LIMO: Less Is More for Reasoning (arxiv.org) | HN Screenshots

YHacker Newsscreenshots

Screenshot of arxiv.org: LIMO: Less Is More for Reasoning

First screen, captured 1 year ago.

Looks similar

Screenshot of arxiv.org: Reasoning LLMs are wandering solution explorers

Reasoning LLMs are wandering solution explorers (arxiv.org)

83 points by Surreal4434 1 year ago | 78 comments

Screenshot of arxiv.org: Language models as compilers: Simulating pseudocode execution

Language models as compilers: Simulating pseudocode execution (arxiv.org)

159 points by milliondreams 2 years ago | 49 comments

Screenshot of arxiv.org: Search-R1: Training LLMs to Reason and Leverage Search Engines with RL

Search-R1: Training LLMs to Reason and Leverage Search Engines with RL (arxiv.org)

87 points by jonbaer 1 year ago | 9 comments

Screenshot of arxiv.org: Gemini in Reasoning: Unveiling Commonsense in Multimodal Large Language Models

Gemini in Reasoning: Unveiling Commonsense in Multimodal Large Language Models (arxiv.org)

49 points by ukuina 2 years ago | 25 comments

Screenshot of arxiv.org: Training LLMs to Reason in a Continuous Latent Space

Training LLMs to Reason in a Continuous Latent Space (arxiv.org)

265 points by omarsar 1 year ago | 93 comments

Screenshot of arxiv.org: Large Language Model Reasoning Failures

Large Language Model Reasoning Failures (arxiv.org)

25 points by T-A 8 months ago | 37 comments

Related stories

Screenshot of arxiv.org: DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL

DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via RL (arxiv.org)

1116 points by gradus_ad 1 year ago | 932 comments

arXiv abstract page for the DeepSeek-AI paper introducing DeepSeek-R1, which uses reinforcement learning to improve reasoning in large language models.

deepseek-r1 deepseek reinforcement-learning llm-reasoning

Screenshot of arxiv.org: Ladder: Self-improving LLMs through recursive problem decomposition

Ladder: Self-improving LLMs through recursive problem decomposition (arxiv.org)

339 points by fofoz 1 year ago | 103 comments

arXiv paper introducing LADDER, a framework where LLMs self-improve by recursively generating simpler problem variants, plus TTRL test-time reinforcement learning.

llm reinforcement-learning self-improvement test-time-compute

Screenshot of simonwillison.net: 2025: The Year in LLMs

2025: The Year in LLMs (simonwillison.net)

823 points by simonw 9 months ago | 451 comments

Simon Willison's annual blog post reviewing major LLM trends of 2025, including reasoning models, coding agents, Chinese open weight models, and Gemini.

llms ai-agents claude-code coding-agents

Screenshot of anthropic.com: Reasoning models don't always say what they think

Reasoning models don't always say what they think (anthropic.com)

369 points by meetpateltech 1 year ago | 246 comments

Anthropic research post on whether reasoning models' Chain-of-Thought faithfully reflects their actual reasoning, linking to the accompanying paper.

chain-of-thought reasoning-models ai-alignment anthropic

Screenshot of shchegrikovich.substack.com: Use Prolog to improve LLM's reasoning

Use Prolog to improve LLM's reasoning (shchegrikovich.substack.com)

343 points by shchegrikovich 2 years ago | 122 comments

Substack post arguing that having LLMs generate Prolog code can improve their reasoning, given the limits of autoregressive transformer generation.

prolog llm reasoning logic-programming

Screenshot of magazine.sebastianraschka.com: Understanding Reasoning LLMs

Understanding Reasoning LLMs (magazine.sebastianraschka.com)

442 points by sebg 1 year ago | 163 comments

Sebastian Raschka's Ahead of AI newsletter article explaining the four main approaches to building and refining reasoning LLMs.

reasoning-models llm inference-time-scaling reinforcement-learning