A comprehensive index of artificial intelligence and machine-learning research with AI-generated summaries, citation metrics, and direct links to papers and code.
Arda Sevinc, Abdurrahman Gumus
AutoReason automatically generates chain-of-thought rationales by decomposing implicit queries into explicit questions, improving multi-step reasoning accuracy in weaker LLMs on StrategyQA and HotpotQA.
Xiaomin Li, Zhou Yu, Zhiwei Zhang, et al.
This paper reveals that chain-of-thought reasoning can significantly degrade instruction-following accuracy in LLMs, and proposes selective reasoning strategies to mitigate the issue.
Violet Xiang, Charles Burton Snell, K. Gandhi, et al.
Proposes Meta Chain-of-Thought (Meta-CoT) to explicitly model the reasoning behind CoT, with a training pipeline and open research questions.
Asifullah Khan, Muhammad Zaeem Khan, Saleha Jamshed, et al.
A comprehensive survey of LLM advances covering reasoning, adaptability, efficiency, and ethics, highlighting techniques like Chain-of-Thought, instruction tuning, and RLHF, and identifying future research directions.
Ayeong Lee, Ethan Che, Tianyi Peng
First systematic study of the relationship between reasoning length and accuracy in LLMs, revealing a universal tradeoff and token complexity threshold.
Unknown
This paper introduces Action Chain-of-Thought (ACoT) for Vision-Language-Action models, decomposing action generation into a step-by-step reasoning process to improve manipulation performance and interpretability.
Unknown
This paper introduces a chain-of-thought prompting method for vision-language models, enabling them to reason step-by-step over visual inputs to improve performance on complex visual reasoning tasks.
Unknown
This paper introduces grounded chain-of-thought (GCoT), a new learning task that enhances multimodal large language models by grounding their reasoning steps in visual evidence.
Unknown
This paper proposes Chain-of-Thought (CoT) supervision to improve long-context language models by training them to reason over dispersed information before generating answers.
Pulkit Verma, Ngoc La, Anthony Favier, et al.
PDDL-Instruct instruction tuning framework enhances LLMs' symbolic planning via logical chain-of-thought reasoning, achieving 94% planning accuracy.
Ruiming Zhu, Tianhao Peng, Tianhao Cheng, et al.
A comprehensive survey on latent reasoning in LLMs, which performs multi-step inference in continuous hidden states to overcome the expressive limits of explicit chain-of-thought.
Shansong Wang, Mingzhe Hu, Qiang Li, et al.
GPT-5 achieves above-human-expert performance on multimodal medical reasoning benchmarks via zero-shot chain-of-thought.