Yizhe Xiong

Yizhe Xiong, Hui Chen, Zijia Lin et al.

Unsupervised domain adaptation aims to transfer knowledge from a fully-labeled source domain to an unlabeled target domain. However, in real-world scenarios, providing abundant labeled data even in the source domain can be infeasible due to the difficulty and high expense of annotation. To address this issue, recent works consider the Few-shot Unsupervised Domain Adaptation (FUDA) where only a few source samples are labeled, and conduct knowledge transfer via self-supervised learning methods. Yet existing methods generally overlook that the sparse label setting hinders learning reliable source knowledge for transfer. Additionally, the learning difficulty difference in target samples is different but ignored, leaving hard target samples poorly classified. To tackle both deficiencies, in this paper, we propose a novel Confidence-based Visual Dispersal Transfer learning method (C-VisDiT) for FUDA. Specifically, C-VisDiT consists of a cross-domain visual dispersal strategy that transfers only high-confidence source knowledge for model adaptation and an intra-domain visual dispersal strategy that guides the learning of hard target samples with easy ones. We conduct extensive experiments on Office-31, Office-Home, VisDA-C, and DomainNet benchmark datasets and the results demonstrate that the proposed C-VisDiT significantly outperforms state-of-the-art FUDA methods. Our code is available at https://github.com/Bostoncake/C-VisDiT.

Zhenpeng Su, Zijia Lin, Xue Bai et al.

Scaling the size of a model enhances its capabilities but significantly increases computation complexity. Mixture-of-Experts models (MoE) address the issue by allowing model size to scale up without substantially increasing training or inference costs. In MoE, there is an important module called the router, which is used to distribute each token to the experts. Currently, the mainstream routing methods include dynamic routing and fixed routing. Despite their promising results, MoE models encounter several challenges. Primarily, for dynamic routing methods, the dispersion of training tokens across multiple experts can lead to underfitting, particularly for infrequent tokens. Additionally, though fixed routing methods can mitigate that issue, they compromise on the diversity of representations. In this paper, we propose \textbf{MaskMoE}, a method designed to enhance token-level learning by employing a routing \textbf{mask}ing technique within the \textbf{M}ixture-\textbf{o}f-\textbf{E}xperts model. MaskMoE is capable of maintaining representation diversity while achieving more comprehensive training. Experimental results demonstrate that our method outperforms previous dominant Mixture-of-Experts models in terms of both perplexity (PPL) and downstream task performance.

Liang Chen, Zekun Wang, Shuhuai Ren et al. · pku

Building on the foundations of language modeling in natural language processing, Next Token Prediction (NTP) has evolved into a versatile training objective for machine learning tasks across various modalities, achieving considerable success. As Large Language Models (LLMs) have advanced to unify understanding and generation tasks within the textual modality, recent research has shown that tasks from different modalities can also be effectively encapsulated within the NTP framework, transforming the multimodal information into tokens and predict the next one given the context. This survey introduces a comprehensive taxonomy that unifies both understanding and generation within multimodal learning through the lens of NTP. The proposed taxonomy covers five key aspects: Multimodal tokenization, MMNTP model architectures, unified task representation, datasets \& evaluation, and open challenges. This new taxonomy aims to aid researchers in their exploration of multimodal intelligence. An associated GitHub repository collecting the latest papers and repos is available at https://github.com/LMM101/Awesome-Multimodal-Next-Token-Prediction

Yizhe Xiong, Wei Huang, Xin Ye et al.

Post-training is essential for adapting Large Language Models (LLMs) to real-world applications. Deploying post-trained models faces significant challenges due to substantial memory overhead and noticeable inference latency. Existing work has identified significant redundancies in LLMs and proposed efficient architectures, namely intra-layer KV sharing and cross-layer KV sharing. However, intra-layer KV sharing still results in high inference costs, while cross-layer KV sharing leads to significant performance degradation. As a result, both methods remain suboptimal for post-training pre-trained LLMs. In this paper, we identify that the \texttt{Softmax} operation is a primary bottleneck for LLM inference and discover that it is actually highly redundant during post-training. We propose Softmax \textbf{Uni}fication in \textbf{Att}e\textbf{n}tion (\textbf{UniAttn}), a novel post-training method that unifies Softmax activations across transformer blocks to reduce LLM inference costs. Additionally, UniAttn adopts a linear projection to compensate for the errors induced by Softmax unification. Experiments show that UniAttn matches the performance of standard post-training while significantly reducing inference costs, outperforming existing efficient architectures during post-training. Our code will be available at \url{https://github.com/Bostoncake/UniAttn}.

Yiwen Liang, Hui Chen, Yizhe Xiong et al.

Vision-language models (VLMs) exhibit remarkable zero-shot capabilities but struggle with distribution shifts in downstream tasks when labeled data is unavailable, which has motivated the development of Test-Time Adaptation (TTA) to improve VLMs' performance during inference without annotations. Among various TTA approaches, cache-based methods show promise by preserving historical knowledge from low-entropy samples in a dynamic cache and fostering efficient adaptation. However, these methods face two critical reliability challenges: (1) entropy often becomes unreliable under distribution shifts, causing error accumulation in the cache and degradation in adaptation performance; (2) the final predictions may be unreliable due to inflexible decision boundaries that fail to accommodate large downstream shifts. To address these challenges, we propose a Reliable Test-time Adaptation (ReTA) method that integrates two complementary strategies to enhance reliability from two perspectives. First, to mitigate the unreliability of entropy as a sample selection criterion for cache construction, we introduce Consistency-aware Entropy Reweighting (CER), which incorporates consistency constraints to weight entropy during cache updating. While conventional approaches rely solely on low entropy for cache prioritization and risk introducing noise, our method leverages predictive consistency to maintain a high-quality cache and facilitate more robust adaptation. Second, we present Diversity-driven Distribution Calibration (DDC), which models class-wise text embeddings as multivariate Gaussian distributions, enabling adaptive decision boundaries for more accurate predictions across visually diverse content. Extensive experiments demonstrate that ReTA consistently outperforms state-of-the-art methods, particularly under real-world distribution shifts. Code: https://github.com/Evelyn1ywliang/ReTA.

Wei Huang, Yizhe Xiong, Xin Ye et al.

Large Language Models (LLMs) have achieved impressive performance across a range of natural language processing tasks. However, recent advances demonstrate that further gains particularly in complex reasoning tasks require more than merely scaling up model sizes or training data. One promising direction is to enable models to think during the reasoning process. Recently, Quiet STaR significantly improves reasoning by generating token-level thought traces, but incurs substantial inference overhead. In this work, we propose Fast Quiet STaR, a more efficient reasoning framework that preserves the benefits of token-level reasoning while reducing computational cost. Our method introduces a curriculum learning based training strategy that gradually reduces the number of thought tokens, enabling the model to internalize more abstract and concise reasoning processes. We further extend this approach to the standard Next Token Prediction (NTP) setting through reinforcement learning-based fine-tuning, resulting in Fast Quiet-STaR NTP, which eliminates the need for explicit thought token generation during inference. Experiments on four benchmark datasets with Mistral 7B and Qwen2.5 7B demonstrate that Fast Quiet-STaR consistently outperforms Quiet-STaR in terms of average accuracy under the same inference time budget. Notably, Fast Quiet-STaR NTP achieves an average accuracy improvement of 9\% on Mistral 7B and 5.7\% on Qwen2.5 7B, while maintaining the same inference latency. Our code will be available at https://github.com/huangwei200012/Fast-Quiet-STaR.

Yizhe Xiong, Hui Chen, Tianxiang Hao et al.

Recently, the scale of transformers has grown rapidly, which introduces considerable challenges in terms of training overhead and inference efficiency in the scope of task adaptation. Existing works, namely Parameter-Efficient Fine-Tuning (PEFT) and model compression, have separately investigated the challenges. However, PEFT cannot guarantee the inference efficiency of the original backbone, especially for large-scale models. Model compression requires significant training costs for structure searching and re-training. Consequently, a simple combination of them cannot guarantee accomplishing both training efficiency and inference efficiency with minimal costs. In this paper, we propose a novel Parallel Yielding Re-Activation (PYRA) method for such a challenge of training-inference efficient task adaptation. PYRA first utilizes parallel yielding adaptive weights to comprehensively perceive the data distribution in downstream tasks. A re-activation strategy for token modulation is then applied for tokens to be merged, leading to calibrated token features. Extensive experiments demonstrate that PYRA outperforms all competing methods under both low compression rate and high compression rate, demonstrating its effectiveness and superiority in maintaining both training efficiency and inference efficiency for large-scale foundation models. Our code is available at https://github.com/THU-MIG/PYRA.

Yizhe Xiong, Xiansheng Chen, Xin Ye et al.

Recently, Large Language Models (LLMs) have been widely adopted in a wide range of tasks, leading to increasing attention towards the research on how scaling LLMs affects their performance. Existing works, termed Scaling Laws, have discovered that the final test loss of LLMs scales as power-laws with model size, computational budget, and dataset size. However, the temporal change of the test loss of an LLM throughout its pre-training process remains unexplored, though it is valuable in many aspects, such as selecting better hyperparameters \textit{directly} on the target LLM. In this paper, we propose the novel concept of Temporal Scaling Law, studying how the test loss of an LLM evolves as the training steps scale up. In contrast to modeling the test loss as a whole in a coarse-grained manner, we break it down and dive into the fine-grained test loss of each token position, and further develop a dynamic hyperbolic-law. Afterwards, we derive the much more precise temporal scaling law by studying the temporal patterns of the parameters in the dynamic hyperbolic-law. Results on both in-distribution (ID) and out-of-distribution (OOD) validation datasets demonstrate that our temporal scaling law accurately predicts the test loss of LLMs across training steps. Our temporal scaling law has broad practical applications. First, it enables direct and efficient hyperparameter selection on the target LLM, such as data mixture proportions. Secondly, viewing the LLM pre-training dynamics from the token position granularity provides some insights to enhance the understanding of LLM pre-training.

Zhenpeng Su, Xing Wu, Zijia Lin et al.

Large language models (LLM) have been attracting much attention from the community recently, due to their remarkable performance in all kinds of downstream tasks. According to the well-known scaling law, scaling up a dense LLM enhances its capabilities, but also significantly increases the computational complexity. Mixture-of-Experts (MoE) models address that by allowing the model size to grow without substantially raising training or inference costs. Yet MoE models face challenges regarding knowledge sharing among experts, making their performance somehow sensitive to routing accuracy. To tackle that, previous works introduced shared experts and combined their outputs with those of the top $K$ routed experts in an ``addition'' manner. In this paper, inspired by collective matrix factorization to learn shared knowledge among data, we propose CartesianMoE, which implements more effective knowledge sharing among experts in more like a ``multiplication'' manner. Extensive experimental results indicate that CartesianMoE outperforms previous MoE models for building LLMs, in terms of both perplexity and downstream task performance. And we also find that CartesianMoE achieves better expert routing robustness.

Haoran Lian, Yizhe Xiong, Jianwei Niu et al.

Byte Pair Encoding (BPE) serves as a foundation method for text tokenization in the Natural Language Processing (NLP) field. Despite its wide adoption, the original BPE algorithm harbors an inherent flaw: it inadvertently introduces a frequency imbalance for tokens in the text corpus. Since BPE iteratively merges the most frequent token pair in the text corpus to generate a new token and keeps all generated tokens in the vocabulary, it unavoidably holds tokens that primarily act as components of a longer token and appear infrequently on their own. We term such tokens as Scaffold Tokens. Due to their infrequent occurrences in the text corpus, Scaffold Tokens pose a learning imbalance issue. To address that issue, we propose Scaffold-BPE, which incorporates a dynamic scaffold token removal mechanism by parameter-free, computation-light, and easy-to-implement modifications to the original BPE method. This novel approach ensures the exclusion of low-frequency Scaffold Tokens from the token representations for given texts, thereby mitigating the issue of frequency imbalance and facilitating model training. On extensive experiments across language modeling and even machine translation, Scaffold-BPE consistently outperforms the original BPE, well demonstrating its effectiveness.

Haoran Lian, Yizhe Xiong, Zijia Lin et al.

The prevalent use of Byte Pair Encoding (BPE) in Large Language Models (LLMs) facilitates robust handling of subword units and avoids issues of out-of-vocabulary words. Despite its success, a critical challenge persists: long tokens, rich in semantic information, have fewer occurrences in tokenized datasets compared to short tokens, which can result in imbalanced learning issue across different tokens. To address that, we propose LBPE, which prioritizes long tokens during the encoding process. LBPE generates tokens according to their reverse ranks of token length rather than their ranks in the vocabulary, granting longer tokens higher priority during the encoding process. Consequently, LBPE smooths the frequency differences between short and long tokens, and thus mitigates the learning imbalance. Extensive experiments across diverse language modeling tasks demonstrate that LBPE consistently outperforms the original BPE, well demonstrating its effectiveness.

Minxuan Lv, Zhenpeng Su, Leiyu Pan et al.

As large language models continue to scale, computational costs and resource consumption have emerged as significant challenges. While existing sparsification methods like pruning reduce computational overhead, they risk losing model knowledge through parameter removal. This paper proposes DSMoE (Dynamic Sparse Mixture-of-Experts), a novel approach that achieves sparsification by partitioning pre-trained FFN layers into computational blocks. We implement adaptive expert routing using sigmoid activation and straight-through estimators, enabling tokens to flexibly access different aspects of model knowledge based on input complexity. Additionally, we introduce a sparsity loss term to balance performance and computational efficiency. Extensive experiments on LLaMA models demonstrate that under equivalent computational constraints, DSMoE achieves superior performance compared to existing pruning and MoE approaches across language modeling and downstream tasks, particularly excelling in generation tasks. Analysis reveals that DSMoE learns distinctive layerwise activation patterns, providing new insights for future MoE architecture design.

Haoran Lian, Junmin Chen, Wei Huang et al.

Recently, Large language models (LLMs) have revolutionized Natural Language Processing (NLP). Pretrained LLMs, due to limited training context size, struggle with handling long token sequences, limiting their performance on various downstream tasks. Current solutions toward long context modeling often employ multi-stage continual pertaining, which progressively increases the effective context length through several continual pretraining stages. However, those approaches require extensive manual tuning and human expertise. In this paper, we introduce a novel single-stage continual pretraining method, Head-Adaptive Rotary Position Encoding (HARPE), to equip LLMs with long context modeling capabilities while simplifying the training process. Our HARPE leverages different Rotary Position Encoding (RoPE) base frequency values across different attention heads and directly trains LLMs on the target context length. Extensive experiments on 4 language modeling benchmarks, including the latest RULER benchmark, demonstrate that HARPE excels in understanding and integrating long-context tasks with single-stage training, matching and even outperforming existing multi-stage methods. Our results highlight that HARPE successfully breaks the stage barrier for training LLMs with long context modeling capabilities.

Yizhe Xiong, Zihan Zhou, Yiwen Liang et al.

Test-Time Adaptation (TTA) has emerged as an effective solution for adapting Vision Transformers (ViT) to distribution shifts without additional training data. However, existing TTA methods often incur substantial computational overhead, limiting their applicability in resource-constrained real-world scenarios. To reduce inference cost, plug-and-play token aggregation methods merge redundant tokens in ViTs to reduce total processed tokens. Albeit efficient, it suffers from significant performance degradation when directly integrated with existing TTA methods. We formalize this problem as Efficient Test-Time Adaptation (ETTA), seeking to preserve the adaptation capability of TTA while reducing inference latency. In this paper, we first provide a theoretical analysis from a novel mutual information perspective, showing that token aggregation inherently leads to information loss, which cannot be fully mitigated by conventional norm-tuning-based TTA methods. Guided by this insight, we propose to \textbf{N}eutralize Token \textbf{A}ggregation \textbf{v}ia \textbf{I}nformation \textbf{A}ugmentation (\textbf{NAVIA}). Specifically, we directly augment the [CLS] token embedding and incorporate adaptive biases into the [CLS] token in shallow layers of ViTs. We theoretically demonstrate that these augmentations, when optimized via entropy minimization, recover the information lost due to token aggregation. Extensive experiments across various out-of-distribution benchmarks demonstrate that NAVIA significantly outperforms state-of-the-art methods by over 2.5\%, while achieving an inference latency reduction of more than 20\%, effectively addressing the ETTA challenge.

Jiahuan Long, Tingsong Jiang, Wen Yao et al.

Vision foundation models (VFMs) have demonstrated remarkable capabilities in learning universal visual representations. However, adapting these models to downstream tasks conventionally requires parameter updates, with even parameter-efficient fine-tuning methods necessitating the modification of thousands to millions of weights. In this paper, we investigate the redundancies in the segment anything model (SAM) and then propose a novel parameter-free fine-tuning method. Unlike traditional fine-tuning methods that adjust parameters, our method emphasizes selecting, reusing, and enhancing pre-trained features, offering a new perspective on fine-tuning foundation models. Specifically, we introduce a channel selection algorithm based on the model's output difference to identify redundant and effective channels. By selectively replacing the redundant channels with more effective ones, we filter out less useful features and reuse more task-irrelevant features to downstream tasks, thereby enhancing the task-specific feature representation. Experiments on both out-of-domain and in-domain datasets demonstrate the efficiency and effectiveness of our method in different vision tasks (e.g., image segmentation, depth estimation and image classification). Notably, our approach can seamlessly integrate with existing fine-tuning strategies (e.g., LoRA, Adapter), further boosting the performance of already fine-tuned models. Moreover, since our channel selection involves only model inference, our method significantly reduces GPU memory overhead.

Leiyu Pan, Zhenpeng Su, Minxuan Lv et al.

Large language models have demonstrated exceptional performance across a wide range of tasks. However, dense models usually suffer from sparse activation, where many activation values tend towards zero (i.e., being inactivated). We argue that this could restrict the efficient exploration of model representation space. To mitigate this issue, we propose Finedeep, a deep-layered fine-grained expert architecture for dense models. Our framework partitions the feed-forward neural network layers of traditional dense models into small experts, arranges them across multiple sub-layers. A novel routing mechanism is proposed to determine each expert's contribution. We conduct extensive experiments across various model sizes, demonstrating that our approach significantly outperforms traditional dense architectures in terms of perplexity and benchmark performance while maintaining a comparable number of parameters and floating-point operations. Moreover, we find that Finedeep achieves optimal results when balancing depth and width, specifically by adjusting the number of expert sub-layers and the number of experts per sub-layer. Empirical results confirm that Finedeep effectively alleviates sparse activation and efficiently utilizes representation capacity in dense models.