相关论文: Bidirectional Attention as a Mixture of Continuous…

Mixture-of-Experts (MoE) models have emerged as a dominant paradigm for efficient LLM scaling, yet adapting them to non-English downstream tasks remains challenging. Existing fine-tuning approaches treat MoE models as monolithic learners,…

Modern large language models become multimodal, analyzing various data formats like text and images. While fine-tuning is effective for adapting these multimodal language models (MLMs) to downstream tasks, full fine-tuning is…

Language is not only a tool for communication but also a medium for human cognition and reasoning. If, as linguistic relativity suggests, the structure of language shapes cognitive patterns, then large language models (LLMs) trained on…

Mixture of experts (MoE) architectures have become a cornerstone for scaling up and are a key component in most large language models such as GPT-OSS, DeepSeek-V3, Llama-4, and Gemini-2.5. However, systematic research on MoE remains…

Multimodal large language models (MLLMs) have garnered widespread attention from researchers due to their remarkable understanding and generation capabilities in visual language tasks (e.g., visual question answering). However, the rapid…

Large language models (LLMs) have demonstrated impressive capabilities in various tasks using the in-context learning (ICL) paradigm. However, their effectiveness is often compromised by inherent bias, leading to prompt brittleness, i.e.,…

In this paper, we demonstrate that an inherent waveform pattern in the attention allocation of large language models (LLMs) significantly affects their performance in tasks demanding a high degree of context awareness, such as utilizing…

Universal Multimodal embedding models built on Multimodal Large Language Models (MLLMs) have traditionally employed contrastive learning, which aligns representations of query-target pairs across different modalities. Yet, despite its…

Mixture-of-Experts (MoE) architectures are often considered a natural fit for continual learning because sparse routing should localize updates and reduce interference, yet MoE Transformers still forget substantially even with sparse,…

As an important part of speech recognition technology, automatic speech keyword recognition has been intensively studied in recent years. Such technology becomes especially pivotal under situations with limited infrastructures and…

We study architectural and optimization techniques for sample-efficient language modeling under the constraints of the BabyLM 2025 shared task. Our model, BLaLM, replaces self-attention with a linear-time mLSTM token mixer and explores…

Mixture of Experts (MoE) architectures have recently advanced the scalability and adaptability of large language models (LLMs) for continual multimodal learning. However, efficiently extending these models to accommodate sequential tasks…

Multi-modal learning has emerged as a crucial research direction, as integrating textual and visual information can substantially enhance performance in tasks such as classification, retrieval, and scene understanding. Despite advances with…

Multilevel data are prevalent in many real-world applications. However, it remains an open research problem to identify and justify a class of models that flexibly capture a wide range of multilevel data. Motivated by the versatility of the…

Large language models allocate uniform computation across all tokens, ignoring that some sequences are trivially predictable while others require deep reasoning. We introduce ConceptMoE, which dynamically merges semantically similar tokens…

Although Transformers with fully connected self-attentions are powerful to model long-term dependencies, they are struggling to scale to long texts with thousands of words in language modeling. One of the solutions is to equip the model…

Transformers and deep state space models (SSMs) sit at opposite ends of a basic design choice: attention routes each query through a growing key-value (KV) cache by content-based matching at quadratic cost, while deep SSMs compress context…

Existing approaches to mapping-based cross-lingual word embeddings are based on the assumption that the source and target embedding spaces are structurally similar. The structures of embedding spaces largely depend on the co-occurrence…

Backdoor attacks creating 'sleeper agents' in large language models (LLMs) pose significant safety risks. This study employs mechanistic interpretability to explore resulting internal structural differences. Comparing clean Qwen2.5-3B…

The attentional mechanism has proven to be effective in improving end-to-end neural machine translation. However, due to the intricate structural divergence between natural languages, unidirectional attention-based models might only capture…