↗

arXiv2025

Dynamic Quadrupedal Legged and Aerial Locomotion via Structure Repurposing

Multi-modal ground-aerial robots have been extensively studied, with a significant challenge lying in the integration of conflicting requirements across different modes of operation. The Husky robot family, developed at Northeastern University, and specifically the Husky v.2 discussed in this study, addresses this challenge by incorporating posture manipulation and thrust vectoring into multi-modal locomotion through structure repurposing. This quadrupedal robot features leg structures that can be repurposed for dynamic legged locomotion and flight. In this paper, we present the hardware design of the robot and report primary results on dynamic quadrupedal legged locomotion and hovering.

cs.RO↗

arXiv2025

Thruster-Enhanced Locomotion: A Decoupled Model Predictive Control with Learned Contact Residuals

Husky Carbon, a robot developed by Northeastern University, serves as a research platform to explore unification of posture manipulation and thrust vectoring. Unlike conventional quadrupeds, its joint actuators and thrusters enable enhanced control authority, facilitating thruster-assisted narrow-path walking. While a unified Model Predictive Control (MPC) framework optimizing both ground reaction forces and thruster forces could theoretically address this control problem, its feasibility is limited by the low torque-control bandwidth of the system's lightweight actuators. To overcome this challenge, we propose a decoupled control architecture: a Raibert-type controller governs legged locomotion using position-based control, while an MPC regulates the thrusters augmented by learned Contact Residual Dynamics (CRD) to account for leg-ground impacts. This separation bypasses the torque-control rate bottleneck while retaining the thruster MPC to explicitly account for leg-ground impact dynamics through learned residuals. We validate this approach through both simulation and hardware experiments, showing that the decoupled control architecture with CRD performs more stable behavior in terms of push recovery and cat-like walking gait compared to the decoupled controller without CRD.

cs.RO↗

arXiv2025

Quadratic Programming-Based Posture Manipulation and Thrust-vectoring for Agile Dynamic Walking on Narrow Pathways

There has been significant advancement in legged robot's agility where they can show impressive acrobatic maneuvers, such as parkour. These maneuvers rely heavily on posture manipulation. To expand the stability and locomotion plasticity, we use the multi-modal ability in our legged-aerial platform, the Husky Beta, to perform thruster-assisted walking. This robot has thrusters on each of its sagittal knee joints which can be used to stabilize its frontal dynamic as it walks. In this work, we perform a simulation study of quadruped narrow-path walking with Husky $β$, where the robot will utilize its thrusters to stably walk on a narrow path. The controller is designed based on a centroidal dynamics model with thruster and foot ground contact forces as inputs. These inputs are regulated using a QP solver to be used in a model predictive control framework. In addition to narrow-path walking, we also perform a lateral push-recovery simulation to study how the thrusters can be used to stabilize the frontal dynamics.

cs.RO↗

arXiv2025

Learning Efficient and Robust Language-conditioned Manipulation using Textual-Visual Relevancy and Equivariant Language Mapping

Controlling robots through natural language is pivotal for enhancing human-robot collaboration and synthesizing complex robot behaviors. Recent works that are trained on large robot datasets show impressive generalization abilities. However, such pretrained methods are (1) often fragile to unseen scenarios, and (2) expensive to adapt to new tasks. This paper introduces Grounded Equivariant Manipulation (GEM), a robust yet efficient approach that leverages pretrained vision-language models with equivariant language mapping for language-conditioned manipulation tasks. Our experiments demonstrate GEM's high sample efficiency and generalization ability across diverse tasks in both simulation and the real world. GEM achieves similar or higher performance with orders of magnitude fewer robot data compared with major data-efficient baselines such as CLIPort and VIMA. Finally, our approach demonstrates greater robustness compared to large VLA model, e.g, OpenVLA, at correctly interpreting natural language commands on unseen objects and poses. Code, data, and training details are available https://saulbatman.github.io/gem_page/

cs.RO↗

arXiv2025

A 3.584 Tbps coherent receiver chip on InP-LiNbO3 wafer-level integration platform

The rapid advancement of the thin-film lithium niobate (LiNbO3) platform has established it as a premier choice for high-performance photonics integrated circuits. However, the scalability and cost-efficiency of this platform are hindered by the reliance on chip-level fabrication and integration for passive and active components, necessitating a robust wafer-level LiNbO3 heterogeneous integration platform. Despite its critical role in enabling ultrahigh-speed optical interconnects, as well as optical mmWave/THz sensing and communication, the realization of ultrahigh-speed photodiodes and optical coherent receivers on the LiNbO_3 platform remains an unresolved challenge. This is primarily due to the challenges associated with the large-scale integration of direct-bandgap materials. To address these challenges, we have developed a scalable, high-speed InP-LiNbO3 wafer-level heterogeneous integration platform. This platform facilitates the fabrication of ultrahigh-speed photodiodes with a bandwidth of 140 GHz, capable of receiving high-quality 100-Gbaud pulse amplitude modulation (PAM4) signals. Moreover, we demonstrate a seven-channel, single-polarization I-Q coherent receiver chip with an aggregate receiving capacity of 3.584 Tbit/s. This coherent receiver exhibits a balanced detection bandwidth of 60 GHz and a common mode rejection ratio (CMRR) exceeding 20 dB. It achieves receiving capacities of 600 Gbit/s/λwith a 100-Gbaud 64-QAM signal and 512 Gbit/s/λwith a 128-Gbaud 16-QAM signal. Furthermore, energy consumption as low as 9.6 fJ/bit and 13.5 fJ/bit is achieved for 200 Gbit/s and 400 Gbit/s capacities, respectively. Our work provides a viable pathway toward enabling Pbps hyperscale data center interconnects, as well as optical mmWave/THz sensing and communication.

physics.optics↗

arXiv2025

Elevating Legal LLM Responses: Harnessing Trainable Logical Structures and Semantic Knowledge with Legal Reasoning

Large Language Models (LLMs) have achieved impressive results across numerous domains, yet they experience notable deficiencies in legal question-answering tasks. LLMs often generate generalized responses that lack the logical specificity required for expert legal advice and are prone to hallucination, providing answers that appear correct but are unreliable. Retrieval-Augmented Generation (RAG) techniques offer partial solutions to address this challenge, but existing approaches typically focus only on semantic similarity, neglecting the logical structure essential to legal reasoning. In this paper, we propose the Logical-Semantic Integration Model (LSIM), a novel supervised framework that bridges semantic and logical coherence. LSIM comprises three components: reinforcement learning predicts a structured fact-rule chain for each question, a trainable Deep Structured Semantic Model (DSSM) retrieves the most relevant candidate questions by integrating semantic and logical features, and in-context learning generates the final answer using the retrieved content. Our experiments on a real-world legal QA dataset-validated through both automated metrics and human evaluation-demonstrate that LSIM significantly enhances accuracy and reliability compared to existing methods.

cs.CL↗

arXiv2024

Conjugate momentum based thruster force estimate in dynamic multimodal robot

In a multi-modal system which combines thruster and legged locomotion such our state-of-the-art Harpy platform to perform dynamic locomotion. Therefore, it is very important to have a proper estimate of Thruster force. Harpy is a bipedal robot capable of legged-aerial locomotion using its legs and thrusters attached to its main frame. we can characterize thruster force using a thrust stand but it generally does not account for working conditions such as battery voltage. In this study, we present a momentum-based thruster force estimator. One of the key information required to estimate is terrain information. we show estimation results with and without terrain knowledge. In this work, we derive a conjugate momentum thruster force estimator and implement it on a numerical simulator that uses thruster force to perform thruster-assisted walking.

cs.RO↗

arXiv2024

Enhanced Capture Point Control Using Thruster Dynamics and QP-Based Optimization for Harpy

Our work aims to make significant strides in understanding unexplored locomotion control paradigms based on the integration of posture manipulation and thrust vectoring. These techniques are commonly seen in nature, such as Chukar birds using their wings to run on a nearly vertical wall. In this work, we developed a capture-point-based controller integrated with a quadratic programming (QP) solver which is used to create a thruster-assisted dynamic bipedal walking controller for our state-of-the-art Harpy platform. Harpy is a bipedal robot capable of legged-aerial locomotion using its legs and thrusters attached to its main frame. While capture point control based on centroidal models for bipedal systems has been extensively studied, the use of these thrusters in determining the capture point for a bipedal robot has not been extensively explored. The addition of these external thrust forces can lead to interesting interpretations of locomotion, such as virtual buoyancy studied in aquatic-legged locomotion. In this work, we derive a thruster-assisted bipedal walking with the capture point controller and implement it in simulation to study its performance.

cs.RO↗