LLM Digest

Story

arxiv_cs_ai · Apr 29, 2026 · paper

Source brief

When to Retrieve During Reasoning: Adaptive Retrieval for Large Reasoning Models

arxiv.orgApr 29, 2026
original source linked

In brief

Large reasoning models such as DeepSeek-R1 and OpenAI o1 generate extended chains of thought spanning thousands of tokens, yet their integration with retrieval-augmented generation (RAG) remains fundamentally misalign...

Read the original at arxiv.org →Open in live feed

When to Retrieve During Reasoning: Adaptive Retrieval for Large Reasoning Models

Earlier in this thread 4 items

Ask HN: What are you doing during inference?

Adaptive Ultrasound Imaging with Physics-Informed NV-Raw2Insights-US AI

SafetyALFRED: Evaluating Safety-Conscious Planning of Multimodal Large Language Models

Evaluating the Progression of Large Language Model Capabilities for Small-Molecule Drug Design