ThinkRetrieve: Retrieval-Augmented Reasoning Traces for Test-Time Scaling
Paper • 2608.10928 • Published
Retrieval-Augmented Test-Time Reasoning
Papers and artifacts on retrieving evidence, worked examples, and reasoning memory during inference.
Note ThinkRetrieve: worked-solution retrieval from an interim reasoning state.
ThinkRetrieve Research Explorer
🔎Explore and compare recorded reasoning retrieval results
Note Recorded ThinkRetrieve examples and failure cases.
Procedural Knowledge at Scale Improves Reasoning
Paper • 2604.01348 • Published • 1Note Retrieval and reuse of procedural knowledge at scale.
ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory
Paper • 2509.25140 • Published • 15Note Experience-derived reasoning memory and memory-aware scaling.
Search-o1: Agentic Search-Enhanced Large Reasoning Models
Paper • 2501.05366 • Published • 102Note Reasoning interleaved with external knowledge retrieval.
Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning
Paper • 2503.09516 • Published • 42Note Training reasoning models to issue searches at inference.
Enhancing Test-Time Scaling of Large Language Models with Hierarchical Retrieval-Augmented MCTS
Paper • 2507.05557 • PublishedNote Hierarchical retrieval-augmented MCTS for test-time scaling.