Writing in the Margins: Better Inference Pattern for Long Context Retrieval
Paper • 2408.14906 • Published • 144
Long context
-
-
DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads
Paper • 2410.10819 • Published • 7 -
LLMtimesMapReduce: Simplified Long-Sequence Processing using Large Language Models
Paper • 2410.09342 • Published • 39 -
PDFTriage: Question Answering over Long, Structured Documents
Paper • 2309.08872 • Published • 55 -
ComplexFuncBench: Exploring Multi-Step and Constrained Function Calling under Long-Context Scenario
Paper • 2501.10132 • Published • 22 -
Can Large Language Models Detect Errors in Long Chain-of-Thought Reasoning?
Paper • 2502.19361 • Published • 28 -
Interleaved Reasoning for Large Language Models via Reinforcement Learning
Paper • 2505.19640 • Published • 15 -
SuperWriter: Reflection-Driven Long-Form Generation with Large Language Models
Paper • 2506.04180 • Published • 34 -
Beyond Homogeneous Attention: Memory-Efficient LLMs via Fourier-Approximated KV Cache
Paper • 2506.11886 • Published • 20 -
Optimizing Length Compression in Large Reasoning Models
Paper • 2506.14755 • Published • 10 -
LongWriter-Zero: Mastering Ultra-Long Text Generation via Reinforcement Learning
Paper • 2506.18841 • Published • 56 -
SpeContext: Enabling Efficient Long-context Reasoning with Speculative Context Sparsity in LLMs
Paper • 2512.00722 • Published • 16 -
Beyond Real: Imaginary Extension of Rotary Position Embeddings for Long-Context LLMs
Paper • 2512.07525 • Published • 60 -
LongCat-Flash-Thinking-2601 Technical Report
Paper • 2601.16725 • Published • 180 -
Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces
Paper • 2601.11868 • Published • 35 -
daVinci-Agency: Unlocking Long-Horizon Agency Data-Efficiently
Paper • 2602.02619 • Published • 53 -
WorldCompass: Reinforcement Learning for Long-Horizon World Models
Paper • 2602.09022 • Published • 21 -
REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agents
Paper • 2602.14234 • Published • 27 -
LMEB: Long-horizon Memory Embedding Benchmark
Paper • 2603.12572 • Published • 73 -
Recursive Language Models Meet Uncertainty: The Surprising Effectiveness of Self-Reflective Program Search for Long Context
Paper • 2603.15653 • Published • 12 -
LongCat-Flash-Prover: Advancing Native Formal Reasoning via Agentic Tool-Integrated Reinforcement Learning
Paper • 2603.21065 • Published • 77 -
ThinkJEPA: Empowering Latent World Models with Large Vision-Language Reasoning Model
Paper • 2603.22281 • Published • 18 -
SpecEyes: Accelerating Agentic Multimodal LLMs via Speculative Perception and Planning
Paper • 2603.23483 • Published • 62 -
Matrix-Game 3.0: Real-Time and Streaming Interactive World Model with Long-Horizon Memory
Paper • 2604.08995 • Published • 47 -
AgentSwing: Adaptive Parallel Context Management Routing for Long-Horizon Web Agents
Paper • 2603.27490 • Published • 17 -
Toward Autonomous Long-Horizon Engineering for ML Research
Paper • 2604.13018 • Published • 34 -
KV Packet: Recomputation-Free Context-Independent KV Caching for LLMs
Paper • 2604.13226 • Published • 9 -
LongAct: Harnessing Intrinsic Activation Patterns for Long-Context Reinforcement Learning
Paper • 2604.14922 • Published • 7