arXiv AI
To Retrieve or To Think? Cross-Boundary Context Evolution for Multi-hop Complex Reasoning
Read the original on arXiv AI →The Flow has not summarised this story yet — read it at arXiv AI.
Related stories
arXiv AI
Aug 7
Contextual Information Policy Optimization for Search Agents
arXiv:2608. 06128v1 Announce Type: new Abstract: Search agents extend large language models beyond static parametric memory by enabling them to acquire and use ex ternal evidence during multi-step reasoning.
arXiv AI
Jul 2
DiscoLoop: Looping Discrete Embeddings and Continuous Hidden States for Multi-hop Reasoning
arXiv:2607. 00341v1 Announce Type: cross Abstract: Large language models achieve strong performance on many reasoning tasks when allowed to externalize intermediate steps as Chain-of-Thought (CoT).
arXiv AI
Aug 6
Chained Recursive Language Models for Multi-Iteration Reasoning
arXiv:2608. 05124v1 Announce Type: cross Abstract: Long context reasoning in large language models (LLMs) is usually constrained by the fact that a single inference trajectory has to simultaneously explore the context, store intermediate state, verify evidence, and produce the final answer.
arXiv AI
Sep 7
A Tree-based RAG Framework for Evidence-Intensive QA via Adaptive Planning and Topology-Aware Evidence Gathering
A new framework called APT‑RAG is introduced for evidence‑intensive question answering. It uses adaptive planning to expand a tree‑based reasoning structure according to question dependencies and evidence needs, and employs topology‑aware evidence gathering that reuses sibling evidence, retrieves directly, and aggregates evidence from child nodes. The approach also includes evidence‑guided batched answer generation to reduce generation overhead, and experimental results show it outperforms existing structured RAG methods on evidence‑intensive QA benchmarks.
arXiv AI
Sep 1
AgenticRag-R1: Agentic Reinforcement Learning with Stack Memory for Multi-Step Reasoning, Retrieval and Memorizing
AgenticRag‑R1 is a reinforcement‑learning framework that integrates reasoning, retrieval, and memory through a stack and fine‑grained action space. It uses hierarchical action‑aware rewards and an information‑aware trajectory rejection strategy to support long‑horizon learning. Experiments on multi‑hop, open‑domain, and agentic reasoning benchmarks show that AgenticRag‑R1 outperforms strong baselines and produces robust, interpretable, memory‑aware reasoning behaviors.
arXiv AI
Aug 11
SAKE: Structured Agentic Knowledge Extrapolation for Complex LLM Reasoning via Reinforcement Learning
arXiv:2505. 15062v5 Announce Type: replace-cross Abstract: Knowledge extrapolation is the process of inferring novel information by combining and extending existing knowledge that is explicitly available.