paper-with-me

홈 › Papers

Loop Neural Networks for Parameter Sharing

2024-09-21 · Kei-Sing Ng, Qingchen Wang

The success of large-scale language models like GPT can be attributed to their ability to efficiently predict the next token in a sequence. However, these models rely on constant computational effort regardless of the complexity of the token they are predicting, lacking the capacity for iterative refinement. In this paper, we introduce a novel Loop Neural Network, which achieves better performance by utilizing longer computational time without increasing the model size. Our approach revisits the input multiple times, refining the prediction by iteratively looping over a subset of the model with residual connections. We demonstrate the effectiveness of this method through experiments comparing versions of GPT-2 with our loop models, showing improved performance in language modeling tasks while maintaining similar parameter counts. Importantly, these improvements are achieved without the need for extra training data.

📄 PDF Abstract BibTeX arXiv:2409.14199

Code (0)

등록된 구현이 없습니다.

Tasks

Language ModelingLanguage ModellingText Generation

Methods 이 논문이 사용한 방법론

Refunds@Expedia|||How do I get a full refund from Expedia? “How do I get a full refund from Expedia? How do I get a full refund from Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Quick Help &…
Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Cosine Annealing Cosine Annealing is a type of learning rate schedule that has the effect of starting with a large learning rate that is relatively rapidly decreased to a minimum value before…
Multi-Head Attention 설명 없음
Weight Decay 설명 없음
Linear Warmup With Cosine Annealing Linear Warmup With Cosine Annealing is a learning rate schedule where we increase the learning rate linearly for $n$ updates and then anneal according to a cosine schedule…
Adam 설명 없음

Similar Papers 제목 키워드 기반

Accelerating Meta-Learning by Sharing Gradients

2023-12-13 · Oscar Chang, Hod Lipson

The success of gradient-based meta-learning is primarily attributed to its ability to leverage related tasks to learn task-invariant information. However, the absence of interactions between different tasks in the inner …

Meta-LearningMulti-Task Learning

Looped GPT-BERT: Trading Parameters for Computation in Small Language Modeling

2026-09-09 · Tingshuo Fan, Hongtao Mu, Tianyu Zhou, Hansen Liu 외 arxiv

When training data are limited, increasing parameter count is not the only way to improve language-model performance. A small parameter set, when repeatedly applied, can also deliver comparable performance. We study Loop…

On the Residual Scaling of Looped Transformers: Stability and Transferability

2026-06-16 · Shaowen Wang, Bingrui Li, Ge Zhang, Wenhao Huang 외 arxiv

Looped (weight-tied) Transformers apply a shared residual block $N$ times ($h \leftarrow h + \varepsilon\,f(h)$, same $f$ at each step), increasing effective depth without adding parameters. Prior depth-scaling analyses …

LoopMoE: Unifying Iterative Computation with Mixture-of-Experts for Language Modeling

2026-06-03 · Wenkai Chen, Tianshu Li, Wenyong Huang, Yichun Yin 외 arxiv

Mixture-of-Experts (MoE) and looped architectures scale models along two orthogonal axes, namely parameter capacity and effective depth. However, mainstream looped architectures rely on dense backbones that couple parame…

Looped State-Space Language Models with Adaptive Exit-State Selection

2026-07-11 · Zhenxuan Yu, Takeshi Kojima, Yutaka Matsuo, Yusuke Iwasawa arxiv

Recent work on looped language models suggests that many reasoning problems benefit from greater computational depth rather than from additional independent parameters. Existing studies, however, focus almost exclusively…