CA - S&T - 240724

SNT CA: Class 1

Large Language Models (LLMs)

What are LLMs? Large Language Models (LLMs) are advanced artificial intelligence systems designed to
understand, generate, and manipulate human language. They are trained on vast amounts of text data and
leverage deep learning techniques to perform various natural language processing (NLP) tasks. Examples
of LLMs include OpenAI's GPT series, Google's BERT.

How LLMs Work LLMs operate using neural network architectures, particularly transformers. The key
components of LLMs include:
1. Training Data: LLMs are trained on extensive datasets, which may include books, articles, websites,
and other text sources.
2. Tokenization: Text is divided into smaller units called tokens. These can be words or subwords.
3. Model Architecture: The core architecture involves transformers, which use self-attention
mechanisms to weigh the importance of different words in a sentence.
4. Training Process: The model learns by predicting the next word in a sentence, adjusting weights
to minimize prediction error across the dataset.
5. Inference: Once trained, the model can generate coherent text, answer questions, summarize
information, and more by analyzing the input and generating appropriate responses.

Role in Facilitating AI:

1. Natural Language Processing: They enhance NLP tasks such as text generation, translation,
summarization, and sentiment analysis.
2. Human-Computer Interaction: LLMs improve interfaces like chatbots and virtual assistants,
making them more responsive and accurate.
3. Information Retrieval: They assist in retrieving and synthesizing information from large text.
4. Automation: LLMs automate content creation, report generation, and data analysis tasks.

Key to Generative AI:

1. Text Generation: LLMs can generate coherent and contextually relevant text for various
applications, including creative writing, news articles, and marketing copy.
2. Creative Applications: They contribute to creative fields by generating poetry, stories, and even
code.
3. Conversational AI: LLMs underpin chatbots and conversational agents, providing human-like
interactions.
4. Personalization: They enable personalized content recommendations and tailored user
experiences.

Role/Applications of LLM (To be discussed in class)

Generative AI
What is Generative AI?
Generative AI is a subset of artificial intelligence that focuses on creating new, original content rather than
simply analyzing or acting on existing data. It leverages various techniques, including Large Language
Models (LLMs), diffusion models, and Generative Adversarial Networks (GANs), to produce a wide range of
outputs like text, images, music, and even videos.

Forum Learning Centre: Delhi - 2nd Floor, IAPL House, 19 Pusa Road, Karol Bagh, New Delhi - 110005 | Patna - 2nd floor, AG Palace, E Boring Canal
Road, Patna, Bihar 800001 | Hyderabad - 1st & 2nd Floor, SM Plaza, RTC X Rd, Indira Park Road, Jawahar Nagar, Hyderabad, Telangana 500020
9311740400, 9311740900 | https://academy.forumias.com | [email protected] | [email protected]

[1]
Tabular Comparison: Generative AI vs. Traditional AI

Feature Generative AI Traditional AI

Purpose To create new content To analyze and make decisions based on existing data

Methods LLMs, diffusion models, GANs Machine learning, rule-based systems, expert
systems

Output Text, images, music, videos Predictions, classifications, recommendations

Examples GPT-3 (text), DALL-E 2 (images), Fraud detection systems, medical diagnosis tools,
MuseNet (music) recommender systems

Tabular Comparison: LLM Vs. Generative AI

Feature Large Language Models (LLMs) Generative AI

Definition A type of artificial intelligence model A broader field of AI that encompasses

specifically designed to understand, various models and techniques to create new
generate, and manipulate human language. content across multiple modalities, including
text, images, music, and more.

Focus Primarily on text-based tasks such as Includes LLMs but also extends to generating
translation, summarization, question- images, music, videos, and other forms of
answering, and text generation. content.

Technology Relies heavily on transformer architecture Employs various models and algorithms
and deep learning techniques to process depending on the type of content being
and generate text. generated, including LLMs, diffusion models,
and GANs.

Applications Chatbots, virtual assistants, language Content creation (text, images, music,
translation tools, content creation tools, videos), drug discovery, material design, art
search engines. generation, and more.

Examples GPT-3, BERT, T5 DALL-E, Midjourney, Stable Diffusion,

ChatGPT

Strengths Excellent at understanding and generating Capable of generating creative and diverse
human-like text, contextually relevant content across various modalities, pushing
responses, and language-based tasks. the boundaries of what's possible.

Limitations May produce inaccurate or biased Can be computationally expensive, require

information, struggle with nuanced specialized hardware, and may raise ethical
understanding of certain topics, and concerns regarding deepfakes and
require large amounts of data for training. misinformation.

Role/Applications of Generative AI (To be discussed in class)

Forum Learning Centre: Delhi - 2nd Floor, IAPL House, 19 Pusa Road, Karol Bagh, New Delhi - 110005 | Patna - 2nd floor, AG Palace, E Boring Canal
Road, Patna, Bihar 800001 | Hyderabad - 1st & 2nd Floor, SM Plaza, RTC X Rd, Indira Park Road, Jawahar Nagar, Hyderabad, Telangana 500020
9311740400, 9311740900 | https://academy.forumias.com | [email protected] | [email protected]

[2]
 Regulating AI

Need for regulating AI/Challenges posed by AI (To be discussed in class)

Key areas of regulation

1. Data Governance: Strict data protection laws (Personal Data Protection Bill), data localization
requirements, and robust cybersecurity measures to safeguard sensitive data.
2. Algorithm Transparency & Accountability: Mandatory audits, explainability of AI decisions, and
mechanisms for redressal in case of harm caused by AI systems. Example: If an AI-powered loan
application denies a loan, the applicant should have the right to know the reasons behind the
decision and have a process for appealing it.
3. Bias Mitigation: Requiring diverse training data, regular bias testing, and fairness-aware
algorithms to prevent discrimination and ensure equitable outcomes.
4. Human Oversight: Defining clear roles for humans in the AI loop, especially for high-risk
applications like healthcare, finance, and law enforcement. Example: In healthcare, a human doctor
should always have the final say in diagnosis and treatment decisions, even when AI provides
recommendations.
5. Liability Framework: Establishing who is responsible for AI harms and how victims can seek
compensation. This is crucial for ensuring accountability in the development and deployment of AI
systems. Example: If an autonomous vehicle causes an accident, regulations should clarify who is
liable (the manufacturer, the software developer, or the owner) and how the victim can be
compensated.
6. Sector-Specific Rules: Tailored regulations for high-impact sectors like healthcare, autonomous
vehicles, and finance, considering their unique risks and benefits.

Indian initiatives to regulate/promote AI (To be discussed in class)

Global initiatives to regulate/promote AI (To be discussed in class)

Quantum Computing and NQM

Introduction
Quantum technology leverages the principles of quantum mechanics, a fundamental theory in physics that
describes nature at the smallest scales of energy levels of atoms and subatomic particles. Quantum
computing is a type of quantum technology that utilizes quantum bits (qubits) to perform calculations that
are significantly more complex than those possible with classical computers.
Key Terms and Definitions
1. Quantum Bit (Qubit): The basic unit of quantum information, analogous to the classical bit. Unlike
a bit, which can be either 0 or 1, a qubit can be in a state of 0, 1, or any quantum superposition of
these states. E.g. An electron's spin, which can be in a state of up, down, or both (superposition).
2. Superposition: A fundamental principle of quantum mechanics where a quantum system can exist
in multiple states at once until it is measured. E.g. A qubit can represent both 0 and 1
simultaneously, unlike a classical bit which is either 0 or 1.
3. Entanglement: A phenomenon where quantum particles become interconnected, such that the
state of one particle instantaneously influences the state of another, regardless of distance. E.g.
Entangled qubits can be used for quantum teleportation and secure communication.

Forum Learning Centre: Delhi - 2nd Floor, IAPL House, 19 Pusa Road, Karol Bagh, New Delhi - 110005 | Patna - 2nd floor, AG Palace, E Boring Canal
Road, Patna, Bihar 800001 | Hyderabad - 1st & 2nd Floor, SM Plaza, RTC X Rd, Indira Park Road, Jawahar Nagar, Hyderabad, Telangana 500020
9311740400, 9311740900 | https://academy.forumias.com | [email protected] | [email protected]

[3]
 4. Quantum Decoherence: The loss of quantum coherence, where quantum systems lose their
quantum behavior and transition into classical states due to interaction with their environment.
E.g. A qubit losing its superposition state due to external noise or interference.
5. Quantum Gate: A basic quantum circuit operating on a small number of qubits. They are the
building blocks of quantum circuits, similar to classical logic gates in classical circuits.

Key Advantages of Quantum Technology/Computing Over Classical Computing

1. Speed and Efficiency: Quantum computers can process complex calculations at speeds
unattainable by classical computers.
Example: Google's quantum computer, Sycamore
2. Parallelism: Due to superposition, quantum computers can handle multiple calculations
simultaneously.
E.g. : A quantum computer with N qubits can represent 2^N states simultaneously.
3. Secure Communication Quantum entanglement enables highly secure communication channels,
making eavesdropping detectable.
Example: Quantum Key Distribution (QKD) protocols, such as BB84.
4. Solving Complex Problems: Quantum computing excels in solving problems related to
optimization, cryptography, and material science that are currently infeasible for classical
computers.
Example: Quantum algorithms like Shor's algorithm can factor large numbers exponentially faster
than the best-known classical algorithms, threatening current encryption methods.
5. Simulation of Quantum Systems: Quantum computers can efficiently simulate quantum systems,
aiding in advancements in chemistry, physics, and biology.
Example. Simulating molecular structures to discover new drugs and materials, which is difficult
for classical computers due to the complexity of quantum interactions.
6. Machine Learning and AI: Quantum computing can significantly enhance machine learning
algorithms, providing faster processing and better optimization.

Role/Applications of QC in development of India (To be discussed in class)

Challenges arising because of QC and in implementing QC (To be discussed in class)
Steps/Initiatives to promote QC in India (To be discussed in class)

National Quantum Mission

The National Quantum Mission (NQM) of India is an initiative aimed at advancing the country's quantum
technology capabilities. The mission focuses on various aspects of quantum science and technology to
position India as a leader in this emerging field.
Key Components
1. Quantum Computing
2. Quantum Communication
3. Quantum Sensing and Metrology
4. Quantum Materials and Devices
Phases
1. Phase 1 (2023-2026):
 Setting up of quantum research and development centers.
 Development of initial quantum computing and communication prototypes.

Forum Learning Centre: Delhi - 2nd Floor, IAPL House, 19 Pusa Road, Karol Bagh, New Delhi - 110005 | Patna - 2nd floor, AG Palace, E Boring Canal
Road, Patna, Bihar 800001 | Hyderabad - 1st & 2nd Floor, SM Plaza, RTC X Rd, Indira Park Road, Jawahar Nagar, Hyderabad, Telangana 500020
9311740400, 9311740900 | https://academy.forumias.com | [email protected] | [email protected]

[4]
  Training and development of skilled quantum scientists and engineers.
2. Phase 2 (2026-2030):
 Scaling up quantum technologies developed in Phase 1.
 Establishing large-scale quantum communication networks.
 Expanding the application of quantum technologies in various sectors.
3. Phase 3 (Beyond 2030):
 Achieving advanced quantum computing capabilities.
 Global deployment of quantum communication and sensing technologies.
 Continued innovation and commercialization of quantum technologies.

Quantum Communication
Introduction
Quantum communication utilizes the principles of quantum mechanics to transmit information securely.
It offers enhanced security over classical communication methods, primarily through Quantum Key
Distribution (QKD).
How Quantum Communication is Done: Step-by-Step
1. Quantum State Preparation
2. Quantum Transmission
3. Quantum Measurement
4. Classical Communication
5. Key Verification
6. Error Estimation
7. Error Correction
8. Message decoded

Advantages of Quantum Communication (To be discussed in class)

Challenges in Quantum Communication (To be discussed in class)

Quantum Dots

1. Definition: Quantum dots are tiny particles, typically 2-10 nanometers in diameter, composed of
semiconductor materials. Their size and shape allow them to exhibit quantum mechanical
properties, leading to discrete electronic states and size-dependent optical properties.
2. Materials: Common materials for quantum dots include cadmium selenide (CdSe), cadmium sulfide
(CdS), indium phosphide (InP), and lead selenide (PbSe).

Key Terms and Technical Specifications

1. Quantum Confinement:
 The phenomenon where the electrons in quantum dots are confined in all three spatial
dimensions, leading to discrete energy levels.
This confinement leads to unique optical properties such as size-dependent emission
wavelengths.
2. Photoluminescence:
 The emission of light from quantum dots when they absorb photons and re-emit them.
 The wavelength of emitted light can be tuned by changing the size of the quantum dots.

Forum Learning Centre: Delhi - 2nd Floor, IAPL House, 19 Pusa Road, Karol Bagh, New Delhi - 110005 | Patna - 2nd floor, AG Palace, E Boring Canal
Road, Patna, Bihar 800001 | Hyderabad - 1st & 2nd Floor, SM Plaza, RTC X Rd, Indira Park Road, Jawahar Nagar, Hyderabad, Telangana 500020
9311740400, 9311740900 | https://academy.forumias.com | [email protected] | [email protected]

[5]
 3. Bandgap Energy:
 The energy difference between the valence band and the conduction band in a
semiconductor.

In quantum dots, the bandgap energy increases as the size of the dots decreases, affecting
their optical properties.
4. Quantum Yield:
 The efficiency of photon emission after excitation.
 High quantum yield means brighter emission, making quantum dots useful for display and
imaging applications.

5. Surface Passivation:
 The process of coating the surface of quantum dots to improve their stability and optical
properties.
 Enhances the performance and longevity of quantum dots in various applications.

Advantages of Quantum Dots

1. Tunable Emission:
 Advantage: The color of light emitted by quantum dots can be precisely controlled by
changing their size, allowing for customized optical properties.
Example: Displays can achieve more accurate color reproduction using quantum dots.
2. High Brightness and Efficiency:
 Advantage: Quantum dots have high quantum yields, making them extremely bright and
efficient light emitters.
Example: Quantum dots in TVs enhance brightness and color accuracy.
3. Narrow Emission Spectra:
 Advantage: Quantum dots emit light in narrow spectral lines, which is beneficial for
applications requiring precise color control.
Example: Improved spectral efficiency in optical devices.
4. Photostability:
 Advantage: Quantum dots are more resistant to photobleaching compared to traditional
dyes, making them suitable for long-term applications.
Example: Long-lasting biomedical imaging markers.

Applications of Quantum Dots (To be discussed in class)

Blockchain – Validation (PoW, PoS)

Introduction
Blockchain technology is a decentralized digital ledger system that records transactions across multiple
computers so that the records cannot be altered retroactively without the alteration of all subsequent
blocks. It ensures transparency, security, and efficiency in digital transactions and has applications across
various sectors.

Key Benefits of Blockchain

1. Transparency: Every transaction is recorded on a public ledger that can be inspected, ensuring full
transparency.
2. Security: Transactions are encrypted and linked to the previous transaction, making it nearly
impossible for hackers to alter data.

Forum Learning Centre: Delhi - 2nd Floor, IAPL House, 19 Pusa Road, Karol Bagh, New Delhi - 110005 | Patna - 2nd floor, AG Palace, E Boring Canal
Road, Patna, Bihar 800001 | Hyderabad - 1st & 2nd Floor, SM Plaza, RTC X Rd, Indira Park Road, Jawahar Nagar, Hyderabad, Telangana 500020
9311740400, 9311740900 | https://academy.forumias.com | [email protected] | [email protected]

[6]
 3. Immutability: Once a transaction is recorded on a blockchain, it cannot be altered or deleted,
ensuring data integrity.
E.g. IBM Food Trust uses blockchain to ensure the immutability of data related to food supply
chains.
4. Efficiency: Blockchain reduces the need for intermediaries, streamlining processes and reducing
transaction times. E.g. Financial services use blockchain for faster cross-border payments and
reduced processing times.
5. Decentralization: The decentralized nature of blockchain eliminates the need for a central
authority, reducing the risk of central points of failure.
E.g. Ethereum’s smart contracts operate on a decentralized network, executing agreements
without intermediaries.

Blockchain Validation
Blockchain validation is the process of verifying and confirming the legitimacy of transactions before they
are added to the blockchain ledger. This ensures the integrity and security of the blockchain network.

How it's done:

1. Transaction Proposal: A new transaction
is proposed by a user.
2. Node Verification: The proposed
transaction is broadcasted to nodes
(computers) within the blockchain
network.
3. Consensus Mechanism: Nodes verify the
transaction's validity based on
predefined rules and the blockchain's
history. Different consensus
mechanisms exist, such as Proof of Work
(PoW) and Proof of Stake (PoS).
4. Block Addition: Once verified, the
transaction is added to a block along
with other verified transactions.
5. Block Validation: The new block is
broadcasted to the network for other
nodes to validate its authenticity.
6. Chain Linking: If validated, the block is
added to the existing blockchain,
creating a permanent and tamper-proof
record.
Key Points:
1. Decentralization: Blockchain validation
is a decentralized process, ensuring no
single entity controls it.
2. Security: The validation process
protects the blockchain from fraud and
double-spending.

Forum Learning Centre: Delhi - 2nd Floor, IAPL House, 19 Pusa Road, Karol Bagh, New Delhi - 110005 | Patna - 2nd floor, AG Palace, E Boring Canal
Road, Patna, Bihar 800001 | Hyderabad - 1st & 2nd Floor, SM Plaza, RTC X Rd, Indira Park Road, Jawahar Nagar, Hyderabad, Telangana 500020
9311740400, 9311740900 | https://academy.forumias.com | [email protected] | [email protected]

[7]
 3. Consensus: Different blockchains use different consensus mechanisms to ensure agreement
among nodes.

Proof of Work (PoW)

1. Mechanism: Miners solve complex cryptographic puzzles, known as hashing problems.
2. Process:
 Puzzle Solving: Miners use computational power to solve a puzzle based on the SHA-256
hashing algorithm.
 Validation: The first miner to solve the puzzle broadcasts the solution.
 Verification: Other miners verify the solution by confirming the hash value meets the
required difficulty target.
 Block Addition: If verified, the block is added to the blockchain and the miner receives a
block reward.
 Energy Consumption: High, due to the intensive computational effort required.

Proof of Stake (PoS)

1. Mechanism: Validators are chosen based on their stake (amount of cryptocurrency held) in the
network.
2. Process:
 Selection: Validators are chosen deterministically, often based on the size of their stake
and sometimes other factors like coin age.
 Validation: Selected validators propose and validate new blocks.
 Confirmation: Once consensus is reached, the block is added to the blockchain. Validators
receive transaction fees as rewards.
 Energy Consumption: Low, as it does not require solving complex puzzles.

Key Differences
Feature Proof of Work (PoW) Proof of Stake (PoS)

Core Action Miners solve complex cryptographic Validators are chosen randomly based on
puzzles (energy-intensive). their stake (energy-efficient).

Validation Miners verify the solution and block's Validators attest to the validity of
validity. proposed blocks.

Reward Block reward (newly minted coins) + Transaction fees (+ sometimes newly
transaction fees. minted coins).

Resource High computational power and energy Low computational power and energy
Intensity consumption. consumption.

Decentralization Highly decentralized, but can lead to Potentially less decentralized if stake is
mining pools and centralization. concentrated among a few.

Need for Proof of Stake algorithm (To be discussed in Class)

Proof of Work (PoW) vs. Proof of Stake (PoS) - Technical Comparison (To be discussed in class)

Forum Learning Centre: Delhi - 2nd Floor, IAPL House, 19 Pusa Road, Karol Bagh, New Delhi - 110005 | Patna - 2nd floor, AG Palace, E Boring Canal
Road, Patna, Bihar 800001 | Hyderabad - 1st & 2nd Floor, SM Plaza, RTC X Rd, Indira Park Road, Jawahar Nagar, Hyderabad, Telangana 500020
9311740400, 9311740900 | https://academy.forumias.com | [email protected] | [email protected]

[8]
 Parameter Proof of Work (PoW) Proof of Stake (PoS)

Mechanism Solving cryptographic puzzles Validators chosen based on stake

Energy Consumption High (e.g., Bitcoin ~100 TWh/year) Low (e.g., Ethereum 2.0 ~1/10,000th of
PoW)

Computing Power High-performance hardware (ASICs) Standard computer hardware

Energy Efficiency Low, energy-intensive High, energy-efficient

Transaction Speed Slower (e.g., Bitcoin ~7 TPS) Faster (e.g., Ethereum 2.0 ~100,000
TPS)

Security High, but vulnerable to 51% attack High, less vulnerable to 51% attack due
to high economic cost

Hardware High-performance, specialized General-purpose hardware

Requirement hardware

Scalability Limited High, capable of handling more

transactions

Environmental Impact High, significant carbon footprint Low, minimal carbon footprint

Cost of Attack High, requires significant energy and High, requires purchasing majority of
hardware investment stake

Reward Structure Block rewards and transaction fees Transaction fees

Complexity of Moderate High, involves complex algorithms

Implementation

Innovation and Slower to adapt due to hardware Faster to adapt due to software
Adaptability reliance flexibility

Answer Writing
Role of AI in development of India
1. Agriculture:
 Application: Precision farming, crop monitoring, and yield prediction.
 AI techniques like machine learning and computer vision analyze data from satellites, drones,
and ground sensors to provide actionable insights for farmers.
 Example: Microsoft's AI for Agriculture program in India uses machine learning algorithms to
analyze soil health, weather conditions, and historical yield data to predict optimal sowing
dates and fertilizer requirements.
2. Healthcare:
 Application: Disease diagnosis, treatment planning, and drug discovery.
 Deep learning algorithms analyze medical imaging data for disease detection, while natural
language processing (NLP) techniques extract insights from electronic health records for
predictive diagnostics. - Earlier disease detection, personalized treatments, and improved
healthcare access in rural areas.

Forum Learning Centre: Delhi - 2nd Floor, IAPL House, 19 Pusa Road, Karol Bagh, New Delhi - 110005 | Patna - 2nd floor, AG Palace, E Boring Canal
Road, Patna, Bihar 800001 | Hyderabad - 1st & 2nd Floor, SM Plaza, RTC X Rd, Indira Park Road, Jawahar Nagar, Hyderabad, Telangana 500020
9311740400, 9311740900 | https://academy.forumias.com | [email protected] | [email protected]

[9]
  Example: Bengaluru-based Niramai uses AI-powered thermal imaging and machine learning
for early-stage breast cancer detection, with an accuracy rate of 90%.
3. Education:
 Application: Personalized learning, automated assessment, and educational accessibility.
 Adaptive learning algorithms process student performance data to create tailored learning
paths, while NLP powers intelligent tutoring systems and automated grading - Improved
learning outcomes, reduced dropout rates, and enhanced skill development;
 Example: AI algorithms to analyze students' learning patterns and provide personalized study
plans, leading to a reported 93% improvement in exam scores.
4. Smart Cities and Urban Planning:
 Application: Traffic management, energy optimization, and public safety.
 Machine learning algorithms process data from IoT sensors for predictive maintenance of
infrastructure, while computer vision analyzes CCTV footage for traffic and safety
management.
 Example: Pune Smart City has implemented an AI-based adaptive traffic control system that
has reduced travel time by 25-30%.
5. Manufacturing and Industry 4.0:
 Application: Predictive maintenance, quality control, and supply chain optimization.
 Justification: Machine learning models analyze sensor data for predictive maintenance, while
computer vision systems perform automated quality checks. Reinforcement learning
algorithms optimize supply chain logistics.
 Example: Tata Steel uses AI-powered predictive maintenance, reducing downtime by 30% and
improving operational efficiency.
6. Financial Services:
 Application: Risk assessment, fraud detection, and personalized financial services.
 Justification: Machine learning algorithms analyze vast amounts of financial data for credit
scoring and fraud detection. NLP powers chatbots for customer service, while deep learning
models process market data for algorithmic trading.
 Example: State Bank of India's AI-powered chatbot, SIA, handles over 10,000 customer
queries per second with 85% accuracy.
7. Environmental Protection and Climate Change Mitigation:
 Application: Environmental monitoring, disaster prediction, and resource optimization.
 Justification: Deep learning models analyze satellite imagery and sensor data to monitor
environmental changes and predict natural disasters. Reinforcement learning algorithms
optimize resource allocation for conservation efforts.
 Example: The Indian Institute of Tropical Meteorology uses AI models to improve monsoon
forecasting accuracy by 60%.
8. E-Governance
 Public Service Delivery: AI enhances the efficiency and transparency of public service delivery,
from issuing documents to handling citizen grievances.
 Example: The Government of Andhra Pradesh uses AI to monitor the performance of
government schemes and services through its Real Time Governance Society (RTGS).
 Predictive Governance: AI tools predict and mitigate issues such as crime, natural disasters,
and infrastructure failures, enabling proactive governance.
 Example: The Telangana government uses AI to predict water levels in reservoirs and manage
water resources effectively.

Forum Learning Centre: Delhi - 2nd Floor, IAPL House, 19 Pusa Road, Karol Bagh, New Delhi - 110005 | Patna - 2nd floor, AG Palace, E Boring Canal
Road, Patna, Bihar 800001 | Hyderabad - 1st & 2nd Floor, SM Plaza, RTC X Rd, Indira Park Road, Jawahar Nagar, Hyderabad, Telangana 500020
9311740400, 9311740900 | https://academy.forumias.com | [email protected] | [email protected]

[10]