arXiv subjects

Chong Li

Publications and source records attributed to Chong Li.

At least 55 records · Page 3Linked to original sources

arXiv2025

Group then Scale: Dynamic Mixture-of-Experts Multilingual Language Model

The curse of multilinguality phenomenon is a fundamental problem of multilingual Large Language Models (LLMs), where the competition between massive languages results in inferior performance. It mainly comes from limited capacity and negative transfer between dissimilar languages. To address this issue, we propose a method to dynamically group and scale up the parameters of multilingual LLM while boosting positive transfer among similar languages. Specifically, the model is first tuned on monolingual corpus to determine the parameter deviation in each layer and quantify the similarity between languages. Layers with more deviations are extended to mixture-of-experts layers to reduce competition between languages, where one expert module serves one group of similar languages. Experimental results on 18 to 128 languages show that our method reduces the negative transfer between languages and significantly boosts multilingual performance with fewer parameters. Such language group specialization on experts benefits the new language adaptation and reduces the inference on the previous multilingual knowledge learned.

cs.CL↗

arXiv2025

TokAlign: Efficient Vocabulary Adaptation via Token Alignment

Tokenization serves as a foundational step for Large Language Models (LLMs) to process text. In new domains or languages, the inefficiency of the tokenizer will slow down the training and generation of LLM. The mismatch in vocabulary also hinders deep knowledge transfer between LLMs like token-level distillation. To mitigate this gap, we propose an efficient method named TokAlign to replace the vocabulary of LLM from the token co-occurrences view, and further transfer the token-level knowledge between models. It first aligns the source vocabulary to the target one by learning a one-to-one mapping matrix for token IDs. Model parameters, including embeddings, are rearranged and progressively fine-tuned for the new vocabulary. Our method significantly improves multilingual text compression rates and vocabulary initialization for LLMs, decreasing the perplexity from 3.4$\text{e}^2$ of strong baseline methods to 1.2$\text{e}^2$ after initialization. Experimental results on models across multiple parameter scales demonstrate the effectiveness and generalization of TokAlign, which costs as few as 5k steps to restore the performance of the vanilla model. After unifying vocabularies between LLMs, token-level distillation can remarkably boost (+4.4% than sentence-level distillation) the base model, costing only 235M tokens.

cs.CL↗

arXiv2025

K12Vista: Exploring the Boundaries of MLLMs in K-12 Education

Multimodal large language models have demonstrated remarkable reasoning capabilities in various visual tasks. However, their abilities in K12 scenarios are still systematically underexplored. Previous studies suffer from various limitations including narrow subject coverage, insufficient data scale, lack of diversity in question types, and naive answer-centric evaluation method, resulting in insufficient exploration of model capabilities. To address these gaps, we propose K12Vista, the most comprehensive multimodal benchmark for Chinese K12 subject knowledge understanding and reasoning to date, featuring 33,000 questions across five core subjects from primary to high school and three question types. Moreover, beyond the final outcome, we are also concerned with the correctness of MLLMs' reasoning processes. For this purpose, we meticulously compiles errors from MLLMs' reasoning processes and leverage an automated data pipeline to construct K12-PEM-800K, the largest process evaluation dataset offering detailed step-by-step judgement annotations for MLLMs' reasoning. Subsequently, we developed K12-PEM, an advanced process evaluation model that integrates an overall assessment of both the reasoning process and answer correctness. Moreover, we also introduce K12-PEBench, the first high-quality, human-annotated benchmark specifically designed for evaluating abilities of reasoning process evaluation.Extensive experiments reveal that current MLLMs exhibit significant flaws when reasoning within K12Vista, providing critical insights for the development of more capable MLLMs.We open our resources at https://github.com/lichongod/K12Vista.

cs.AI↗

arXiv2025

Evaluating Menu OCR and Translation: A Benchmark for Aligning Human and Automated Evaluations in Large Vision-Language Models

The rapid advancement of large vision-language models (LVLMs) has significantly propelled applications in document understanding, particularly in optical character recognition (OCR) and multilingual translation. However, current evaluations of LVLMs, like the widely used OCRBench, mainly focus on verifying the correctness of their short-text responses and long-text responses with simple layout, while the evaluation of their ability to understand long texts with complex layout design is highly significant but largely overlooked. In this paper, we propose Menu OCR and Translation Benchmark (MOTBench), a specialized evaluation framework emphasizing the pivotal role of menu translation in cross-cultural communication. MOTBench requires LVLMs to accurately recognize and translate each dish, along with its price and unit items on a menu, providing a comprehensive assessment of their visual understanding and language processing capabilities. Our benchmark is comprised of a collection of Chinese and English menus, characterized by intricate layouts, a variety of fonts, and culturally specific elements across different languages, along with precise human annotations. Experiments show that our automatic evaluation results are highly consistent with professional human evaluation. We evaluate a range of publicly available state-of-the-art LVLMs, and through analyzing their output to identify the strengths and weaknesses in their performance, offering valuable insights to guide future advancements in LVLM development. MOTBench is available at https://github.com/gitwzl/MOTBench.

cs.LG↗

arXiv2025

Towards a Vision-Language Episodic Memory Framework: Large-scale Pretrained Model-Augmented Hippocampal Attractor Dynamics

Modeling episodic memory (EM) remains a significant challenge in both neuroscience and AI, with existing models either lacking interpretability or struggling with practical applications. This paper proposes the Vision-Language Episodic Memory (VLEM) framework to address these challenges by integrating large-scale pretrained models with hippocampal attractor dynamics. VLEM leverages the strong semantic understanding of pretrained models to transform sensory input into semantic embeddings as the neocortex, while the hippocampus supports stable memory storage and retrieval through attractor dynamics. In addition, VLEM incorporates prefrontal working memory and the entorhinal gateway, allowing interaction between the neocortex and the hippocampus. To facilitate real-world applications, we introduce EpiGibson, a 3D simulation platform for generating episodic memory data. Experimental results demonstrate the VLEM framework's ability to efficiently learn high-level temporal representations from sensory input, showcasing its robustness, interpretability, and applicability in real-world scenarios.

q-bio.NC↗

arXiv2025

Classical Interfaces for Controlling Cryogenic Quantum Computing Technologies

Quantum processors have the potential to revolutionise computing on a scale unseen since the development of semiconductor technology in the middle of the 20th century. However, while there is now huge activity and investment in the field, there are a number of challenges that must be overcome before the technology can be fully realised. Of primary concern is the development of the classical technology required to interface with quantum systems, as we push towards a new era of high-performance, large-scale quantum computing. In this review, we briefly discuss some of the main challenges facing the development of universally useful quantum computers and the different architectures being investigated. We are primarily concerned with cryogenic quantum systems. These systems are among the most mature quantum computing architectures to date, and are garnering a lot of both industrial and academic attention. We present and analyse the leading methods of interfacing with quantum processors, both now and for the next generation of larger, multi-qubit systems. Recent advancements in control cryoelectronics, both semiconducting and superconducting, are covered, while a view towards newer methods such as optical and wireless qubit interfaces are also presented.

quant-ph↗

arXiv2025

DIMT25@ICDAR2025: HW-TSC's End-to-End Document Image Machine Translation System Leveraging Large Vision-Language Model

This paper presents the technical solution proposed by Huawei Translation Service Center (HW-TSC) for the "End-to-End Document Image Machine Translation for Complex Layouts" competition at the 19th International Conference on Document Analysis and Recognition (DIMT25@ICDAR2025). Leveraging state-of-the-art open-source large vision-language model (LVLM), we introduce a training framework that combines multi-task learning with perceptual chain-of-thought to develop a comprehensive end-to-end document translation system. During the inference phase, we apply minimum Bayesian decoding and post-processing strategies to further enhance the system's translation capabilities. Our solution uniquely addresses both OCR-based and OCR-free document image translation tasks within a unified framework. This paper systematically details the training methods, inference strategies, LVLM base models, training data, experimental setups, and results, demonstrating an effective approach to document image machine translation.

cs.CV↗

arXiv2025

Beyond Words: AuralLLM and SignMST-C for Sign Language Production and Bidirectional Accessibility

Sign language is the primary communication mode for 72 million hearing-impaired individuals worldwide, necessitating effective bidirectional Sign Language Production and Sign Language Translation systems. However, functional bidirectional systems require a unified linguistic environment, hindered by the lack of suitable unified datasets, particularly those providing the necessary pose information for accurate Sign Language Production (SLP) evaluation. Concurrently, current SLP evaluation methods like back-translation ignore pose accuracy, and high-quality coordinated generation remains challenging. To create this crucial environment and overcome these challenges, we introduce CNText2Sign and CNSign, which together constitute the first unified dataset aimed at supporting bidirectional accessibility systems for Chinese sign language; CNText2Sign provides 15,000 natural language-to-sign mappings and standardized skeletal keypoints for 8,643 vocabulary items supporting pose assessment. Building upon this foundation, we propose the AuraLLM model, which leverages a decoupled architecture with CNText2Sign's pose data for novel direct gesture accuracy assessment. The model employs retrieval augmentation and Cascading Vocabulary Resolution to handle semantic mapping and out-of-vocabulary words and achieves all-scenario production with controllable coordination of gestures and facial expressions via pose-conditioned video synthesis. Concurrently, our Sign Language Translation model SignMST-C employs targeted self-supervised pretraining for dynamic feature capture, achieving new SOTA results on PHOENIX2014-T with BLEU-4 scores up to 32.08. AuraLLM establishes a strong performance baseline on CNText2Sign with a BLEU-4 score of 50.41 under direct evaluation.

cs.CV↗

arXiv2025

DecoFuse: Decomposing and Fusing the "What", "Where", and "How" for Brain-Inspired fMRI-to-Video Decoding

Decoding visual experiences from brain activity is a significant challenge. Existing fMRI-to-video methods often focus on semantic content while overlooking spatial and motion information. However, these aspects are all essential and are processed through distinct pathways in the brain. Motivated by this, we propose DecoFuse, a novel brain-inspired framework for decoding videos from fMRI signals. It first decomposes the video into three components - semantic, spatial, and motion - then decodes each component separately before fusing them to reconstruct the video. This approach not only simplifies the complex task of video decoding by decomposing it into manageable sub-tasks, but also establishes a clearer connection between learned representations and their biological counterpart, as supported by ablation studies. Further, our experiments show significant improvements over previous state-of-the-art methods, achieving 82.4% accuracy for semantic classification, 70.6% accuracy in spatial consistency, a 0.212 cosine similarity for motion prediction, and 21.9% 50-way accuracy for video generation. Additionally, neural encoding analyses for semantic and spatial information align with the two-streams hypothesis, further validating the distinct roles of the ventral and dorsal pathways. Overall, DecoFuse provides a strong and biologically plausible framework for fMRI-to-video decoding. Project page: https://chongjg.github.io/DecoFuse/.

cs.CV↗

arXiv2025

Baichuan-Omni-1.5 Technical Report

We introduce Baichuan-Omni-1.5, an omni-modal model that not only has omni-modal understanding capabilities but also provides end-to-end audio generation capabilities. To achieve fluent and high-quality interaction across modalities without compromising the capabilities of any modality, we prioritized optimizing three key aspects. First, we establish a comprehensive data cleaning and synthesis pipeline for multimodal data, obtaining about 500B high-quality data (text, audio, and vision). Second, an audio-tokenizer (Baichuan-Audio-Tokenizer) has been designed to capture both semantic and acoustic information from audio, enabling seamless integration and enhanced compatibility with MLLM. Lastly, we designed a multi-stage training strategy that progressively integrates multimodal alignment and multitask fine-tuning, ensuring effective synergy across all modalities. Baichuan-Omni-1.5 leads contemporary models (including GPT4o-mini and MiniCPM-o 2.6) in terms of comprehensive omni-modal capabilities. Notably, it achieves results comparable to leading models such as Qwen2-VL-72B across various multimodal medical benchmarks.

cs.CL↗

arXiv2025

Magnon Blockade with Skyrmion Qubit-Magnon Coupling in a Hybrid Quantum System

Magnon blockade is a fundamental quantum phenomenon for generating single-magnon state, which gradually becomes one of the candidates for quantum information processing. In this paper, we propose a theoretical scheme to generate the magnon blockade in a hybrid system consisting of a YIG micromagnet and a skyrmion. Considering weak probing of the magnon and driving of the skyrmion qubit, the second-order correlation function is analytically derived, and the optimal condition for realizing the magnon blockade is identified. Under the optimal condition, we systematically analyze the behavior of the second-order correlation function $g^{(2)}(0)$ under different parameter regimes. Our analysis shows that with appropriate driving and probing field intensities, the magnon blockade effect can be significantly enhanced, effectively suppressing multi-magnon states and facilitating the generation of high-purity single-magnon states exhibiting pronounced antibunching. Furthermore, we explore the physical mechanisms underlying the magnon blockade, revealing the coexistence and interplay of conventional and unconventional magnon blockade. This scheme provides a versatile all-magnetic platform for generating high purity single-magnon sources.

quant-ph↗

arXiv2025

Examining Turbulence in Galactic Molecular Clouds -- I: A Statistical Analysis of Velocity Structures

We present a systematic analysis of the velocity structure functions (VSFs) of 167 molecular clouds with angular sizes greater than $\sim$176 arcmin$^2$ in three sectors of the Galactic mid-plane. We calculated the 1st- to 3rd-order VSFs and found that 60\% of the VSFs exhibit power-law distributions. The relative power-law exponents are consistent with predictions from intermittent turbulence models. Column density weighting reduces the proportion of power-law VSFs and steepens the VSF slopes, implying a reduction of turbulent energy in high-density regions. All clouds show small-scale intermittency, with slightly stronger intermittency in those molecular clouds showing none power-law VSFs. Negative VSF exponents that may indicate gravitational collapse are not observed in our sample. The scaling exponents of the observed VSFs do not correlate with the virial parameters of the molecular clouds. These two observations suggest that gravity-dominated scales in molecular clouds still need further investigation. Consistent VSF scaling exponents for the molecular clouds with significant power-law VSFs suggest large-scale external driving of turbulence in these molecular clouds. However, the driving mechanisms are likely not universal, as the power-law scaling coefficients in our results show relatively large scatter. The fact that nearly 40\% of the VSFs deviate to some extent from power-law distributions suggests that the influence of local environments on the internal turbulence of molecular clouds may not be negligible.

astro-ph.GA↗

arXiv2025

Decoding the Flow: CauseMotion for Emotional Causality Analysis in Long-form Conversations

Long-sequence causal reasoning seeks to uncover causal relationships within extended time series data but is hindered by complex dependencies and the challenges of validating causal links. To address the limitations of large-scale language models (e.g., GPT-4) in capturing intricate emotional causality within extended dialogues, we propose CauseMotion, a long-sequence emotional causal reasoning framework grounded in Retrieval-Augmented Generation (RAG) and multimodal fusion. Unlike conventional methods relying only on textual information, CauseMotion enriches semantic representations by incorporating audio-derived features-vocal emotion, emotional intensity, and speech rate-into textual modalities. By integrating RAG with a sliding window mechanism, it effectively retrieves and leverages contextually relevant dialogue segments, thus enabling the inference of complex emotional causal chains spanning multiple conversational turns. To evaluate its effectiveness, we constructed the first benchmark dataset dedicated to long-sequence emotional causal reasoning, featuring dialogues with over 70 turns. Experimental results demonstrate that the proposed RAG-based multimodal integrated approach, the efficacy of substantially enhances both the depth of emotional understanding and the causal inference capabilities of large-scale language models. A GLM-4 integrated with CauseMotion achieves an 8.7% improvement in causal accuracy over the original model and surpasses GPT-4o by 1.2%. Additionally, on the publicly available DiaASQ dataset, CauseMotion-GLM-4 achieves state-of-the-art results in accuracy, F1 score, and causal reasoning accuracy.

cs.CL↗

arXiv2024

VASA-1: Lifelike Audio-Driven Talking Faces Generated in Real Time

We introduce VASA, a framework for generating lifelike talking faces with appealing visual affective skills (VAS) given a single static image and a speech audio clip. Our premiere model, VASA-1, is capable of not only generating lip movements that are exquisitely synchronized with the audio, but also producing a large spectrum of facial nuances and natural head motions that contribute to the perception of authenticity and liveliness. The core innovations include a holistic facial dynamics and head movement generation model that works in a face latent space, and the development of such an expressive and disentangled face latent space using videos. Through extensive experiments including evaluation on a set of new metrics, we show that our method significantly outperforms previous methods along various dimensions comprehensively. Our method not only delivers high video quality with realistic facial and head dynamics but also supports the online generation of 512x512 videos at up to 40 FPS with negligible starting latency. It paves the way for real-time engagements with lifelike avatars that emulate human conversational behaviors.

cs.CV↗

arXiv2024

LLM$\times$MapReduce: Simplified Long-Sequence Processing using Large Language Models

Enlarging the context window of large language models (LLMs) has become a crucial research area, particularly for applications involving extremely long texts. In this work, we propose a novel training-free framework for processing long texts, utilizing a divide-and-conquer strategy to achieve comprehensive document understanding. The proposed LLM$\times$MapReduce framework splits the entire document into several chunks for LLMs to read and then aggregates the intermediate answers to produce the final output. The main challenge for divide-and-conquer long text processing frameworks lies in the risk of losing essential long-range information when splitting the document, which can lead the model to produce incomplete or incorrect answers based on the segmented texts. Disrupted long-range information can be classified into two categories: inter-chunk dependency and inter-chunk conflict. We design a structured information protocol to better cope with inter-chunk dependency and an in-context confidence calibration mechanism to resolve inter-chunk conflicts. Experimental results demonstrate that LLM$\times$MapReduce can outperform representative open-source and commercial long-context LLMs, and is applicable to several different models.

cs.CL↗

arXiv2024

NeuroPictor: Refining fMRI-to-Image Reconstruction via Multi-individual Pretraining and Multi-level Modulation

Recent fMRI-to-image approaches mainly focused on associating fMRI signals with specific conditions of pre-trained diffusion models. These approaches, while producing high-quality images, capture only a limited aspect of the complex information in fMRI signals and offer little detailed control over image creation. In contrast, this paper proposes to directly modulate the generation process of diffusion models using fMRI signals. Our approach, NeuroPictor, divides the fMRI-to-image process into three steps: i) fMRI calibrated-encoding, to tackle multi-individual pre-training for a shared latent space to minimize individual difference and enable the subsequent multi-subject training; ii) fMRI-to-image multi-subject pre-training, perceptually learning to guide diffusion model with high- and low-level conditions across different individuals; iii) fMRI-to-image single-subject refining, similar with step ii but focus on adapting to particular individual. NeuroPictor extracts high-level semantic features from fMRI signals that characterizing the visual stimulus and incrementally fine-tunes the diffusion model with a low-level manipulation network to provide precise structural instructions. By training with about 67,000 fMRI-image pairs from various individuals, our model enjoys superior fMRI-to-image decoding capacity, particularly in the within-subject setting, as evidenced in benchmark datasets. Our code and model are available at https://jingyanghuo.github.io/neuropictor/.

cs.CV↗

arXiv2024

The existence of solutions for a Schrodinger equation with jumping nonlinearities crossing the essential spectrum

In this paper, we establish the existence of one solution for a Schrödinger equation with jumping nonlinearities: $-Δu+V(x)u=f(x,u)$, $x\in \mathbb {R}^N$, and $u(x)\to 0$, $|x|\to +\infty$, where $V$ is a potential function on which we make hypotheses, and in particular allow $V$ which is unbounded below, and $f(x,u)=au^-+bu^++g(x,u)$. No restriction on $b$ is required, which implies that $f(x,s)s^{-1}$ may interfere with the essential spectrum of $ -Δ+V$ for $s\to +\infty$. Using the truncation method and the Morse theory, we can compute critical groups of the corresponding functional at zero and infinity, then prove the existence of one negative solution.

math.AP↗

arXiv2024

Improving In-context Learning of Multilingual Generative Language Models with Cross-lingual Alignment

Multilingual generative models obtain remarkable cross-lingual in-context learning capabilities through pre-training on large-scale corpora. However, they still exhibit a performance bias toward high-resource languages and learn isolated distributions of multilingual sentence representations, which may hinder knowledge transfer across languages. To bridge this gap, we propose a simple yet effective cross-lingual alignment framework exploiting pairs of translation sentences. It aligns the internal sentence representations across different languages via multilingual contrastive learning and aligns outputs by following cross-lingual instructions in the target language. Experimental results show that even with less than 0.1 {\textperthousand} of pre-training tokens, our alignment framework significantly boosts the cross-lingual abilities of generative language models and mitigates the performance gap. Further analyses reveal that it results in a better internal multilingual representation distribution of multilingual models.

cs.CL↗