Papers with attention computation
RoBSA: RoPE-based Blockwise Sparse Multi-head Latent Attention (2026.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have advanced in recent years, scaling up in both parameter count and context length. |
| Approach: | They propose a method to compute attention over a subset of context tokens and to implement token selection in a blockwise manner. |
| Outcome: | The proposed method reduces end-to-end inference latency by up to 2.55x with minimal accuracy loss compared to full attention in long-context scenarios for very large models. |
On-the-Fly Attention Modulation for Neural Generation (2021.findings-acl)
Copied to clipboard
Yue Dong, Chandra Bhagavatula, Ximing Lu, Jena D. Hwang, Antoine Bosselut, Jackie Chi Kit Cheung, Yejin Choi
| Challenge: | Degeneration of neural text is associated with insufficient learning of task-specific characteristics by the attention mechanism. |
| Approach: | They propose to use attention modulation to inject priors into inference to improve fluency, creativity, and commonsense reasoning in neural text generation models. |
| Outcome: | The proposed method improves fluency, creativity, and commonsense reasoning, and significantly reduces sentence-level repetition. |
Conv-Basis: A New Paradigm for Efficient Attention Inference and Gradient Computation in Transformers (2025.findings-emnlp)
Copied to clipboard
| Challenge: | a large computational cost for attention computation in large language models is a major obstacle . |
| Approach: | They propose a convolution-like structure for attention computation using convolution matrices . they then propose an efficient approximation method to approximate the attention matrix . |
| Outcome: | The proposed method achieves nearly linear time complexity in n1+o(1) time. |
Data-Informed Global Sparseness in Attention Mechanisms for Deep Neural Networks (2024.lrec-main)
Copied to clipboard
| Challenge: | Attention pruning techniques have been developed to identify and exploit sparseness . previous work has taken pioneering steps to discover and explain the sparsity in attention patterns . |
| Approach: | They propose a framework that observes attention patterns in a fixed dataset and generates a global sparseness mask. |
| Outcome: | The proposed approach saves 90% of computations and maintains quality of results. |
The Devil in Linear Transformer (2022.emnlp-main)
Copied to clipboard
| Challenge: | Existing linear transformers suffer from performance degradations on various tasks and corpus. |
| Approach: | They propose a new linear attention that replaces scaling with a normalization to stabilize gradients and confine attention to neighbouring tokens in early layers. |
| Outcome: | The proposed model outperforms vanilla transformers on the long-range arena benchmark while being significantly more space-time efficient. |
Evolving Sparsity: Leveraging Token Importance Dynamics for Efficient LLM Decoding with Sparse Attention (2026.acl-long)
Copied to clipboard
| Challenge: | Efficient long-context inference remains a major challenge for large language models (LLMs), as the cost of attention computation during auto-regressive decoding grows linearly with the context length. |
| Approach: | They propose to model token importance as a dynamic process that evolves over decoding steps and propagates through model layers. |
| Outcome: | The proposed method outperforms baseline sparse attention methods and achieves speedups of up to 5.36 for attention latency and 2.33 for end-to-end decoding. |
Focus-dLLM: Accelerating Long-Context Diffusion LLM Inference via Confidence-Guided Context Focusing (2026.acl-long)
Copied to clipboard
| Challenge: | Existing methods for estimating attention importance for tokens are ineffective . dLLMs require bidirectional attention, which limits inference efficiency . |
| Approach: | They propose a training-free attention sparsification framework for efficient long-context inference . they propose 'sink-aware pruning strategy' to accurately estimate and remove redundant computation . |
| Outcome: | The proposed approach offers 29 lossless speedup under 32K context length. |
DELTA: Dynamic Layer-Aware Token Attention for Efficient Long-Context Reasoning (2026.findings-acl)
Copied to clipboard
| Challenge: | Large reasoning models generate long chains of intermediate steps, but their inference cost is dominated by decoding, where each new token must attend to the entire growing sequence. |
| Approach: | They propose a training-free sparse attention mechanism that reduces inference cost by evicting entries from the key-value cache. |
| Outcome: | The proposed model matches or surpasses full attention on reasoning benchmarks . it reduces the number of attended tokens by up to 4.25 and delivers 1.54 speedup . |
Back Attention: Understanding and Enhancing Multi-Hop Reasoning in Large Language Models (2025.emnlp-main)
Copied to clipboard
| Challenge: | Increasing the ability of large language models to perform latent multihop reasoning is crucial for reducing the cost and deployment challenges. |
| Approach: | They propose an interpretability method that traces how logits propagate across layers and positions toward the final prediction. |
| Outcome: | The proposed method improves accuracy on five reasoning datasets. |
APrompt: Attention Prompt Tuning for Efficient Adaptation of Pre-trained Language Models (2023.emnlp-main)
Copied to clipboard
Qifan Wang, Yuning Mao, Jingang Wang, Hanchao Yu, Shaoliang Nie, Sinong Wang, Fuli Feng, Lifu Huang, Xiaojun Quan, Zenglin Xu, Dongfang Liu
| Challenge: | Existing prompt tuning methods only introduce prompts at the input layer, limiting performance and leaving large room for improvement. |
| Approach: | They propose a method that involves tuning a small set of soft prompts for pre-trained language models. |
| Outcome: | The proposed method outperforms state-of-the-art methods with pre-trained models on the SuperGLUE benchmark. |
Pruning Large Language Models to Intra-module Low-rank Architecture with Transitional Activations (2024.findings-acl)
Copied to clipboard
| Challenge: | Structured pruning is a feasible solution for end-side LLM deployment . however, achieving a high compression ratio for scaled-up LLMs remains a challenge . |
| Approach: | They propose a task-agnostic structured pruning approach coupled with a compact Transformer architecture to prune LLMs into an intra-module low-rank architecture. |
| Outcome: | The proposed approach reduces transitional activations inside multi-head attention (MHA) and multi-layer perceptron (MLP) modules while preserving inter-module activations sensitive to perturbations. |
TokenSelect: Efficient Long-Context Inference and Length Extrapolation for LLMs via Dynamic Token-Level KV Cache Selection (2025.emnlp-main)
Copied to clipboard
| Challenge: | Rapid advances in Large Language Models have spurred demand for processing extended context sequences . however, performance degradation due to sequence lengths out-of-distribution and excessively long inference times are limiting LLMs in long-context scenarios. |
| Approach: | They propose a training-free method for efficient and accurate long-context inference . they selectively involves a few critical KV cache tokens in attention calculation . |
| Outcome: | The proposed method speeds up attention computation and accelerates inference time while reducing selection overhead. |
EQUIP: EQUivariant preserving In-Place updates for Efficient Token Pruning (2026.acl-long)
Copied to clipboard
| Challenge: | Token-pruning methods cause "holes" in KV tensors, posing major challenges . equip reduces recomputation of rotation operations through in-place update, caching and re-indexing . |
| Approach: | They propose an EQUIP-based in-place token update mechanism that preserves the equivariance property of the operations performed in the attention computation. |
| Outcome: | EQUIP reduces recomputation of rotation operations and reduces eviction overheads . it achieves geomean speedups of 1.62 (or 1.47) over StreamingLLM and 3.45 ( or 1.86) |