🔬 Applied Scientist & ML Specialist938 Problems · 127 Topics🌐 Linked to Tech Vault

TalentMe Science Depth: Universal Applied Science & ML Depth Problem Bank

938 end-to-end science depth, system design, transition bridge & BQ problems covering derivations, production trade-offs, pitfalls, and follow-ups.

🎯

Science Depth Mastery Tracker

0% Mastered

Locally synced in your browser. Track, review, and master all 938 scientific & design topics.

Mastered: 0
Reviewing: 0
Unstarted: 938
🚀 Start Next Unpracticed →
🌐
Linked to TalentMe Tech Vault & MindmapsGlobal System
All 938 questions are directly linked to TalentMe's Tech Vault guides, atomic cards, and interactive mindmaps.
🧭Math Foundations Mindmap & Cards→📖Bayesian Inference, Entropy & KL Divergence→🎯DS Track→
Found 24 in-depth questions
M5-012NLP & Large Language ModelsPretraining Objectives & Data Curation
Hard

Data Contamination and Its Impact on Benchmark Reliability

⚡ One-Sentence Answer

Data contamination occurs when benchmark test sets leak into pre-training or fine-tuning corpora, inflating evaluation scores through verbatim memorization and compromising the scientific validity of model comparisons.

Math & Tradeoffs→
M5-013NLP & Large Language ModelsPretraining Objectives & Data Curation
Medium

Pre-training Domain Mixtures and the Role of the Annealing Phase

⚡ One-Sentence Answer

The pre-training annealing phase decays the learning rate to near-zero while shifting the data mixture toward ultra-high-quality math, code, and curated reasoning corpora, achieving rapid benchmark uplift in the final 5% of training.

Math & Tradeoffs→
M5-014NLP & Large Language ModelsScaling Laws & Compute Allocation
Easy

Core Conclusions of Kaplan vs. Chinchilla Scaling Laws

⚡ One-Sentence Answer

Kaplan et al. concluded that model parameters should scale faster than training tokens ($N \propto C^{0.73}, D \propto C^{0.27}$), while Chinchilla proved that parameters and tokens should scale in equal proportion ($N \propto C^{0.5}, D \propto C^{0.5}$, $\approx 20$ tokens per parameter).

Math & Tradeoffs→
M5-015NLP & Large Language ModelsScaling Laws & Compute Allocation
Easy

Why Inference Costs Alter Compute-Optimal Scaling Choices

⚡ One-Sentence Answer

Chinchilla optimality minimizes training compute alone, but production systems minimize total lifecycle cost (training + inference), justifying massive over-training of compact models to achieve low serving latency and high throughput.

Math & Tradeoffs→
M5-016NLP & Large Language ModelsScaling Laws & Compute Allocation
Medium

Emergent Abilities in Large Language Models and the Metric Discontinuity Debate

⚡ One-Sentence Answer

Emergent abilities describe capabilities that appear abruptly at scale, but Schaeffer et al. revealed that this apparent sharp discontinuity is largely an artifact of non-linear step-function evaluation metrics rather than a discontinuous shift in underlying model representations.

Math & Tradeoffs→
M5-017NLP & Large Language ModelsScaling Laws & Compute Allocation
Medium

Scaling Strategies Under Data-Constrained Regimes

⚡ One-Sentence Answer

As high-quality public human text approaches exhaustion, scaling strategies shift toward multi-epoch pre-training, aggressive synthetic data synthesis, multimodal data expansion, and test-time reasoning compute.

Math & Tradeoffs→
M5-018NLP & Large Language ModelsScaling Laws & Compute Allocation
Hard

Guiding Hyperparameter and Architecture Choices via Scaling Laws

⚡ One-Sentence Answer

Scaling laws guide pre-training hyperparameter tuning by using Maximal Update Parametrization ($\mu$P) to transfer optimal learning rates from tiny proxy models directly to hundred-billion-parameter models without expensive hyperparameter sweeps.

Math & Tradeoffs→
M5-019NLP & Large Language ModelsScaling Laws & Compute Allocation
Hard

Post-Training Scaling (RL and Test-Time Compute) vs. Pre-Training Scaling

⚡ One-Sentence Answer

Pre-training scaling expands general world knowledge via next-token prediction over static text, whereas post-training scaling expands complex reasoning through reinforcement learning exploration and test-time search computation.

Math & Tradeoffs→
M5-020NLP & Large Language ModelsInstruction Tuning & Supervised Fine-Tuning
Easy

Supervised Fine-Tuning (SFT) Objectives and Data Construction Principles

⚡ One-Sentence Answer

SFT adapts pre-trained base models into conversational assistants by optimizing next-token cross-entropy exclusively over target assistant responses, requiring high instruction diversity, clear formatting, and strict factual quality.

Math & Tradeoffs→
M5-021NLP & Large Language ModelsInstruction Tuning & Supervised Fine-Tuning
Easy

Loss Masking and Sequence Packing in SFT Training

⚡ One-Sentence Answer

Loss masking restricts gradient backpropagation exclusively to response tokens, while sequence packing concatenates multiple conversations into a single sequence with block-diagonal attention masks to eliminate padding waste.

Math & Tradeoffs→
M5-022NLP & Large Language ModelsInstruction Tuning & Supervised Fine-Tuning
Medium

Catastrophic Forgetting in SFT and Mitigation Strategies

⚡ One-Sentence Answer

Catastrophic forgetting occurs when fine-tuning on narrow instruction datasets overwrites pre-trained general knowledge and reasoning circuits, mitigated through pre-training data replay, low learning rates, and parameter-efficient tuning.

Math & Tradeoffs→
M5-023NLP & Large Language ModelsInstruction Tuning & Supervised Fine-Tuning
Medium

Chain-of-Thought (CoT) Distillation: Benefits and Hallucination Risks

⚡ One-Sentence Answer

CoT distillation transfers multi-step reasoning capabilities from teacher models to compact students by training on intermediate reasoning steps, but risks superficial format mimicry and unverified hallucinations if reasoning tokens are not anchored in verification.

Math & Tradeoffs→
M5-024NLP & Large Language ModelsInstruction Tuning & Supervised Fine-Tuning
Hard

SFT Data Mixture and Multi-Task Balancing

⚡ One-Sentence Answer

SFT multi-task balancing allocates data proportions across dialogue, coding, mathematics, reasoning, and safety to prevent over-fitting to single modalities while maintaining generalist conversational capabilities.

Math & Tradeoffs→
M5-025NLP & Large Language ModelsInstruction Tuning & Supervised Fine-Tuning
Hard

Division of Labor Between SFT and RLHF/DPO

⚡ One-Sentence Answer

SFT establishes foundational instruction following, domain knowledge, and response formatting, whereas RLHF/DPO refines preference boundaries, suppresses low-probability tail errors, and aligns model outputs with human values.

Math & Tradeoffs→
M5-026NLP & Large Language ModelsInstruction Tuning & Supervised Fine-Tuning
Medium

Choosing Learning Rates and Epochs in SFT

⚡ One-Sentence Answer

Supervised fine-tuning requires conservative learning rates ($10\times$ to $50\times$ smaller than pre-training) and low epoch counts (1 to 3 epochs) with cosine decay to adapt conversational formatting without destroying pre-trained knowledge representations.

Math & Tradeoffs→
M5-027NLP & Large Language ModelsInstruction Tuning & Supervised Fine-Tuning
Medium

Chat Template Formatting and Multi-Turn Conversation Construction

⚡ One-Sentence Answer

Chat templates serialize multi-turn conversational roles into unambiguous token streams using special delimiter tokens and Jinja2 formatting, ensuring strict boundary separation between system, user, and assistant turns.

Math & Tradeoffs→
M5-028NLP & Large Language ModelsInstruction Tuning & Supervised Fine-Tuning
Hard

Rejection Sampling Fine-Tuning (RFT / STaR) Mechanisms

⚡ One-Sentence Answer

Rejection Sampling Fine-Tuning (RFT / STaR) samples multiple candidate reasoning paths from a model, filters them using automated verifiers, and fine-tunes the model on its own successful generations to bootstrap reasoning capabilities.

Math & Tradeoffs→
M5-029NLP & Large Language ModelsInstruction Tuning & Supervised Fine-Tuning
Medium

SFT Data Sources and Engineering Trade-offs

⚡ One-Sentence Answer

SFT datasets combine four complementary sources: human expert annotation for high-quality instruction adherence, strong model distillation for scalable volume, open-source benchmarks for baseline coverage, and verified synthetic generation for complex reasoning.

Math & Tradeoffs→
M5-030NLP & Large Language ModelsAlignment & RLHF
Easy

The Three-Stage RLHF Pipeline

⚡ One-Sentence Answer

The canonical RLHF pipeline consists of three sequential phases: Supervised Fine-Tuning to establish instruction following, Reward Modeling on human preference pairs, and PPO Reinforcement Learning with a KL penalty to optimize policy generation.

Math & Tradeoffs→
M5-031NLP & Large Language ModelsAlignment & RLHF
Easy

Bradley-Terry Preference Model and Reward Model Loss Formulation

⚡ One-Sentence Answer

The Bradley-Terry model parameterizes pairwise preference probability as the sigmoid of reward differences, training the reward model via binary cross-entropy loss where only relative reward margins are statistically identifiable.

Math & Tradeoffs→
M5-032NLP & Large Language ModelsAlignment & RLHF
Medium

PPO Clipped Surrogate Objective and the Rationale for Min and Clip

⚡ One-Sentence Answer

PPO limits policy updates by clipping the probability ratio to $[1-\epsilon, 1+\epsilon]$ and taking the minimum against the unclipped objective, forming a pessimistic lower bound that prevents destructive policy collapse.

Math & Tradeoffs→
M5-033NLP & Large Language ModelsAlignment & RLHF
Medium

Generalized Advantage Estimation (GAE) and Advantage in RLHF

⚡ One-Sentence Answer

GAE calculates the advantage function by computing an exponentially weighted average of multi-step temporal difference residuals, balancing bias and variance via hyperparameter $\lambda$ to guide policy updates.

Math & Tradeoffs→
M5-034NLP & Large Language ModelsAlignment & RLHF
Hard

Reward Hacking and Mitigation Strategies in RLHF

⚡ One-Sentence Answer

Reward hacking occurs when a policy exploits flaws in the proxy reward model to score high rewards without improving true response quality, mitigated through KL penalties, reward ensembles, length penalties, and iterative reward model updates.

Math & Tradeoffs→
M5-035NLP & Large Language ModelsAlignment & RLHF
Hard

Engineering Challenges and Stability Issues in RLHF

⚡ One-Sentence Answer

RLHF is notoriously complex to scale because it requires orchestrating four large models across distributed GPU clusters while managing high policy variance, value network divergence, and reward model exploitation.

Math & Tradeoffs→