TRAMS: Training-free Memory Selection for Long-range Language Modeling
Paper
•
2310.15494
•
Published
•
1
Long context
-
-
A Long Way to Go: Investigating Length Correlations in RLHF
Paper • 2310.03716 • Published • 9 -
YaRN: Efficient Context Window Extension of Large Language Models
Paper • 2309.00071 • Published • 65 -
Giraffe: Adventures in Expanding Context Lengths in LLMs
Paper • 2308.10882 • Published • 1 -
LM-Infinite: Simple On-the-Fly Length Generalization for Large Language Models
Paper • 2308.16137 • Published • 39 -
Scaling Transformer to 1M tokens and beyond with RMT
Paper • 2304.11062 • Published • 2 -
Investigating Answerability of LLMs for Long-Form Question Answering
Paper • 2309.08210 • Published • 12 -
DeepSpeed Ulysses: System Optimizations for Enabling Training of Extreme Long Sequence Transformer Models
Paper • 2309.14509 • Published • 17 -
LongLoRA: Efficient Fine-tuning of Long-Context Large Language Models
Paper • 2309.12307 • Published • 87 -
PoSE: Efficient Context Window Extension of LLMs via Positional Skip-wise Training
Paper • 2309.10400 • Published • 26 -
CLEX: Continuous Length Extrapolation for Large Language Models
Paper • 2310.16450 • Published • 9 -
Code Llama: Open Foundation Models for Code
Paper • 2308.12950 • Published • 22 -
CAT-LM: Training Language Models on Aligned Code And Tests
Paper • 2310.01602 • Published • 1 -
LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding
Paper • 2308.14508 • Published • 2 -
BTLM-3B-8K: 7B Parameter Performance in a 3B Parameter Model
Paper • 2309.11568 • Published • 10 -
XGen-7B Technical Report
Paper • 2309.03450 • Published • 8 -
Effective Long-Context Scaling of Foundation Models
Paper • 2309.16039 • Published • 30 -
LongLLMLingua: Accelerating and Enhancing LLMs in Long Context Scenarios via Prompt Compression
Paper • 2310.06839 • Published • 3 -
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 -
Long-range Language Modeling with Self-retrieval
Paper • 2306.13421 • Published • 16 -
Can Retriever-Augmented Language Models Reason? The Blame Game Between the Retriever and the Language Model
Paper • 2212.09146 • Published • 3 -
Knowledge-Augmented Reasoning Distillation for Small Language Models in Knowledge-Intensive Tasks
Paper • 2305.18395 • Published • 1 -
LLM+P: Empowering Large Language Models with Optimal Planning Proficiency
Paper • 2304.11477 • Published • 3 -
SayCanPay: Heuristic Planning with Large Language Models using Learnable Domain Knowledge
Paper • 2308.12682 • Published • 2 -
Combiner: Full Attention Transformer with Sparse Computation Cost
Paper • 2107.05768 • Published • 1 -
Memorizing Transformers
Paper • 2203.08913 • Published • 2 -
Adapting Language Models to Compress Contexts
Paper • 2305.14788 • Published • 1 -
Lost in the Middle: How Language Models Use Long Contexts
Paper • 2307.03172 • Published • 36 -
L-Eval: Instituting Standardized Evaluation for Long Context Language Models
Paper • 2307.11088 • Published • 4 -
A Unified View of Long-Sequence Models towards Modeling Million-Scale Dependencies
Paper • 2302.06218 • Published • 1 -
Blockwise Parallel Transformer for Long Context Large Models
Paper • 2305.19370 • Published • 3 -
Blockwise Self-Attention for Long Document Understanding
Paper • 1911.02972 • Published • 1 -
LSG Attention: Extrapolation of pretrained Transformers to long sequences
Paper • 2210.15497 • Published • 1 -
Efficient Long-Text Understanding with Short-Text Models
Paper • 2208.00748 • Published • 1 -
Cure the headache of Transformers via Collinear Constrained Attention
Paper • 2309.08646 • Published • 12 -
Bird-Eye Transformers for Text Generation Models
Paper • 2210.03985 • Published • 1 -
Memoria: Hebbian Memory Architecture for Human-Like Sequential Processing
Paper • 2310.03052 • Published • 1 -
Efficient Streaming Language Models with Attention Sinks
Paper • 2309.17453 • Published • 13 -
LightSeq: Sequence Level Parallelism for Distributed Training of Long Context Transformers
Paper • 2310.03294 • Published • 2 -
Ultra-Long Sequence Distributed Transformer
Paper • 2311.02382 • Published • 2 -
In-Context Pretraining: Language Modeling Beyond Document Boundaries
Paper • 2310.10638 • Published • 28 -
Retrieval meets Long Context Large Language Models
Paper • 2310.03025 • Published • 4 -
AWESOME: GPU Memory-constrained Long Document Summarization using Memory Mechanism and Global Salient Content
Paper • 2305.14806 • Published • 1 -
mLongT5: A Multilingual and Efficient Text-To-Text Transformer for Longer Sequences
Paper • 2305.11129 • Published • 2 -
LongT5: Efficient Text-To-Text Transformer for Long Sequences
Paper • 2112.07916 • Published • 2 -
Unleashing Infinite-Length Input Capacity for Large-scale Language Models with Self-Controlled Memory System
Paper • 2304.13343 • Published • 1 -
Vcc: Scaling Transformers to 128K Tokens or More by Prioritizing Important Tokens
Paper • 2305.04241 • Published • 1 -
Are We Falling in a Middle-Intelligence Trap? An Analysis and Mitigation of the Reversal Curse
Paper • 2311.07468 • Published • 1 -
Never Lost in the Middle: Improving Large Language Models via Attention Strengthening Question Answering
Paper • 2311.09198 • Published • 3 -
SpanDrop: Simple and Effective Counterfactual Learning for Long Sequences
Paper • 2208.02169 • Published • 1 -
System 2 Attention (is something you might need too)
Paper • 2311.11829 • Published • 39 -
Attention Sorting Combats Recency Bias In Long Context Language Models
Paper • 2310.01427 • Published • 1 -
CoLT5: Faster Long-Range Transformers with Conditional Computation
Paper • 2303.09752 • Published • 2 -
Cached Transformers: Improving Transformers with Differentiable Memory Cache
Paper • 2312.12742 • Published • 12 -
Axiomatic Preference Modeling for Longform Question Answering
Paper • 2312.02206 • Published • 7 -
TextGenSHAP: Scalable Post-hoc Explanations in Text Generation with Long Documents
Paper • 2312.01279 • Published • 3 -
Extending Context Window of Large Language Models via Semantic Compression
Paper • 2312.09571 • Published • 12 -
Zebra: Extending Context Window with Layerwise Grouped Local-Global Attention
Paper • 2312.08618 • Published • 11 -
LongAlign: A Recipe for Long Context Alignment of Large Language Models
Paper • 2401.18058 • Published • 21 -
Extending LLMs' Context Window with 100 Samples
Paper • 2401.07004 • Published • 15 -
The What, Why, and How of Context Length Extension Techniques in Large Language Models -- A Detailed Survey
Paper • 2401.07872 • Published • 2 -
E^2-LLM: Efficient and Extreme Length Extension of Large Language Models
Paper • 2401.06951 • Published • 25 -
Exploring Transformer Extrapolation
Paper • 2307.10156 • Published • 1 -
Gated Linear Attention Transformers with Hardware-Efficient Training
Paper • 2312.06635 • Published • 6 -
LLM Maybe LongLM: Self-Extend LLM Context Window Without Tuning
Paper • 2401.01325 • Published • 26 -
LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens
Paper • 2402.13753 • Published • 112 -
Training-Free Long-Context Scaling of Large Language Models
Paper • 2402.17463 • Published • 19 -
LOCOST: State-Space Models for Long Document Abstractive Summarization
Paper • 2401.17919 • Published -
Megalodon: Efficient LLM Pretraining and Inference with Unlimited Context Length
Paper • 2404.08801 • Published • 63 -
Orchid: Flexible and Data-Dependent Convolution for Sequence Modeling
Paper • 2402.18508 • Published -
HMT: Hierarchical Memory Transformer for Long Context Language Processing
Paper • 2405.06067 • Published • 1 -
LLoCO: Learning Long Contexts Offline
Paper • 2404.07979 • Published • 20 -
LongHeads: Multi-Head Attention is Secretly a Long Context Processor
Paper • 2402.10685 • Published • 1 -
XL3M: A Training-free Framework for LLM Length Extension Based on Segment-wise Inference
Paper • 2405.17755 • Published -
Base of RoPE Bounds Context Length
Paper • 2405.14591 • Published -
SinkLoRA: Enhanced Efficiency and Chat Capabilities for Long-Context Large Language Models
Paper • 2406.05678 • Published -
LongSkywork: A Training Recipe for Efficiently Extending Context Length in Large Language Models
Paper • 2406.00605 • Published • 2 -
Equipping Transformer with Random-Access Reading for Long-Context Understanding
Paper • 2405.13216 • Published -
THEANINE: Revisiting Memory Management in Long-term Conversations with Timeline-augmented Response Generation
Paper • 2406.10996 • Published • 32 -
InfLLM: Unveiling the Intrinsic Capacity of LLMs for Understanding Extremely Long Sequences with Training-Free Memory
Paper • 2402.04617 • Published • 4 -
Farewell to Length Extrapolation, a Training-Free Infinite Context with Finite Attention Scope
Paper • 2407.15176 • Published -
RazorAttention: Efficient KV Cache Compression Through Retrieval Heads
Paper • 2407.15891 • Published -
Writing in the Margins: Better Inference Pattern for Long Context Retrieval
Paper • 2408.14906 • Published • 138 -
Retrieval Augmented Generation or Long-Context LLMs? A Comprehensive Study and Hybrid Approach
Paper • 2407.16833 • Published -
ReMamba: Equip Mamba with Effective Long-Sequence Modeling
Paper • 2408.15496 • Published • 10 -
General-purpose, long-context autoregressive modeling with Perceiver AR
Paper • 2202.07765 • Published -
MINI-SEQUENCE TRANSFORMER: Optimizing Intermediate Memory for Long Sequences Training
Paper • 2407.15892 • Published -
ContextCite: Attributing Model Generation to Context
Paper • 2409.00729 • Published • 13 -
UIO-LLMs: Unbiased Incremental Optimization for Long-Context LLMs
Paper • 2406.18173 • Published -
MemLong: Memory-Augmented Retrieval for Long Text Modeling
Paper • 2408.16967 • Published • 2 -
Efficient LLM Training and Serving with Heterogeneous Context Sharding among Attention Heads
Paper • 2407.17678 • Published -
E2LLM: Encoder Elongated Large Language Models for Long-Context Understanding and Reasoning
Paper • 2409.06679 • Published -
InfiniGen: Efficient Generative Inference of Large Language Models with Dynamic KV Cache Management
Paper • 2406.19707 • Published