MassAlloc Attention: Let Attention Allocate Its Own Compute
cs.AI, cs.CL
Submitted: 2026-09-26
Updated: 2026-09-26
Code: https://github.com/huggingface/open-r1
Terminology
Sources
- Why do LLMs attend to the first token?
- Longformer: The Long-Document Transformer
- Generating Long Sequences with Sparse Transformers
- Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge
- Training Verifiers to Solve Math Word Problems
- Attention Sinks and Compression Valleys in LLMs are Two Sides of the Same Coin
- DeepSeek-V3.2: Pushing the Frontier of Open Large Language Models
- Flex Attention: A Programming Model for Generating Optimized Attention Kernels
- Sliding Window Attention Training for Efficient Large Language Models
- SeerAttention: Learning Intrinsic Sparse Attention in Your LLMs
- When Attention Sink Emerges in Language Models: An Empirical View
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Measuring Mathematical Problem Solving With the MATH Dataset
- RULER: What's the Real Context Size of Your Long-Context Language Models?
- NOSA: Native and Offloadable Sparse Attention
- Scaling Laws for Neural Language Models
- FlexPrefill: A Context-Aware Sparse Attention Mechanism for Efficient Long-Sequence Inference
- Predictable Scale: Part I, Step Law -- Optimal Hyperparameter Scaling Law in Large Language Model Pretraining
- Twilight: Adaptive Attention Sparsity with Hierarchical Top-$p$ Pruning
- ClusterKV: Manipulating LLM KV Cache in Semantic Space for Recallable Compression
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection