KV Shifting Attention Enhances Language Modeling

• URL: http://arxiv.org/abs/2411.19574v2
• Date: Thu, 05 Dec 2024 12:19:38 GMT
• Title: KV Shifting Attention Enhances Language Modeling
• Authors: Mingyu Xu, Wei Cheng, Bingning Wang, Weipeng Chen,
• Abstract summary: Current large language models are mainly based on decode-only structure transformers, which have great in-context learning capabilities.<n>We propose a KV shifting attention to more efficiently implement the ability of the model's induction.<n>Our experimental results demonstrate that KV shifting attention is beneficial to learning induction heads and language modeling.
• Score: 10.265219156828907
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: The current large language models are mainly based on decode-only structure transformers, which have great in-context learning (ICL) capabilities. It is generally believed that the important foundation of its ICL capability is the induction heads mechanism, which requires at least two layers attention. In order to more efficiently implement the ability of the model's induction, we revisit the induction heads mechanism and proposed a KV shifting attention. We theoretically prove that the KV shifting attention reducing the model's requirements for the depth and width of the induction heads mechanism. Our experimental results demonstrate that KV shifting attention is beneficial to learning induction heads and language modeling, which lead to better performance or faster convergence from toy models to the pre-training models with more than 10 B parameters.

Related papers

• Explicit Multi-head Attention for Inter-head Interaction in Large Language Models [70.96854312026319]
  Multi-head Explicit Attention (MEA) is a simple yet effective attention variant that explicitly models cross-head interaction.<n>MEA shows strong robustness in pretraining, which allows the use of larger learning rates that lead to faster convergence.<n>This enables a practical key-value cache compression strategy that reduces KV-cache memory usage by 50% with negligible performance loss.
  arXiv  Detail & Related papers  (2026-01-27T13:45:03Z)
• How LLMs Learn to Reason: A Complex Network Perspective [14.638878448692493]
  Training large language models with Reinforcement Learning from Verifiable Rewards exhibits a set of puzzling behaviors.<n>We propose that these seemingly disparate phenomena can be explained using a single unifying theory.<n>Our work provides a new physical intuition for engineering the emergent reasoning capabilities of future AI systems.
  arXiv  Detail & Related papers  (2025-09-28T04:10:37Z)
• What One Cannot, Two Can: Two-Layer Transformers Provably Represent Induction Heads on Any-Order Markov Chains [64.31313691823088]
  In-context learning (ICL) is a capability of transformers, through which trained models learn to adapt to new tasks by leveraging information from the input context.<n>We show that a two-layer transformer with one head per layer can indeed represent any conditional k-gram.
  arXiv  Detail & Related papers  (2025-08-10T07:03:01Z)
• KV-Latent: Dimensional-level KV Cache Reduction with Frequency-aware Rotary Positional Embedding [72.12756830560217]
  Large language models (LLMs) based on Transformer Decoders have become the preferred choice for conversational generative AI.<n>Despite the overall superiority of the Decoder architecture, the gradually increasing Key-Value cache during inference has emerged as a primary efficiency bottleneck.<n>By down-sampling the Key-Value vector dimensions into a latent space, we can significantly reduce the KV Cache footprint and improve inference speed.
  arXiv  Detail & Related papers  (2025-07-15T12:52:12Z)
• Neural Attention: A Novel Mechanism for Enhanced Expressive Power in Transformer Models [0.0]
  We propose a technique that replaces dot products with feed-forward networks, enabling a more expressive representation of relationships between tokens. This work establishes Neural Attention as an effective means of enhancing the predictive capabilities of transformer models across a variety of applications.
  arXiv  Detail & Related papers  (2025-02-24T14:39:40Z)
• Which Attention Heads Matter for In-Context Learning? [41.048579134842285]
  Large language models (LLMs) exhibit impressive in-context learning (ICL) capability. Two different mechanisms have been proposed to explain ICL: induction heads that find and copy relevant tokens, and function vector (FV) heads whose activations compute a latent encoding of the ICL task. We study and compare induction heads and FV heads in 12 language models.
  arXiv  Detail & Related papers  (2025-02-19T12:25:02Z)
• Scalable Language Models with Posterior Inference of Latent Thought Vectors [52.63299874322121]
  Latent-Thought Language Models (LTMs) incorporate explicit latent thought vectors that follow an explicit prior model in latent space. LTMs possess additional scaling dimensions beyond traditional LLMs, yielding a structured design space. LTMs significantly outperform conventional autoregressive models and discrete diffusion models in validation perplexity and zero-shot language modeling.
  arXiv  Detail & Related papers  (2025-02-03T17:50:34Z)
• Tensor Product Attention Is All You Need [53.69820973900921]
  Product Attention (TPA) is a novel attention mechanism that uses tensor decompositions to represent queries, keys, and values compactly.<n>TPA achieves improved model quality alongside memory efficiency.<n>Based on TPA, we introduce the Product Attention Transformer,(T6), a new model architecture for sequence modeling.
  arXiv  Detail & Related papers  (2025-01-11T03:37:10Z)
• Efficient Machine Translation with a BiLSTM-Attention Approach [0.0]
  This paper proposes a novel Seq2Seq model aimed at improving translation quality while reducing the storage space required by the model. The model employs a Bidirectional Long Short-Term Memory network (Bi-LSTM) as the encoder to capture the context information of the input sequence. Compared to the current mainstream Transformer model, our model achieves superior performance on the WMT14 machine translation dataset.
  arXiv  Detail & Related papers  (2024-10-29T01:12:50Z)
• A Primal-Dual Framework for Transformers and Neural Networks [52.814467832108875]
  Self-attention is key to the remarkable success of transformers in sequence modeling tasks. We show that the self-attention corresponds to the support vector expansion derived from a support vector regression problem. We propose two new attentions: Batch Normalized Attention (Attention-BN) and Attention with Scaled Head (Attention-SH)
  arXiv  Detail & Related papers  (2024-06-19T19:11:22Z)
• Boosting Vision-Language Models with Transduction [12.281505126587048]
  We present TransCLIP, a novel and computationally efficient transductive approach for vision-language models. TransCLIP is applicable as a plug-and-play module on top of popular inductive zero- and few-shot models.
  arXiv  Detail & Related papers  (2024-06-03T23:09:30Z)
• DistiLLM: Towards Streamlined Distillation for Large Language Models [53.46759297929675]
  DistiLLM is a more effective and efficient KD framework for auto-regressive language models. DisiLLM comprises two components: (1) a novel skew Kullback-Leibler divergence loss, where we unveil and leverage its theoretical properties, and (2) an adaptive off-policy approach designed to enhance the efficiency in utilizing student-generated outputs.
  arXiv  Detail & Related papers  (2024-02-06T11:10:35Z)
• Interpreting and Improving Attention From the Perspective of Large Kernel Convolution [51.06461246235176]
  We introduce Large Kernel Convolutional Attention (LKCA), a novel formulation that reinterprets attention operations as a single large- Kernel convolution. LKCA achieves competitive performance across various visual tasks, particularly in data-constrained settings.
  arXiv  Detail & Related papers  (2024-01-11T08:40:35Z)
• Improving In-context Learning via Bidirectional Alignment [41.214003703218914]
  Large language models (LLMs) have shown impressive few-shot generalization on many tasks via in-context learning (ICL) We propose Bidirectional Alignment (BiAlign) to fully leverage the models' preferences for ICL examples to improve the ICL abilities of student models. Specifically, we introduce the alignment of input preferences between student and teacher models by incorporating a novel ranking loss.
  arXiv  Detail & Related papers  (2023-12-28T15:02:03Z)
• Fast-ELECTRA for Efficient Pre-training [83.29484808667532]
  ELECTRA pre-trains language models by detecting tokens in a sequence that have been replaced by an auxiliary model. We propose Fast-ELECTRA, which leverages an existing language model as the auxiliary model. Our approach rivals the performance of state-of-the-art ELECTRA-style pre-training methods, while significantly eliminating the computation and memory cost brought by the joint training of the auxiliary model.
  arXiv  Detail & Related papers  (2023-10-11T09:55:46Z)
• Learning to Grow Pretrained Models for Efficient Transformer Training [72.20676008625641]
  We learn to grow pretrained transformers, where we learn to linearly map the parameters of the smaller model to initialize the larger model. Experiments across both language and vision transformers demonstrate that our learned Linear Growth Operator (LiGO) can save up to 50% computational cost of training from scratch.
  arXiv  Detail & Related papers  (2023-03-02T05:21:18Z)
• In-context Learning and Induction Heads [5.123049926855312]
  "Induction heads" are attention heads that implement a simple algorithm to complete token sequences. We find that induction heads develop at precisely the same point as a sudden sharp increase in in-context learning ability.
  arXiv  Detail & Related papers  (2022-09-24T00:43:19Z)
• Low-Rank Bottleneck in Multi-head Attention Models [74.83235382203604]
  We argue that the scaling between the number of heads and the size of each head in the current architecture gives rise to a low-rank bottleneck in attention heads. We propose to set the head size of an attention unit to input sequence length, and independent of the number of heads, resulting in multi-head attention layers with provably more expressive power.
  arXiv  Detail & Related papers  (2020-02-17T16:16:40Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.