SlideShare a Scribd company logo
LLM Agents
Contents
1. Prerequisites
2. Why LLM Agents?
3. LLM’s vs Agentic Response
4. LLM Agent overview
a. Component One: Planning
b. Component Two: Memory (Human Agent analogy)
c. Component Three: Tool Use
5. ReACT: An agent technique for “Component One: Planning”
Prerequisite: Retrieval Augmented Generation (RAG)
Why LLM Agents?
In a human body brain (specifically
consciousness, excluding memory) in itself
can’t do anything without hands/sense organs
memory!
Consider LLM as brain and Agent using LLM as
complete body
Why (What is) LLM Agents?
Consider a LLM application that is designed to help financial analysts.
Simple question “What was X corporation’s total revenue for FY
2022?” -> RAG pipeline with company's data can answer
Real life question which the analyst would ask. -> “What were the
three takeaways from the Q2 earnings call from FY 23? Focus on the
technological moats that the company is building”
- this information requires more than a simple lookup from an
earnings call. It requires planning, tailored focus, memory, using
different tools, and breaking down a complex question into
simpler sub-parts….
- These concepts assembled together are essentially what we have
come to refer to as an LLM Agent.
LLM’s vs Agentic Response
ReAct is one type of Agent Technique
LLM Agent overview
Agent uses LLM as brain’s cerebrum to perform multiple
decision making
Component One: Planning
- Task Decomposition: A complicated task usually involves many
steps. An agent needs to know what they are and plan ahead.
- Self-Reflection: Allows autonomous agents to improve iteratively by
refining past action decisions and correcting previous mistakes.
Component One: Planning
- Task Decomposition: A complicated task usually involves many
steps. An agent needs to know what they are and plan ahead.
- Self-Reflection: Allows autonomous agents to improve iteratively by
refining past action decisions and correcting previous mistakes.
- one notable technique: ReACT: SYNERGIZING REASONING
AND ACTING IN LANGUAGE MODELS
Component Two: Memory
Categorization of human memory.
Component Two: Memory
We can roughly consider the following mappings, for Human memory to
Agent’s memory,
● Sensory memory -> embedding representations for raw inputs,
including text, image or other modalities;
● Short-term memory -> in-context learning. It is short and finite, as it
is restricted by the finite context window length of Transformer.
● Long-term memory -> external vector store that the agent can
attend to at query time, accessible via fast retrieval
Component Three: Tool Use
Utilizing LLM to know which tool to use when, and how to use that tool
- TALM (Tool Augmented Language Models; Parisi et al. 2022)
- Toolformer by Meta
- HuggingGPT
The format of API calls in TALM
ReACT

More Related Content

Similar to 社内勉強会資料_LLM Agents                              .

Cmsc411(Pascuappt Report)
Cmsc411(Pascuappt Report)Cmsc411(Pascuappt Report)
Cmsc411(Pascuappt Report)
Mannilou Pascua
 
Software project estimation
Software project estimationSoftware project estimation
Software project estimation
inayat khan
 
software project management.lpu.slide.ansh.gupta
software project management.lpu.slide.ansh.guptasoftware project management.lpu.slide.ansh.gupta
software project management.lpu.slide.ansh.gupta
yitoxay869
 
Mandarkulkarni 111003065827-phpapp01
Mandarkulkarni 111003065827-phpapp01Mandarkulkarni 111003065827-phpapp01
Mandarkulkarni 111003065827-phpapp01
PMI_IREP_TP
 
Loan Prediction System Using Machine Learning.pptx
Loan Prediction System Using Machine Learning.pptxLoan Prediction System Using Machine Learning.pptx
Loan Prediction System Using Machine Learning.pptx
BhoirRitesh19ET5008
 
Sad considerations for-candidate_system
Sad considerations for-candidate_systemSad considerations for-candidate_system
Sad considerations for-candidate_system
Swapnil Walde
 
Intelligent Algorithm for Assignment of Agents to Human Strategy in Centraliz...
Intelligent Algorithm for Assignment of Agents to Human Strategy in Centraliz...Intelligent Algorithm for Assignment of Agents to Human Strategy in Centraliz...
Intelligent Algorithm for Assignment of Agents to Human Strategy in Centraliz...
Reza Nourjou, Ph.D.
 
Management Information Systems – Week 7 Lecture 2Developme.docx
Management Information Systems – Week 7 Lecture 2Developme.docxManagement Information Systems – Week 7 Lecture 2Developme.docx
Management Information Systems – Week 7 Lecture 2Developme.docx
croysierkathey
 
Hms project report
Hms project reportHms project report
Hms project report
Simranjitkaur89
 
Machine learning
Machine learningMachine learning
Machine learning
osman ansari
 
Matrix organisation
Matrix organisationMatrix organisation
Matrix organisation
9947338518
 
Software engg. pressman_ch-6 & 7
Software engg. pressman_ch-6 & 7Software engg. pressman_ch-6 & 7
Software engg. pressman_ch-6 & 7
Dhairya Joshi
 
Hotel management
Hotel managementHotel management
Hotel management
Arman Ahmed
 
reqsforlearningagents.ppt
reqsforlearningagents.pptreqsforlearningagents.ppt
reqsforlearningagents.ppt
butest
 
Appendix AProof of effectiveness of some of the agile methods us.docx
Appendix AProof of effectiveness of some of the agile methods us.docxAppendix AProof of effectiveness of some of the agile methods us.docx
Appendix AProof of effectiveness of some of the agile methods us.docx
armitageclaire49
 
Managing Technical Debt
Managing Technical DebtManaging Technical Debt
Managing Technical Debt
Andre Perkins
 
Improving Effort Estimation in Agile Software Development Projects
Improving Effort Estimation in Agile Software Development ProjectsImproving Effort Estimation in Agile Software Development Projects
Improving Effort Estimation in Agile Software Development Projects
Gedi Siuskus
 
ERP - Implementation is The Challenge
ERP - Implementation is The ChallengeERP - Implementation is The Challenge
ERP - Implementation is The Challenge
vinaya.hs
 
Program concep sequential statements
Program concep sequential statementsProgram concep sequential statements
Program concep sequential statements
ankurkhanna
 
Report.pdf
Report.pdfReport.pdf
Report.pdf
hezamgawbah
 

Similar to 社内勉強会資料_LLM Agents                              . (20)

Cmsc411(Pascuappt Report)
Cmsc411(Pascuappt Report)Cmsc411(Pascuappt Report)
Cmsc411(Pascuappt Report)
 
Software project estimation
Software project estimationSoftware project estimation
Software project estimation
 
software project management.lpu.slide.ansh.gupta
software project management.lpu.slide.ansh.guptasoftware project management.lpu.slide.ansh.gupta
software project management.lpu.slide.ansh.gupta
 
Mandarkulkarni 111003065827-phpapp01
Mandarkulkarni 111003065827-phpapp01Mandarkulkarni 111003065827-phpapp01
Mandarkulkarni 111003065827-phpapp01
 
Loan Prediction System Using Machine Learning.pptx
Loan Prediction System Using Machine Learning.pptxLoan Prediction System Using Machine Learning.pptx
Loan Prediction System Using Machine Learning.pptx
 
Sad considerations for-candidate_system
Sad considerations for-candidate_systemSad considerations for-candidate_system
Sad considerations for-candidate_system
 
Intelligent Algorithm for Assignment of Agents to Human Strategy in Centraliz...
Intelligent Algorithm for Assignment of Agents to Human Strategy in Centraliz...Intelligent Algorithm for Assignment of Agents to Human Strategy in Centraliz...
Intelligent Algorithm for Assignment of Agents to Human Strategy in Centraliz...
 
Management Information Systems – Week 7 Lecture 2Developme.docx
Management Information Systems – Week 7 Lecture 2Developme.docxManagement Information Systems – Week 7 Lecture 2Developme.docx
Management Information Systems – Week 7 Lecture 2Developme.docx
 
Hms project report
Hms project reportHms project report
Hms project report
 
Machine learning
Machine learningMachine learning
Machine learning
 
Matrix organisation
Matrix organisationMatrix organisation
Matrix organisation
 
Software engg. pressman_ch-6 & 7
Software engg. pressman_ch-6 & 7Software engg. pressman_ch-6 & 7
Software engg. pressman_ch-6 & 7
 
Hotel management
Hotel managementHotel management
Hotel management
 
reqsforlearningagents.ppt
reqsforlearningagents.pptreqsforlearningagents.ppt
reqsforlearningagents.ppt
 
Appendix AProof of effectiveness of some of the agile methods us.docx
Appendix AProof of effectiveness of some of the agile methods us.docxAppendix AProof of effectiveness of some of the agile methods us.docx
Appendix AProof of effectiveness of some of the agile methods us.docx
 
Managing Technical Debt
Managing Technical DebtManaging Technical Debt
Managing Technical Debt
 
Improving Effort Estimation in Agile Software Development Projects
Improving Effort Estimation in Agile Software Development ProjectsImproving Effort Estimation in Agile Software Development Projects
Improving Effort Estimation in Agile Software Development Projects
 
ERP - Implementation is The Challenge
ERP - Implementation is The ChallengeERP - Implementation is The Challenge
ERP - Implementation is The Challenge
 
Program concep sequential statements
Program concep sequential statementsProgram concep sequential statements
Program concep sequential statements
 
Report.pdf
Report.pdfReport.pdf
Report.pdf
 

More from NABLAS株式会社

社内勉強会資料_Two Papers Contribute to Faster Python.pdf
社内勉強会資料_Two Papers Contribute to Faster Python.pdf社内勉強会資料_Two Papers Contribute to Faster Python.pdf
社内勉強会資料_Two Papers Contribute to Faster Python.pdf
NABLAS株式会社
 
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
NABLAS株式会社
 
【NABLAS Inc.】Recruitment materials - Ver. 2024
【NABLAS Inc.】Recruitment materials - Ver. 2024【NABLAS Inc.】Recruitment materials - Ver. 2024
【NABLAS Inc.】Recruitment materials - Ver. 2024
NABLAS株式会社
 
【NABLAS株式会社】採用ピッチ資料 Ver. 2024           .
【NABLAS株式会社】採用ピッチ資料 Ver. 2024           .【NABLAS株式会社】採用ピッチ資料 Ver. 2024           .
【NABLAS株式会社】採用ピッチ資料 Ver. 2024           .
NABLAS株式会社
 
社内勉強会資料  Mamba - A new era or ephemeral
社内勉強会資料   Mamba - A new era or ephemeral社内勉強会資料   Mamba - A new era or ephemeral
社内勉強会資料  Mamba - A new era or ephemeral
NABLAS株式会社
 
社内勉強会資料_Object Recognition as Next Token Prediction
社内勉強会資料_Object Recognition as Next Token Prediction社内勉強会資料_Object Recognition as Next Token Prediction
社内勉強会資料_Object Recognition as Next Token Prediction
NABLAS株式会社
 

More from NABLAS株式会社 (6)

社内勉強会資料_Two Papers Contribute to Faster Python.pdf
社内勉強会資料_Two Papers Contribute to Faster Python.pdf社内勉強会資料_Two Papers Contribute to Faster Python.pdf
社内勉強会資料_Two Papers Contribute to Faster Python.pdf
 
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
【社内勉強会資料_Octo: An Open-Source Generalist Robot Policy】
 
【NABLAS Inc.】Recruitment materials - Ver. 2024
【NABLAS Inc.】Recruitment materials - Ver. 2024【NABLAS Inc.】Recruitment materials - Ver. 2024
【NABLAS Inc.】Recruitment materials - Ver. 2024
 
【NABLAS株式会社】採用ピッチ資料 Ver. 2024           .
【NABLAS株式会社】採用ピッチ資料 Ver. 2024           .【NABLAS株式会社】採用ピッチ資料 Ver. 2024           .
【NABLAS株式会社】採用ピッチ資料 Ver. 2024           .
 
社内勉強会資料  Mamba - A new era or ephemeral
社内勉強会資料   Mamba - A new era or ephemeral社内勉強会資料   Mamba - A new era or ephemeral
社内勉強会資料  Mamba - A new era or ephemeral
 
社内勉強会資料_Object Recognition as Next Token Prediction
社内勉強会資料_Object Recognition as Next Token Prediction社内勉強会資料_Object Recognition as Next Token Prediction
社内勉強会資料_Object Recognition as Next Token Prediction
 

Recently uploaded

一比一原版(Glasgow毕业证书)格拉斯哥大学毕业证如何办理
一比一原版(Glasgow毕业证书)格拉斯哥大学毕业证如何办理一比一原版(Glasgow毕业证书)格拉斯哥大学毕业证如何办理
一比一原版(Glasgow毕业证书)格拉斯哥大学毕业证如何办理
g4dpvqap0
 
My burning issue is homelessness K.C.M.O.
My burning issue is homelessness K.C.M.O.My burning issue is homelessness K.C.M.O.
My burning issue is homelessness K.C.M.O.
rwarrenll
 
The Ipsos - AI - Monitor 2024 Report.pdf
The  Ipsos - AI - Monitor 2024 Report.pdfThe  Ipsos - AI - Monitor 2024 Report.pdf
The Ipsos - AI - Monitor 2024 Report.pdf
Social Samosa
 
ViewShift: Hassle-free Dynamic Policy Enforcement for Every Data Lake
ViewShift: Hassle-free Dynamic Policy Enforcement for Every Data LakeViewShift: Hassle-free Dynamic Policy Enforcement for Every Data Lake
ViewShift: Hassle-free Dynamic Policy Enforcement for Every Data Lake
Walaa Eldin Moustafa
 
一比一原版(爱大毕业证书)爱丁堡大学毕业证如何办理
一比一原版(爱大毕业证书)爱丁堡大学毕业证如何办理一比一原版(爱大毕业证书)爱丁堡大学毕业证如何办理
一比一原版(爱大毕业证书)爱丁堡大学毕业证如何办理
g4dpvqap0
 
Influence of Marketing Strategy and Market Competition on Business Plan
Influence of Marketing Strategy and Market Competition on Business PlanInfluence of Marketing Strategy and Market Competition on Business Plan
Influence of Marketing Strategy and Market Competition on Business Plan
jerlynmaetalle
 
Natural Language Processing (NLP), RAG and its applications .pptx
Natural Language Processing (NLP), RAG and its applications .pptxNatural Language Processing (NLP), RAG and its applications .pptx
Natural Language Processing (NLP), RAG and its applications .pptx
fkyes25
 
一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理
一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理
一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理
bopyb
 
State of Artificial intelligence Report 2023
State of Artificial intelligence Report 2023State of Artificial intelligence Report 2023
State of Artificial intelligence Report 2023
kuntobimo2016
 
Global Situational Awareness of A.I. and where its headed
Global Situational Awareness of A.I. and where its headedGlobal Situational Awareness of A.I. and where its headed
Global Situational Awareness of A.I. and where its headed
vikram sood
 
Beyond the Basics of A/B Tests: Highly Innovative Experimentation Tactics You...
Beyond the Basics of A/B Tests: Highly Innovative Experimentation Tactics You...Beyond the Basics of A/B Tests: Highly Innovative Experimentation Tactics You...
Beyond the Basics of A/B Tests: Highly Innovative Experimentation Tactics You...
Aggregage
 
Palo Alto Cortex XDR presentation .......
Palo Alto Cortex XDR presentation .......Palo Alto Cortex XDR presentation .......
Palo Alto Cortex XDR presentation .......
Sachin Paul
 
Intelligence supported media monitoring in veterinary medicine
Intelligence supported media monitoring in veterinary medicineIntelligence supported media monitoring in veterinary medicine
Intelligence supported media monitoring in veterinary medicine
AndrzejJarynowski
 
End-to-end pipeline agility - Berlin Buzzwords 2024
End-to-end pipeline agility - Berlin Buzzwords 2024End-to-end pipeline agility - Berlin Buzzwords 2024
End-to-end pipeline agility - Berlin Buzzwords 2024
Lars Albertsson
 
University of New South Wales degree offer diploma Transcript
University of New South Wales degree offer diploma TranscriptUniversity of New South Wales degree offer diploma Transcript
University of New South Wales degree offer diploma Transcript
soxrziqu
 
Challenges of Nation Building-1.pptx with more important
Challenges of Nation Building-1.pptx with more importantChallenges of Nation Building-1.pptx with more important
Challenges of Nation Building-1.pptx with more important
Sm321
 
Enhanced Enterprise Intelligence with your personal AI Data Copilot.pdf
Enhanced Enterprise Intelligence with your personal AI Data Copilot.pdfEnhanced Enterprise Intelligence with your personal AI Data Copilot.pdf
Enhanced Enterprise Intelligence with your personal AI Data Copilot.pdf
GetInData
 
一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理
一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理
一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理
slg6lamcq
 
Predictably Improve Your B2B Tech Company's Performance by Leveraging Data
Predictably Improve Your B2B Tech Company's Performance by Leveraging DataPredictably Improve Your B2B Tech Company's Performance by Leveraging Data
Predictably Improve Your B2B Tech Company's Performance by Leveraging Data
Kiwi Creative
 
Everything you wanted to know about LIHTC
Everything you wanted to know about LIHTCEverything you wanted to know about LIHTC
Everything you wanted to know about LIHTC
Roger Valdez
 

Recently uploaded (20)

一比一原版(Glasgow毕业证书)格拉斯哥大学毕业证如何办理
一比一原版(Glasgow毕业证书)格拉斯哥大学毕业证如何办理一比一原版(Glasgow毕业证书)格拉斯哥大学毕业证如何办理
一比一原版(Glasgow毕业证书)格拉斯哥大学毕业证如何办理
 
My burning issue is homelessness K.C.M.O.
My burning issue is homelessness K.C.M.O.My burning issue is homelessness K.C.M.O.
My burning issue is homelessness K.C.M.O.
 
The Ipsos - AI - Monitor 2024 Report.pdf
The  Ipsos - AI - Monitor 2024 Report.pdfThe  Ipsos - AI - Monitor 2024 Report.pdf
The Ipsos - AI - Monitor 2024 Report.pdf
 
ViewShift: Hassle-free Dynamic Policy Enforcement for Every Data Lake
ViewShift: Hassle-free Dynamic Policy Enforcement for Every Data LakeViewShift: Hassle-free Dynamic Policy Enforcement for Every Data Lake
ViewShift: Hassle-free Dynamic Policy Enforcement for Every Data Lake
 
一比一原版(爱大毕业证书)爱丁堡大学毕业证如何办理
一比一原版(爱大毕业证书)爱丁堡大学毕业证如何办理一比一原版(爱大毕业证书)爱丁堡大学毕业证如何办理
一比一原版(爱大毕业证书)爱丁堡大学毕业证如何办理
 
Influence of Marketing Strategy and Market Competition on Business Plan
Influence of Marketing Strategy and Market Competition on Business PlanInfluence of Marketing Strategy and Market Competition on Business Plan
Influence of Marketing Strategy and Market Competition on Business Plan
 
Natural Language Processing (NLP), RAG and its applications .pptx
Natural Language Processing (NLP), RAG and its applications .pptxNatural Language Processing (NLP), RAG and its applications .pptx
Natural Language Processing (NLP), RAG and its applications .pptx
 
一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理
一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理
一比一原版(GWU,GW文凭证书)乔治·华盛顿大学毕业证如何办理
 
State of Artificial intelligence Report 2023
State of Artificial intelligence Report 2023State of Artificial intelligence Report 2023
State of Artificial intelligence Report 2023
 
Global Situational Awareness of A.I. and where its headed
Global Situational Awareness of A.I. and where its headedGlobal Situational Awareness of A.I. and where its headed
Global Situational Awareness of A.I. and where its headed
 
Beyond the Basics of A/B Tests: Highly Innovative Experimentation Tactics You...
Beyond the Basics of A/B Tests: Highly Innovative Experimentation Tactics You...Beyond the Basics of A/B Tests: Highly Innovative Experimentation Tactics You...
Beyond the Basics of A/B Tests: Highly Innovative Experimentation Tactics You...
 
Palo Alto Cortex XDR presentation .......
Palo Alto Cortex XDR presentation .......Palo Alto Cortex XDR presentation .......
Palo Alto Cortex XDR presentation .......
 
Intelligence supported media monitoring in veterinary medicine
Intelligence supported media monitoring in veterinary medicineIntelligence supported media monitoring in veterinary medicine
Intelligence supported media monitoring in veterinary medicine
 
End-to-end pipeline agility - Berlin Buzzwords 2024
End-to-end pipeline agility - Berlin Buzzwords 2024End-to-end pipeline agility - Berlin Buzzwords 2024
End-to-end pipeline agility - Berlin Buzzwords 2024
 
University of New South Wales degree offer diploma Transcript
University of New South Wales degree offer diploma TranscriptUniversity of New South Wales degree offer diploma Transcript
University of New South Wales degree offer diploma Transcript
 
Challenges of Nation Building-1.pptx with more important
Challenges of Nation Building-1.pptx with more importantChallenges of Nation Building-1.pptx with more important
Challenges of Nation Building-1.pptx with more important
 
Enhanced Enterprise Intelligence with your personal AI Data Copilot.pdf
Enhanced Enterprise Intelligence with your personal AI Data Copilot.pdfEnhanced Enterprise Intelligence with your personal AI Data Copilot.pdf
Enhanced Enterprise Intelligence with your personal AI Data Copilot.pdf
 
一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理
一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理
一比一原版(Adelaide毕业证书)阿德莱德大学毕业证如何办理
 
Predictably Improve Your B2B Tech Company's Performance by Leveraging Data
Predictably Improve Your B2B Tech Company's Performance by Leveraging DataPredictably Improve Your B2B Tech Company's Performance by Leveraging Data
Predictably Improve Your B2B Tech Company's Performance by Leveraging Data
 
Everything you wanted to know about LIHTC
Everything you wanted to know about LIHTCEverything you wanted to know about LIHTC
Everything you wanted to know about LIHTC
 

社内勉強会資料_LLM Agents                              .

  • 2. Contents 1. Prerequisites 2. Why LLM Agents? 3. LLM’s vs Agentic Response 4. LLM Agent overview a. Component One: Planning b. Component Two: Memory (Human Agent analogy) c. Component Three: Tool Use 5. ReACT: An agent technique for “Component One: Planning”
  • 4. Why LLM Agents? In a human body brain (specifically consciousness, excluding memory) in itself can’t do anything without hands/sense organs memory! Consider LLM as brain and Agent using LLM as complete body
  • 5. Why (What is) LLM Agents? Consider a LLM application that is designed to help financial analysts. Simple question “What was X corporation’s total revenue for FY 2022?” -> RAG pipeline with company's data can answer Real life question which the analyst would ask. -> “What were the three takeaways from the Q2 earnings call from FY 23? Focus on the technological moats that the company is building” - this information requires more than a simple lookup from an earnings call. It requires planning, tailored focus, memory, using different tools, and breaking down a complex question into simpler sub-parts…. - These concepts assembled together are essentially what we have come to refer to as an LLM Agent.
  • 6. LLM’s vs Agentic Response ReAct is one type of Agent Technique
  • 7. LLM Agent overview Agent uses LLM as brain’s cerebrum to perform multiple decision making
  • 8. Component One: Planning - Task Decomposition: A complicated task usually involves many steps. An agent needs to know what they are and plan ahead. - Self-Reflection: Allows autonomous agents to improve iteratively by refining past action decisions and correcting previous mistakes.
  • 9. Component One: Planning - Task Decomposition: A complicated task usually involves many steps. An agent needs to know what they are and plan ahead. - Self-Reflection: Allows autonomous agents to improve iteratively by refining past action decisions and correcting previous mistakes. - one notable technique: ReACT: SYNERGIZING REASONING AND ACTING IN LANGUAGE MODELS
  • 11. Component Two: Memory We can roughly consider the following mappings, for Human memory to Agent’s memory, ● Sensory memory -> embedding representations for raw inputs, including text, image or other modalities; ● Short-term memory -> in-context learning. It is short and finite, as it is restricted by the finite context window length of Transformer. ● Long-term memory -> external vector store that the agent can attend to at query time, accessible via fast retrieval
  • 12. Component Three: Tool Use Utilizing LLM to know which tool to use when, and how to use that tool - TALM (Tool Augmented Language Models; Parisi et al. 2022) - Toolformer by Meta - HuggingGPT The format of API calls in TALM
  • 13. ReACT