ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning
Understanding and reasoning over long contexts has become a key requirement for deploying large language models (LLMs) in realistic applications. Although recent LLMs support increasingly long context windows, they often fail to use relevant evidence that is already present in the input, revealing a gap between context access and effective context utilization. In this work, we propose Recursive Evidence Replay as LLM Harness for Long-Context Reasoning (RECONTEXT), a training-free inference method for improving long-context reasoning. RECONTEXT uses model-internal relevance signals to construct
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
- Linked via arxiv authorYanjun Zhao →
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning
- Linked via arxiv authorRuizhong Qiu →
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning
- Linked via arxiv authorTianxin Wei →
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning
- Linked via arxiv authorYuanchen Bei →
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning
- Linked via arxiv authorZhining Liu →
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning
- Linked via arxiv authorLingjie Chen →
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning
- Linked via arxiv authorIsmini Lourentzou →
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning
- Linked via arxiv authorHanghang Tong →
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning
- Linked via arxiv authorJingrui He →
ReContext: Recursive Evidence Replay as LLM Harness for Long-Context Reasoning