LLM-Agents

Timeline for the first release of related work.

Reasoning

1. CoT (NIPS-22)

   [2201.11903] Chain-of-Thought Prompting Elicits Reasoning in Large Language Models (arxiv.org) (opens new window)

2. SC-CoT (ICLR-23)

   [2203.11171] Self-Consistency Improves Chain of Thought Reasoning in Language Models (arxiv.org) (opens new window)

3. Zero-Shot CoT (NIPS-22)

   [2205.11916] Large Language Models are Zero-Shot Reasoners (arxiv.org) (opens new window)

4. Selection-Inference (ICLR-23 top5%)

   [2205.09712] Selection-Inference: Exploiting Large Language Models for Interpretable Logical Reasoning (arxiv.org) (opens new window)

5. Least-to-Most (ICLR-23)

   [2205.10625] Least-to-Most Prompting Enables Complex Reasoning in Large Language Models (opens new window)

6. MRKL Systems

   [2205.00445] MRKL Systems: A modular, neuro-symbolic architecture that combines large language models, external knowledge sources and discrete reasoning (arxiv.org) (opens new window)

7. Self-Critique

   [2206.05802] Self-critiquing models for assisting human evaluators (arxiv.org) (opens new window)

8. Auto-CoT (ICLR-23)

   [2210.03493] Automatic Chain of Thought Prompting in Large Language Models (arxiv.org) (opens new window)

9. Self-Refine

   [2303.17651] Self-Refine: Iterative Refinement with Self-Feedback (opens new window)

10. Self-Polish

    [2305.14497] Self-Polish: Enhance Reasoning in Large Language Models via Problem Refinement (opens new window)

11. ToT (NIPS-23 Oral)

    [2305.10601] Tree of Thoughts: Deliberate Problem Solving with Large Language Models (opens new window)

12. ChatCoT

    [2305.14323] ChatCoT: Tool-Augmented Chain-of-Thought Reasoning on Chat-based Large Language Models (opens new window)

13. RAP

    [2305.14992] Reasoning with Language Model is Planning with World Model (arxiv.org) (opens new window)

14. ReWOO

    [2305.18323] ReWOO: Decoupling Reasoning from Observations for Efficient Augmented Language Models (arxiv.org) (opens new window)

15. Thought decomposition

    [2305.00633] Decomposition Enhances Reasoning via Self-Evaluation Guided Decoding (arxiv.org) (opens new window)

16. Plan-and-Solve Prompting (ACL-23)

    [2305.04091] Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models (opens new window)

17. SoT

    [2307.15337] Skeleton-of-Thought: Large Language Models Can Do Parallel Decoding (arxiv.org) (opens new window)

18. SelfCheck

    [2308.00436] SelfCheck: Using LLMs to Zero-Shot Check Their Own Step-by-Step Reasoning (opens new window)

19. GoT

    [2308.09687] Graph of Thoughts: Solving Elaborate Problems with Large Language Models (arxiv.org) (opens new window)

20. AoT

    [2308.10379] Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language Models (arxiv.org) (opens new window)

21. ReConcile

    [2309.13007] ReConcile: Round-Table Conference Improves Reasoning via Consensus among Diverse LLMs (arxiv.org) (opens new window)

22. LEMA

    [2310.20689] Learning From Mistakes Makes LLM Better Reasoner (arxiv.org) (opens new window)

23. Step-Back Prompting

    [2310.06117] Take a Step Back: Evoking Reasoning via Abstraction in Large Language Models (arxiv.org) (opens new window)

24. Analogical Reasoners

    [2310.01714] Large Language Models as Analogical Reasoners (arxiv.org) (opens new window)

Planning

1. Language Models as Zero-Shot Planners (ICML-22)

   [2201.07207] Language Models as Zero-Shot Planners: Extracting Actionable Knowledge for Embodied Agents (opens new window)

2. SayCan (CoRL-22 Oral)

   [2204.01691] Do As I Can, Not As I Say: Grounding Language in Robotic Affordances (opens new window)

3. Inner Monologue (CoRL-22)

   [2207.05608] Inner Monologue: Embodied Reasoning through Planning with Language Models (opens new window)

4. ReAct (ICLR-23 top5%)

   [2210.03629] ReAct: Synergizing Reasoning and Acting in Language Models (opens new window)

5. ICPI (NIPS-23)

   [2210.03821v2] Large Language Models can Implement Policy Iteration (opens new window)

6. Re-Prompting(NIPS-22 Workshop)

   [2211.09935] Planning with Large Language Models via Corrective Re-prompting (arxiv.org) (opens new window)

7. LLM-Planner (ICCV-23)

   [2212.04088] LLM-Planner: Few-Shot Grounded Planning for Embodied Agents with Large Language Models (opens new window)

8. Don’t Generate, Discriminate?�?ACL-23?��

   https://aclanthology.org/2023.acl-long.270.pdf

9. DECKARD?�?ICML-23?��

   [2301.12050] Do Embodied Agents Dream of Pixelated Sheep: Embodied Decision Making using Language Guided World Modelling (arxiv.org) (opens new window)

10. Describe, Explain, Plan and Select (NIPS-23)

    [2302.01560] Describe, Explain, Plan and Select: Interactive Planning with Large Language Models Enables Open-World Multi-Task Agents (opens new window)

11. Refleion (NIPS-23)

    [2303.11366] Reflexion: Language Agents with Verbal Reinforcement Learning (opens new window)

12. PaLM-E

    [2303.03378] PaLM-E: An Embodied Multimodal Language Model (arxiv.org) (opens new window)

13. Plan4MC

    [2303.16563] Plan4MC: Skill Reinforcement Learning and Planning for Open-World Minecraft Tasks (arxiv.org) (opens new window)

14. Chat with the Environment

    [2303.08268] Chat with the Environment: Interactive Multimodal Perception Using Large Language Models (opens new window)

15. Text2Motion

    [2303.12153] Text2Motion: From Natural Language Instructions to Feasible Plans (arxiv.org) (opens new window)

16. LLM+P

    [2304.11477] LLM+P: Empowering Large Language Models with Optimal Planning Proficiency (opens new window)

17. Self-Debug

    [2304.05128] Teaching Large Language Models to Self-Debug (arxiv.org) (opens new window)

18. LLM-MCTS (NIPS-23)

    [2305.14078] Large Language Models as Commonsense Knowledge for Large-Scale Task Planning (arxiv.org) (opens new window)

19. LLMs-World-Models-for-Planning (NIPS-23)

    [2305.14909] Leveraging Pre-trained Large Language Models to Construct and Utilize World Models for Model-based Task Planning (arxiv.org) (opens new window)

20. Voyager

    [2305.16291] Voyager: An Open-Ended Embodied Agent with Large Language Models (opens new window)

21. SwiftSage (NIPS-23)

    [2305.17390] SwiftSage: A Generative Agent with Fast and Slow Thinking for Complex Interactive Tasks (opens new window)

22. Plan, Eliminate, and Track

    [2305.02412] Plan, Eliminate, and Track -- Language Models are Good Teachers for Embodied Agents (opens new window)

23. AdaPlanner

    [2305.16653] AdaPlanner: Adaptive Planning from Feedback with Language Models (opens new window)

24. Language Models Meet World Models

    [2305.10626] Language Models Meet World Models: Embodied Experiences Enhance Language Models (arxiv.org) (opens new window)

25. Ghost in the Minecraft

    [2305.17144] Ghost in the Minecraft: Generally Capable Agents for Open-World Environments via Large Language Models with Text-based Knowledge and Memory (opens new window)

26. AlphaBlock

    [2305.18898] AlphaBlock: Embodied Finetuning for Vision-Language Reasoning in Robot Manipulation (arxiv.org) (opens new window)

27. RecAgent

    [2306.02552] When Large Language Model based Agent Meets User Behavior Analysis: A Novel User Simulation Paradigm (opens new window)

28. Sayplan (CoRL-23 Oral)

    [2307.06135] SayPlan: Grounding Large Language Models using 3D Scene Graphs for Scalable Robot Task Planning (arxiv.org) (opens new window)

29. Robot-Help (CoRL-23 Oral)

    [2307.01928] Robots That Ask For Help: Uncertainty Alignment for Large Language Model Planners (arxiv.org) (opens new window)

30. TaPA

    [2307.01848] Embodied Task Planning with Large Language Models (arxiv.org) (opens new window)

31. A Unified Agent (ICLR-23 Workshop)

    [2307.09668] Towards A Unified Agent with Foundation Models (arxiv.org) (opens new window)

32. ExpeL

    [2308.10144] ExpeL: LLM Agents Are Experiential Learners (opens new window)

33. TPTU

    [2308.03427] TPTU: Task Planning and Tool Usage of Large Language Model-based AI Agents (opens new window)

34. LLM-DP

    [2308.06391] Dynamic Planning with a LLM (arxiv.org) (opens new window)

35. Retroformer

    [2308.02151] Retroformer: Retrospective Large Language Agents with Policy Gradient Optimization (opens new window)

36. CodePlan

    [2309.12499] CodePlan: Repository-level Coding using LLMs and Planning (opens new window)

37. Hierarchical Planning (NIPS-23)

    [2309.08587] Compositional Foundation Models for Hierarchical Planning (opens new window)

38. Agents

    [2309.07870] Agents: An Open-source Framework for Autonomous Language Agents (arxiv.org) (opens new window)

39. Self-driven Grounding

    [2309.01352] Self-driven Grounding: Large Language Model Agents with Automatical Language-aligned Skill Learning (opens new window)

40. LATS

    [2310.04406] Language Agent Tree Search Unifies Reasoning Acting and Planning in Language Models (arxiv.org) (opens new window)

Tool

Reference Material

zjunlp/Prompt4ReasoningPapers: Repository for the ACL2023 paper "Reasoning with Language Model Prompting: A Survey". (github.com) (opens new window)

AGI-Edgerunners/LLM-Planning-Papers: Must-read Papers on Large Language Model (LLM) Planning. (github.com) (opens new window)