Hacker News
- Study: Large language models still lack general reasoning skills https://santafe.edu/news-center/news/study-large-language-models-still-lack-general-reasoning-skills 4 comments
- Can large language models reason? https://www.arnaldur.be/writing/about/large-language-model-reasoning 60 comments
- PhD Knowledge Not Required: A Reasoning Challenge for Large Language Models https://arxiv.org/abs/2502.01584 80 comments
- Does Reasoning Emerge? Probabilities of Causation in Large Language Models https://arxiv.org/abs/2408.08210 192 comments
- Reasoning in Large Language Models: A Geometric Perspective https://arxiv.org/abs/2407.02678 170 comments
- Large Language Models Are Neurosymbolic Reasoners https://arxiv.org/abs/2401.09334 164 comments
- Gemini in Reasoning: Unveiling Commonsense in Multimodal Large Language Models https://arxiv.org/abs/2312.17661 38 comments
- Procedural knowledge in pretraining drives reasoning in large language models https://arxiv.org/abs/2411.12580 101 comments
- Reasoning skills of large language models are often overestimated https://news.mit.edu/2024/reasoning-skills-large-language-models-often-overestimated-0711 36 comments
- Can large language models reason about medical questions? https://arxiv.org/abs/2207.08143 110 comments
- Understanding the Limitations of Mathematical Reasoning in Large Language Models https://machinelearning.apple.com/research/gsm-symbolic 3 comments
Lobsters
- Alice in Wonderland: Simple Tasks Showing Complete Reasoning Breakdown in State-Of-the-Art Large Language Models https://arxiv.org/abs/2406.02061 14 comments ai
- [R] Alice in Wonderland: Simple Tasks Showing Complete Reasoning Breakdown in State-Of-the-Art Large Language Model https://arxiv.org/pdf/2406.02061 69 comments machinelearning
- [R] Large Language Models as Analogical Reasoners https://arxiv.org/abs/2310.01714 2 comments machinelearning
- Large language models have a reasoning problem https://bdtechtalks.com/2022/06/27/large-language-models-logical-reasoning/ 6 comments artificial
- [2403.04642] Teaching Large Language Models to Reason with Reinforcement Learning https://arxiv.org/abs/2403.04642 2 comments reinforcementlearning
- Reasoning skills of large language models are often overestimated https://news.mit.edu/2024/reasoning-skills-large-language-models-often-overestimated-0711 8 comments technology
- [R] NExT: Teaching Large Language Models to Reason about Code Execution https://arxiv.org/abs/2404.14662 9 comments machinelearning
- [D] Transforming Large Language Models from Fact Databases to Dynamic Reasoning Engines: The Next Paradigm https://www.workbyjacob.com/thoughts/from-llm-to-rqm-real-time-query-model 17 comments machinelearning
- [R] Large Language Models trained on code reason better, even on benchmarks that have nothing to do with code https://arxiv.org/abs/2210.07128 50 comments machinelearning
- Fin-R1: A Specialized Large Language Model for Financial Reasoning and Decision-Making https://www.marktechpost.com/2025/03/22/fin-r1-a-specialized-large-language-model-for-financial-reasoning-and-decision-making/ 2 comments machinelearningnews
- [R] Procedural Knowledge in Pretraining Drives Reasoning in Large Language Models https://arxiv.org/abs/2411.12580 3 comments machinelearning
- Large Language Models and the Socratic Method: Exploring how LLMs can simulate Socratic dialogues to stimulate critical thinking. Introducing the Tree of Thoughts method to improve AI’s performance on complex reasoning tasks and emphasizing the importance of critical thinking in the era of AI. https://www.cbrincoveanu.com/posts/large-language-models-and-the-socratic-method/ 2 comments learnmachinelearning
- [R] Large Language Models are Zero-Shot Reasoners. My summary: Adding text such as "Let’s think step by step" to a prompt "elicits chain of thought from large language models across a variety of reasoning tasks". https://arxiv.org/abs/2205.11916 55 comments machinelearning
- Reasoning skills of large language models are often overestimated | MIT News | Massachusetts Institute of Technology https://news.mit.edu/2024/reasoning-skills-large-language-models-often-overestimated-0711 15 comments computerscience
- [R] Chameleon: Plug-and-Play Compositional Reasoning with Large Language Models https://arxiv.org/abs/2304.09842 4 comments machinelearning
- [R] MathPrompter: Mathematical Reasoning using Large Language Models. New State of the Art on MultiArith ( 78.7% to 92.5%) with Text-Davinci 002 https://arxiv.org/abs/2303.05398 17 comments machinelearning
- [N][R] Reality check on the over hyped reasoning ability of large language models. I think it's high time we stop and spend more time in understanding what we're building rather than blindly scaling up to larger and larger models. https://medium.com/geekculture/cutting-through-the-hype-around-reasoning-ability-of-large-language-models-f96ad1d31e59 3 comments machinelearning
- [R] "Contrastive Decoding Improves Reasoning in Large Language Models", O'Brien & Lewis 2023 (boosts LLaMA-8B to >GPT-3.5/PaLM-540B on GSM8K) https://arxiv.org/abs/2309.09117#facebook 24 comments machinelearning