Computer Science · Literature

Research papers on Reasoning in large language models

Recent and highly-cited academic work on reasoning in large language models, gathered from Semantic Scholar, CrossRef and OpenAlex.

1. Chain of Thought Prompting Elicits Reasoning in Large Language Models

   Jason Wei, Xuezhi Wang, Dale Schuurmans, et al. · 2022 · ArXiv · 20,292 citations

   We explore how generating a chain of thought -- a series of intermediate reasoning steps -- significantly improves the ability of large language models to perform complex reasoning. In particular, we show how such reasoning abilities emerge naturally in sufficiently large language models via a simple method called chain of thought prompting, where a few chain of thought demonstrations are provided as exemplars in prompting. Experiments on three large language models show that chain of thought prompting improves performance on a range of arithmetic, commonsense, and symbolic reasoning tasks. The empirical gains can be striking. For instance, prompting a 540B-parameter language model with just

2. Self-Consistency Improves Chain of Thought Reasoning in Language Models

   Xuezhi Wang, Jason Wei, Dale Schuurmans, et al. · 2022 · ArXiv · 7,215 citations

   Chain-of-thought prompting combined with pre-trained large language models has achieved encouraging results on complex reasoning tasks. In this paper, we propose a new decoding strategy, self-consistency, to replace the naive greedy decoding used in chain-of-thought prompting. It first samples a diverse set of reasoning paths instead of only taking the greedy one, and then selects the most consistent answer by marginalizing out the sampled reasoning paths. Self-consistency leverages the intuition that a complex reasoning problem typically admits multiple different ways of thinking leading to its unique correct answer. Our extensive empirical evaluation shows that self-consistency boosts the

3. LLaMA: Open and Efficient Foundation Language Models

   Hugo Touvron, Thibaut Lavril, Gautier Izacard, et al. · 2023 · arXiv (Cornell University) · 3,926 citations

   We introduce LLaMA, a collection of foundation language models ranging from 7B to 65B parameters. We train our models on trillions of tokens, and show that it is possible to train state-of-the-art models using publicly available datasets exclusively, without resorting to proprietary and inaccessible datasets. In particular, LLaMA-13B outperforms GPT-3 (175B) on most benchmarks, and LLaMA-65B is competitive with the best models, Chinchilla-70B and PaLM-540B. We release all our models to the research community.

4. Performance of ChatGPT on USMLE: Potential for AI-assisted medical education using large language models

   Tiffany H. Kung, Morgan Cheatham, Arielle Medenilla, et al. · 2023 · PLOS Digital Health · 3,638 citations

   We evaluated the performance of a large language model called ChatGPT on the United States Medical Licensing Exam (USMLE), which consists of three exams: Step 1, Step 2CK, and Step 3. ChatGPT performed at or near the passing threshold for all three exams without any specialized training or reinforcement. Additionally, ChatGPT demonstrated a high level of concordance and insight in its explanations. These results suggest that large language models may have the potential to assist with medical education, and potentially, clinical decision-making.

5. Large language models encode clinical knowledge

   Karan Singhal, Shekoofeh Azizi, Tao Tu, et al. · 2023 · Nature · 3,347 citations

   Abstract Large language models (LLMs) have demonstrated impressive capabilities, but the bar for clinical applications is high. Attempts to assess the clinical knowledge of models typically rely on automated evaluations based on limited benchmarks. Here, to address these limitations, we present MultiMedQA, a benchmark combining six existing medical question answering datasets spanning professional medicine, research and consumer queries and a new dataset of medical questions searched online, HealthSearchQA. We propose a human evaluation framework for model answers along multiple axes including factuality, comprehension, reasoning, possible harm and bias. In addition, we evaluate Pathways Lan

6. Evaluating Large Language Models Trained on Code

   Mark Chen, Jerry Tworek, Heewoo Jun, et al. · 2021 · arXiv (Cornell University) · 1,449 citations

   I created a basic proof of concept of LMVM (Language Model Virtual Machine), a command line toolchain that English instructions into native binaries without intermediary high-level language compilation. The command line uses a large language model (LLM) as IR generation backend, producing LLVM Intermediate Representation (IR) directly from user instructions coding. The IR is compiled and linked to machine code (-o files) via `clang` (with an `llc`/`lld` fallback on POSIX hosts) and executed on bare metal, within Docker sandboxes, on AWS Lambda, or as standalone HTTP servers. How It Works (5 Steps) 1. You type an instruction - e.g. "Create a rest api running on localhost port 3000 and return

7. A Survey of Large Language Models

   Wayne Xin Zhao, Kun Zhou, Junyi Li, et al. · 2026 · Frontiers of Computer Science · 1,420 citations

   Abstract The rapid evolution of large language models (LLMs) has driven a transformative shift in artificial intelligence (AI), reshaping both research paradigms and practical applications. Distinguished from their predecessors by unprecedented scale and advanced capabilities, LLMs necessitate new frameworks for understanding their development, behavior, and societal impact. This survey systematically reviews recent advancements in LLM techniques across four key dimensions: (1) pre-training methodologies, which establish core model capabilities through large-scale self-supervised training, architectural innovations, and data curation strategies; (2) post-training techniques, including superv

8. Large Language Models are Zero-Shot Reasoners

   Takeshi Kojima, Shixiang Gu, Machel Reid, et al. · 2022 · arXiv (Cornell University) · 1,112 citations

   Pretrained large language models (LLMs) are widely used in many sub-fields of natural language processing (NLP) and generally known as excellent few-shot learners with task-specific exemplars. Notably, chain of thought (CoT) prompting, a recent technique for eliciting complex multi-step reasoning through step-by-step answer examples, achieved the state-of-the-art performances in arithmetics and symbolic reasoning, difficult system-2 tasks that do not follow the standard scaling laws for LLMs. While these successes are often attributed to LLMs' ability for few-shot learning, we show that LLMs are decent zero-shot reasoners by simply adding "Let's think step by step" before each answer. Experi

9. Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models

   Lei Wang, Wanyu Xu, Yihuai Lan, et al. · 2023 · 781 citations

   Large language models (LLMs) have recently been shown to deliver impressive performance in various NLP tasks. To tackle multi-step reasoning tasks, Few-shot chain-of-thought (CoT) prompting includes a few manually crafted step-by-step reasoning demonstrations which enable LLMs to explicitly generate reasoning steps and improve their reasoning task accuracy. To eliminate the manual efforts, Zero-shot-CoT concatenates the target problem statement with “Let’s think step by step” as an input prompt to LLMs. Despite the success of Zero-shot-CoT, it still suffers from three pitfalls: calculation errors, missing-step errors, and semantic misunderstanding errors. To address the missing-step errors,

10. A survey on multimodal large language models

    Shukang Yin, Chaoyou Fu, Sirui Zhao, et al. · 2024 · National Science Review · 602 citations

    Recently, the multimodal large language model (MLLM) represented by GPT-4V has been a new rising research hotspot, which uses powerful large language models (LLMs) as a brain to perform multimodal tasks. The surprising emergent capabilities of the MLLM, such as writing stories based on images and optical character recognition-free math reasoning, are rare in traditional multimodal methods, suggesting a potential path to artificial general intelligence. To this end, both academia and industry have endeavored to develop MLLMs that can compete with or even outperform GPT-4V, pushing the limit of research at a surprising speed. In this paper, we aim to trace and summarize the recent progress of

11. Graph of Thoughts: Solving Elaborate Problems with Large Language Models

    Maciej Besta, Nils Blach, Ales Kubicek, et al. · 2024 · Proceedings of the AAAI Conference on Artificial Intelligence · 426 citations

    We introduce Graph of Thoughts (GoT): a framework that advances prompting capabilities in large language models (LLMs) beyond those offered by paradigms such as Chain-of-Thought or Tree of Thoughts (ToT). The key idea and primary advantage of GoT is the ability to model the information generated by an LLM as an arbitrary graph, where units of information ("LLM thoughts") are vertices, and edges correspond to dependencies between these vertices. This approach enables combining arbitrary LLM thoughts into synergistic outcomes, distilling the essence of whole networks of thoughts, or enhancing thoughts using feedback loops. We illustrate that GoT offers advantages over state of the art on diffe

12. Towards Reasoning Era: A Survey of Long Chain-of-Thought for Reasoning Large Language Models

    Qiguang Chen, Libo Qin, Jinhao Liu, et al. · 2025 · ArXiv · 330 citations

    Recent advancements in reasoning with large language models (RLLMs), such as OpenAI-O1 and DeepSeek-R1, have demonstrated their impressive capabilities in complex domains like mathematics and coding. A central factor in their success lies in the application of long chain-of-thought (Long CoT) characteristics, which enhance reasoning abilities and enable the solution of intricate problems. However, despite these developments, a comprehensive survey on Long CoT is still lacking, limiting our understanding of its distinctions from traditional short chain-of-thought (Short CoT) and complicating ongoing debates on issues like"overthinking"and"inference-time scaling."This survey seeks to fill this

13. Automatic Chain of Thought Prompting in Large Language Models

    Zhuosheng Zhang, Aston Zhang, Mu Li, et al. · 2022 · arXiv (Cornell University) · 235 citations

    Large language models (LLMs) can perform complex reasoning by generating intermediate reasoning steps. Providing these steps for prompting demonstrations is called chain-of-thought (CoT) prompting. CoT prompting has two major paradigms. One leverages a simple prompt like "Let's think step by step" to facilitate step-by-step thinking before answering a question. The other uses a few manual demonstrations one by one, each composed of a question and a reasoning chain that leads to an answer. The superior performance of the second paradigm hinges on the hand-crafting of task-specific demonstrations one by one. We show that such manual efforts may be eliminated by leveraging LLMs with the "Let's

14. Graph Chain-of-Thought: Augmenting Large Language Models by Reasoning on Graphs

    Bowen Jin, Chulin Xie, Jiawei Zhang, et al. · 2024 · 154 citations

    Large language models (LLMs), while exhibiting exceptional performance, suffer from hallucinations, especially on knowledge-intensive tasks. Existing works propose to augment LLMs with individual text units retrieved from external knowledge corpora to alleviate the issue. However, in many domains, texts are interconnected (e.g., academic papers in a bibliographic graph are linked by citations and co-authorships) which form a (text-attributed) graph. The knowledge in such graphs is encoded not only in single texts/nodes but also in their associated connections. To facilitate the research of augmenting LLMs with graphs, we manually construct a Graph Reasoning Benchmark dataset called GRBench,

15. Chain-of-Thought Hub: A Continuous Effort to Measure Large Language Models' Reasoning Performance

    Yao Fu, Litu Ou, Mingyu Chen, et al. · 2023 · ArXiv · 134 citations

    As large language models (LLMs) are continuously being developed, their evaluation becomes increasingly important yet challenging. This work proposes Chain-of-Thought Hub, an open-source evaluation suite on the multi-step reasoning capabilities of large language models. We are interested in this setting for two reasons: (1) from the behavior of GPT and PaLM model family, we observe that complex reasoning is likely to be a key differentiator between weaker and stronger LLMs; (2) we envisage large language models to become the next-generation computational platform and foster an ecosystem of LLM-based new applications, this naturally requires the foundation models to perform complex tasks that

16. Enhancing Zero-Shot Chain-of-Thought Reasoning in Large Language Models through Logic

    Xufeng Zhao, Mengdi Li, Wenhao Lu, et al. · 2023 · ArXiv · 83 citations

    Recent advancements in large language models have showcased their remarkable generalizability across various domains. However, their reasoning abilities still have significant room for improvement, especially when confronted with scenarios requiring multi-step reasoning. Although large language models possess extensive knowledge, their reasoning often fails to effectively utilize this knowledge to establish a coherent thinking paradigm. These models sometimes show hallucinations as their reasoning procedures are unconstrained by logical principles. Aiming at improving the zero-shot chain-of-thought reasoning ability of large language models, we propose LoT (Logical Thoughts), a self-improvem

17. Can Large Language Models Detect Errors in Long Chain-of-Thought Reasoning?

    Yancheng He, Shilong Li, Jiaheng Liu, et al. · 2025 · ArXiv · 49 citations

    Recently, o1-like models have drawn significant attention, where these models produce the long Chain-of-Thought (CoT) reasoning steps to improve the reasoning abilities of existing Large Language Models (LLMs). In this paper, to understand the qualities of these long CoTs and measure the critique abilities of existing LLMs on these long CoTs, we introduce the DeltaBench, including the generated long CoTs from different o1-like models (e.g., QwQ, DeepSeek-R1) for different reasoning tasks (e.g., Math, Code, General Reasoning), to measure the ability to detect errors in long CoT reasoning. Based on DeltaBench, we first perform fine-grained analysis of the generated long CoTs to discover the ef

18. Stepwise Perplexity-Guided Refinement for Efficient Chain-of-Thought Reasoning in Large Language Models

    Yingqian Cui, Pengfei He, Jingying Zeng, et al. · 2025 · ArXiv · 44 citations

    Chain-of-Thought (CoT) reasoning, which breaks down complex tasks into intermediate reasoning steps, has significantly enhanced the performance of large language models (LLMs) on challenging tasks. However, the detailed reasoning process in CoT often incurs long generation times and high computational costs, partly due to the inclusion of unnecessary steps. To address this, we propose a method to identify critical reasoning steps using perplexity as a measure of their importance: a step is deemed critical if its removal causes a significant increase in perplexity. Our method enables models to focus solely on generating these critical steps. This can be achieved through two approaches: refini

19. On the Hardness of Faithful Chain-of-Thought Reasoning in Large Language Models

    Sree Harsha Tanneru, Dan Ley, Chirag Agarwal, et al. · 2024 · ArXiv · 36 citations

    As Large Language Models (LLMs) are increasingly being employed in real-world applications in critical domains such as healthcare, it is important to ensure that the Chain-of-Thought (CoT) reasoning generated by these models faithfully captures their underlying behavior. While LLMs are known to generate CoT reasoning that is appealing to humans, prior studies have shown that these explanations do not accurately reflect the actual behavior of the underlying LLMs. In this work, we explore the promise of three broad approaches commonly employed to steer the behavior of LLMs to enhance the faithfulness of the CoT reasoning generated by LLMs: in-context learning, fine-tuning, and activation editi

20. Unlocking General Long Chain-of-Thought Reasoning Capabilities of Large Language Models via Representation Engineering

    Xinyu Tang, Xiaolei Wang, Zhihao Lv, et al. · 2025 · 35 citations

    Recent advancements in long chain-of-thoughts(long CoTs) have significantly improved the reasoning capabilities of large language models(LLMs). Existing work finds that the capability of long CoT reasoning can be efficiently elicited by tuning on only a few examples and can easily transfer to other tasks. This motivates us to investigate whether long CoT reasoning is a general capability for LLMs. In this work, we conduct an empirical analysis for this question from the perspective of representation. We find that LLMs do encode long CoT reasoning as a general capability, with a clear distinction from vanilla CoTs. Furthermore, domain-specific representations are also required for the effecti

21. ThinkSound: Chain-of-Thought Reasoning in Multimodal Large Language Models for Audio Generation and Editing

    Huadai Liu, Jialei Wang, Kaicheng Luo, et al. · 2025 · ArXiv · 34 citations

    While end-to-end video-to-audio generation has greatly improved, producing high-fidelity audio that authentically captures the nuances of visual content remains challenging. Like professionals in the creative industries, this generation requires sophisticated reasoning about items such as visual dynamics, acoustic environments, and temporal relationships. We present ThinkSound, a novel framework that leverages Chain-of-Thought (CoT) reasoning to enable stepwise, interactive audio generation and editing for videos. Our approach decomposes the process into three complementary stages: foundational foley generation that creates semantically coherent soundscapes, interactive object-centric refine

22. Long-Short Chain-of-Thought Mixture Supervised Fine-Tuning Eliciting Efficient Reasoning in Large Language Models

    Bin Yu, Hang Yuan, Haotian Li, et al. · 2025 · ArXiv · 31 citations

    Recent advances in large language models have demonstrated that Supervised Fine-Tuning (SFT) with Chain-of-Thought (CoT) reasoning data distilled from large reasoning models (e.g., DeepSeek R1) can effectively transfer reasoning capabilities to non-reasoning models. However, models fine-tuned with this approach inherit the"overthinking"problem from teacher models, producing verbose and redundant reasoning chains during inference. To address this challenge, we propose Long-Short Chain-of-Thought Mixture Supervised Fine-Tuning (LS-Mixture SFT), which combines long CoT reasoning dataset with their short counterparts obtained through structure-preserved rewriting. Our experiments demonstrate tha

23. Enhancing Chain of Thought Prompting in Large Language Models via Reasoning Patterns

    Yufeng Zhang, Xuepeng Wang, Lingxiang Wu, et al. · 2024 · 31 citations

    Chain of Thought (CoT) prompting can encourage language models to engage in multi-step logical reasoning. The quality of the provided demonstrations significantly influences the success of downstream inference tasks. Current unsupervised CoT methods primarily select examples based on the semantics of the questions, which can introduce noise and lack interpretability. In this paper, we propose leveraging reasoning patterns to enhance CoT prompting effectiveness. Reasoning patterns represent the process by which language models arrive at their final results. By utilizing prior knowledge and prompt-based methods from large models, we first construct task-specific pattern sets. We then select di

24. Interactive Reasoning: Visualizing and Controlling Chain-of-Thought Reasoning in Large Language Models

    Rock Yuren Pang, K. Feng, Shangbin Feng, et al. · 2025 · Proceedings of the 31st International Conference on Intelligent User Interfaces · 15 citations

    The output quality of large language models (LLMs) can be improved via “reasoning”: generating segments of chain-of-thought (CoT) content to further condition the model prior to producing user-facing output. While these chains contain valuable information, they are verbose and lack explicit organization, making them tedious to review. Moreover, they lack opportunities for user feedback, such as removing unwanted considerations, adding desired ones, or clarifying unclear assumptions. We introduce Interactive Reasoning, an interaction design that visualizes chain-of-thought outputs as a hierarchy of topics and enables user review and modification. We implement interactive reasoning in Hippo, a