In-Context Learning Creates Task Vectors
Paper • 2310.15916 • Published • 44
Context compression
-
-
When can transformers reason with abstract symbols?
Paper • 2310.09753 • Published • 3 -
Improving Length-Generalization in Transformers via Task Hinting
Paper • 2310.00726 • Published • 1 -
In-context Autoencoder for Context Compression in a Large Language Model
Paper • 2307.06945 • Published • 29 -
Adapting Language Models to Compress Contexts
Paper • 2305.14788 • Published • 1 -
Context Compression for Auto-regressive Transformers with Sentinel Tokens
Paper • 2310.08152 • Published • 1 -
Learning to Compress Prompts with Gist Tokens
Paper • 2304.08467 • Published • 3 -
Dynamic Context Pruning for Efficient and Interpretable Autoregressive Transformers
Paper • 2305.15805 • Published • 1 -
Compress, Then Prompt: Improving Accuracy-Efficiency Trade-off of LLM Inference with Transferable Prompt
Paper • 2305.11186 • Published • 1 -
Self-slimmed Vision Transformer
Paper • 2111.12624 • Published • 1 -
Deja Vu: Contextual Sparsity for Efficient LLMs at Inference Time
Paper • 2310.17157 • Published • 14 -
RECOMP: Improving Retrieval-Augmented LMs with Compression and Selective Augmentation
Paper • 2310.04408 • Published • 1 -
Dynamic Token Pruning in Plain Vision Transformers for Semantic Segmentation
Paper • 2308.01045 • Published • 1 -
Adaptive Token Sampling For Efficient Vision Transformers
Paper • 2111.15667 • Published • 1 -
Dynamic Token-Pass Transformers for Semantic Segmentation
Paper • 2308.01944 • Published • 1 -
Multi-Scale And Token Mergence: Make Your ViT More Efficient
Paper • 2306.04897 • Published • 1 -
Sparsifiner: Learning Sparse Instance-Dependent Attention for Efficient Vision Transformers
Paper • 2303.13755 • Published • 1 -
Nugget 2D: Dynamic Contextual Compression for Scaling Decoder-only Language Models
Paper • 2310.02409 • Published • 1 -
LM-CPPF: Paraphrasing-Guided Data Augmentation for Contrastive Prompt-Based Few-Shot Fine-Tuning
Paper • 2305.18169 • Published • 1 -
ComputeGPT: A computational chat model for numerical problems
Paper • 2305.06223 • Published • 1 -
XPrompt: Exploring the Extreme of Prompt Tuning
Paper • 2210.04457 • Published • 1 -
Reducing Sequence Length by Predicting Edit Operations with Large Language Models
Paper • 2305.11862 • Published • 2 -
Diet Code Is Healthy: Simplifying Programs for Pre-trained Models of Code
Paper • 2206.14390 • Published • 1 -
Split, Encode and Aggregate for Long Code Search
Paper • 2208.11271 • Published • 1 -
Recursively Summarizing Enables Long-Term Dialogue Memory in Large Language Models
Paper • 2308.15022 • Published • 4 -
Vcc: Scaling Transformers to 128K Tokens or More by Prioritizing Important Tokens
Paper • 2305.04241 • Published • 1 -
Latency Adjustable Transformer Encoder for Language Understanding
Paper • 2201.03327 • Published • 1 -
Block-Skim: Efficient Question Answering for Transformer
Paper • 2112.08560 • Published • 1 -
Learned Token Pruning for Transformers
Paper • 2107.00910 • Published • 1 -
Zero-TPrune: Zero-Shot Token Pruning through Leveraging of the Attention Graph in Pre-Trained Transformers
Paper • 2305.17328 • Published • 2 -
Learned Thresholds Token Merging and Pruning for Vision Transformers
Paper • 2307.10780 • Published • 1 -
Can the Inference Logic of Large Language Models be Disentangled into Symbolic Concepts?
Paper • 2304.01083 • Published • 1 -
System 2 Attention (is something you might need too)
Paper • 2311.11829 • Published • 43 -
CoLT5: Faster Long-Range Transformers with Conditional Computation
Paper • 2303.09752 • Published • 2 -
Random-LTD: Random and Layerwise Token Dropping Brings Efficient Training for Large-scale Transformers
Paper • 2211.11586 • Published • 1 -
TCRA-LLM: Token Compression Retrieval Augmented Large Language Model for Inference Cost Reduction
Paper • 2310.15556 • Published • 1 -
Extending Context Window of Large Language Models via Semantic Compression
Paper • 2312.09571 • Published • 16 -
LLoCO: Learning Long Contexts Offline
Paper • 2404.07979 • Published • 22 -
SelfCP: Compressing Long Prompt to 1/12 Using the Frozen Large Language Model Itself
Paper • 2405.17052 • Published • 2 -
Equipping Transformer with Random-Access Reading for Long-Context Understanding
Paper • 2405.13216 • Published • 1